Turquoise Energy Ltd. News #121
covering June 2018 (Posted July 3rd)
Lawnhill BC Canada
by Craig Carmichael

www.TurquoiseEnergy.com = www.ElectricCaik.com = www.ElectricHubcap.com = www.ElectricWeel.com

* Stable Zinc Electrodes, The "Holy Grail" of battery making... hundreds have tried and (more or less) failed: Grail TAKEN by new electrolyte! 
(See Month in Brief, Electricity Storage)
* New Bandsaw Mill Concept: Self-Correcting Band Angle always cuts straight! (See Month in Brief, Other "Green" Projects)

Month In Brief (Project Summaries etc.) - Band mill - Band - Batteries! - "Electrafest":Portable 6 KW J1772 EV charger

In Passing (Miscellaneous topics, editorial comments & opinionated rants)
 - Battery Commercializations?...Useless SDTC Again-Changhong-DIY Crowd-Factory? - Crystal Sponge? - Moles and hydrocortosone cream: Update - Feeding the 9 Billion? - After the Thugs (bad poetry that doesn't rhyme) - Age of World?

- Project Reports -

Electric Transport - Electric Hubcap Motor Systems (no reports)

Other "Green" Electric Equipment Projects
* Carmichael Mill ("Handheld Bandsaw Alaska Mill")
  - "railway" band guide wheels: Mounting, adjustments & Cuts - Self Correcting Band Guides?
* Proposed New Electrical Standards "RFC": a new standard Voltage (38/36 VDC) + Standard Connectors for 12 VDC, 38 VDC
  - Idea: If making 38/36 volt connectors, why not market 38 volt lights and appliances along with them? - Other notes

Electricity Generation (no reports)

Electricity Storage - Turquoise Battery Project (NiMn, NiNi, O2-Ni), etc.
* Zinc: notorious battery electrode substance, TAMED by the new electrolyte!
* Nickel-Zinc Batteries - Compacting electrodes - Glue & Jell - Why, suddenly, Zinc instead of Nickel?
* Manganese-zinc & Lead-zinc cells & experiments (They all work!)
* Zinc Sheet Electrodes work great - Shaping Plates to gain More Surface Area
* Mn-Zn cell from scratch: works great!

June in Brief

    This month is mainly about battery experimentation and development. Now that I'm having real successes it seemed like a good thing to concentrate on and I spent most of the month on it. (More below.)

   Notwithstanding that, I did just a bit of work on the bandsaw mill guide wheels near the end -- and on the 30th conceived of what I believe will be a real breakthrough in band mill design: place the guide wheel pivots in front of the blade (instead of behind it) and have spring mounted adjustments instead of solidly fixed. Then the band will continuously adjust its angle of attack to keep cutting straight and never veer off up or down (even if the blade is dull), which is the main and very serious complaint I've been hearing for all band mills whenever the subject is mentioned. With a single stroke, this should raise the general liking for band mills among those who've milled wood from maybe a 3 to an 8 out of 10. It's going to "Wow!" people.

   I also said I'd play my Supercorder (the flute instrument I developed 2003-2006 - www.saers.com/craig/recorder/ ) in a couple of songs with a small band on Canada day, July first. Practice and daily rehearsals took up much time in the last week - with a half hour commute each way (in the electric Nissan Leaf, of course) adding to the time. They didn't want me to use sheet music in the concert and I've never played a "fixed script" without reading the music before, so they were trying to teach an old dog new tricks and it didn't go very smoothly. Then in the concert things went really awry when the sound guy came up and wanted to adjust my microphone while we were playing, talking to me and breaking my concentration in my most prominent song. I missed my entry and played some wrong notes in the rest. But the audience seemed to like it anyway, and (from the movie "Star Wars")... "It's not my fault!"


              Assembling nickel-zinc cell
   I started the month by looking at some charts - Pourbaix diagrams. I forget why. I hit zinc (just before zirconium) and suddenly realized that zinc would work as an everlasting negative electrode with the electrolyte I had come up with. A long life zinc electrode has been a "holy grail" of battery making ever since batteries were first made. Many have tried, but no one has had real success. Batteries with zinc are single use or notorious for their short cycle life. Thomas Edison picked nickel-iron instead of nickel-zinc to get a battery that would last, in spite of iron's lower energy storage and poorer performance. A web search will reveal many past and current attempts to make zinc electrolytes work, or at least to last a few hundred charge-discharge cycles - mostly in pH 14 alkali. For example, a Journal of the Electrochemical Society article from 1991 cites ten previous articles on the subject and below it is a list of 21 newer articles citing it.
   But in my pH 12 electrolyte zinc should last forever. Zinc has high voltage and energy per kilogram. In developing the pH 12 electrolyte with oxalate to use with nickel (or possibly manganese), I had taken the grail without realizing it! In fact, the electrolyte is key to a number of potential better chemistries, mostly best using zinc for a negative side.
   So I made a zinc electrode with nano-zinc flake/powder and traded out the nickel negative electrode for it. But the voltage being higher meant that oxygen had to be kept out of the cell, or it would spontaneously convert ("discharge") the zinc to zinc oxide. And I wanted to get better compaction on the plus electrodes. I found a compactor I had made previously that would do that. It made square electrodes instead of round. So next I made a whole new, square battery cell. (I could use the 3D printed cases I developed quite a while back along with this compactor.)
   Concurrently with these great chemical finds and results, I started putting "stuff" into the electrodes to help hold them together as thickeners, binding agents and gels. With successful electrodes to compare with I could tell if the additive(s) were causing a problem. They helped.

   There was still a self discharge problem in each cell I made, including in the manganese-zinc, and that chemistry is noted for low self discharge. So it could hardly be inherent in the electrode chemistry. By process of elimination with the various cells I made, it finally seemed it must be my cheap "pottery supply" calcium carbonate, converted by me to calcium oxide in my mini kiln 6 years ago (TE News #47, #66). It pretty much had to be contaminated, probably with nitrates. I ordered some guaranteed pure stuff from Westlab chem/lab supply. I look forward to batteries that hold their charge for weeks or months instead of hours. (How many years has this been an ongoing concern of mine, which I've attributed to other things or had no explanation for? Finally the light dawns...)

   But the main thing was I had a great working chemistry [and soon three of them] that could be produced one way or another. Fantastic! Now, how to get production happening?...

   On the 5th I e-mailed SDTC (Sustainable Development Technology Canada) to tell them Canada could freely have this fabulous new "created in Canada" battery chemistry. But Canada didn't want this once in a lifetime opportunity. An inventor with an invention, however Earth-shattering, doesn't fit their Procrustian profile of a "well organized and supported innovative company" for funding. They don't even pay lip service to inventions and they don't want to talk to inventors. Is that how to foster development of new sustainable technologies? I have written of how they put the cart before the horse before in TE News #80, and I've written more again below in In Passing.

   The next day I e-mailed to Sichuan Changhong Batteries in China (to "the president") and pointed out the new chemistry they could freely have. Given their current obsolescent line of flooded alkaline cells - nickel-iron, nickel cadmium and nickel-metal hydride, made on their production line bought from Varta in Sweden or Germany many years ago when they ceased production there, I felt this was one company that might be in need of a new battery like this. They would be much more likely to adopt it than a lithium or lead-acid battery maker (who would more likely just wish it would go away!), and their "green" markets would really appreciate them. (Of course, their markets would then start expanding, too, instead of shrinking!) So far I have received no reply. They'll get a pointer to this newsletter, too.

   Nickel plus electrodes would be great for a mass production line, but so far seem hard to do for a homemade battery because of their high voltage. (more experiments!...) So, what about other elements? The ones that come readily to mind are lead and manganese. I soon had a battery that worked well using the manganese electrode of an old "F" size dry cell (found in 6 volt lantern batteries). And I bought a new lead-acid battery to get lead positive plates from. I got some good results from these with zinc negatives in the new electrolyte.

L: Remaking manganese-zinc "F" cell with new zinc nano powder electrode and the new oxalate electrolyte
R: Testing first lead-zinc cell in original lead-acid battery case

Manganese-zinc flat plates cell from scratch.
(Outside padding is because cell didn't fill case.)
This worked very well and will probably be
the pattern for battery production - if I do it.

   But I started thinking rechargeable, "everlasting" Mn-Zn would be the batteries to make at home. Much the easiest to make, CHEAP materials, and higher energy density than lithium types. It would seem that all that's been missing for 150 years to have a lightweight, high energy, low cost and long cycle life battery was an appropriate electrolyte. And I probably wouldn't have found it either if it hadn't been for someone giving me an old kid's chemistry set with a small bottle of oxalic acid in it. I started with that name, years ago.

   I put this out on two "DIY" battery related e-mail lists, and I'll post a link to this newsletter as well.

   But maybe, since most people obviously would rather buy batteries than make them, I should set up a small production plant here? Along with the regulated voltage chargers they would need. "Prismatic" cells made with multiple plates stacked end to end (similar to lead-acid batteries) yielded excellent results in my experiments. They would be 1.5 volts. A factory is easier to accomplish with funding. And someone tells me he saw on BBC that battery factories are presently the world's biggest factories. There's stiff competition even if one has the "better mouse trap". It would be typical these days of well established companies to pull some sort of hi-jinks (bribing government to pass some prohibitive regulation or tie the company up in court on some pretext with a "stop work" order in place, devious legal and business maneuverings, criminal threats and violence in various forms...) to have such an upstart shut down before it got far onto the public radar screen.
   But one can't worry too much about such possibilities. Hopefully nothing will happen. I'll fill out an application with a place in BC that has "Executives in Residence" to help manage a new business, and see how far I get with that.

Saltspring Island "Electrafest" & Portable 6 KW Charger

   In other news, Tom Sawyer sent me pictures from the "Electrafest" electric transport gathering on Saltspring Island. He said an entire ferry trip to the island was fully booked with electric cars and not a single gasoline vehicle. He also says that the price of a 2015 Nissan Leaf like mine has gone from 18000$ to 23000$ (in 6 months). People are catching on and there are few electric cars to be had at the dealerships. Mostly they're pre-sold, right into next year. Of course, in Victoria driving distances are generally pretty short compared to on the sprawling mainland so they're more practical. (When the better batteries start being incorporated into cars, the continent will want them too!) On Saltspring Island, from the sounds of things most everybody has an electric car; mostly Nissan Leafs.

A donated Chevy Firefly (2 door version of my Sprint)
converted to electric by Saltspring Island high school class.
I'm not sure about the motor but it looks like they've
kept the original manual transmission and gear shifting.
Back seat removed - lead-acid batteries
are heavy and take up a lot of space.

   Probably the most interesting thing there was a 6 KW EV charger that instead of installing, one plugs into a 240 volt dryer plug. I'm not sure I want to buy one because those plugs aren't everywhere and are usually hard to get at. But if there was one available anywhere in Masset, perhaps at the town hall, I'd be able to drive there in the Leaf, charge a while while I did my shopping or business, and then have the range to return home. Then they would only need to buy the unit - no electrician and costly retrofit installation.
   (And maybe I could drop the insurance on my Toyota Echo for part of the year - bigger savings on this "administrative" item than on fuel. Of course, I don't always know in advance when I'll want to pull a trailer. ...Maybe I should ignore recommendations and put a hitch on the Leaf for my lightweight trailers? Insurance for two vehicles, one of which is hardly ever used, really costs.)

   Also remember that dryer plugs and 240 volts can be hazardous. People get electrocuted plugging/unplugging dryers in damp laundry rooms. That's another reason not to carry one around in the car: if they get condensation on them around the plug they could be lethal. It's best if they're kept in a dry, heated space. And this of course is why the J1772 charging plug was created in the first place. The power doesn't come on until and unless the socket is securely plugged into the car, and there's a ground fault detector. (It still galls me that there's no neutral provided for plugging in 120 volt chargers. It can probably be tricked, perhaps by using the metal box or actual ground as a neutral.)
   Maybe I'll volunteer to put up some of the money for Masset to buy one for the town hall or other conveniently situated building.

"Portable" J1772 - 6 KW charging unit plugs
into 240 volt, 30 amp (dryer/welder) wall outlet.

In Passing
(Miscellaneous topics, editorial comments & opinionated rants)

Battery Commercializations? - Useless SDTC Again - Changhong - DIY Crowd - Factory?

   Regardless of my own troubles making nickel electrodes, I'm sure there are people out there who know the best ways to make them, and also how to set up mass production lines (actually I know someone). With funding one can hire talent. The chemistry worked, and everything else would surely come together as needed if there was commercial interest. But putting that together by myself seemed beyond me.

   On the 5th I e-mailed SDTC (Sustainable Development Technology Canada) to tell them Canada could freely have this fabulous new battery chemistry and technology, created in Canada. But the chance of a lifetime for starting a fine new Canadian industry ahead of the world was passed up on a technicality: An inventor with an invention, however Earth-shattering it is, doesn't fit their profile of a "well organized and supported innovative company". I said I could help set up the production facility, but that I was an inventor, not an entrepreneur. But they don't want inventions. It is my belief that to do this sort of thing is exactly why parliament created SDTC in the first place - the very sort of thing the legislators had hoped would happen, and that parliament probably left them a pretty free hand as to how to go about it.
   Their self-imposed (I believe) rules don't allow them to simply say (after evaluating the technology, of course), "Parliament created us for the purpose of seeing that sustainable energy technologies are developed and commercialized within Canada. Here is a fine new invention that Canada should want. Let's make it happen!" If they won't do that, 99% of sustainable energy inventions will simply not be developed, or commercialized if they are, just as was the case before SDTC existed. With a new enterprise the funds can certainly be committed piecemeal based on progress, so large sums aren't being gambled on a single impression. And once an enterprise is profitable, the funding can be repaid directly or through taxation of an industry that wasn't there before.

   In fact, why are they not out there simply hiring the people for a "technology development park(s)" to move desirable sustainable energy technologies forward from ideas to prototypes, so they can then go to production? Their funds would go toward product development; their staff would have inventors, innovators and engineers who could work together to accomplish more, instead of just bureaucrats. This would soon attract a retinue of business people, investors and entrepreneurs looking for a product or products to base new businesses on and invest in. Instead the agency has no goals or road map, no plan... no clue. It doesn't foster development of desirable technologies or recognize golden opportunities when they are shoved in its face. SDTC is supposed to take the lead for fostering renewable energy in Canada, but not a trace of leadership seems to emanate from it.

   I have written of this before, especially in TE News #80 after attending an SDTC seminar that supremely disappointed a whole room full of some of Victoria, BC's most innovative and active green energy inventors and product developers. To meet their criteria for funding, it seemed one would have to have a million dollars to start with, and would be blowing money out the windows left and right to maintain a facade of a prosperous company, a "going concern". The truth is it isn't a "going concern" until the product is for sale. The funding (as I've seen happen before) would be used up creating and maintaining the image complete with superfluous office staff, without the proponent ever being able to put it to its intended use of developing the product. Most of the room would have felt it was a dream come true if they had anything like a million dollars to develop their product, and would have been off working on it instead of attending a funding seminar in the vain hope that someone might actually provide even a little seed money.

