Turquoise Energy Ltd. News #122
covering July 2018 (Posted August 4th)
Lawnhill BC Canada
by Craig Carmichael

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

: Bandmill Design Breakthrough: Self-Adjusting Band Guides for reliable straight cutting

              (see Month in Brief, Other "Green" Electric Equipment Projects)

Month In Brief (Project Summaries etc.)
 - Scenery! - Batteries - Solar - Reluctance Motor - Development Centers - Bandmill breakthrough

In Passing (Miscellaneous topics, editorial comments & opinionated rants)
  - Election Reform: Corrupting Influence - Keeping/Restoring Your Hair - A Hydrocortisone Cream Healing - Bad Grammar: "Mine" versus "Yours" - Peace Between USA and Russia - Quinoa as an emergency food grain? & Vegetable Gardening - Birth Control: Helping Ourselves by Helping Underdeveloped Countries - South Africa: No genocide after all? - More on Bankers Creating the Money Supply, by a banker - What is News? - Double Agent Strzok

- Project Reports -
Electric Transport - Electric Hubcap Motor Systems
* Revisiting Reluctance Motors and Unipolar Motor Controllers - Considering the 2015 AFSRM Motors & Studies - Note: Current density - First project: Back to the old Motor, with new "Rotor Poles"

Other "Green" Electric Equipment Projects
* Carmichael Mill ("Handheld Bandsaw Alaska Mill")
  - "Front Pivot" Self Correcting Band Guides - blade protector - Conclusions (in short: yes, it works!)

Electricity Generation
* Solar Panels Installation - nasty cable run job - Grid Tie Inverter: NOT as advertised! -  -  - Synchronous Rectification?

Electricity Storage - Turquoise Battery Project (Now Mn-Zn, Ni-Zn or Pb-Zn)
* Mn-Zn rechargeable 50/75 amp-hour cell with ABS sheet plastic case (1.5" t x 3.25" w x 5.75" h)
* Pocket Electrodes: Perforating Sheet Metal with a "Bed of Nails"
* Lead-Zinc Cell From Scratch With Rolled-up Sheet Metal Electrodes (not finished)

July in Brief

   I had friends visit for a week and took a bit of a holiday myself.

Tow Hill on the north coast of Graham Island, from the beach.
It's a "volcanic plug" the center core of lava that came up,
while the softer outer parts were washed away by the ocean.
Similar hexagonal columns can be found at Fingal's Cave in Scotland.
On the land side it can be climbed.

View from the top looking west toward Masset.

   I did some work on the bandmill, which (at that point) didn't seem to cut as I'd expected. I didn't get very much else started until the 10th, when my guests left, the [pure] calcium oxide arrived, and I did some more battery experiments.

The components of a new Mn-Zn Oxalate cell with "pocket" electrodes.
Apparently there weren't enough perforations in the "pockets" to conduct
the electrolyte freely, greatly limiting the current capacity in a cell that
should have had at least a few amps and 25 amp-hours capacity.
(and later I'll add more electrodes to double that.)

   In the middle of the month I put four of my solar panels on the roof of the house (1000 watts). I thought of simply wiring them to a grid-tie inverter. But that won't help in a power failure. I'd rather have a separate system - especially if it can have good, high capacity rechargeable batteries that last forever to store the power. But I had a grid-tie inverter and I tried it out. It proved to put out only 400 watts. It was advertised as, and said on it "1000 Watts". I think I was gypped. So it was effectively using only two panels. I discovered it might take about 8 panels (2 KW, with 2 KW of inverter(s)) to put things in reverse and stop the utility company meter from racking up KWH on my bill, even on sunny days in the summer. But when it got cloudy, the panels only put out 200 or 300 watts anyway and the inverter was big enough.

I climbed the left ladder, which was tied to a hook.
I hauled the panels up the right one with a rope.
(Setting them on the block of wood kept them from snagging under the bottom of the ladder.)

   I'm not happy that I still can't drive the Chevy Sprint car on the highway. I started thinking about reluctance motors again, with their good torque and very high RPM capacity. A reluctance motor could be the best solution to outboard motor conversion as well as for cars. Gas outboards gear down too much at the propeller shaft, so they need a higher RPM electric motor and a reluctance motor should be ideal. At first I decided to more or less copy the layout of the axial flux, two rotor design done in Guelph, Ont. in 2014. In addition to the Sprint under-hood motor and an outboard that could make a small boat plane, if enough power can be packed into a light enough unit, the "Electric Hubcap" wheel motor plan (with a compact planetary or fixed gear reduction) could be "on" again. And what about for a ground effect craft? Surely the ducted fan design needs a higher RPM than a 'regular' airplane propeller?
   Finally what I took away from the papers, which I read more carefully than in 2015, was that (a) the axial flux motor is the right layout and (b) - what I'd missed - that it needs, bug, chunky iron magnetic components for good force, not the fine stuff I made in 2015. (After all, it's much harder to pull a big wrench off a supermagnet than a small screwdriver.)  Ideally for the car, "slice of cake" coil cores might be 3 inches thick, and similar shape "rotor poles" maybe half that thickness, with thick "back iron" behind them all.
   I decided to try first the motor I'd made in 2015 with its 6 donut coil cores, but to match that with a rotor that used the same iron powder cores as "rotor pole" elements. I found I could use the same rotor by cutting some bits out and inserting the cores, another good simplification for a test. And it just might have the power to run the outboard once it was working well.

   On the 19th I took a day off and drove over the rough logging roads to Rennell Sound on the west coast near the south end of Graham island. The scenery was rugged and spectacular. The road rose so steeply from the waterfront into the mountains that I had to drive the Toyota Echo in first gear part of the way to climb out!

Rennel Sound, opening to the west coast of Graham Island (looking SW)

There was a camping area with a boat ramp (looking NW)

and a little farther on, a "log op" - with warnings to wait and follow a truck with a radio to avoid running into logging trucks (you'll come off worst).
This is a small island. You run into people you know...
A friend from Tlell who runs big machines got a job working here later in July.
(looking W)

   On the 22nd I finally decided to send an application to ForesightCAC.com, "cleantech accelerator centre" in Burnaby BC. But did I really want to apply as a cleantech startup business? When after quite some length I had finished filling out the online form which had many involved business questions, I hit "send" and it said "failed". Then it said "Prove you're not a robot", but there was nothing to click on or type for that purpose. It said "failed" again, and I lost all my work, which was considerable.
   So the next day I telephoned. I had to leave a message, and I said I had created new and 'better than lithiums' batteries and a breakthough in bandsaw mill design among other things, and that "Now I want to get them commercialized." That bit of grammar I stumbled on is really exactly what I want. I don't really want to convert myself from a talented inventor and product developer into a mediocre entrepreneur. ("It's too bad we have to buy it this Canadian technology from China. He had some great products but he sure didn't know how to run a business!") Surely they must be in touch with talented and enthusiastic entrepreneurs looking for products who would be delighted to have some great newly developed cutting edge technology(s), theirs to make and sell?
   I managed to put this across quite well to the person who returned my call. I hope they may manage to find a way to help get things going.

   I don't know how in our society that helps inventors to put food on the table when patents seem to be worse than useless to inventors (exactly who they were theoretically created for) and there's nothing else, but division of labor is a fundamental principle of advanced societies. Perhaps there's something to be learned from the music industry where those who write music as well as those who perform it and record it are entitled to royalties whenever that music earns commercial revenue. The composer/songwriter doesn't have to sell the rights to his work for peanuts before someone will perform it and then perhaps it becomes a top hit. Or defend it on his own budget in court against those who've used it without telling him.

   On the 25th an e-mail from United Inventors' Association in the USA said they are trying to have passed a new "Inventor Protection Act". I applaud this effort! If successful inventors can make a living, there will be more of them, and new products and technologies to take us off fossil fuels.
   To conclude this topic, on August 2nd I had an inspiration as to how an inventor, even me, might after all get paid for his discoveries and all the work that usually went into finding them. As it's a brand new thought and as it involves others it may evolve rapidly, so I'll leave off explaining it until next month. I've been groping for something for 10 years, not to mention for another decade before that. Maybe this is it.

   26th: I note that I seem to be spending more and more time writing - seemingly more than I spend on the projects these days. Does writing more make up for not getting projects done? I've known what I want to do on the bandmill for a couple of weeks now, but never seem to get to it. Hmm. Not that nothing is happening!

   Near the end of the month it didn't just get warm. It got hot. There were much warmer days than any last summer, and lots of them. The garden needed watering every day, or things started to wilt. Temperatures approached or exceeded 30 °C in the shade. Smaller creeks stopped flowing. I actually appreciated having a real summer after the cool and wet of last summer. But I know people over most of the northern hemisphere this summer won't raise their eyebrows over our little "heat wave" here. (It ended August 1st.)

   In the last 4 days of the month I got back to the bandmill. The cuts kept veering upward in spite of the band being aimed downward by the self-adjusting configuration. Finally on the 31st with a strong magnifying glass I found a problem if not the problem: in addition to being dull, the 'set' of the lower teeth was zero, no clearance at all. I did some rough sharpening and tooth setting. That evening I finally cut a 6" wide board, 13' long, that wasn't bad. And for once, without making any adjustments either before or during the cutting.

   I finally added a cooling water system on August 2nd to keep the band from heating up and thus expanding and getting slack. With that I cut a couple more boards even with the dull band. A new blade/band will probably help a lot.

Reorganizing SDTC to meet real needs?

   Early in the month I e-mailed SDTC with a link to my uncomplimentary piece about them in the last newsletter. I got the idea of writing my MP to suggest that SDTC needs to be completely reorganized on a new basis since it doesn't do what it was created to do - support sustainable energy inventions and new product development - or else be scrapped altogether as being a poor use of citizens' money. Then I started thinking that perhaps several of us who were doing sustainable energy projects and getting no support from them should get together and write to parliament instead of just me.
   Then I thought it would be even better if we can get a few people together - a social sustainability design team - and work out how we would set up the organization to meet the real needs for fostering development of sustainable technologies. A plan sketched out in some detail for recommendation to parliament would probably be better than just criticism or even vague suggestions. My now rather remote location works against organizing and meeting with people, unless it is done by e-mail and perhaps conference calls.

   Any takers on this? I welcome ideas and suggestions along these lines, especially from fellow inventors and product developers in Canada (but also anywhere), and especially from those with any sort of experience with organizational structures, who think they might have good ideas for how such an organization might ideally be put together.
   And what does anyone think of the idea of having them run a "Sustainable Technology Product Development Park" or even 2 or 3 or more across the country? Look how we got the graphical user interface and the whole workings of the internet from Xerox's "Park Pacific" technology park. (Xerox didn't even understand what they had achieved or know what to do with the results.) One technology park for a few short years. What fabulous advances could we get from an ongoing one or more than one?
   This/these park(s) would hire creative people and guide them as to what was to be created, instead of just waiting for whatever off-target, high budget proposals drift their way from well established corporations who meet some Procrustean organizational criteria. They could hold patents on behalf of the inventors, and being an organization of government instead of one pretty defenseless inventor, should be able to successfully extract royalties from all manufacturers both for the inventors and to become a self-sustaining organization itself.

   If a few people are interested it's not impossible we might get something useful done.


"Don't worry about people stealing your ideas. If your ideas are any good you'll have to ram them down peoples' throats."

- Howard Aiken [Mathematician, physicist & computer pioneer]

A nest of swallows messing up my front door light.
Just about ready to 'fly the coop' I should think (Aug 2nd).

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

Election Reform: Corrupting Influence

   In Canada we were promised a new electoral system by the present government if we elected them. Then they held meetings across the country to help decide how best to set it up. I went to the one in Victoria, and the one thing that was abundantly clear was that few people who attended liked our current "illiterate's 'X'" single mark, single ballot system. When those meetings were done, they dropped it and will make no changes. And apparently this has happened at least a couple of times in the past in Canadian history.
   In BC we are now to vote on a new two-question "proportional representation" system referendum in the next election. While I'm not entirely happy with the choices being made available, almost anything would be better than what we have. As soon as the BC government announced it, a lawsuit was taken out to try to have the referendum blocked, complaining about almost every detail. And Google ads started appearing telling us that our present system is great and "proven*", the new one (regardless of what is chosen?) would be a disaster, that once we changed it we would be lost forever, and so on.
   The .01% of corrupt, selfish, filthy rich and anti-democratic people don't want a system they can manipulate so easily and buy politicians to help them get their way to be replaced with one that would make it hard for them, and will spare no effort or expense trying to nip it in the bud. I suspect they will succeed again, as they have for over a century and most recently federally. There are just so many apathetic people who won't take the time to see and to think and perceive and make up their minds for themselves, who will let their minds be made up for them by those who squawk loudest - those with the money to hire PR firms and run plenty of ads. (One of their complaints in the lawsuit was about the campaign financing rules - because the government supplied a little money to counter their smear ad campaign!) Having given it no prior thought, all many of these people may remember at the voting booth when they are surprised to see these questions actually there to vote on is all the negative ads warning of doom and gloom should they vote for change, and reminding them of how rosy the present system is - the one that keeps people rising up in ineffective protest against its repeated unilateral acts, its giveaways of public assets and its free-spending management of the public's money.
   My own views on what sort of systems of governance might be best are laid out in detail at HandsOnDemocracy.org

* "Proven" only to be unfair to all and socio-politically polarizing.