   In spite of a few improvements since that time, SDTC is then merely a supplemental funding source for large established companies, from which tens or even hundreds of millions of dollars may be sucked by having a "rent seeking" department to put together carefully tailored fund seeking proposals for things the company was probably going to do regardless. Is this good use of our tax money? It certainly uses it up in bulk for the benefit of just a few projects that already have other funding.

   The best the contact I reached at SDTC could do was point me to a couple of other groups. I checked out one, a "green chemistry" group. It was set up with pretty much the same vague, undefinable goals as SDTC: it would help connect "academic research" with "innovative companies". Again not even lip service was paid to inventors or inventions or to getting a new invention from prototype to market stage - the "pre-corporate" stage of development. Pathetic!
   The slow, flaky internet here somehow deserted me each time I tried to check out the other one, 2 or 3 occasions. I finally got there on the 24th. By then I wasn't expecting much, but in fact it looks potentially promising. (ForesightCAC.com - Foresight Cleantech Accelerator Center) So I sent off an e-mail.

   No doubt agencies such as SDTC and NRC would have been proud to support an "innovative company" like (as an on topic example) Cobasys (AKA "Ovonics" - the name seems to flip-flop). But "Cobasys" didn't invent the good nickel-metal hydride car batteries that made the GM EV1 and other EVs of the time fabulous. Stanford Ovshinsky did... after he retired, living and funding the project just on his pension. The whole basis for the company would never have existed without first the individual inventor - working in his garage after retirement for want of any outside support. Then Cobasys was organized. (No doubt Ovshinsky had to do that pretty much by himself, too, before he could get any help.) No amount of money then thrown at Cobasys would have changed the fact that the chief "innovation" of the "innovative company", the one that gave it its reason to exist, had already taken place beforehand. The time funding was most needed was when he was working in his garage to develop the product, the "pre-commercial" period. He might well have got there sooner. That this was the intent of parliament in creating SDTC,  is shown SDTC's own graph:

The time there's no other funding available ("pilot to full scale"),
the time of "Development and Demonstration:", might in most cases
be better described as "development of an elaborated concept to a
production prototype". The "D" of "R & D" - where it takes place
at all - is usually done in peoples' garages and basements owing
to there generally being no funding from any source.
STDC has done nothing to address that.

   By "elaborated concept", I mean that it isn't just a fuzzy idea or one that can't be explained clearly and understandably. It's one in which the parameters and plans are reasonably well defined, and perhaps some preliminary work or study has been done, and there seems to be a reasonable chance that the development if pursued long enough through the vicissitudes of various trials and some inevitable failures will lead to a successful outcome.

   SDTC in its current form (and ditto for most funding agencies as currently constituted) would have been proud to have funded "Cobasys" -- but it would never have funded Ovshinsky the individual inventor to do his battery development or to get Cobasys started. And that notwithstanding the fact that Ovshinsky had already blessed the world with the thin film transistor (TFT) and LCD advances that has allowed us to trade in our picture tubes for hi-rez flat screens and quite a number of related and other advances to computer and energy technologies. The stupendous benefits already accrued to everyone from these entitled him to nothing. Credentials and accomplishments in the inventive world count for nothing when trying to start the next project. It has to be a "business corporation" that already has money coming in to qualify for funding to develop a product. Why on Earth does one have to put in money to qualify to receive funding? Who is supposed to be funding who? Does anybody not think this is putting the cart before the horse? And are corporations not composed of individuals? Why not support the individual inventor, especially those with proven track records, the rare and key player in new technology, even just for the lowest cost so he can eat while he develops the product when there is no other support available to him, instead of trying to simply add support to a whole corporation that already has products and production in the two last stages of the SDTC chart?

   And Cobasys is Ovshinsky's success story. Far more often the inventor is left out in the cold after all his time and energy have been spent. Either the invention is never used or corporations pick up on it and have no reason to tell the inventor they have done so. He can find out when he sees it for sale in big box stores, and having spent his funds in development and having never made any money off of it, he can try at his own expense on his own time and his own burden of proof to sue the perhaps large, well funded transnational company or companies that have adopted his technology, who can use some of their profits from the invention to have lawyers do their work in court -- without those who made the decision to use without a contract (steal) the usually patented invention having to spend any of their time on it, answer embarrassing questions or face personal criminal prosecution. That's not worth trying!

   It reminds me too of the company "Taligent", created by Apple, Motorola and IBM to write a BIOS and operating system for their new "Power PC" CPU chip. (1994?) I looked at their website. Management wanted the prestige of hiring only university trained "computer science" graduates. These are people who can put an operating system to use with application software, not build one. It had no clue what needed to be done to accomplish the simple things they needed, and they weren't hiring anyone who did either. They wanted a "sophisticated" image. They didn't want the real "creative hackers" who could have easily accomplished what they wanted. They would never have hired Steve Wozniak, without whose technical genius Apple would never have existed, nor even the people who designed the Power PC chip itself who would have had a very good grasp of what was needed at the base level. Instead the whole clueless company failed to even begin to create the simple operating system (their raison d'etre) for a marvelous chip, and was finally dissolved. One person, the right person, could have got them well underway in a matter of weeks. SDTC too has failed to acheive the simple results for which they were created, but lumbers on year after year on taxpayers' money.

   The net effect of everyone closing the door in the face of society's small elite of even the most gifted inventors when they have a good and practical plan - or even a developed prototype - for something people would want is not to foster vital progress and innovation but to prevent them from taking place. One way or another seed money is needed - sometimes even just a "pension" to live on so they can do the work instead of taking some mundane job. How many potential inventors with paradigm changing product ideas are out there who haven't got the luxury of both time and income to pursue them? Why not fund a few of the best who seem to have the most promising plans with the greatest potential for changing the world in desired directions -- or directly employ them to do their work? They are certainly entitled to pick and choose the most promising. (This brings us to management having to determine what the "desired directions" are - what are the intended outcomes, preferably in accordance with the seven core values. What should Canada be working toward? Without there being aims and intentions, organizations can only drift. That's another whole subject.)

   We all know not every plan works out. But the ones that do can change the world. That should be duly noted on the web site so as not to provide fuel for the adversarial elements of society who would use anything as an excuse for bad publicity. As one corporate president (?Toyota? ...it was before the internet) once said: "90% of my advertising budget is a complete waste. If only I knew which 90%." That didn't stop the company from advertising. Likewise, we don't know which projects will bear fruit and which won't. Past accomplishments of the inventor presenting the plan might give some indication of the prospects for success. For the pittance of the fraction of 1% of the GDP it would cost, the inevitability of some percentage of failures shouldn't stop us as a society and a nation from having some promising developments on the go instead of supporting none.


   Following this summary dismissal by my native Canada I e-mailed to Sichuan Changhong Batteries in China (to "the president") and pointed out the new nickel-zinc chemistry they could freely have. Given their current obsolescent line of flooded alkaline cells - nickel-iron, nickel cadmium and nickel-metal hydride, made on their production line bought from Varta in Sweden or Germany many years ago when they ceased production there, I felt they might be in need of a new battery like this and would be much more likely to adopt it than a lithium or lead-acid battery maker, and that their "green" markets would really appreciate them. (Of course, their markets would then start expanding, too!)
   About 10 days later, having come up with the simpler rechargeable manganese-zinc and lead-zinc ideas, I wrote to two "DIY" (Do It Yourselfers) e-mail lists, "Edison NiFe cells" and "BatteryConversions" (@yahoogroups.com) and let them know about them. I'll follow up with a link to this newsletter, and mention the other two possibilities to Changhong. Mostly it's the electrolyte that enables all these things. That's the big thing that's been overlooked for 150 years.

   There are still ways I might start manufacturing batteries. It's a lot harder with no support and one must start at the smallest scale to make basic production equipment. Do I want to do it? I've achieved the breakthroughs and can add them to my resume of inventive accomplishments that gives me no credit or 'in' with anyone and no entitlement to any sort of funding for accomplished or present work for future benefits to society. Others can put the advances to use for their corporate profits if they so choose, or it can sit on the shelf for a century and be of no help to anyone. In our western society it seems inventors are supposed to be satisfied with that.
   Or let's go back a little farther, to 2006... Should I have kept up making and selling my new 'Supercorder' woodwind instruments and never have started doing renewable energy stuff? I might have become well known in music circles for it and I would probably have been making a decent or even a good living over the last decade. No green energy developments. No Turquoise Energy News. No battery breakthroughs. No bandsaw mill breakthrough. No potential future breakthroughs (eg, projects as detailed in my "Fantasy Budget" in TE News #118). But I have no rooms to rent now. Canada Revenue's SR & ED tax credit program won't work for me any longer, and they've even denied my 10000$ 2016 application, which money I was counting on. How long can I go on with no real income? Not much longer I expect. (Next step: I've applied for a reverse mortgage on my new house, having already liberated - and used up - nearly half the value of my old house by selling it and buying this one. Sigh!)

Crystal Sponge?

   Some years ago an oceanographic survey found a "sponge reef" in Hecate Strait. Something different than a "coral reef", in water too cold for corals. This caused quite a stir in paleontology as it had been thought that sponge reefs had been extinct since the days of the dinosaurs.
   Now I'm going on memory, but it seems to me the sponges in question were called "crystal sponges". These things wash up on the beach here during the summer. I suspect this is the creature here. Is this what all the fuss is about?

Moles and hydrocortosone cream: Update

   Some months ago I mentioned the idea of putting something on moles and other similar skin blemishes that would treat them as injured skin that needed to heal rather than as a foreign entity that needed to be eradicated. (TE News #111, April-August 2017) I tried it out and said I would write an update if I got any results.

   Attacking the blemish is what one thinks of first, but it rarely works out well. (Usually it's best to leave them alone, although this is not invariably true and it can occasionally mean life or death to make the right choice. It's best to consult a doctor if something on your skin is changing and you are concerned.)

   But I decided to try putting something on them to see if some moles or other blemishes might be coaxed to heal. The only substance I know of that helps to heal instead of attacking is hydrocortosone cream. I very gradually shrank a cyst on my arm since childhood from a considerable lump and irritant to almost nothing with it. It's still shrinking with application, but so far has refused to entirely disappear.

   At first I applied the cream every evening, but not noticing any results, I soon dropped that to every 2nd (even numbered) evening. I did that for the many months it's been since I first wrote about it, and wasn't seeing much change. Then a few weeks ago I stopped. At first I didn't see any change, but then a few of them started getting dark spots and crusty on the surface. Then I remembered that these had been that way previously. The lightening and softening had been so gradual I hadn't been aware of it. Only a very few moles with a smoother texture had stayed dark. So applying the cream had eliminated the dark crusty spots and prevented them from reappearing as long as I was using it. They stayed relatively soft and light in color. It's not a cure, but I'm sure it's significant. It's at least a healthy and innocuous general treatment that can be applied to moles where there hasn't been much of one of any sort available that I've heard of.
   I presently plan to keep it up one day a week, Saturday or Sunday. I'll increase that if the crusty and dark spots don't go and stay away. I'll think of it as something that just needs to be done for better health, like brushing and flossing your teeth.

   I hope that some day someone will invent a cream (or something) that actually heals such things and makes them vanish. That would be a factor in people starting to live 2 or 3 hundred years in the coming decades and centuries. Today people already live substantially longer, on average, than they did 70 or 80 years ago, when the percentage of the population over age 65 or 70 was very small. (whether because of improving living standards and medicine since then, or just because everyone smoked tobacco back then...?) Go back just 300 (or maybe a little farther) years and living to 50 was an accomplishment not shared by the majority. In principle the potential for cultural improvement goes up with life span, as more people gain time to learn and understand what's better and have opportunities to apply and share that knowledge before they exit the planet. (And now we can share much better, via the internet.)

Feeding the 9 Billion?

   A news article asked how we were going to feed 9 billion people "by 2050". I came up with two possible answers which may be seen by some as impractical:
1) Start feeding them now. You'll have fed them all long before 2050.
2) Deposit food in a food bank. By 2050 everyone should be able to eat off the accrued interest.

   Well I might be more concerned if I thought there would ever be that many people. Signs of a crash well before mid century are all around us. 47% of Americans are now having trouble paying for food and rent. Most will never own their own house, and most of those who do have a huge mortgage on it - banks and investors have bought them all with 'free money' - wealth transferred from citizens to the financial 'services' community mainly via unprecedented new money printing, which will sooner or later cause hyperinflation. The whole middle east is in turmoil. Venezuela has crashed. Brazil and Argentina are crashing. China is in bad shape. More refugees than ever before in history are on the move or stuck in refugee camps, and they are destabilizing Europe too. The heat keeps rising so gradually people don't notice it. The "news" doesn't cover it. And now the environment and the climate are bringing natural disasters (some of them actually man caused, I'm sure) into the picture and there are predictions for famine and devastating epidemics. Food production today is adequate at the best of times, but is being hit by climate change, and things beyond human control are happening with increasing frequency and severity. Consider if for example a volcano should darken the skies for a year or two as occasionally happens, it would spark a major catastrophe in food production.
   We are surely on the eve of the gigantic collapse predicted as the inevitable outcome of uncontrolled growth on a finite planet ever since the earliest computer modeling in the 1960s. And the more people there are, the lower everyones' quality of life. Most of our problems have their root in overpopulation. If we were suddenly down to 2 billion people (as we just might be by 2050), there would be almost four times the resources to support each person as there are today. Imagine the relief to the environment. Imagine how the cost of housing would drop if the population was gradually shrinking instead of gradually expanding. In a rapid drop in which homes are not destroyed en masse too, there will quickly be no more homeless people. What is an optimum population for this planet? Surely under half of what we have today (ie, 3.5 billion) and probably closer to 3 billion, or even under that.
   It has been predicted that even a single year may come when the "world as we know it" will be there at the start, but will no longer exist after it. It will be incredibly transformed. The aware may sense that year approaching... or not!


After the Thugs

When the thugs have come to rule the roost,
when psychopathy is the most rewarded trait,

And those in economic power, behind the scenes,
grasp control from the people the reigns of political power
to manipulate them for selfish ends,
and now resent attempts to address imbalances,
or to improve the static status quo,
When the rights and interests of the citizens take a back seat to everything else,
and their attempts to participate in decision making are repeatedly rejected
with indifference, or they are beaten by police,
When a few live beyond the reach of the laws imposed on everyone else,
Such a society is doomed.

And this becomes the best outcome,
because it would never be possible to start again
unless these covert kings were toppled from their hidden thrones,
with the tilted cards laid face up on the table and in the light for all to see,
and the people exercise their democratic rights,
to fairly distribute new cards, change the games and level the tables,
in favor of the seven cosmic values of Social Sustainability
to sustain families and society, everyone equitably, for all future time:
Life, Equality, Growth, and Quality of Life, all executed in
understanding Empathy, Compassion and Love for humanity and the world.

And with these
to create sustainable moralities anew in each progressing generation,
the planet and its peoples,
with an ever advancing populace and society, and a stable population,
headed toward light and life instead of away,
towards our cosmic evolutionary destiny of Universal Perfection,
even as the Father in Heaven,
the first source and center of infinity and reality,
is Perfect.


   We know from the Bible that the world is only 6000 years old. (um, where does it say that, again?) Therefore those early Christians who first postulated that the world is 6000 years old were wrong, since it must have been less than 5000 years old back when they did!