Keeping/Restoring Your Hair

   A couple of years ago there was a piece of "click bait" going around that showed a picture of a man brushing his partly bald head with a brush, with the words something like "To keep or restore your hair, do this daily." In fact this was complete within itself, because brushing with a brush to stimulate the hair follicles is exactly the right thing to do in (I believe the majority of) cases of thinning hair or baldness as we age. Doubtless the makers were sure no one would believe it was so simple and would follow the link. (Or they may have given some finer details in the links.)

   Now I've found a brush in a dollar store (4$ ... yay, we now have a dollar store!) that's much better than any other I've tried. I could feel the difference the first time I used it and since. And after 3 or 4 months, I can see (in two mirrors) that the thinning hair at the top-back of my scalp has gotten thicker and if not "perfect" (yet?) no longer looks half way to bald. And that's where I most felt that it had a very good stimulating effect where other brushes I've tried (which I admit isn't very many) had seemed pretty listless by comparison in that area.
   I knew there was a difference between combs and brushes, but I didn't realize brushes could be so markedly different. I don't know what it is about this particular brush that makes it seem so good, but here's a pictures of it. (I bought an extra to send to someone.)

A Hydrocortisone Cream Healing

   Something I missed last month writing about applying hydrocortisone cream... When I was maybe 12 I was trying to hacksaw something. The piece wasn't held well and in the vibration the saw suddenly jumped up and over as I cut, and cut into the back of my thumb. Trying again brought a second slice. I didn't think much about these at the time. Weren't they just cuts that would heal? But they never did heal properly. The hacksaw teeth had ripped up the skin pretty well and it hadn't gone back into place. They were always two rough, scaly lines across the top of my thumb.
   In rubbing hydrocortisone cream onto all the moles and a couple of cysts, I started applying it to those rough lines too. They seemed to get worse at first and irritated, then - after 50 years - they healed over. They are smooth and just visible as slightly raised lines. Wow!
   (I never thought to take any "before" pictures, and "after" there's nothing anyone would notice.)

   The labels on hydrocortisone cream tubes all seem to say to 'discontinue after 7 days'. But these sorts of skin healings - the cuts and especially the cyst (never mind keeping the moles light) required continuing application over much longer periods of time.

Bad Grammar: "Mine" versus "Yours".

   Somebody's "cutesy" but crazy idea from the 1980s(?) has got parroted by all sorts of organizations. Even the government (CRA) speaks of "your my account", a complete oxymoron in three words. It bothers me that any organization should think it somehow enhances customer or public perception of them or makes them seem friendlier or more sophisticated to perpetrate such confusing misuse of language. Since I wrote the following almost 20 years ago things have only got worse, and now I've had the urge to dig it out of my e-mail records and reprint it.


Dear xx Bank VISA:

Our use of the words 'I' and 'my', and 'you' and 'your' in the new cardholder's agreement which we sent you do not make sense to you!

It is common ground, implicit in grammar in every language that the word 'you' refers to the person intended to read a document, and 'me' or 'us' refers to the party which wrote it.

But we have elected to reverse the meaning of these common words, which, of course, makes our cardholder's agreement difficult for you to read. If you had written the agreement, there would be no need for us to send it to you to read, would there? Instead, we there at VISA would be the ones to have to receive it, check it over and see if your terms were agreeable.

And we are not even consistent in our approach, since all the correspondence we send to you, including the cover letter "Special Notice To Cardholders" on the very same piece of paper as the new agreement, says 'WE are writing to tell YOU', instead of 'YOU are writing to tell ME'! Because if we did that, our letter to you would be even more confusing than your cardholder agreement, would it not?

Who on earth writes our stuff, anyway? I should be shot!
You would be greatly relieved if we chose to restore these words to their proper meanings in the future. Thank us very much.

Mine truly,
VISA cardholder

Peace Between USA and Russia

   In mid July, US president Donald Trump met with Russian president Vladimir Putin in Helsinki, Finland. Trump said relations between the two countries 'have never been worse', and that that was thanks to stupidity in Washington. He wanted to improve them and to have peace between the two democratic republics.
   Trump was attacked for this, not by Russia but by his own country. He was accused of "being in Putin's pocket" and "what did Putin have on Trump that he could be blackmailed in such a way?" One malcontent said Trump's promotion of peace between them was "the highest treason". The funny thing is, Trump said before the meeting that regardless of the outcome that is what would be said of it. And that no possible outcome would be good enough to satisfy these critics. But one notable person said Washington had "criminalized diplomacy".

   The "deep state" - largely the civil service bureaucracy including the military and the many security agencies, as well "cross-connects" from the largest corporations - that holds real power hidden away in private from public scrutiny isn't used to having a president who, whatever his faults may be, is trying to make state policies in public and apart from their lies and conniving back room deals to keep the military-industrial-banking-oil-monsanto-pharmaceuticals complex in business and receiving way over 50% of the US national budget, regardless of the needs of the people and the potential for peace in the world. They will discredit him with any lies and unfounded insinuations they can think of, and they dearly wish someone would shoot him and get him out of their way. In fact some of them are actually advocating it - real treason not only against the peoples' chosen leader but against democracy itself. And they will feel the same way about any president who follows him who has the temerity to try to lead the nation instead of to follow their dictates.
   Dwight D. Eisenhower warned that the need to maintain a large "military industrial complex" after world war two was itself a grave threat to American liberty. Liberty and equality minded John F. Kennedy was shot to get rid of him, apparently by the mafia but with at least complicity by federal agencies. (See "Oswald's last phone call" on youtube.) Robert Kennedy was shot beforehand to prevent him from ever taking up the reins. Then Richard Nixon was thrown out. Why? After being a perfect "deep state" stoogie for several years, Nixon got fed up. He ended the Vietnam war, Accorded China long overdue diplomatic recognition, and started nuclear arms limitation talks with the Soviet Union. That was why the Watergate manipulations, in which he had of course played a part, suddenly came to light: deliberately to arouse public indignation to get rid of him. Dr. Paul Craig Roberts says they repeatedly tried to block Ronald Reagan in most everything he tried to do in the 1980s, including rejecting or stalling on approval of his chosen appointees (as with Trump), "Reaganomics" (of which Roberts as assistant treasury secretary was the co-author), and most especially of his trying to end the cold war. After his second election Reagan became senile, and instead of promoting the vise-president it was covered up for the rest of his term. What a perfect president to let them have their way with no argument! (The senility was just fortuitous... right?)

   I know Nevil Chamberlain once said this and was quickly proven wrong, but personally I don't believe any important nation in the world today wants war... apart from the US 'deep state' actors. If it was up to them, the world could well be a desolate radioactive wasteland by now. Unfortunately these types of people too often get their way. They won't take "NO!" for an answer and will keep trying different tactics until they do. But here, finally there is just too much at stake to let them and they haven't.

   Which brings up another topic... One of the worst problems (besides overpopulation) is that our justice systems are so gutless (if not corrupt). Some of the worst of the selfish people in government and in corporations who commit egregious crimes, predating on and working against their own people year after year and decade after decade should be given an overdose of morphine and be peacefully removed from the planet to get them out of high places and to demonstrate that society will not give in to crimes of betrayal against itself that prejudice the whole future of civilization and freedom. It's not that "jail is too good for them": it's that they will simply continue to predate whenever they are free to do so and that many of them can call on their friends or henchmen to bribe, threaten or do away with those who are causing them trouble even while they are in jail. Someone is going to die: should it be the troublemaker, or whoever opposes the troublemaker? Gutlessly allowing the latter course to occur leads to problem after problem and often victim after victim since the source of the trouble is still there.
   Just for one familiar example, how different might history have been if Adolf Hitler had been executed for attempting to violently overthrow democratic government in the "beer hall putsch", instead of simply being jailed for a time? There are many and various other examples of ways in which the world has been markedly damaged by single individuals who should have been permanently stopped, but the consequences of that one particular failure to act resolutely at the critical moment are the most stark. Civilization was shaken to its core and tens of millions died.

   The only two ways people are safe are (a) knuckle under to unreasonable demands and conditions - and ultimately live in a totalitarian dictatorship or (b) eradicate the predators. The only way the network of associates and henchmen won't bribe or threaten or murder is if their boss is no longer there to give the orders and to pay them and protect them from the law themselves. Not there to ensure there are no investigations into briberies or threats, or that the suspicious death of an opponent will be deemed a "heart attack", "suicide" or "accident" likewise without any investigation. How many judges, prosecuting attorneys, witnesses and others have been murdered by those they sent away for more limited punishment? It makes obvious sense: only one person, the predator, is killed instead of potentially numerous innocent and usually upright victims who try to stand up to him, know too much or just get in the way. We will not have social stability, much less sustainability, while people are, with good reason, afraid of violence.

   The media is now owned and paid for by the same "deep state" that holds the real power hidden away in the back rooms and they trumpeted this "Trump's treason" nonsense to the American public. I'm not sure how many are still buying it. But no major station or network (all owned by the same 5 or 6 people) will depart from those narratives. Interviewees (and even news announcers) saying the wrong thing have been known to be cut off in mid sentence as the network "has technical difficulties". The CIA has an agent now in every major newsroom to vet what stories must, may or can't be aired.
   But they haven't been able to stop the internet. Trust in the western mainstream media is more and more being eroded as more and more people start to realize they're being fed crap that doesn't match facts and leaves major issues and events uncovered. And as opposing information, opinions and viewpoints from many lands present themselves on the internet. In spite of accidental or deliberate misinformation here and there, a different and more coherent picture of events is starting to be discerned by increasing numbers of people. The 'establishment' wages real, serious war on real journalism as exemplified by The car booby-trap murder of Michael Hastings and Hillary "Can't we just drone strike him?" Clinton regarding Julian Assange as well as the whole Wikileaks organization. They seem to have cowed Equador into agreeing to turn Assange over to their tender mercies.

   Russia was once the home base of communist zealots who thought the world would be a better place if they ruled it in accordance with their ideology and who thought any and all means to achieve that end including violence and war was justified. They ended up dictating to the whole of eastern Europe for decades. (And if Victor Suvarov is right, the rest of Europe and perhaps the world got off lucky.) Today Russia is a democratic republic where there is probably more freedom than in North America, which since 2001 is rapidly becoming increasingly unlivable by government decree. (Toronto psychology professor Jordan Peterson [see on youtube] has attacked one Canadian 2016 bill in particular as giving the government sweeping power pretty much to prosecute - persecute - anybody they decide isn't being "politically correct" or is saying things someone disagrees with even if it's an honestly held opinion. People have become afraid to voice disagreement about anything, however disagreeable, for fear of official retaliation without recourse: fines, loss of job, loss of rights, eg, to fly or leave the country, even jail. Freedom of speech and freedom in general is under increasing assault.) Even Iran, another special target of American venom, has an elected president, who was peacefully changed without so much as honorable mention or a hiccup in the "Iran needs regime change" rhetoric.

   Get over it!

   Look to home for where "regime change" is really needed. We have democracies. We need better communication channels with our elected representatives in order that they do our collective will instead of that of corrupt special interests or vocal minorities. And we need to better know who we are really electing. What have been their past actions and activities? Are they capable, experienced, honest, sincere? Or are they a social predator in disguise? Knowing like Hitler how to present themselves in public as being good and reasonable, knowing how to tailor their "beliefs" and "principles" to the audience, makes them more dangerous unless they are exposed, which to date rarely happens.
   And whatever most people who aren't themselves a special target of some social predator may think today, we desperately need to have a death penalty and we need to use it on those who prey on us all and it has become a long continued, habitual pattern for them or a present menace to the future of society. Permitting social predation sets the stage to wreck the future. The tentacles start spreading pretty soon after the unpunished, uncorrected crime, and encourage more and more and larger and larger such crimes. (Anybody remember what happened to Hitler after he got out of jail? Did it reform him and he went about doing good? He just changed his tactics that hadn't worked to ones that did.)
   We are all tacitly or indirectly responsible for decisions by leaders. Employees and shareholders of corporations are at least tacitly complicit in immoral decisions made by management. When the great majority plays their part, each person helping to prevent wrong things including the wrongs behind the other wrongs, or to "regime change" what has already gone wrong, we will we have peace and stability - and perhaps even continuity and social sustainability.

Quinoa as an emergency food grain? - Salal - Vegetable Gardening

Quinoa, mid July. By the end of the
month it was taller and bushier, and buzzing with pollinating
insects, flowers with seeds forming and growing everywhere.
   Potatoes is still the easiest way to get calories from a garden. If I plant wheat here in a large enough patch to be useful, that would be too large for a tall fence and the deer would probably eat it all. But here's something else where wheat (etc) may be impractical for a home gardener. A couple of years ago I grew just 5 plants of quinoa in my garden in Victoria BC. They grew tall and bushy and produced a whole jar of seeds. This year I'm trying a larger patch.
   Only one of the 6 seedlings planted early indoors survived outdoors. I planted a bunch of seeds in the garden - twice - and for a while things looked rather sparse and unpromising. I even planted beans and things between the rows in case the quinoa didn't do anything. The one from indoors is still the largest, but others finally started catching up, and in spite of some being eaten by slugs when they were small, it now appears they are pretty crowded (just one notable bare spot could have used another plant) and I should have followed the spacing guidelines or thinned them. But it looks like there'll be a good crop.
   Since quinoa is rather grainy unless soaked, on the 24th I tried grinding up some into a flour in the coffee grinder. It's probably more like rice flour than wheat flour. It added some grittiness to a cake and cookies, but not so crunchy as the whole seeds. A real flour mill would doubtless make it quite smooth.