   "in depth reports" for each project are below. I hope they may be useful to anyone who wants to get into a similar project, to glean ideas for how something might be done, as well as things that might have been tried or thought of... and even of how not to do something - why it didn't work or proved impractical. Sometimes they set out inventive thoughts almost as they occur - and are the actual organization and elaboration in writing of those thoughts. They are thus partly a diary and are not extensively proof-read for literary perfection and consistency before publication. I hope they add to the body of wisdom for other researchers and developers to help them find more productive paths and avoid potential pitfalls.

Other "Green" Electric Equipment Projects

Carmichael Mill ("Bandsaw Alaska Mill")

Band Guides - Adjustments & Cuts

   After spending the whole month on one exciting (and mostly successful) battery experiment after another, I finally mounted the "railway wheel" band guide wheels on the saw on the evening of the 27th .

   On the exit side I did the simplest thing: I used the main wheel as the 'other' guide wheel by not putting the new one exactly in line with it, so the band went over it in a very slight arc. (Okay very slight.)

   On the entry side I figured that wasn't good enough and put in the offset wheels design as planned in the last issue: The "railway" wheel in line with the main wheels and the offset wheel pushing the band down a bit against it, again the band making a slight arc over it. (When tension was put on it it swiveled a bit and again the arc was very slight.)
   The attack angles to aim the blade up and down are again both adjustable by the hinged assemblies with springs.

   This brings up the idea of possibly changing the design a bit: The main wheels could be any size, eg, 9 inches, with the guide wheels lower and the band arched under them, making it a "four wheel bandsaw". With the guide wheels lower than the main wheels they can make extra cutting depth. On the minus side, to change the cut width the band would have to be slackened off and retensioned.

   The next morning I tried cutting. I hadn't got the positions set very well. There wasn't enough bend in the band on the entry side, and if it decided to go up, it pulled away from the main wheel entirely. That pretty much upset the applecart. The band would then also pull completely away from tracking on the exit wheel - which also didn't form enough of a bend in the band to be secure. After one crappy board and starting in a couple of feet into another it started raining, plus I had other things to do (band rehearsals).
   But the fixes were in my mind. Both sides would have accompanying top wheels, and they would both cause more bending of the band so a good arc was stretched over each 'railway' wheel. Either that, or I should set both guide wheels down lower per the potential design change above. But if I did that, the guide wheel adjustment springs were upside down. They would have to push up instead of down, with a solid adjustment for pushing down, since they would then be operating against the band's tension.

   On the 29th I fixed up the entry side. I had been having problems from the start with the threads being stripped in the adjustment hole. I had bashed the thin end to squash the threaded hole down a bit more than once, but now again to my surprise, the bolt just skipped threads when I tried to tighten it beyond "easy". It occurred to me to make another threaded hole in my add-on piece, and let the bolt come right through into that. Using that hole as one support as well actually made it easier to make. But once again by then it was getting dark. I didn't get to doing anything with the exit side.
   The next evening (30th) I took the saw out and tried cutting. For some reason it seemed the guide would adjust too far down or even farther down. The spring was fully compressed. (maybe at the top it was about centered.) And I couldn't understand why it seemed so hard to push the saw, even while the cut seemed about right... and the band didn't stop instantly when I released the switch. Finally I realized that the rough cedar board had a lot more friction because it was damp. There was nothing wrong with the cutting, it wasn't the friction. I got a washer and inserted it so the whole adjustment mechanism was aimed up a bit higher. It seemed I goofed on that because when I went farther the cut was veering up. Then it seemed very hard to get it to go level. Then I started cutting slower, and the cut started going straight. I didn't think I was really pushing it before, but slowing down definitely helped.
   When the cut was veering up, the exit side wheel, being just a single wheel, simply let it lose contact and go up. That was actually useful to see when it wasn't going straight. This board didn't have any of the side to side bow shape of many of this saw's cuts, from poor adjustment. But then also I had considerably tightened the band, which doubtless made a big difference.

   I started thinking maybe my neighbor was wrong about the band speed being too high. There's probably a relationship between band speed and the speed you can push the saw without it starting to veer off up or down. If the relationship is as it seemed to be this day, it should go faster so you can cut faster, instead.

   Another possibility I though of for design change would be to have the guide wheels fixed in position (if it's possible to align them very exactly and alike), and have the 'skis' (or whatever part of the saw rests on the board) adjust front to back, so the whole saw angles up or down a bit until it's cutting straight. That would also be the easiest and most obvious thing to adjust from an operator's point of view.

   Then I had a whole new inspiration that changed everything...

Self Correcting Band Guides - Concept

   Most people, including me, just copy what has been done before. We "innovate" but within the existing basic framework. Only occasionally do we manage to step back outside that box and see that the box itself is the problem. Then we "invent".

   Finally, on the night of the 30th, I thought that the cut direction of all bandsaws - all the ones I've ever seen - has a positive feedback loop in it: Once it starts veering off the line, the farther the band strays from the intended cut line, the more it wants to veer still farther off. Freehand in a shop, the operator can compensate by moving the back of the board left or right as he feeds it. On a mill or saw where the saw to board cutting angle is fixed, once it has veered off the line far enough, the back of the band tries to stay on target, while the front leads the way off target and the tilt or attack angle of the band gets worse and worse. The board is ruined, and finally the saw jams.
   A negative feedback loop leads to stability. How to get that? Instead of manual adjustments, the band angle needs to adjust itself to straighten out the cut as it cuts.

   What if it was made so the front of the guide wheels, at the front of the band, was the pivot point and the back of the wheels could be moved up and down, instead of the pivot point being behind the band or at best in the middle with guide blocks? Instead of trying to carefully fix the band aim as a solid constant, there should be some way to make it so that if the cut was going above the set line, the back of the band would be pushed upward further than the front was up, and if it was below, the back would be pushed downward more than the front. Thus instead of veering off and going off worse and worse the more it veered, the band's attack angle would adjust to compensate. Instead of being a finicky adjustment, the cut direction would be self correcting. In fact, if the pivot point was a little in front of the band, this correction effect should be attained smoothly and automatically with little deviation from the path.
   The one manual adjustment point would be to set the springs to hold the back up so the blade enters the wood level at the beginning of the cut.

   Here, I think, is a real breakthrough in bandsaw mill design. I could see this one being the first handheld band mill that will be commercially successful, and wanted by everyone. The serious objection to all band mills, that the cuts sometimes veer off, the one that keeps people looking for a better mill even after they have a band mill of some sort, would be eliminated. In retrospect it seems like a stupidly simple idea - as so many good ideas do in retrospect - yet as far as I've discovered no one ever seems to have made such a guide system for a bandsaw or bandmill before. (This will be fantastic! Now... how to build it?)

   On July 2nd I went and looked at the Woodmizer mill again. I remembered some unusual and elaborate things about how it held the guide wheels. Could I have possibly come up with something that already exists? But no, the elaborate things were to hold the wheels as stiffly and as precisely as possible, not to make them self correcting. If Woodmizer, the premium maker of bandsaw mills, hadn't figured it out, no one had.

Woodmizer band guide wheels and mountings.
Note the elaborate means provided for fine up-down and left-right adjustments,
and for keeping the adjustments solidly in exact place once made...
nothing for self-correction of the cutting path.

Proposed New Electrical Standards: A Low Voltage Standard (36/38 volts DC) and Standard Connectors

   I got no feedback on my ideas from last month. I don't suppose people (even most TE News readers) had or have given such an area much thought. A new idea that occurs to me for the 38 (AKA "36") volt standard is to make not only plugs, sockets, wall plates and other connection components, but to also make (or more likely have others make) 36 volt appliances sold with these connectors - especially LED light fixtures and lamps. Such products would kick-start that voltage as a new standard. They would be marketed to solar and off-grid equipment suppliers and distributors.
   (BTW I vaguely remember some new car a few years ago that was going to use 36 volts for the lights and all. (At least, I think it was 36... lights, starter, alternator, fans, stereo...?) I wonder if that was ever produced, and what 36 volt equipment might have become available? Or did they change their minds because of the lack of available 36 volt equipment?)

   Something I didn't think to remark on in the last issue: When I ran the coffee maker from a 36 VDC to 120 VAC inverter, I had to wire the inverter to the car batteries. If the coffee maker was 1500 watts and thus took 12.5 amps at 120 volts, it must have drawn about 39.5 amps at 38 volts. Thus a 40 amp (or maybe it should be 50 amp?) HAT standard socket in the car and matching plug on the inverter would have made for a simple plug-in solution - no wrenches and no chance of miswiring or shorting something out, which often destroys inverters among other potential problems.
   On further consideration, it might be worth making the 40/50 amp sockets all "click-latch" type. Sudden loss of power from the supply coming unplugged is another cause of delicate equipment failure, and latching plugs to sockets should eliminate that. For simple equipment where it might be a nuisance the plug need not have the latch. (Which gets me to the thought that I haven't defined the latching spec. For now see the CAT standard click-lock plug and socket designs on thingiverse.com .)
   Another thing I might remark on is that a 12 VDC to 120 VAC inverter would have drawn 125 amps from the 12 volts. That's pretty serious current needing very heavy wires, and a 1 volt line drop is 8% power loss. 38 volts DC is a much better voltage for things needing any significant amount of power, yet it still doesn't need a big bank of batteries to attain and no one is in serious electrical danger.

Electricity Storage

Rechargeable Battery Making
with oxalate electrolyte:

* Nickel-Zinc *
* Manganese-Zinc *
* Lead-Zinc *

   I consider that after 10 years I finally achieved the main objective of the whole "better battery" project when I got a nickel-zinc cell to work almost at the start of June.
   However, that particular chemistry seems rather hard to make at home, and I thought it would be nice to find some means for home battery making, or at least for making them with a minimal production setup. So I continued experimenting. Making good electrodes has always seemed to be a difficult and obscure problem beyond the chemistry, and I started coming to grips with that in June.

It's the Electrolyte

   Perhaps the most exciting invention has been the electrolyte rather than any particular electrode or electrode developments. Realizing that an everlasting zinc negative electrode is enabled by this electrolyte is certainly the second most prominent development. Then three fine possibilities appear for positive electrodes: nickel, manganese, and lead.

   The voltage of the nickel, or mixed valence nickel manganates, at pH 12 instead of 14 is such that it appears special techniques must be employed to use it, but the cells have the highest voltage (~1.8v) and energy. I may experiment further, but it looks right now as if it's best to leave this to commercial battery makers.

   Manganese (1.4-1.5v cells) is simpler. Much has been made of manganese forming permanganate and dissolving, limiting cycle life, but at pH 12 there's a .45 volt spread between MnO2 (manganese dioxide) and KMnO4 (potassium permanganate), and I think it's just a matter of strictly regulating the charge voltage to stay about 1/4 volt below the permanganate forming threshold. An everlasting manganese-zinc battery would be very exciting because it's so low cost and such high energy - substantially higher by weight than lithiums. And if there was ever a battery that might be made from scratch at home, this is probably it. I made one that, while not perfect, worked well before the end of the month.

   Lead (1.6v) is also simpler. If the permanganate turns out to be a serious issue in spite of the various improvements, everlasting lead-zinc could be a great alternative. The easy way to get lead dioxide plus electrodes is to take them from a lead-acid battery, but it should be a new battery that has never been filled with acid to avoid sulfate contamination, or somehow the sulfates must be completely removed. The energy density depends on how much energy can be coaxed out of the lead electrode at good voltage compared with its apparently much higher theoretical value. I also made more than one such lead-zinc test cell that worked with lead oxide electrodes taken from a new, unfilled lead-acid battery.

   While I note that a metallic manganese negative electrode, enabled by trace additives (successfully made - see previous issues from 2011-2013) at -1.5v has a higher voltage than zinc at -1.2v and with a nickel plus makes cells of 2.4-2.6 volts, the zinc is easier to use and apparently yields a higher percentage of its theoretical amp-hour capacity.


   Almost since batteries were invented, a cell with a zinc negative electrode that doesn't degrade has been a "holy grail" of battery making. Zinc has soluble ions at any acid pH and also at the alkaline battery pH of 14. There is a range of pH from 7 to 13.5 where zinc doesn't inherently form a soluble ion, but in neutral salt electrolytes such as the ammonium chloride of "good old" single use manganese-zinc (AKA "carbon-zinc") dry cells, zinc chloride is soluble, so the zinc still dissolves. Acid has also been tried. In all cases the zinc electrode deteriorates over time and cycling, giving batteries with a zinc electrode a notoriously short cycle life. Many have tried with no better than marginal success to tame zinc for battery negative electrodes.
   On the same day I put out the last newsletter I was also looking at some Pourbaix diagrams. I looked at zinc and suddenly realized I had done it, almost by accident. My electrolyte was about pH 12, and the diagram showed zinc didn't form the soluble zincate ion below about pH 13.5. Furthermore, unlike zinc chloride, zinc oxalate isn't soluble. The three forms the zinc might take then, metallic zinc, zinc oxide and zinc oxalate are all solid, so the electrode won't degrade. It would work just as well as nickel and has a much higher voltage (-1.2 V versus -.5 V).
   (This reminds me of creating my "Supercorder" alto recorder. It could just as well have been created in the 19th century when so many other woodwind instruments and their key systems were being much improved, but nobody made a truly practical modern recorder until I did in 2003-2006. Likewise, someone could have made this electrolyte even 150 years ago (oxalate salts were known before batteries), and we'd have had cheap rechargeable batteries ever since. But nobody did. The grail was left to me in 2018!)

Positive Electrode Compaction

   I was right onto this at the turn of the month and I kept working on it and thinking about it. By the evening of June 2nd I had realized that the only way I could get the extreme pressure that seemed to be required to condense the loose powder into an electrode "briquette" was to hammer heavily on a very small surface area. Say, the thickness of the electrode instead of the width, and only a thin strip instead of a wide electrode. Maybe 3 mm by 10 mm. It could be any depth because a little powder would be poured in at a time.
   Next morning I looked at an "edge loading" compactor I'd made 3 or 4 years ago to make electrodes for my 3D printed nickel-manganese cells, and started remembering details of its operation. Its compaction using the hydraulic press had been insufficient too, but far superior to most of my other results. The cells at least had something like an amp-hour (out of 5 or 6 expected) instead of milliamp-hours, and could charge at and put out at least somewhat respectable current.
   I would fill it with powder and then compress it 5 or 6 times before it was filled. Each layer was thinner than the one below since it fit less powder each time as it filled. And the resistance measured much lower at the top of each layer than at the bottom, so the material wasn't sliding along the sides of the compactor very well.

   I conceived that the required compactor would have to be made of something tougher than mild steel, which was bound to bend and bulge out, and sure enough this one was tougher 3/8" stainless steel. But it had to be narrower to get the pressure, eg, 1 cm wide instead of 6. Then I thought, what if I put in 6 narrow "plungers" to pound on, each 1 cm wide? If pounding on each one didn't make its neighbors bounce out and just shift the powder back and forth, they could be pounded on individually in sequence, more gradually doing the whole width. Suddenly the whole plan started seeming more possible, at least for test electrodes.
   In practice I ended up cutting up an old file that was the right thickness and using the three pieces as a bottom for the compactor, a filler for one side, both filling in space to make it a much smaller electrode, and the handle as the plunger to pound on. The handle was wider than I had thought of, but it could do one side or the other and it could be angled to hit harder on one corner than the other.

   I also thought about different mixtures for this electrode and whether some might need less pressure to compact than others. First I measured out 30 grams of the mix I had been using (15 cc?) and added 3 cc of Veegum, then I weighed 30 grams of a nickel-manganese mix that I found in the cupboard. This one was much less dense and filled 24 cc, plus again 3 cc Veegum. I didn't use it. (The description of Veegum is in TE News #21... Just ignore the parts of the project where I didn't know what I was doing back then!)
   I compacted the second substance first, adding a bit at a time and a bit of the dishsoap here and there, and pounding it with a 20 ounce hammer until it felt like I was simply pounding on metal. It made a nice "briquette". Resistance from the surface to the back was around 70 kilohms. That's pretty high, but two orders of magnitude better than 7 megohms. I only squeezed in about 60% of the powder or maybe 18 grams to fill the space, about 4 x 40 x 40 mm. I did the second one from the denser original mix and put in the whole 30 grams to get the same size. I pounded pretty hard and was hoping for better results, but it was again ones of megohms. I fear something would bust or bend if I used the 6 pound maul. Then the other briquette started giving very similar figures. Then this one gave the lower values. Perhaps there's some ionic conductivity going on. Simply reversing the test leads always gave widely different results, so one couldn't just say "It's X ohms" as a repeatable figure. All very frustrating!

18 grams in (.4 * 4 * 4)cm = 2.8 g/cc
30 grams in (.4 * 4 * 4)cm = 4.6 g/cc

From these figures it can be seen that if the two powders give the same energy per gram, the second one will make for a more compact battery, with savings in space and case material. IIRC, both of them are higher than I remember seeing for typical nickel hydroxide electrodes: 1.8 g/cc (or was it 2.2?).

   Perhaps I should start experimenting with mixtures for best performance, which would really mean lowest resistance with the pressure I can attain.

Glue & Jell

   Concurrently with the chemical and compaction trials, I again started putting "stuff" into the electrodes to help hold them together as thickeners, binding agents and gels. Pretty much the same ones I was using years ago. With the successful electrodes to compare with I could tell if the additive(s) were causing a problem. They helped. As I proposed and later research also discovered, jelled electrodes can last forever. (I wonder if they got the idea to try it from Turquoise Energy News?)
 See TE News #99: http://www.theweathernetwork.com/news/articles/scientists-accidentally-make-batteries-that-last-a-lifetime/66934/
 -- 200000 cycles with no deterioration, to the surprise of the researchers.

  The first was something called "Veegum", a powdered smectite (bentonite) clay: Magnesium-Aluminum Silicate. I put in by volume maybe 15-20% Veegum into the electrode. It thickens and adds viscosity. The powder is very light, so it wasn't much by weight. The next was Sunlight original formula dishsoap as a "glue" or "gel" with its organic sulfate and sulfonate compounds. (I mentioned it from as early as TE News #6 in 2008.) I just added drops of this as I compacted electrodes.
   The clay seemed to help hold them together without wrecking the electrodes chemically. I used both of these in both the plus and minus electrodes. I considered using "CMC gum" as noted by other battery makers, but I didn't add any. I note that in the literature on "Veegum" there are several varieties, and "Veegum Plus" has CMC gum as an ingredient, probably better dispersed in the clay than by adding it separately.

Zinc !

   I was searching through my collection of Pourbaix diagrams for reaction voltages and characteristics that might make a good metal hydride with nickel but wouldn't corrode. The second last one in the folder was zinc. I had early cast zinc aside as an electrode material because it is notorious for deteriorating rapidly. Regular manganese-zinc dry cells are pH 7 but in a chloride electrolyte in which zinc readily dissolves. In pH 14 alkali, potassium hydroxide, it's better but it temporarily forms a soluble ion before it turns into insoluble zinc oxide. In either case, the electrode gradually loses substance, and worse, it grows "dendrites" when attempting to recharge it - tentacles of zinc between the electrodes. These tend to short out the cell sooner or later. (Not so different from zinc's heavier cousin, cadmium.)

   Now I stared at zinc's very simple diagram. It showed no soluble states between pH 7 or so to over 13. When I had looked before I was using potassium chloride electrolyte which zinc could dissolve in. Now I was using potassium oxalate, and zinc oxalate is insoluble. (That most potential battery elements including zinc were insoluble as oxalates was one of my original reasons for noticing oxalate, going through the "solubility table" in Wikipedia. But I was thinking of oxalic acid at the time.)

   Many have tried to 'tame' zinc for rechargeable cells without more than partial success. Some nickel-zinc "AA" alkaline cells are (or recently were) available which claim they have 500 cycle capacity (I suspect only under ideal conditions). I finally concluded the researchers were wasting their time, and looked for other metals. But every reported attempt I've looked at - several patents or studies - was in pH 14 hydroxide electrolyte except one. That one, in acid, had zinc intentionally dissolve as the cell discharged, and reform as it recharged. It still grew dendrites.

   Here is something quite different. As with nickel, I don't think anyone else has tried oxalate or "mid alkaline pH" other than chloride, with zinc before. Apparently zinc/zinc oxide/zinc oxalate would stay solid at all times up to over pH 13. And higher voltage negative electrodes might be better in ethaline DES than in water, too. Nickel-zinc-oxalate cells would give around 1.8 volts. Whatever nickel's extra amp-hours (as a negative electrode), it wasn't likely to match the energy of zinc with this high voltage when zinc is also a good energetic electrode. Nickel-zinc would take fewer cells to reach a desired voltage: probably 7 cells for 12 volts instead of 8 or 9 or 10.
   As an added benefit zinc is cheap, much cheaper than nickel, which will mean, in principle, pretty cheap batteries.

   So I started pondering: Would zinc insist on growing dendrites and deteriorate at pH 12 to 13 in oxalate? Or would it really last forever as this Pourbaix diagram indicates it should?
   Since my cells are already just right in every other respect I decided it would definitely be worth an experiment or two with zinc powder instead of nickel powder.

   I decided to make a square cell just the size of the above electrodes. I finally recalled that the positive electrodes, however well they are compacted, swell up once wetted (at least in water - I'm not sure about ethaline DES. Either that or it's with charging and discharging.) In a cell the right size, they'd have little room to swell. I used the 30 gram electrode.

   The zinc electrode turned out to be just 20 grams, about the same size because it was compacted in the same compactor, but I nipped a bit off the edges so it would fit inside the paper basket, so it was about 37 x 38 mm. (theoretically 20 g * 820 AH/g = 16.4 amp-hours... in such a tiny, lightweight cell? How well will that actually be utilized? Prospects seemed good for high utilization because the zinc powder, unlike most metal powders, seems to be of "nano" size, providing potentially the highest surface area.) After working with zinc powder I knew exactly how they gave the tin woodsman his color in the Wizard of Oz movie. (Did you know the entire plot of the book was a metaphor for the whole twisted financial system? "Follow the Yellow Brick Road to the Emerald City"... does that not sound like the promise dangled in front of everyone? The "silver slippers", silver having been until recently the money of the common innocent person (Dorothy), turned into "ruby slippers" in the movie. It didn't matter because nobody knew the underlying plot anyway. After this book families demanded more kids fairy tales about Oz, and the author had found his calling. They are however apparently not as memorable as the first. I only read one other, "The Land of Oz". It was okay but the plot wasn't the equal of "Wizard of Oz".)

Making a New Cell

I went to put the new cell together and did everything wrong, out of sequence. As a result, the plus electrode was pretty much broken up into chunks and powder.

Here's how it ought to have been done:

1. Torch the positive electrode with a strong  propane torch for about 10 seconds to sinter the particles together a little better. (I finally remembered about this. After so long doing no battery work I had forgotten on the last cells.)

2. Make the case but leave one side and the top (front) face off.

3. Cut the graphite current collector piece and paint it with the osmium doped acetal ester. When it's dry, put it in. Put a small bolt and washer through to the plastic support before the tab gets broken off.

4. Slide the plus electrode into the case - easier with one side off the case - then glue the last side on with methylene chloride (or plumbing ABS solvent glue).

5. Make a [watercolor, thick] paper 'basket' to cover the bottom electrode and go up the sides so the electrodes won't short against each other.

6. Put the zinc minus electrode in the paper basket.

7. Cut a piece of nickel mesh to cover the back of the zinc electrode for a current collector. (Unless the zinc electrode is totally discharged to zinc oxide, the nickel will remain in metallic form.) I made one with 3 folds and a fourth bit.

8. Stick a piece of cupro-nickel sheet in for a tab. I tucked it under the last fold in the mesh for better contact.

9. Fill in any remaining empty space in the cell. I used bits of cupro-nickel sheet.

10. Take out enough bits to expose the mesh and add a little calcium hydroxide powder. Add some DES with potassium oxalate dissolved in it. (Put the bits back in.)

11. Put the top/front face on. (I just held it shut with a C-clamp so I could open it again. Later I put heat glue around the edges and the terminals.)

Ready to charge!

   After I torched the plus electrode I couldn't get any resistance readings. It was off the scale. I hoped it would improve with charging and discharging. When I put the cell on charge, it only took .3 mA at 2.2 volts. It was going to take forever (approximately) to charge up to a few amp-hours!
   But maybe if the plus electrode has no room to bloat out, its conductivity will go up over time instead of peaking and then going down? Then I discovered the filler pieces sitting on the counter. somehow I hadn't put them back in when I put a C-clamp on to clamp the front cover closed. So it was pretty loose inside until I fixed it.
   After a while, after that, it was drawing 2.3 mA. (forever / 8) Then I got out the heat glue and sealed it up better around the terminal tabs and the clamped-on front. While a nickel negative charging to lower than the voltage of hydride wouldn't react to oxygen entering the cell, zinc would certainly self discharge.

   How ironic that after a little demo with nickel-zinc in oxalic acid a decade ago and then casting it aside, I now end up doing nickel-zinc in oxalate. I little knew what I almost had, how close I was.

HOMEMADE! Manganese-Zinc Batteries

   Many readers will recognize this chemistry - the most familiar of all, so-called "carbon-zinc". I thought about how to make a manganese-zinc cell because the reaction voltage of manganese is lower than nickel, so the ethaline DES shouldn't be required, and graphite or conductive carbon black can be used to make the electrode conductive. That would solve the biggest problem for making them at home.
   We already know MnZn works in water both at neutral pH (standard dry cell) and at pH 14 (alkaline dry cell). And onto that topic... there are already MnZn cells everywhere! Is it necessary to make your own at all?

   So since they already exist and much is known about them... How would MnZn be for performance? Wikipedia says MnZn alkaline "D" cells have 12 to 18 amp-hours at a nominal 1.5 volts. (Capacity decreases as current rate increases. A cell might have 18 amp-hours if discharged at 50 mA and only 12 amp-hours at 500 mA.) The old alkaline D cells I tried put out 7 to 8 amps short circuit current. Admittedly it would take a lot of cells in parallel to get 300 amps - about 60 or 70 of them. It takes 8 cells to make 12 volts, so 24 for the 36 volt car: around 1560 cells to run the car. 65 cells * 12 AH/cell * 36v = 28080 WH - quite a respectable EV battery. 147 g/cell * 1560 = 229 Kg = 506 pounds. Now we compare that with the lithiums in the Sprint: 38 V * 300 AH = 11400 WH, at 260 pounds. To get the 28080 WH they would weigh more. (640 pounds)
   Apparently even manganese-zinc "D" cells are better than lithiums - if they can put out the required current and if they can be recharged reliably. If they were made differently (eg, rolled-up electrodes like Ni-MH, Ni-Cd - or flat plates) they would put out higher current and fewer would be required to get the 300 amps. (Nickel-zinc is somewhat higher voltage and so still better.)

Oops, "Charged" refers to MnO2 state area and "Discharged" to
the Mn2O3 region. These words do not refer to the lines with
arrows that they are immediately above, which represent the
"overcharge" and "charge" electrode voltages.
   The one caveat to potentially making rechargeable cells with manganese positive electrodes is that if the charge voltage is a little too high, the manganese dioxide turns into soluble permanganate ion (MnO4-), which degrades the electrode. If it can be charged slowly and kept under that voltage, how long might it last? We don't really know, because it's always been used with zinc electrodes, and even charging slowly the zinc electrode degrades so the cell doesn't last.
   So we're in uncharted territory here, but it appears that using a less alkaline pH also prevents soluble Mn(OH)3- ion from forming even if the cell is "overdischarged". So it just might last a long time as long as the rather exact charging voltage condition is always met. (With jelled electrodes it just might last forever, again if charged exactly each time.)
   An interesting aspect of "overdischarge" with manganese is that if there's enough zinc, the cell will continue working well putting out about 1.1 volts while the Mn2O3 converts to Mn3O4, and then at about .9 volts as it goes to Mn(OH)2. It has as many amp-hours in these two states together as it has at the full 1.5 volts, so double amp-hours but at ever decreasing voltages.

   From the Pourbaix diagram we see that at pH 12.7 the manganese electrode needs to hit about .25 volts in order to charge back to MnO2. But if it gets above about .7 volts, soluble ion MnO4- (makes the electrolyte purple) is created.
   Combining with the -1.2 volts of the zinc (Pourbaix diagram in article above this one), we need to charge at somewhere over 1.45 volts to charge at all, but under 1.9 volts to avoid forming  permanganate. If we charge the cell at say 1.7 volts, in theory it will last a long time. If it doesn't we could try 1.6 volts with a longer charge. If 1.7 seems flawless, 1.8 could be tried but it would be more likely to cause eventual trouble. 1.9 volts would be too much and would cause deterioration.
   In charging (eg) a 12 volt battery (of, say, 8 cells, 8 * 1.5 = 12.0; charge at 8 * 1.7 = 13.6 volts) one will have to be careful that all the cells in series are at the same state of charge. One can't have (eg) two cells at 1.6 volts and another forming permanganate at 1.9. It may or may not be necessary to have some sort of BMS that bleeds off any excess above 1.8 volts or even 1.7. But I'm getting into too much supposition for an idea that hasn't been tried yet.

   Once I thought about using MnZn, I thought about just using already made dry cells and changing the electrolyte from chloride or hydroxide to oxalate. That simple step might make them rechargeable.

   The alkaline version - the whole existing battery - would be perfect except that the can, which is the plus electrode current collector, would probably oxidize away. Unless the oxalate formed a surface layer that protected the rest underneath, as happens with aluminum and titanium oxides. It was worth a try. But how to get into the cell to change the electrolyte, without wrecking it? (Probably take the bottom off of it.)

   In the non-alkaline version, the standard dry cell, the plus electrode goes to a carbon rod in the middle. That would work great! It would be perfect except for the zinc - the outside can. It's just a piece of sheet metal with little surface area. In the chloride electrolyte it works by dissolving and continually exposing a fresh layer during discharge. (until the zinc has a hole in it and the electrolyte leaks out of the cell all over the inside of your device.)
   So to use a 'standard' dry cell, we would want to peel off the zinc outer can (eg, needlenose pliers). Put some cardboard or thick paper over any breaks in the separator paper and then spread some zinc powder around it. The less damage done to the inside the better. Then tape the outer can (or something else?) back on over the powder with plastic tape. The zinc outer can is sometimes an inner layer surrounded by a steel outer can. (Accept only the cheapest cells, ones without the steel!)