Salal, Aug 1st.  Hmm... Picking season is hardly starting...
but these ones growing near a creek have some notably bigger berries, already ripe.
Maybe I should water mine?
   I also (finally) read up on salal, a berry native to the west coast of North America from about Washington state north to the southern islands of Alaska. It didn't seem to be related to anything, but I discovered it's a member of the heather family and has been imported to the UK and Europe. It's very plentiful in my neighborhood, especially in an area of my field and by the highway. Stems with leaves are very popular in floral arrangements. It has a small blue-black berry that makes good pies and jam. Being so small, they're as tedious to pick as wild blueberries, huckleberries or saskatoons; maybe more so. Then usually you have to pick out bits of stems and petals before using them. So far it hasn't been available commercially, but after picking saskatoons wild in Edmonton over half a century ago and hardly seeing them since, I never expected to see them available commercially either. (The wild ones are bursting with flavor. I've been unimpressed by the commercial offerings except the last jar of commercial saskatoon jam I had was very good.)

   If you suspect there'll ever be food supply problems, the time to learn how to garden vegetables is before it happens. A sort of insurance, but it does take a certain amount of know-how. It's easier to expand if necessary by already knowing what's needed than to start learning to recognize a weed from a cabbage seedling when hunger is around. And you'll need some basic seeds and supplies in advance too. And of course you always get the best, freshest and most nutritious vegetables when you grow them. Commercial frozen peas don't have the flavor of 'real ones' fresh from the garden.
   And various things to learn are different in different areas. Despite still being on the coast I note that the bugs buzzing around the flowers here are much more like the ones in Edmonton AB than those of Victoria BC. (must be a latitude thing?) And slugs were never a real pest in town in Victoria. I never bought or needed "slug bait" there, but here they can eat whole rows of seedlings and I lost lots of crop "in the bud" last year before I caught on, and still some this year. (Must stock up on slug bait in advance of any supply disruptions!)

Helping Ourselves by Helping Underdeveloped Countries

   I've always thought it was irresponsible of all those undeveloped countries to keep having so many children when they can't feed them all. In the "developed" nations we have been having children under the rate needed to maintain the population for decades now. This problem continually comes to light as people from more crowded places with few opportunities keep flooding into the lands that should be enjoying a great standard of living for everyone, keeping them pressed to support growing populations. This influx of mostly unwanted immigrants is reaching crisis proportions. We are everywhere hitting the limits. We can't maintain our quality of life as long as the population keeps growing, especially when the migrants don't have our culture and values or even language, and so are a drain rather than a benefit for quite some years after arrival. Opportunities even in the developed nations are shrinking to the point of vanishing especially for the young, and inequality has become egregious.

   Now it has come to my attention that it's not that the undeveloped regions want all those children. But people everywhere do want sex. The problem boils down mainly to just one thing: in the more developed nations we have ready access to birth control products. In the less developed places they don't. So we don't have babies unless we want them, whereas in the more primitive places the babies keep coming, wanted or not.

   Instead of sending food or other immediate material aid, we would surely do ourselves as much as them a much greater service by offering to send free birth control products so the people there can choose not to have more children than their land can support. Solving the refugee and migrant problem is worth the price - especially when the aid won't simply be fostering a larger and larger population who then need more and more aid, and people who are compelled to migrate to seek meaningful lives.

South Africa: No genocide after all?

   In a later news article, the president of South Africa seems to have backed off on his earlier "get rid of the whites" stance, saying that they weren't out to kill off the whites and other minorities after all. Hopefully and likely some voices of reason even from within the black community have induced or forced him to change the attitude of a few months back, and hopefully this improved attitude will continue. Perhaps they look next door to Zimbabwe where the whites were killed or driven out and the country has become primitive and barbaric. "Is that really what we want?" Still from what I've heard South Africa is a relatively violent place and I'm glad I don't live there. Doubtless so are the 15% or more of the entire white population which has already departed.

More on Bankers Creating the Money Supply

"Banking was conceived in iniquity and born in sin. ... Bankers own the Earth. Take it away from them but leave them the power to create money, and, with the flick of a pen, they will create enough money to buy it back again.

"Take this great power away from them and all the great fortunes like mine will disappear and they ought to disappear, for then this would be a better and happier world to live in.

"But, if you want to continue to be a slave of the bankers and pay the cost of your own slavery, then let the bankers continue to create money and control credit."

 - Josiah Stamp, director of the Bank of England, 1928

   Just how much longer are we going to let this admitted pyramid scheme continue? Why don't nations simply say "Our treasury department will print whatever new money is needed, and the public then won't owe it - with interest - to anyone." Admittedly the bankers kill or try to kill any national leader who tries this approach and perhaps it's small wonder our politicians seem to have no courage. As I understand it Argentina was letting its provinces print their own money some years back and having an economic boom, then the IMF stepped in and told them the nation had to trade in all these local currencies for national currency (wait for it...), borrowed from the IMF. I don't know what coercion was used but it probably involved some pretty serious threats. No such prosperity making alternative was to be permitted. Today Argentina's currency is inflating away and the nation is in a depression again.

   We need people everywhere in lesser places too to say "Whatever happens, even to our leader, the new money policy will still be law and the nation will never again borrow money created out of thin air by a bank instead of creating it itself."

What is News?

   On TheWeatherNetwork.com was a headline about a teen infested with hookworm parasite while playing on a beach. Such a thing catches our eye. But why? And why include such a story in national news?
   Where was it? Most likely far, far away from you or I. Unfortunate though it was, just for example, millions die annually of cancer, and in our own neighborhoods. Wouldn't it be more useful to talk about how you can cut your chances of getting cancer at least in half by getting either vitamin D or some sunshine daily or at least very frequently? Cite some study or a bunch of them - they all say it. That gives people good news and something positive they can do, instead of just making them needlessly anxious over something that virtually never happens to anyone. (Okay... just had to look... it was Memphis TN, and he didn't just walk across something bad in bare feet, he had been buried by friends in the sand, where the hookworms apparently lived. (I bet it looked dirty, too.) And TN is inland, so it was a freshwater beach, not ocean. I.E: this will never, ever happen to me or you.)
   And if one wanted to focus on bad news, there was plenty of more newsworthy bad news without dragging in such drivel.

Double Agent Strzok

   Hah! He was supposedly an FBI agent, now it seems Peter Strzok was actually a CIA agent planted in the FBI! Impersonating an FBI agent is a federal offense. So is espionage against the government, and so is CIA activity within the USA. So is he classed as a foreign spy or a domestic terrorist? How many more are infiltrating how many departments? Does the FBI infiltrate the CIA? Does the IRS know Strzok is collecting two paychecks? Who should investigate this clearly illegal activity? Does the IRS infiltrate the CIA? Does the CIA infiltrate the DHS? Is the snake eating its own tail?

   "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.