   With either type, once it's open, immerse it in pure water to dissolve out the chloride or hydroxide electrolyte. (Be especially careful with the caustic hydroxide in the alkalines. Wash any off your skin immediately and wear safety glasses. DO NOT get any in your eyes.) After a while drain it/them on something to blot up the water, and change the water in the tray to further dilute the electrolyte, at least three times over several hours.

   I decided to try the regular dry cells first. I didn't find any "D" cells in my junk, so I broke open an old square 6 V lantern battery with four "F" cells, a little taller than "D". The outer zinc in these particular cells had completely corroded away and fell off in flakes when disturbed. That was perfect! On closer examination, much of the separator paper was gone. But the inner plus electrodes were there complete with carbon rod current collector. I had ready made "half cells"!


   Then I went for an alkaline "Duracell" "D" cell. I peeled off the outer plastic around the bottom. This exposed a cardboard(?) separator and the bottom metal cover. I stuck in a knife and pried, and the cover started coming out, revealing the rod of zinc powder/paste. I started thinking it might be really hard to get it back in once it was out, or even very far out, and suddenly remembered that I only had to get the cell open a bit. I put it in pure water to dilute out the KOH (potassium hydroxide) electrolyte, however quickly or slowly that happened. I tried an "Eveready" cell. The bottom cover was visible but recessed and would have been harder to get open. A second "Duracell" was different from the first with a smaller metal piece on the bottom so the cardboard was exposed around it once the plastic was removed, but it came open nicely too. Several "Evereadys" all looked the same. Soon I had 4 "D" Duracells cracked open in the water along with the four "F" manganese electrodes. If the alkaline cells worked, all one would have to do was collect them from recycling bins to make rechargeable batteries easily and almost for free.
   The voltage read a bit over 1.5 volts. These discarded batteries put out about 7 or 8 amps. Not huge, but better than anything I had made! Let's see... I just needed 8 for 12 volts. How many were in the bag? 7. That figures! I tried to get an "Eveready" open with a chisel and hammer, but even once I got the chisel under it wouldn't come up. I finally ripped the bottom apart with the chisel to get it off. Even then I still wasn't sure I was into the cell because there was another metal bottom inside. It wasn't connected to anything and only the little brass center pin was the negative terminal. (Later I pried a Duracell farther open and found it also had another metal piece inside blocking access to the bottom.)

Lead-Zinc Batteries

   Another interesting possibility occurred to me. Lead hydroxide in both oxidations (Pb(OH)2 & Pb(OH)4) is insoluble and so is lead oxalate (PbC2O4). A Lead electrode would probably last for ages in 'mild' alkali oxalate. (Even more valuable if manganese proves not to work very well in spite of the better electrolyte and with careful charge voltage control.)
   The voltage of about +.49 volts at pH 12 is virtually the same as nickel at pH 14, and higher than manganese dioxide's +.25 volts at pH 12.

Mn-Zn: +.25 - -1.2 = 1.45 volts
Pb-Zn: +.49 - -1.2 = 1.69 volts (Hmm, not bad at all!)

To make lead-zinc cells, one would open a lead-acid battery, dilute out the acid, and collect the positive plates, discarding the negative ones. (Presumably they wouldn't work very well if charged backward to become positives? Then again...)
   Make zinc negative electrodes. Reassemble the battery with them instead of the lead negatives. If there's a bunch of space under the electrodes, fill it in with something inert like plastic so you don't need so much electrolyte.

   Presumably the battery will be lighter. Zinc is a lighter element (65.4) so it will take less of it to match the number of atoms in the Pb (207.2) electrode. The voltage per cell is less so instead of 12 volts for 6 cells it'll be 10.14 volts. 7 cells would make 11.83 volts - close enough? (But maybe not very convenient for arranging or packaging - and not for using a regular Pb-Pb battery charger.)

12 volts - with _?_ cells
Lithium/?       - 4 (12.8v)
Ni-Mn/OxDES - 5 (12.5?) [Ox=the new oxalate/lime electrolyte. DES = in ethaline DES instead of water.]
Pb-Pb/H2SO4 - 6
Ni-Zn/OxDES  - 7 (12.6v?)
Pb-Zn/Ox       - 7 (11.8v?)
Mn-Zn/Ox      - 8 (12.5v?)
Ni-Ni/OxDES - 10
Ni-MH/KOH  - 10
Ni-Cd/KOH   - 10

   I decided this would be worth a try. But at least for the first try, I would get a new battery that hasn't been filled with acid yet. It will be safer to cut open and the electrodes won't be contaminated with sulfate, which zinc dissolves in. There are several electrode plates to be made, and it would all be a waste of time if they decayed because of that.
   On the 12th I ended up buying a very small motorcycle battery because those were the only ones around that weren't pre-filled with acid. It weighed about 1670 grams (empty), or 278 g/cell.
   I broke the case open just under the lid with a hammer and chisel. The works stayed with the case rather than with the lid. That was because the lead terminals were punched (melted?) together through small holes in the walls between cells. Hitting the terminals with a chisel broke the connections and I pried a cell out.

   The 2 plus plates were about 2.7 mm thick and together weighed 125 grams. The 3 minuses were 1.65 mm and weighed 122 grams. The pluses were 58.5 mm x 62.5 mm. The minuses were 56.5 x 62.5. The fibreglass mats were folded over the plus plates to cover both sides. There were thus 4 interface faces, totaling about (57.5 mm * 62.5 mm) * 4 = 144 square centimeters.

Lead dioxide-Lead [lead acid] cell.
The brown ones are the lead oxide "+" electrodes


   Some calculations, with much speculation: It fits that both electrode sets were similar weight because both sides move two electrons per reaction, and the lead is way heavier than the oxygen. If they put out, say, 200 mA per square cm with zinc negatives, that would be 28 amps.
   It was probably under about 4 amp-hours(?) (@ 12 V = 48 watt-hours; a whopping 48/1.670 Kg = 29 WH/Kg), holding about the energy of two alkaline "D" cells (15 amp-hours @ 1.5 V = 22.5 watt-hours; *2=45 WH; 45/294 g = 153 WH/Kg). (Typical of lead-acids, it didn't actually say. But it said "Fast charge at .8 amps for 3 hours"... or "Slow charge at .4 amps for 5 to 10 hours" - somewhat vague clues! That's where the "much speculation" comes in, because that's not enough to charge 4 amp-hours if it's actually completely discharged.) But metallic lead is theoretically 259 AH/Kg. Both sets of electrodes in a cell weighed 247 grams: 1/4 Kg... * 259 = 64 amp-hours. Of course, not all of that weight was active substance, obviously these little cells were nothing like even 30 amp-hours. Maybe the charging spec wasn't for "completely dead to full". If it was for 50% to full, they would be 4 or 5 amp-hours and 28 or 36 watt-hours per kilogram. Those would be typical lead-acid figures. Evidently actual lead amp-hours in a battery is under about 10% of theoretical.

   A 100% effective set of zinc electrodes would need to be: 5 amp-hours / 820 AH/Kg =  6.1 g. I expect it would be overconservative to assume only 10% effectiveness for zinc. If one assumed 20% (probably still overly conservative), the electrodes would need to be about 30 grams. 30/122 = 1/4 the weight of the lead electrode. So even with the lead plus side the electrode weight is dropped from 125+122 = 247 g down to 125+30 = 155 g, or 63% of its original weight. But then it's only ~1.65 volts instead of 2. If it was 30 WH/Kg, it becomes 40. If it was 40, the highest figure I've seen for lead-acid, it becomes 53 watt-hours/kilogram. So lead-zinc is an improvement in energy density by weight over lead-lead. And zinc can probably deliver well over 20% of its theoretical, in which case the battery would be headed for just over 50% of its original weight and 60 to 75 WH/Kg.

   The 62.5 mm height of the electrodes was fortuitously the same as the width of my edge compactor. For the zinc electrodes I could get enough pressure. But these figures seemed disappointing for energy density.

Manganese Dioxide again

   I started thinking that in spite of lower voltage, manganese dioxide would be better than lead -- assuming it could be kept from forming permanganate. The prospects for that seemed quite good. After all, at pH 12 there is .45 volts to play with between MnO2 and KMnO4 - and once again that spread is higher than at pH 14. Surely in the middle of that it somewhere could take a full charge yet not make permanganate? One would charge with a steady regulated voltage near 1.7 - NO pulse charging and never reaching the 1.9 volts where permanganate should start to form. I suspect that reputed poor results come from sloppy work and never really doing what it takes because of the expectation that "it won't work anyway": charging with unregulated voltages or pulse charging, the degradation of the zinc electrode is actually to blame for cell degradation, over-discharge at pH 14 causing another soluble ion to form (like zinc, it doesn't form one at pH 12-13). Also the voltage spread between "charge" and "overcharge" is a little less at pH 14 than at pH 12 to 13.

   Since the alkaline cells were only open a crack at the bottom, and since after days they were still bubbling out brown stuff, I looked again at the "F" size standard dry cells. I measured them as 29.5 mm diameter and 66 mm tall and 92 grams total weight. One might estimate that about 70 grams was MnO2. IIRC the capacity of Mn metal as a negative electrode is 975 AH/Kg. The plus side only moves 1 electron instead of 2, so 487. And MnO2 has 2 oxygen, weight 55 (Mn) + (16*2=32) (O2) = 87. Then, 55/87*487=308 AH/Kg of MnO2.
   So the 70 grams of MnO2 in the "F" electrodes should theoretically give 302*.07 Kg = 21 amp-hours. And this is surely more manganese than is in alkaline "D" cells, for which 12 to 18 amp-hours are claimed. But, per the Pourbaix diagram, when the cell voltage drops from 1.5 to 1.2, it has only given half its potential amp-hours and still has half as much charge to give as it had at 'full' voltage, and another half as much at .9 volts. So if there is enough zinc to keep it going, it can put out double amp-hours - at lower voltages.

   The conclusion might be that manganese dioxide actually yields a far greater percentage of its theoretical amp-hours than lead does. I wouldn't bank on that conclusion until and unless I test both, but that's how it looked as of June 13th.
   Then I found that if I put an "F" on a paper towel, it would absorb the water. That gave a way to be certain of diluting out the chloride fairly quickly, since the removed water would carry it off. 3 or 4 baths in water with drying out between should leave it .999 chloride free.

Circumference = pi*d. So 29.5 mm * pi = 93 mm. 9.3 * 6.6 = 612 sq.cm. If one could coax 20 mA/sq.cm from the cell it would put out 12 amps.
   From a thinner electrode than the fat centers of big dry cells perhaps one could get 50 mA/sq.cm or more, ie 30 amps and up. If I went back to thinking of my long ago original intended cell size of 4" x 6" (3.5 x 5.5 I.D.) that would be 124 sq.cm. 50 mA/sq.cm*124=62 amps. NOW we're talking EV batteries again!

   Back to today (13th). I sprinkled and then brushed some lime onto/into one of the "F" MnO2 electrodes all around the outside. I wrapped a piece of watercolor paper around one of the four MnO2 electrodes. Then I cut a zinc strip to cover it. Then I cut a piece of nickel foam to fit in that. Then I measured out 36 g of zinc flake/powder. Again I was taken by surprise how light and fluffy this metal powder is. It made a pile about 3/8" deep on top of the sheet. I nonetheless rolled it up, zinc spilling out on all sides. (lost at least 12 grams.) Owing to the thickness now enclosed, the zinc strip which should have overlapped didn't meet at the ends by 1/4". I held it on with 3 small cable ties. It was definitely fatter than a regular "F" or "D" cell. I wrapped it up with packaging tape. By some miracle there were no shorts.

Wrapped in paper and preparing to cut zinc sheet;
Wrapping up zinc powder inside zinc sheet.

   Most of the electrolyte just ran through and out the bottom. I just set it in (more like on) a beaker and added more. The voltage was about 1.1 volts. Since the zinc was metallic, that presumably meant the manganese was discharged to Mn3O4. It only charged at a milliamp - just like all my homemade cells! At first I thought well, maybe Mn3O4 is a high resistance semiconductor and it'll get way better when it's charged up toward MnO2. But how long would that take at such a slow rate?
   Then I thought that it should be far more conductive because of the graphite or conductive carbon black in the electrode. After all, the biggest reason for using manganese was to get that conductivity.
   After an hour it was up to 2 mA, and short circuit current was up from 3 mA to 6. Voltage was still 1.1 tho. Until it started to rise, there was no telling. I went out to cut and burn tree branches. When I came back (<1Hr) it was drawing 4.5 mA (still fluctuating a lot, to almost 5.5). was 1.46 volts very soon after taking it off charge, and short circuit current was up to 80 mA and 70 mA after 10 seconds - again fluctuating considerably.
   I went out and came back. It was drawing under 5 mA again. I started wriggling wires. When I touched the one on the plus terminal current jumped to 22 mA. I've never been impressed with the performance of alligator clip leads. They're better than nothing for temporary hook-ups, but sometimes not much. Soldering the crimped-on wire to the clip is part of the answer. Maybe gold plated clips would work better? (Gold does have real uses.)
   By an hour later it held 1.500 volts for 6 minutes. Short circuit currents had gone from 250 to 330 to as high as 400 mA. Now it was starting to seem like a battery! It was still charging at a paltry 20, 22 then 25 mA. At least it was going the right direction.
   In another hour it was only charging at 18 mA again, but it held over 1.56 volts for at least 4 minutes and delivered over 1/2 an amp for almost 10 seconds. At its best it was delivering well over an amp. So if it started out looking as crappy as my cells made from scratch and then gradually got good, how close were my cells to working well?

   The next morning (14th) something seemed to be wrong. A wire may have become disconnected? I played a bit with the wires and it ended up charging at 110-130 mA. If I took it off charge the voltage started dropping rapidly. I thought a short must have developed between the electrodes. That would be no surprise with the separator sheet just wide enough and all that zinc powder going everywhere when I put it together - I was more surprised that it had worked in the first place. I took it off longer and discovered that the voltage only dropped to 1.386 volts and stayed there. That didn't seem like a short, so I put the charge back on. A while later it was charging at 140 mA and the "stop" voltage was up to 1.392. It seemed as if most of the battery had been still "asleep", somehow disconnected, and now much more of it was coming on line and starting to charge. In another hour it was drawing 150 mA and the voltage was down again (1.360). Another segment "coming on line"? Then I tried load testing it and after that charge current dropped to 100 mA again. Mysteries! Another 3 hours and it was at 170 mA and dropped to 1.26v. Huh? Apparently my suppositions weren't right.
   A short and yet not a short?... at first I thought it might be a trace of zinc getting around the separator sheet. Or, was it possible my conclusions from the solubility table and the chart were wrong, that zinc could still grow dendrites and short out the cell in oxalate at pH 12?
   Then another possibility occurred to me - that maybe I hadn't diluted out the ammonium chloride electrolyte well enough. After all, the alkaline cells were still putting out brown stuff (and still several amps of current if shorted) after several days in water, if not a week. I had no way of actually knowing the salt was all gone. (I'm not saying there's no chemical way to test it, but I'm not going to try to figure it out.) It seemed like the sort of symptoms one might get. Charge and zinc tentacles grow, let it discharge and the tentacle dissolves again as zinc chloride, so it starts charging again... a vicious circle and the one that has prevented zinc electrodes from working well.

   I had planned to put together a second similar cell with a wider separator sheet and hope for better results. But thinking perhaps not all the chloride was gone I put the three remaining MnO2 electrodes back in new water for a while to further dilute it in case there still was any.

   I also checked and found that the same '1-1/4" PVC irrigation system' pipe I used for making NiMH battery tubes was about the right size for a case for my new, fatter, Mn-Zn cell(s), originally "F" or "D" cells. I could already make end pieces with the hole saw that was just about the right size. That was if just using the (Duracell) alkaline cells didn't work out. With them it would just be a matter of putting in the new electrolyte and closing them up again. That was the path of least resistance if it worked, but they were still oozing brown stuff and not ready to try yet. The "standard" dry cell MnO2 electrodes did nothing visible in water.

   On the 16th I started thinking that even if the alkaline cells worked, the electrodes weren't jelled and so would have a limited life span. I didn't want to spend my time on something that might have a shorter life than lead-acid. The "F" cell Mn electrodes were also pre-made, but they were accessible from all around the outside. I came up with a new plan for making the next cell:

First process the electrode so it lasts (hopefully) forever:
1. Dilute Sunlight dishsoap, Veegum, and calcium hydroxide in some water. Not too dilute, but so some of it would absorb deep into the porous electrode. (I added more and more water, but it was always a pretty stiff mix.)
2. Paint this mix onto/into the electrode.
3. Paint in some less diluted lime around the outside.
4. Wait for it to dry (dry in oven - oops, it melted the plastic "washers". Electrode however is still good.)
5. Lightly torch the electrode from all sides. (shield the plastic if it's still good.)

   I have all the raw ingredients for MnO2 electrodes. The only difference between this and a completely homemade cell is it's already done: The MnO2 and graphite are fully mixed and compacted, with a carbon rod with a metal cap sticking out to connect to. ...And the clay and dishsoap have to be added separately, painted on. The torching is after, and the CaO (Ca(OH)2) ia painted on after anyway.

6. Tape on a big enough separator paper that there's no chance of any zinc getting around the edges anywhere.

7. Make a zinc or nickel-brass or cupro-nickel shell (tube) for the outside. This time, make it about 120 mm long instead of 100 so there's an overlap instead of a gap. Have a tab to connect to, again on the top at the outside.

8. Make a zinc electrode (per below?) and set it on the shell piece.

9. Wrap it up and stuff it into a PVC shell. Hmm... this one was pretty loose. I thought of stuffing in an extra roll or two of zinc sheet to push the outer electrode in, but then I decided to leave it. Currents might be low, but it should take a lot to short out this cell!

10. Pour in some electrolyte.

11. Seal the top. (Modeling clay [or wax, heat glue...])

12. Rats - the bottom leaks around the edges!

   This second cell started out at 1.4 volts and would put out about 270 mA into a 1 ohm load. Charge settled in to about 40 mA. It was different from the first, first in working right away, then in not getting much better as the hours of charging passed except that the voltage went up. First, being rather loose, the zinc wasn't much pressed against the separator paper. Then, because in spite of my efforts to seal the bottom it leaked, I couldn't fill it with electrolyte to make a "wet cell". So it was no surprise that you couldn't get more than about 1/3 of an amp out of it.
   Off of charge (which had dropped below 10 mA after some hours) it held over 1.6 volts for a few minutes, and over 1.55 for quite a long time. I resolved to try again to fix the leak(s) and fill the cell. But first I thought I'd do a load test as it was. I also decided maybe I shouldn't be trying to make cells using this pipe and end plug system. Maybe use old soup cans? They're much fatter than I planned, but at least they wouldn't leak. (Yikes, they might hold something like 150 amp-hours each - just 48 big cells for the 36 volt, 300 AH car battery. Instead of 36 much larger and heavier lithium cells. I'm starting to see where the "drive for two days without recharging" could actually be a possibility. But the cells will have to be made so they can put out very heavy currents for use with EV cars. Maybe "prismatic plates" cells (like most lead-acids) instead of just fat, round electrodes. Nevertheless with lower but still reasonable currents they'd be good for off-grid electricity storage. Perhaps that's a good first target market?)

   I put a 100 ohm resistor on it - 15 mA at 1.5 volts. It was almost 1.6 volts, but was quickly down to 1.5 when the load was connected. After 1/2 an hour it was 1.410 V, and after an hour, 1.350 V. Two hours: 1.275 V. 2.5 Hrs, 1.25 V. 3 Hrs, 1.225 V. Occasionally it would go up 10 mV or so for no apparent reason and start down again from there. I stopped after 5 hours at 1.142 volts. (bedtime!)
   Okay, tens of milliamp-hours. (63 by my count.) Better than some of the "milliamp-minutes" results I've been getting. Now, what would it do when flooded with the electrolyte so everything was in direct contact between electrodes? And with the top covered so oxygen wasn't getting in?

   The next morning I found a separate cause for low maximum currents - and jumpy voltages: as I disconnected the cell to try out the new lead-zinc one, the metal cap fell off the carbon rod. Or at least half of it did. It was corroded and making poor contact. I did a quick short circuit direct to the rod and found it would put out 1/2 an amp momentarily.

Zinc Electrodes

   On the 14th I thought, instead of trying to compact the zinc powder, why not put a nickel plated copper mesh on a steel plate, pile zinc powder on it and smooth it off, and then torch it not just to make a "skin" on it, but to melt all the zinc particles together so it becomes a single thin "briquette"? So I did that. The torching sort of worked, and the top sort of held together. But under the screen it was still loose powder. I put a plate on top and flipped it over.

   I guess I torched the second side a bit too long or too hot. Suddenly the zinc powder caught fire. I tried to blow it out 3 or 4 times, to no avail. It burned fiercely with flames, with flames as I recall a few inches tall. What was left in 10(?) seconds was yellowish calcined zinc oxide.
   My first conclusion was that zinc powder has a lot of energy in it - it should make a good battery electrode!

   I did another one but I only torched the one surface, more lightly. Then on the 17th I was going to put together a cell and my brain came to life. "What about the jell? That's what's supposed to make electrodes last forever."

   I took the soap and Veegum mix and added it to a small ointment jar full of zinc. This made a zinc paste. If I had used a bigger jar or less zinc I wouldn't have made a mess. But once it was mixed, no more zinc powder going everywhere. I took the nickel plated copper screen and pasted it up. It was about 30 grams. (Hopefully well over 20 amp-hours of zinc. Since the largest use of Veegum is in cosmetics and the zinc powder color is identical, I think I've uncovered the secret of the Tin Woodsman's makeup in The Wizard of Oz movie.)
   I put this in the oven for an hour at 275°F to dry before (one more time) torching it to put a "skin" on it next to the separator paper. (Oops, no picture.)

I put it together into a second Mn-Zn cell. Drying the MnO2 electrode in the oven at 270°F caused the plastic top piece to shrivel. It performed better than the first one, but the plastic pipe leaked and after running some load tests I got tired of it.

   Meanwhile still on the 17th, after I had done the Mn-Zn cell, I thought I would try out the lead-zinc, Pb-Zn. That should be a high current cell.

   I cut a sheet of zinc flashing the size of the Pb motorcycle battery with a tab to connect to. How was the zinc paste going to stick to that? I took the plate, the large flower pin frog and some foam out to the hydraulic press. With the foam for a cushion, I pressed the zinc into the frog's (?)teeth. I stopped at 2 tons pressure. I was already risking having the pins suddenly bend over. The pins never go as far into the material as I want, even soft zinc, but it did manage to raise ragged edges and make small holes.
Then I buttered the toast with zinc butter; in this case, one piece of toast, buttered both sides. It weighed 20 grams, mostly "butter" - more than a match in amp-hours for the two lead plates (125 g), I suspect. I'll put this one "-" zinc electrode between the two "+" lead oxide plates. If I don't think it's enough current, I'll make two more single sided plates and put them on the outsides of the lead plates. Four faces interface instead of two should double the current.

   I put it in the oven at 260°F for an hour to dry the paste. That ended the day.

Lead-Zinc Cell

Right: Lead dioxide electrodes painted with CaO
Left, testing: Not much to see, cell 'sealed' with modeling clay

   I torched it the next morning (18th) and painted calcium hydroxide on all four faces of the lead plates. [Later: Impure chemical! Apparently the long unsuspected cause of self-discharge in all my cells.] Then I put the Pb-Zn cell together in its original compartment in the Pb-Pb case and added what seemed like quite a lot of KC2O4 electrolyte to fill it and flood the plates.
   It started off reading 1.888 volts. That seemed like a good start! Of course both electrodes should have been pretty much fully in their charged states. I put a 1 ohm resistor across it and it started with about 1.6 volts (1 ohm, so volts=amps), but dropped in 10 seconds to somewhere just over an amp. Then I put it on charge for a few minutes at 2.0 volts input and it was drawing 90 mA. (With 2.1 volts input it was over 200 mA. With lead and zinc, there is no "too high" voltage as with manganese, just "too much current" - however much current that is, making bubbles and heat.) Then I left it off for a few minutes and it was sitting at about 1.9 volts. With the 1 ohm load it held 1.53 volts steady after 10 seconds. When I put it back on charge it drew 220 mA, dropping to 110 in a few minutes.
   For a second time I took it off charge and after a few minutes it was 1.93 volts. With the 1 ohm load after 10 seconds it was doing 1.55 V. I kept it on for 30 seconds, at which time it hit 1.50 V and the 5 watt resistor started getting warm. Back on charge it started off drawing 270 mA. (Hmm, I had the power supply limited to that value!) Soon it was less, but I raised the current limit on the supply to 1/2 an amp.

   These several tests occupied less than an hour. Lead might be heavy, but I was starting to like the performance of the chemistry! Or was it the performance of the configuration?: relatively thin plates of lead and zinc instead of absurdly fat electrode cylinders. Or was it the combination? I decided I should try out an Mn-Zn cell with several thin plates instead of thick electrodes. OTOH, Mn is very slightly soluble in oxalate. And how sure was I that it wouldn't form permanganate? Might it not last 'forever'? If not, how long would it last? I looked up the figures again in the Wikipedia Solubility Table. Well, really it's only twice the solubility of nickel hydroxide, and nickel [oxy]hydroxide electrodes in potassium hydroxide last a very long time.

Solubility of Metal Oxalate
(units: grams per 100 mL)
Solubility of Metal
Oxide or Hydroxide
6.5 * 10^-4
1.6*10^-4 (II)
7.2*10^-11 (IV)
(no figure given)
8.1*10^-7 (I)
1.7*10^-6 (II)

   I note that the solubility of potassium oxalate as an electrolyte is similar to that of potassium chloride (34.2). Also that any calcium that forms oxalate is likely to be pretty inert. That will probably help further stabilize the Mn oxides in the electrode.

   The voltage was also appealing (call it 1.71429 volts nominal under load?), thus needing 7 cells to make 12 volts instead of 8.

   I left it off charge for a couple of hours. The voltage was down to 1.773. Bubbles were coming up every few seconds from somewhere under the negative terminal. The voltage of the zinc being higher than that of hydrogen dissociation from water (the lower dotted line in the Pourbaix diagram), any air entering the electrolyte should be causing the zinc to discharge itself and make hydrogen gas. The cell needed to be, if not sealed, at least capped off so air wasn't circulating into it. (This also applied to the Mn-Zn cells.) (Also note that zinc wouldn't work in water except for the "hydrogen overvoltage" effect of the zinc itself that keeps the water stable.)
   After a couple more hours, it was down to 1.730 volts. I covered the top with modeling clay. I put it back on charge and at first it drew .5 amps. It soon started going down, but I raised the current limiting to 3/4 amp.
   In spite of the modelling clay "seal", it still seemed to drop down to 1.73 volts in a while. But why not? According to the diagrams it should be Pb +.48 and Zn -1.2 = 1.68 volts. Why should I expect it to stay up at 1.8 or higher? Just because it was 1.888 V when I first filled the cell, I guess.
   I tried a .5 ohm load. It held 1.25 volts (2.5 amps) for 10 seconds, and was still about 1.1 V after 30 seconds with the resistors getting warm. Then I tried a plain short circuit. It started out a little over 5 amps and dropped to about 4.85 after 10 seconds. I can't help but reflect that that's less than the 8 amp alkaline "D" cells. However, it also probably has less common surface between electrodes. Current should double if I put two more plates on, even if they had only a few grams of zinc.
   Later I put it back on charge from 1.677 volts. I lowered the charge voltage to 1.9 volts, and it was drawing .43 amps. Later on it was down to around 60 mA and when the charge was removed the voltage drop was much slower, taking somewhere around 1/2 an hour to drop to 1.8 volts. I left it off overnight and in the morning it was 1.711 volts. With a 1 ohm load it then put out just over 1.5 volts and 1.44 after 10 seconds. Back on charge it started out at .32 amps. I cut the charge voltage from 1.9 to 1.8 (actually about 1.83 - the power supply has no fine adjust) and that current cut in half to .16 amps. In an hour it was down to 40 mA and I could hear bubbling from inside the cell. I 'unplugged' it and the voltage dropped only very slowly. By 1.79 volts it was only dropping 2 mV/minute. I think I'd call that "fully charged". Toward the end of the day it was at 1.693 V. 1 ohm load, 10 seconds: 1.395 V. A little later it read 1.685 V. I connected .5 ohms again, and again it put out 1.25 V for 10 seconds.
   I put it back on charge at 1.8 volts at 7 PM. It started out at 250 mA and worked its way down to 90 in 15 minutes.

   By this point, on the 19th just the day after making it, I tentatively decided that nickel-zinc was pretty hard to do and the current drive and longevity of manganese-zinc were somewhat questionable. Lead-zinc seemed to offer the most certain way forward for small production.

Load Test

   A while later I took it off and did a 5 hour load test with an 11 ohm resistor. (I couldn't find a 10 ohm except 1/4 watt.) So at 1.6 volts, that's 145 mA. It sat at 1.774 volts and dropped to 1.726 when I connected the load. It was down to 1.7 in 3 minutes, and 1.6 in 23 minutes. It hit 1.5 in an hour and 1.4 in a little over 2 hours. In 3 hours it was around 1.35 volts. After that I forgot about it for 2 hours, and found it dropping pretty quickly. It crossed below 1.000 volts almost right on the 5 hour mark. I'm sorry I missed around the 4 hour mark, when it might have been around 1.25 volts - and I might have turned it off then.
   Adding up the current for each hour using the average (the mean) voltage during that hour: 1.6v/11ohms=.145 A (*1 hr = 1.45 AH); 1.45/11=.132 AH; 1.375/11=.125; 1.25/11=113; 1.15/11=.105. Total .145+.132+.125+.113+.105=.620 AH. Another disappointing result. But it wasn't entirely dead yet, and it was a lot more than I've had from anything else so far. Theoretically if the 20 grams zinc electrode was even 15 grams of zinc powder, 15*.820 mAH/g = 12.3 AH.