Electric Transport Projects

ARM Reluctance Motor & Unipolar Motor Controller

   I was working on my "ARM" (Axial flux Reluctance Motor) and "unipolar motor controller" in 2014 and 2015, then other things caught up with me and I stopped. Now having spent some months to get new chemistry batteries more or less created and 'tamed' so to speak, I thought that getting a good reluctance motor and controller working would give the best of everything:

* It would in and of itself be a superior motor, and it would (presumably) provide a motor controller for various reluctance motors.
* It would provide a platform for trying out the "permanent magnet assisted" motors idea, since I already had some magnets the right size.
* It would get the electric outboard going without trying to solve the cooling problem of the present Electric Caik BLDC motor. And it would be better because it could exceed the RPM of the original gasoline engine removed from any outboard where most electric motor RPM.s limit boat speed.
* It would provide a motor to get the Chevy Sprint - and by extension most any light car - on the road and on the highway, efficiently with simple fixed ratio gears.

   Note: I have recently found there are two entirely different types of "PM assisted" reluctance motors. In one the permanent magnets are in the coils as previously described. In the other they are in the rotor. I haven't come to grips with how this second type works yet myself.

   So I decided to locate the components of the project and pretty much start again, but with certain things already made. I would try out the same unipolar motor controller and the same motor stator, but with a new "more conventional" reluctance rotor on the motor. My attempted "flower" pattern rotor matching the donut coils ran, but it was nothing like it was intended to be, perhaps at least partly because when made, the "rings" didn't seem to line up quite the same as when I had cut them out of cardboard and simply set them in place for a rough estimate. As it was, there was some braking force as well as the forward force at certain points of rotation.
   In some reading I found out also that the rotor poles each need way more iron for the stator to attract. That's probably the biggest key to getting a motor with good torque. Changing the toroidal coils to rectangular or pie shaped ones to maximize the magnetic attraction along a rotary line of force would also be very useful.

   But the first question was, where? My electronics lab here is tiny. The big workshop seems not very suitable. But it's summer and the workshop is probably the place until it gets too cool to work out there - if I can figure out how to set up a good workstation somewhere - and get to the project. I started by freeing up some space by the only wall without shelves built in in front of it - by installing the solar panels leaning against it.

   Then other things caught up again. By the 16th I thought I had decided to start afresh on the motor and just use the controller. I would look up a few more reluctance motor projects by others on line, and perhaps find a pattern of coils and rotor design that "meshed" really well together. Donut coils were great for BLDC motors, but without supermagnets on the rotor, merely soft magnetic material, reluctance motors require far closer and more exact magnetic coupling to get high torque - or even "good" torque. Practically everything that could contribute to that coupling seemed worthwhile incorporating into the design.
   In the evening I tried to find more info on line but I wasn't finding much besides things I already knew. Finally I thought of the two papers on axial flux reluctance motors I'd downloaded in 2015, and finally I found the folder I'd put them in.


The first paper was (Sorry, I don't have the URL) [Hmm, extra words appeared when I copied and pasted the text of the title and authors!]:

Preparation of a Formatted Technical Work for the ICEM [the above words are not visible on the page] A Design of Axial-gap Switched Reluctance Motor for In-Wheel Direct-Drive EV
by J. W. Haggle, L. L. Grigsby [the above names aren't visible on the page] Tohru Shibamoto, Kenji Nakamura, Hiroki Goto and Osamu Ichinokura Elec. and Comm. Eng. Dept., Tohoku University

   This motor was very high torque, but it had a problem that the torque dropped off much too quickly with RPM above 200. The graph stopped at just 800 RPM. I thought reluctance motors were supposed to be able to easily spin really really fast? What about 8000 RPM? Was this steep torque drop-off inherent, or was it programmed in as part of the operating characteristics they desired for their specific "in wheel" motor in their specific test car? (I confess I don't think I'd want to be doing highway speeds in it. Aviator goggles required!)
   OTOH, they actually built and tested this motor. It turned out the other paper from U of Guelph was just a theoretical study.

   In what they actually built, their "stator support link" looked in the diagram like a solid steel plate with a coil above and a coil below. If that was so, wasn't it just two mirror image motors with a common shaft? Then wouldn't a single motor with one rotor perform just as well?


The other paper was:

Machines 2015, 3, 27-54; doi:10.3390/machines3010027
machines ISSN 2075-1702 www.mdpi.com/journal/machines/


A Novel Approach to the Design of Axial-Flux Switched-Reluctance Motors
Tim Lambert *, Mohammad Biglarbegian and Shohel Mahmud
School of Engineering, University of Guelph, Guelph, ON N1G 2T6, Canada


   Both papers said that with reluctance motors, axial flux gave best performance. Their motor was pretty similar, but with fewer coils and apparently a different scheme for attaching the stator. (Some things weren't clear and I started thinking of the "Stator/Case Key Insets" as perhaps being slippery plastic bits that the rotors would hit during vibration and jarring suspension changes. But that's probably wrong. I don't have a clear picture in my head of how they intended to attach the stator. A problem with a theoretical study where nothing was actually built is that they may not have thought it through completely themselves.)


   Both motors had two rotors, one on each side of the stator. There, apparently, was a chief key to maximizing torque, more or less doubling it over one rotor, as well as supposedly improving energy conversion efficiency. Each stator coil core was a pie shape with the edges radial to the axle. The rotor poles of this motor appeared to be similar, but of course there were fewer of them. All very well, but... With a rotor on each end, it looked to me like it would be difficult to mount the stator coils. (Gosh, just like the axial flux generators with two rotors!) The Tohoku motor showed a single stator in the initial diagrams, but apparently when built was made pretty much as two stators with a plate ("link") between them, which was required for sufficient stiffness to handle the calculated magnetic forces. Would it not then perform just like two single rotor motors? That would certainly be easier to make and to mount, even if it had to be a little bigger and or be geared down more.

   I thought of an alternative: putting "horseshoe" electromagnets around the outside, whose arms would equally attract the top and bottom the poles of a single central rotor. But it would need very fat "arms" going to the stator to carry the flux. These authors seem to refer to that as "the double stator" layout. Their main complaint was it took extra volume.

   At first, knowing so much (ie so little) about all this, I decided that the best thing to get the highest torque was to more or less copy designs that worked. The axial flux seemed to be superior, but the two rotor design seemed superfluous or difficult. I e-mailed the authors of the paper in case any of them might still be reached at U of Guelph. But somehow I hadn't realized at that point that theirs was merely a theoretical paper and nothing had been built. There was no reply. And only the Tohoku one had actually been made and tested.