   If the 125 gram lead electrodes were maybe 70 grams of lead dioxide we should have had .820*(65.4/(207.2+32))=15.7 AH. Well...! It was certainly never going to be 15.7 AH as a lead acid cell. Typically lead-acid is around 30 WH/Kg, and each cell had 1/4 Kg total of lead, so 7.5 WH. Since it would be 2 volts, that's 3.25 AH. If I have all that right, apparently lead only yields about 15% of its theoretical capacity in practice. I'm sure zinc is higher. I decided to fill one of the cells with acid and test it as originally made.

   I did that and within 15 minutes concluded that it was going to put out way over an amp-hour, with low voltage drop, so my zinc electrode must be the weak link. Either that or the electrolyte doesn't work very well... but that would explain low currents, not low amp-hours. The next test would be to use the oxalate and calcium in a lead-lead cell. Lead-Lead at neutral to alkaline pH is just one volt instead of two, but it should work fine. I ran the lead-acid cell 5 hours and having delivered about a whole amp-hour, the voltage wasn't down much at all. It was obvious it was going to run for at least several more hours no problem.
   So at least that proved definitively that my zinc electrodes were the problem.

   I didn't get around to trying the Pb-Pb in alkaline oxalate.

   I wonder if the lead can be coaxed into putting out more of its theoretical amp-hours per kilogram in the alkaline cell than in acid? Maybe not likely, but not impossible either. And it certainly seems likely it'll last far more cycles.

My Crappy Electrodes

   Unless I could make better electrodes, I wouldn't be making real batteries! Well, perhaps I should have known even a zinc powder electrode has to be compacted? Or was it the other way around? Should I have a coarser zinc powder so the electrolyte can get in better?  I had noted in a paper I read that their zinc powder was much coarser than mine - "20 to 200 mesh" where mine was virtually a nano powder. These people had experience making their "AA" cells. Maybe my powder was so fine it was behaving like a solid and the electrolyte wasn't getting in. In fact, maybe that's been my main problem trying to make good electrodes all along. Don't most electrodes look grainier than mine? I concentrate on "super fine" to get the most surface area per volume of particle, and I compact the heck out of them, but if there are no spaces for the electrolyte, none of the inside is surface area! I might as well be using solid sheets!

   I took a chunk of cast zinc from an old ship electrode, that had big coarse crystals, to bash - or file - into bits. I ended up filing it. I used a fairly fine file and a very coarse one to hopefully get variety. I don't suppose that with so many rather coarse particles this one'll give the full 8.9 theoretical amp-hours from the 10.9 grams of zinc either. But if the electrolyte wasn't penetrating and now is, performance should be substantially improved. (Perhaps a certain amount of the fine powder would be beneficial too, as long as it wasn't enough to clog things up, but I didn't want to try that in first exploring the "grainy electrode" idea.)

Fine "nano" powder and grainy powder zinc electrodes

Hydraulic pressed: the nano powder one worked worse, the coarse powder one worked better.

   I put the new electrode in and the cell back in its 'slot'. It read 1.6xx volts. I put it on charge at 1.8 volts first (=1.83 today). It started drawing 250 mA. A .5 ohm load test showed it to be not as good as the other one. It dropped below 1 volt before 10 seconds. I suppose the key is mixed sizes including both the coarse and the very fine in the right proportions.
   Or... maybe it still is in compacting, even with an electrode of conductive zinc particles? The new electrode underperforming the old one, I decided to try just squashing the old one down with a few tons of force once it was dry again. For that I put it in the oven at 270F again. Then I pressed it between two pieces of stainless steel at 12 tons. I swapped the electrodes back. But it wasn't as good as it had been before I pressed it. That seemed to confirm that the powder was too fine and not letting electrolyte in, because squashing the particles made it worse.
   I took the grainy electrode and pressed it at 10 tons. After that it had that sheen and didn't look grainy except where the lower areas were. It also had excess electrolyte oozing out of it as I had forgotten to dry it, and now it had less volume to hold liquid. I fear I might have overdone it. It worked better than it had before.
   But neither one was even as good as the first one with the fine powder just made into paste and pasted onto the zinc sheet. This wasn't going the way I expected or hoped!
   Might they improve with cycling? I put on a 1 ohm load for a minute (it dropped to 1.21 V=A) then 'blasted' it with 2 volts of charge, from which it at first drew an amp and was still doing .4 some minutes later. Perhaps it was just starting from the high charge, but with the 1 ohm load it held 1.28 volts after 1 minute the second time. Third time I let it run 4 minutes and it crossed below 1 volt at the end. I turned the charge voltage down to 1.9 but it still did 1.5 amps when reconnected and stayed over an amp for a while. The fourth time proved almost identical to the third, except I ran it for another minute (5 minutes) and down to .94 volts. (Hah! Around 6 amp-minutes: 1/10th of an amp-hour.)  Fifth try was almost the same as the fourth -  15 or 20 millivolts higher toward the end of the 5 minutes. But it was probably just a bit more charged to start. The sixth (and last for the night) was virtually identical to the fourth. Not the sort of promising cycle-by-cycle improvement I was hoping for. It's disheartening to watch the voltage drop and drop by the second when the lead acid stayed over 2 volts for hours. But I was actually cycling it, at least a bit (and rather quickly). Significantly it seemed to indicate stability: it wasn't getting worse with each cycle, which my earlier cells in potassium chloride usually if not always always seemed to do.

   So I thought, what about one more try at making a practical electrode? I could use some coarse powder and some fine powder, and compact it but much less, 3 or 4 tons. (and that's over the whole electrode, 35 sq.cm. So 4000/35 = 115 Kg/sq.cm.) Was there anything better than perforating the zinc sheets as a backing? I had no zinc mesh or 'zinc foam'. Then I remembered I had some zinc wire... Nope. Zinc coated wire. I decided that wasn't good enough. There were also little bits of the zinc sheet metal. I could cut them into really thin strips and throw them in. I can't think of anything I have that would be as good as the powder being just right. Maybe some "U" shaped zinc strips well squashed down around the edges ("coined", like the lead electrodes) to help hold things together?
   Or, seeing that I had managed to perforate the zinc sheets, maybe a flat "pocket electrode": zinc powder inside a shell of perforated sheet zinc? The pin frog, even pressed in 3 or 4 times into each sheet, couldn't make anything like the density of holes of commercial pocket electrodes. Still, with thin strips of sheet thrown in and different sizes of powder, it seemed to have promise. (And don't forget to etch the sheets this time!)
   It would seem that what's really needed are a lot more tiny pores. What if I added some sort of tiny "sandy" grains to the electrode that would dissolve out in a solvent, leaving little holes? That just might make a tremendous difference. But what substance and what solvent?

Zinc Sheet Electrodes: Best Ones Yet!

   On the 21st I thought: I wasn't getting anything like the theoretical amp-hours from the zinc... what would I get if I just used etched zinc sheets? I folded up my perforated, etched sheet that I had intended to make into a "pocket electrode" cover and stuck it in the cell. The voltage started at 1.70 and rose in a few minutes to 1.76 by itself.
   With a 1 ohm load it dropped to a volt in one minute and to .85 in two. That wasn't very good, but it wasn't so much worse than my others, either. And it soon sprang back to 1.625 volts, so it was at least partly a question of insufficient convoluted/fractal interface area rather than insufficient material.

   So... what about several perforated, etched zinc sheets simply stacked together? I also thought about running some through the rolling mill and making them thinner - same volume but more surface area with less "dead weight" interior. Then too, one might add a little powder or coarser filings between the sheets, too. I didn't see any way that could be anything but helpful. And then press it down to make it thinner and less lumpy, but not to the extent of forcing the sheets into close contact.

   So I tried it. I put a textured roller on the rolling mill to give yet more surface area. It wasn't practical to put the 3" wide sheets through the 3" wide rolling mill because they didn't stretch evenly so they were always running over the edge. So I cut the strip lengthways. Even with 1.5" width I sometimes had to stop, unscrew the pressure, and realign the sheet. That made it narrower than the Pb electrodes. I simply folded it up, and I did it a bit crooked so it came out wider than the individual pieces- about the width of the Pb electrodes.

Before rolling: 13.5" x 1.5"
After rolling: ~19" x 1.5" (460 mm x 37 mm = 170 sq.cm; 170 * 2 faces = 340 sq.cm)
Etched and not yet brushed off zinc sheet                        
Before etching: 20.3 g
After etching: 18.3 g

   I tried a 1 ohm load without charging first. It wasn't as good as the first electrode, dropping to 1.18 volts in one minute. But it didn't drop below a volt for well over 3 minutes and it ended 5 minutes at .86 instead of .84 or less. When I put it on charge at 1.8 volts it drew over an amp at first. I left it until it was down to 90 mA before I tried again. It wasn't nearly as good. Later it was only drawing 40 mA. I tried an 11 ohm load and compared it to the figures from the first 11 ohm test. It started out almost 90 mV lower and kept underperforming until at 1/2 hour it was well over 100 mV lower. I stopped there and recharged. I started 11 ohms again for a few minutes and it was only 60 mV below - an improvement. So I tried 1 ohm. It was better than the second time. It did put out over 5 amps monetarily when shorted, but that dropped quickly to 3.5 in 10 seconds and under 2.5 in 30.

   Somewhere after that I made a dozen more perforated zinc sheets, but I haven't used them.

   Well, it didn't change much over a couple of days. What about the ultimate in simplicity: one plain sheet of zinc next to each face of the lead electrodes? With just one thickness, they didn't need to be perforated to let electrolyte through to the sheets behind. On the 22nd I cut just three sheets, rolled them just enough to texture the surface and then etched them. (hmm... no pictures)

before etching: 21.4 g
after: 20.25 (6.75 g each)

   Now there were 3 plates instead of 1, covering all 4 faces of the lead oxide electrodes. (BTW I used a different cell and Pb electrode, next to the first one, which I didn't take apart.) The voltage was way over 1.8, so I tried a short and got well over 6 amps, briefly. A .5 ohm load test showed it was much overloaded, soon dropping under a volt. I tried 1 ohm. It dropped to just over a volt within a minute... then it rose back up to 1.11 volts, and dropped very slowly from there. After 5 minutes it was still 1.043 volts. Then it went up over 1.1 volts and I shifted the aligator clip leads around. (Darn aligator clip test leads - so often poor connections!) It rose to 1.179 by the 7 minute mark, and I turned it off at 8 minutes doing 1.092 volts=amps. It was apparent it would run for quite a while - at least substantially longer than the others - even if it was somewhat overloaded.
   I left it on charge some hours and it had dropped from 1.3 amps to to around 30 mA charge current. A short circuit yielded over 7 amps for a moment. I didn't leave it shorted because I wanted to try an 11 ohm load test. That started off discharging 25 mV below where the first and best one that I recorded showed, for the first 10 minutes. But this time I wanted to see where it was after an hour or two. For the first hour it was dropping about as fast as the first one. But after an hour and 15 minutes or so, the rate of drop decreased so that by the two hour mark, it was virtually the same voltage. with 3 zinc sheets no heavier in total than the first single electrode.  By the 3 hour mark it was starting to drop faster. The performance didn't seem very different. But by 4 hours the voltage was probably above. At 5 hours the other one had been down to 1.000 volts, but this one was still 1.160 so I decided to keep it running. However it was 1 AM. Knowing that batteries recover voltage but quickly drop back to where they were when a load was resumed, I disconnected the resistor and left it idle overnight. In the morning it read 1.474 volts. When I put it back on, it took a while dropping back down, starting at 1.348 volts and still being 1.245 after 10 minutes. It took 1/2 an hour to get back down to where it had been. To make up for it, it continued dropping faster. And as I surmised, once below 1.1 volts it dropped rapidly. This electrode ran for 6 hours and 11 minutes instead of 5 hours 0 minutes - over 20% longer. Amp-hours delivered: .140+.130+.125+.115+.109+.105+.019=.743. Even if that's under 5% of the theoretical 16 amp-hours of 20 grams of zinc, it's still a working electrode. And 743/620=20% more than the first uncompacted fine zinc powder electrode, the previous best. It recovered to 1.318 in half an hour or so and drew all but 2 amps when placed on charge.

   So even while they got nothing like the theoretical amp-hours of zinc per weight, the 3 simple plates seemed to outperform any of my attempts to make a zinc powder electrode. Yet surely most of the zinc is locked up in the interior of the sheets. Only the [textured and etched] surface wetted by the electrolyte reacts. Interface area of the first electrode was 52*64mm*2 sides = 70 sq.cm, and for the second one with 4 sides facing the PbO2 electrodes, 140. So to put out, say, 1.5 amps, would be 21 mA/sq.cm for the first electrode but only 10.7 mA/sq.cm for the plates. It's not high, but over 10 is at least in the right order of magnitude for a "real" battery. Zinc is cheap. It depended what one wanted to do with the battery. If weight and size were no object (ie, off-grid electricity storage), one merely needed to use enough of it for the current and storage requirements. That would be a good start for a market, then if the electrodes could be improved sufficiently one might go for the EV market.

24th: Further 1 ohm load tests (far faster than 11 ohm tests!) showed gradual improvement in the zinc plate electrodes. Running it until a little under one volt, the cell discharge went from 6 minutes to 7 to 6 to 8. 8 is 33% better than 6. Then I upped the charge to 1.9 volts in order to decrease the time between runs. Maybe it would get up to 10% of the zinc's theoretical capacity or better by itself with cycling?

   The lesson I took was that solid plates are "good enough" electrodes - better than anything else I've made. And they're the easiest to make and zinc is cheap. The thing to do then, if the plates don't have much "oompf" per square centimeter of interface area is to get more square centimeters. Naturally the first thing I thought of was to have lots of big plates all stacked up - and of course with positive plates in between. But what about a plate with an accordian fold or similar? That could have a lot more surface area, albeit by being substantially thicker. That would be easier to make if the zinc sheet was thinner and softer. If it was (eg) copper I would roll it thinner and then heat it red hot to anneal it, then roll it thinner again. But the last time I heated some zinc - powder - too much it caught fire. Would sheet fare better? What was the best way to do this? And just how thick was it, anyway? The calipers said .0075" or about .2 mm. Actually, that's pretty thin already. There has to be a certain amount of inside strength when the surface is being converted to oxide and back, to hold the electrode together. But what is a realistic minimum? Is .0075 already there? .00375" electrodes would be half the weight of .0075".
   Perhaps the thing to do is to buy some different thicknesses to try out. OTOH, zinc "moss killer" strips for roofs is cheap - and probably all the same. I decided to see what was available locally (nothing) before looking on line.
   Or... what if I just etched them longer? (or in new ferric chloride instead of 'worn out'? or both?) Perhaps they'd develop more convolutions at a micro or nano scale... and the longer they're etched, the thinner they'll get. So it's also another way to try out thinner pieces.