   What about the number and size of the active components? I figured that the fewer poles there were, the higher the RPM could likely be, so 6 or 8 coils should be better than the large number in the Japanese design. If I made it 8 like the Ontario design, it would need a 4-phase motor controller instead of 3-phase and I'd immediately have to do a new motor controller as well as the motor. I expect a microcontroller based motor controller is the way to go, and I may end up doing 4 phases if I don't like the behavior of the 3-phase motor, but for now the working controller, which I also wanted to test further before starting a new one, made 6 coils the simple choice.
   Also, the fewer poles, the fewer the magnetic reversals for a given RPM, and so the lower the hysteresis losses in the rotor. (Remembering that I already tried like poles and to avoid magnetic reversals in 2015, and the motor had even less torque. That could change with two rotors, with the flux going through the axle, but for now I'll assume it won't.) Note that only the rotor poles reverse polarity, not the stator coils, which already avoids the bulk of the hysteresis losses. This is another advantage of the simple 3 phase design: the coils have no reason not to actuate in pairs, the same pairs each time, and so polarity doesn't matter.

I started thinking about actual dimensions and placements:

Phases: 3 (6 coils, 4 rotor poles)
Stator Thickness: 3"             --- 1" made enough flux depth for coils for supermagnet rotors, but we need much more flux with just a steel rotor.
                                               Without going completely nuts, why not greatly increase the depth of the field?
Rotor Salient Pole Thickness: 1/2" plus rotor disk --- (Or should it be thicker, in keeping with increased depth of field?)
Rotor/Magnetic Diameter: 7" --- We get to go very high RPM, so why use a very large diameter? Keep it compact.
                                               7" should do it? (Or maybe 7.8"/200 mm, and then I can use Princess Auto "brake rotor disks"?)

   I found that the U of Guelph motor said 100 mm stator thickness (4") and 50 mm rotor thickness (2"). The diameter is a very large 400 mm (16"). In spite of intending a "wheel motor", it seems a bit of a monster at 40 Kg. I wouldn't want that as unsprung weight on my car wheel!
   Since my plan is just a motor to mount anywhere, not to go inside a wheel, it can spin higher revs with less torque and doesn't need to be as big. If I maintain the proportions with 3" (75 mm) thick, the rotors poles should be 1.5" (37 mm) thick. Thus the motor is already 6" thick (or 4.5" with one rotor. But if I use an effective diameter of 200 mm instead of 400, it has just 1/4 of the interface area. That shouldn't going too small if it's geared down enough - eg, 4 times more.

BTW, the idea of allowing the rotor to move off center compared to the stator so it can handle wheel suspension movements didn't work for me in 2008. Instead of turning the rotor, the stator jumped around wildly in front of the rotor (car wheel) with the magnetic forces. I had to put in a "lazy susan" bearing as a thrust bearing to keep them on a common center. (how little I knew about mechanical components and supply sources back then!) I didn't see that this issue was dealt with anywhere in this theoretical paper. How well did they really know what they were doing as far as making a real, practical motor? All the forces in this motor are attractive, none repulsive, and the stator is much heavier. Those things might make a difference, but I have my doubts.

   The control scheme, position sensing and flux paths are markedly simplified in the 6 coil, 4 rotor pole design compared to 8 separate coils individually driving 3 rotor poles. For one thing the coils simply come on in pairs making for just three phases instead of eight separate activations. When the coils come on in pairs with one magnetized 'north' and the other 'south', there is an even strength flux path across the rotors between the two driving coils and the four driven poles - two on each rotor at opposite ends.

   I'd rather have a stationary case than a spinning one, with only the shaft spinning externally. It seems to me a 3" thick stator whose outside is the outside of the motor will also provide more outside surface for cooling. Maybe some heatsink fins around the outside? OTOH, if the outside of a spinning case was centrifugal fan blades with the air drawn in through the ends by holes in the rotors, that might do a good job of keeping the motor cool too. Then the stator coils, and the motor itself, would have to attach to the stationary center shaft.

   There was no reply from any of the U of Guelph team. The lead author, Tim Lambert, seemed to have moved to a company selling BLDC motors and equipment for scooters, etc. So it didn't sound like they had made a real success of the axial reluctance motor themselves, or they'd be for sale. (Well, if they never even built one!...) Here's a last table from their paper:

The motor referenced as [13] is the one from the Tohoku University paper (First paper above).
It appears only the Japanese weren't using an insanely dangerous voltage.

Note: Current density

   How is it I've been making motors for so long and yet paid so little attention to this parameter? How many amps does one want to ram through what size of wire, as in the table above? Perhaps it's just that I'd never taken note of what others were spec'ing, as in the table above.
   Of course in the absence of room temperature and above superconductors, the more the amps, the hotter the wire will get. [what's happened to those, anyway? somebody's research was getting results a few years ago. For the first time ever, they had to raise the temperature to find the superconducting point.]
   My Electric Hubcap on the Sprint car, with the Kelly controller, occasionally used up to 150 amps. It got hot fast. The cross section of the #11 AWG wire I used is 4.17 sq.mm. If I restricted current to the figure spec'ed for the Japanese motor, 22.6 amps/mm^2, I should have limited the current to 94 amps. 94 A * 36 V = 3400 watts or 4.5 HP. (It still would have got hot pretty fast, I think.) I think the U of Guelph team were unrealistic spec'ing 30 amps/sq.mm - that's about like my 150 amps that got so hot so fast.

   If I want the reluctance motor to have more horsepower at the same 36 volts, it would appear I should wind two strands of #11, or equivalent to keep it from getting too hot. As it is, the one I have now with a single strand at 24 volts should be limited (again using the 22.6 amp figure) to (22.6 A/sq.mm * 4.17 sq.mm) * 24 V = 2260 W or 3 HP. And that's only if it has quite good cooling.

   One can balance by putting coils in series or parallel. If instead of having two coils of 21 turns (2 layers or wire) in series, it had two of 42 turns (4 layers) in parallel, each coil would handle 94 amps current, total 188 amps. That's theoretically double the available current and horsepower. (And it looks like there would actually be enough room for the extra wire.) Of course, with four layers of copper wire on each coil instead of two, each coil will be harder to cool, so one may have to limit the current to a lower value and the overall improvement, except perhaps for short burst capacity, may be more limited than expected.
   Or I could use two coils in series with ~32 winds (3 layers) instead of 21 and change it to 36 volts. There again is the theoretical 4.5 HP but probably somewhat less than that in practice.

Back to the old Motor, with a new Rotor

   Well, double RPM makes up for half the torque via a different gear ratio, so if there was actually little or no real advantage to having double rotors in practice, the stator plate I had made in 2015 should be as good as any other. So before I got too carried away, I decided to try the previously unused four pole rotor I cut in 2015 with the original stator, just to see what I might learn. On the 25th I found some various washers and spacers that let me put that rotor in. That just left mounting the optical interrupters. The spacings for those would be completely different.
   Then I thought about how fat the rotor poles were in the other motors, and how at least in the Guelph design, the poles were the same size and shape as the stator coils. A big chunk of steel is harder to pull from a magnet than a small one. That made the thin 3/16" steel of my rotor seem rather pathetic. In fact, whatever the other failings of the 1/4" thick "flower petals" rotor, too little iron in the poles was doubtless the biggest one. If I was to use this motor with its 2" O.D. round coil cores, probably what it should have for rotor poles was something like 2" steel hockey pucks, at least 1/2" thick. That might actually make a motor that would run the outboard. ...What about using four of the actual iron powder toroidal cores? I wasn't sure the rotor needed to be made of laminates, but these would be even better than laminates. (Not physically stronger for high RPM.s, though!)
   The diameters said these "pucks" would stick out slightly past the outside of my rotor. So the next question was how to mount the optical interrupter parts. It might be necessary to cut some pieces of sheet metal and bolt them onto the outside of the rotor to make "solids" and "gaps" or "slots". Maybe I should use a different rotor after all?
    In fact, it just might be worth getting another special one cut for me at Victoria Waterjet.
   And in the stator, I still have the cylinder supermagnets and can still try out "permanent magnet assisted" configuration. That's a valuable reason to pursue the existing motor design. So on the night of the 26th I decided that was the plan.