Shaping Plates to gain More Surface Area
(about 2x desired sizes)
   I could imagine some high-pressure stamp that imparts a convoluted shape and texture to the plates. A bit like the pin frog, but just texture rather than holes. (Though holes wouldn't hurt.) Or perhaps a convoluted roller to run the sheets through... my jewellers rolling mill was the right idea but it doesn't have the really convoluted textures I'm after, and the lower roller is always smooth - only the upper one has textured rollers that can be swapped around. It would be nice to get the amp-hours figures up to 15 or 20% of theoretical - or even 10%, rather than the 5% of the set of 3 plates I started with.
   Then (23rd) I thought of a stamp that might rather readily be made: thin plates of hard steel bolted together, every second one sticking up a little farther making "fins" and "troughs", maybe 2mm wide and 2 or 3 mm deep. Two of those would be pressed together with the zinc between them to make up-down ridges. The zinc would (hopefully) stretch out to make the edges of the troughs, so it would be the same zinc with more surface area for its volume. (2 mm deep with straight sides would double the surface area. Pushing it even further together (somehow), somewhat closing the tops of the troughs on both sides, would bring still more plate substance and surface into the same space.)

   But I'll leave this topic now. [And later I found another idea: etch in lots of little holes and have two plates.] The plates even flat are sufficient for batteries. So we have electrolyte and negative electrodes that can be done at home to make practical working batteries. Next: what matching electrodes might go between those plates on the positive side?

Zinc Plates and What?

   What about lead oxide electrodes? Buying new batteries to salvage from for homemade is costly and seems self defeating. One might make them of course, but taking them out of old batteries is probably a much easier, and cheap, way to get the best electrodes. But how to get ALL the sulfate out of the plates? Zinc sulfate is soluble, so there can't be any sulfate left in the cell. I started thinking about chemicals. But what about just hot water? Might the sulfates be boiled off?

   But I started thinking of manganese electrodes again. Lead electrodes are heavy. Even if I only care about getting working batteries at first, I'd certainly like to keep improving them incrementally until they're good for EV batteries, and Mn-Zn would certainly be lighter even if the zinc in Pb-Zn was weightless.

Manganese-Zinc 'Prismatic' Cell with Copper

   If I was making low capacity plates from zinc , why not try making matching low capacity plates of manganese oxide and stack them all together? The question is first what to use for current collectors and second how to 'glue' the oxide and graphite powder mix to them.

Copper as a current collector?                
Similar to zinc the "Cu(OH)2-" ion shown here only develops
above pH 13 at room temperature. But transitions near "zero
volts" don't make copper a very appealing electrode substance.
Around manganese's +.25 volts transition region however, or
even if "overdischarged" to about 0 volts, copper doesn't
change state, so if a surface oxide layer forms and stays,
it won't be subject to change over time with cycling.
   When I used copper, nickel or stainless steel metal (also solder) in a positive electrode before, they corroded away in no time. That seemed to leave only graphite. But back then I was using potassium chloride for electrolyte, and most metals are soluble in chlorides. In potassium oxalate, might a metal form an oxide skin (eg, CuO) and stay solid inside, as aluminum does in air? It seemed worth trying. If I used a copper mesh and pasted the powders on, they should have a fair chance of staying put. Then if the copper didn't corrode away, it should be a great electrode. I could also try the nickel foam and stainless steel mesh. [Thought of Later: as well as, and maybe especially, monel or cupro-nickel, which is noted for very strong resistance to corrosion.]

The area of the electrodes was 35 x 70 mm for 24.5 sq.cm interface, times two zinc plates is 49 sq.cm total.

I pressed the electrode, the copper mesh with the powder sprinkled on and with a little of the dishsoap squirted on top from the bottle, simply between two flat pieces of stainless steel with 10 tons force. It came out just a couple of millimeters thick and I trimmed the edges, which flaked off easily. The copper mesh with a wire for a terminal weighed 3.8 g, and when the electrode was pressed onto it, 13.3 g.
   The composition of the MnO2 electrode was 20 cc of Mn oxides mixed with graphite or carbon black from dry cells (unknown proportions), mixed with 4 cc of Veegum and wetted with a couple of grams of Sunlight dishsoap.

Etched Zinc sheet electrodes.
Just the left one has been brushed off after etching

   On the other side, two pieces of zinc weighed 8.8 g before etching. I forgot to weigh them after but I etched them for 30 seconds instead of 20. I'll estimate 7.5 g.

   The body of the cell was a square box of ABS plastic, made quite a long time ago. I dumped the previous test cell pieces that were in it and cleaned it out.

I put the cell together with rayon cleaning rag as a separator (fat material like the lead-acid cell's separators), and then I added a couple more similar pieces around the outsides to fill in the excess space in the cell.

   The cell started out around 1.3 volts and drew over 100 mA when I hooked it up to charge. That was more encouraging than the previous Mn-Zn dry cells. After a while on charge, it would put out an amp momentarily if shorted. But unlike most of my cells, that current soon stopped rapidly dropping off, at .76 amps. It was still doing .75 amps after 10 seconds... and .72 amps after 30. This is why I wanted an electrode substance that conductive graphite could be added to. It's so much easier to get a good conductive electrode.
.75 A / 49 sq.cm = 15 mA/sq.cm. That's not high, but it's in the target range. Later it gave an amp for 10 seconds, still later 1.6 amps. Then disaster struck... The copper terminal wire appeared to disconnect from the mesh and currents went way down. Wiggling it around brought it back, but the currents were never as high again. It needed a rivet or something to secure the connection.
   I found a bag of little rivets (thankfully it was stashed in a "battery related" box). I took the cell apart to see what could be done at this point... and discovered something remarkable. There was only one zinc electrode instead of two! I had somehow not installed the other one when I finally assembled the cell with the separator and filler cloths. That meant it had been doing this fine job with half the interface area I thought, 24.5 sq.cm, and that meant 30 mA/sq.cm current density. And the 1.6 amp short circuit current was 65 mA/sq.cm. These figures are definitely in good, practical battery current density range.
   Now, about that connection... It was too damp and messy to think of soldering even if solder wouldn't corrode away. I poked a hole through the electrode next to the wire and stuck a rivet through. I managed to squeeze it flat with pliers without the electrode falling to pieces. I reconnected the cell, this time with both zinc electrodes, and put it on charge for a short while. Then I shorted it again. It only gave 1.3 amps, but it stayed up at well over an amp for 10 seconds and was still .90 amps at the end of 30 seconds. It was after 1 "AM" (really PM - not yet midnight with the clocks 2 hours ahead of the sun) so further tests with a full charge awaited morning.

   25th: In the morning it would barely put out an amp, and charging currents seemed to be down too. My idea of the most likely cause was swelling of the MnO2 electrode because the cell had extra space in it that would allow it to expand. Perhaps the felt-like cloth wasn't a very good idea? But later I found it was just the copper wire again making poor connection to the screen, as wiggling it made currents jump around. Once surface oxide has formed it's hard to get the solid connection needed to the terminal. It could be that the copper was corroding away, too. It didn't seem to be when I had the cell apart. But if it was, it would become obvious in a day or two(?) when the terminal fell off. If that was the case, I could (a) try another metal (b) paint the copper with the osmium doped film (c) go back to graphite positive electrode current collectors. There's nothing there that can't be made to work one way or another. But I just don't think - so far - that any special measures will be necessary.
   I also tried refilling the electrolyte, but it hardly took any. Later with the lid still off and liquid to the top I could see it was bubbling. It may be that the DES will prove to be a better electrolyte for any zinc electrode regardless of the plus side. Or I should be adding hydrogen overvoltage increasing materials to the zinc. (How to add antimony sulfide to a zinc plate? Maybe I should reread that patent, where they made various zinc alloys and ran various experiments with them?)

   After weeks of pretty intense battery experiments and developments in May and June, I had other things to get done.

   On the 26th I disassembled the cell and inspected the copper mesh and terminal wire. They were of course covered with black copper oxide, but they didn't seem to be corroding away. In chloride they'd have been long gone. It looks like I've found the metal (or at least one that works) to use for pH 12 with oxalate, in place of the nickel used at pH 14 with strong hydroxide. This is another important find, albeit subsidiary to the electrolyte itself. A metal current collector is better than graphite both for mechanical strength and for current capacity.
   And I didn't see any dissolved purple "dye" color that would indicate the manganese was charging to permanganate, which would probably cut short the life of the electrode.
   The disassembly however didn't do the electrode any good. Some of it fell off the mesh and stayed on the separator cloth and I rinsed much black powder out of it. I put what was left back together in fresh(er) electrolyte. The short circuit current was down to under 1/2 an amp - and it still varied if the wire was moved around, since I hadn't tried to fix it.

   It appears this battery, and by extension future ones of similar design, is a success needing just a single piece copper current collector. Or maybe cupro-nickel, so noted for high corrosion resistance? (Can you melt two pieces of copper together with a torch if you put some flux on them?) I'm pretty sure they now just need techniques for limited manufacture, then put them to use and see how they fare over weeks and months - and years - of time.

Self Discharge Problem Solved (I think)

   27th (rev.30th): There is, after all this time, still a self discharge problem with every cell I make. It's quite variable so sometimes I've thought it had become trivial, but it's always unacceptable. However I believe at long last I've found the culprit by process of elimination. I've tried a number of chemistries now and they all have the same problem. Especially we know for sure that both regular and alkaline Mn-Zn cells normally have very low rates of self discharge, having a shelf life of a couple of years, so it isn't inherent in the electrochemistry. And the lead-zinc cells with the commercial lead oxide electrodes have it too, so it couldn't be the Veegum or the dishsoap because I didn't add any to those, or to the first Mn-Zn "F" cell. The problem could have been the separator paper, but the lead cells (lead-zinc) came with their own separator sheets like "blotter paper" or "felt paper"(?) that must work fine (at least in sulfuric acid). And I tried some felt/cloth separators in the last cell instead - it didn't help.
   So by process of elimination I'm attributing it to impurities in the electrolyte. I've had the problem for a long time even when I was using potassium chloride, and the potassium oxalate came from a chemistry supply. But the calcium oxide was calcium carbonate from a pottery supply, which I converted in the mini kiln to calcium oxide in 2012. (And I've been having self-discharge problems since when? ...Yup.) Of course if the calcium is contaminated, when changing the electrolyte I'm just replacing it generously with more of the same. Yes, it all fits. Any hint of nitrates or nitrites would do it. Chlorides or sulfates might do it. This is where one might want a chemist on staff to figure it out. Or maybe I just did. Perhaps I'll order some pure chem supply calcium oxide.
   28th: Ooh, look! In addition to CaO, Westlab.com also has "nitrate/nitrite test strips 0-500 PPM". And I see there's a test for chlorides with silver nitrate. Maybe I don't need a chemist after all, just the right stuff to test it myself? And they have microscopes... Hmm, do I want to spend 200$ or more on a microscope at this point? I've made out pretty well with just the x4, x10 and x40 microscope lenses I somehow have so far. I passed on the microscope.
   I found I hadn't ordered enough, by 25$, to avoid a 30$ shipping charge. ("Free shipping over 70$") But silver nitrate was over 100$. Unable to think of anything else in particular, I perused their list of chemicals and ordered some nitric acid (the one common acid I don't have) for 35$. Only after I placed the order did I realize I could use it (and a bit of silver) to make a little silver nitrate to test for chlorides, so it was after all an "on target" thing to order. (And if you're not using much, a little silver nitrate is better than a lot of it. As I recall from telescope making in college it can be explosive stuff.) The calcium oxide, the original and main reason for the order, was the cheapest item (13$).
   I'll remark too that it's great - after all these years - to have found a chem/lab supply store that doesn't demand 500$ orders and whose prices don't start at 50$ for every little thing. The other ones must cater to big institutions with lavish budgets. (read: your tax dollars?) But why did it take me so long to find out about it, and then only when someone else gave me their URL? How can it be that it was so hard to find the on-line store I needed in this day and age of the internet?
   (Awg! I already had a nitrate/nitrite test set! It's in my aquarium stuff from having fish! I never made the connection.)

   I looked back to find out when I first started with the calcium hydroxide and came to TE News #48, February 2012. Yup. There was the original idea to use it to get the electrolyte to pH 12-13. (If only I had thought of the oxalate back then, too!) But I've used that same (doubtless impure) calcium all along, and that was when I started having self discharge problems. (I knew I was doing better on self discharge in a few cells in the earlier years!) After about 2013 my battery ideas became very occasional work, since I didn't seem to be making much headway. Working so much in the last three months or so, with a number of different cells and chemistries, has finally helped me zero in on the problem, which probably should have been obvious far sooner. One problem was having only found Westlab for reasonably priced lab chemicals a few months ago, or I might have tried buying a jar and trying it out just on spec long since.

   I played around in my head with some alternate cylindrical cell design ideas, too. If copper makes a good current collector, instead of a carbon rod current collector, use a copper one. Then there's a multitude of possible designs.
   But in order to have plenty of contact area between the manganese and the copper (really the copper oxide surface layer), instead of a copper wire use a copper tube or pipe. It can be filled with something lightweight for strength. That gives the large contact surface area without a really fat, heavy piece of wire in the middle of the cell. In fact, a copper mesh "impregnated" solidly around the inert center might work as well or better. Or maybe a copper mesh filled inside with MnO2 powder as well as the powder surrounding it? That could be even larger diameter, with the mesh even just inside the electrode, near the zinc surrounding. That would have the highest current capacity.
   Or it could even have: an inner hollow plastic pipe covered with copper mesh for a moderately thin (6 mm?) MnO2 powder electrode surrounding it, then the outer zinc sheet electrode. This would give a larger surface interface area with more zinc sheet than a smaller diameter solid center cell with the same amount of MnO2. Of course the air space inside the the center pipe would be wasted. (unless it's actually valuable to help cool the high current cell?)
   OTOH... once all these cylindrical cell options and potential improvements are considered, for unpressurized flooded cells it's surely easier just to make a flat cell with flat electrode plates just as big as the cylindrical cell would have. And then one can start multiplying plates and get higher currents and capacities by just making thicker cases, without multiplying cells for each electrode pair.
   So I ended up deciding cylindrical cells are probably best saved for making "standard" (but rechargeable) cells, "AAA", "AA", "Sub-C" (battery power tool size), "C" and "D". For bigger, [probably flooded] cells with real capacity, the flat plates are better.

   Then I came up with what might be a great idea for multiple plates: fold up a single long piece of zinc, "zigzag", into many plates each going across  to the next one at the tops and bottoms, with sufficient room for the other electrode between each plate. Or "zigzagging" at the sides. Line both faces with separators. This could save a lot of internal interconnections. (Internal joins inside a cell seem like potential trouble.)

   While looking up calcium, TE News #66 also mentioned it, and there I found an idea I'd long forgotten about: To make a perforated metal sheet (zinc, copper, nickel), one might use something like a silkscreen with the desired holes in it, screen it onto the solid metal sheet, and then etch the metal in (eg) ferric chloride until the holes were eaten away as and where desired. That might be a good way to make a one piece metal current collector for a paste electrode, complete with a solid (no holes) terminal/tab. Or with very fine holes, a "pocket electrode" perforated outer case.
   I've often racked my brain trying to think of a good way to perforate a sheet of metal with thousands of fine holes. The next best thing I came up with was the "pin frog" and a hydraulic press. How could I have thought of such a potentially useful and valuable idea and then forgotten all about it? Two (or more) sheets of zinc with holes in the front one(s) and spaced just slightly apart could replace the "plate shaping" idea and would probably be easier - and more compact. In fact, the edges of tiny holes themselves add surface area to the sheet, while reducing its weight.

Haida Gwaii, BC Canada