Step 1: design another rotor and have it cut for me. I figured I'd put the coils into four 2" round holes made for them. Then they'd be aligned and I could center them vertically for ideal balance. I could have two outer diameters for the optical slots and solids areas, 4 of each.

   Then I looked at a rotor with four cores sitting on it. A big problem (all along) with using them was the there was nothing to grip to hold them onto anything without having something extend past the end. Well, really they were just iron powder with an epoxy coating. What was the difference between that and sintered iron or cast iron? Could I turn them on the lathe so they would have an inner mounting surface that would leave the outside flush? How about a fat washer that didn't stick out, with a flat head bolt? In spite of it being after 1 AM I took one out to the shop and turned it so it could be held in place that way. It milled easily and it just took a minute.
   Great! That opened up new possibilities for using them. Move to "Plan B": attach the "pucks" to this rotor with four specially made "washers". Then make four pieces to divide the edge into optical solids and gaps. Once the optical parts were mounted and the flux gap was set by putting just the right spacers on the shaft, it would be ready to run. That would seem to be the path of least reluctance. Too bad the coils were already made and glued down - I could have done similar for the permanent magnet "keepers". But first, I should get it to run as a simple reluctance motor. The performance of that would be the benchmark.

   Then I got the "flower pattern" rotor off the shelf. If I just made a few cuts and did some filing, I could put the four "pucks" into holes in it, all perfectly aligned. Okay, "Plan C". On the 28th I did one using a jigsaw with a metal cutting blade. That seemed good. Then I realized it needed support in both directions - it could slide either way. If I turned off say 1/32" from the outside but only went just over half way across, and left a bit more material in the rotor holes, they could slide in to there and no further. Then they'd only need to be kept from sliding out the way they went in.
   Even better would be to turn that 1/32" deep channel 1/4" wide (the thickness of the rotor) in the center so they couldn't move either way. But that would take some special and larger cuts in order to insert the donut. (It would slide into a bigger hole closer to the center of the rotor and then slide out to where it was held securely. A piece would be bolted on to prevent the donut from sliding back toward the center and the bigger hole. Hmm...)

   Anyway, on the 29th and 30th I cut and filed out the other three holes. The rotor would be the "back iron" of the toroidal rotor poles and one edge of them would be flush with its surface - never mind ideal balance for this one. Then I would mill a 1/8" holding ring slot just below and make four 1/8" steel pieces to (a) hold the poles in place and (b) form outside "optical solids" for the optical interrupters. The shape of these pieces to do this will be easier to figure out by cutting some pieces of cardboard. These will be bolted to the rotor plate. (...Or is that really how I'll do it?)

Other "Green" Electric Equipment Projects

Carmichael Mill ("Bandsaw Alaska Mill")

Front Pivot Self Adjusting Band Guides

   After thinking of this idea on June 30th, I made and mounted them on the 4th and 5th. I wasn't quite sure how to do it. The supports were behind the blade, not in front. Well... the saw was somewhat symmetrical. I took the blade off, turned it inside out, and put it on backwards facing the other way. Now the back was the front and the motor was on the left instead of the right.
   The "pickup truck canopy clamps" now seemed unsuitable. I made two bars of aluminum .5" x 1.25" x 2.5", each with a 1/4" hole going across the width in the middle as the pivot point. A 1/4" bolt went through the sides of a piece of 1.25" wide "U" channel steel, holding the pivoting bar in the middle. I mounted the wheels and bearings such that the band crossed over the center line along the bolt. I drilled bolt holes in the "U" channel pieces holding everything to match the holes already in the mounting arms on the saw. So I got the guide wheels mounted nicely. As I was turning the saw to adjust the band tracking, my finger discovered a sharp shard of aluminum on the pulley. It cut a fair slit and I quit for the day to nurse it.

   I tried the saw out the next morning (6th). I was pretty confident it would be great, but in fact it didn't behave well at all. It started out about right, but angled up a bit to what should have been above the cut depth and straightened out. Then it cut straight for a foot. I stopped to look and it seemed good. But after that it wouldn't cut straight. It seemed that if the band started going up, the guides indeed angled the band to cut downward, but the band twisted in the two inches between the wheel and the wood and continued its upward climb. Apparently it needed either more tension or a stiffer band. (and I suppose that the wider the cut, the more tension is needed to keep everything straight.)

With the band and cut too high, the guides aim the
band down to bring the cut back to the intended line.
But the band twisted and kept cutting too high.
(The problem turned out to be an unsymmetrical band
with dull teeth and no set at all to the lower teeth.)

I increased tension on the band twice, and each time it veered upward a little less than the previous time.

   Then it did the one thing I really feared: it tracked forward on all the wheels and started cutting into one of the steel "U" channels holding the pivoting guide wheels, which of course would immediately dull the teeth. I quit cutting.
   That this was possible - perhaps even likely sooner or later - was my biggest concern about the arrangement, but with all the action going on one doesn't necessarily see it as it's happening. One could hardly put a block of anything in front of the teeth to prevent it from tracking/moving forward. It would either be cut up itself or would dull the blade. But perhaps a wheel could be put in front somewhere - plastic or aluminum. If the blade hit it, it wouldn't dull the teeth. It would doubtless get gouged up some, but it would spin with the blade and not be quickly sliced into very far. That would be better than the blade hitting the working parts of the saw. And with any luck it would make enough noise and vibration to alert the operator.

   On the 12th (my guests having left) I looked around and found a piece of 2" diameter UHMW plastic with a 5/16" hole in the middle. A 5/16" bolt would suffice for an axle. (After all it was just a safety... it wasn't supposed to spin.) Next, where and how to mount it? I decided it should go just before the entry wheel, and could be attached near or to its mounting.

Left and right pivoting band guides with wheels grooved to shed sawdust,
and a roller to protect the blade teeth if the band should track forward.

The band still veered up, and the protection roller saw use at some point.

   At the same time I had noted that the guide wheels had a layer of sawdust caked onto them. In fact, it was glued on with spruce pitch. There were patches glued onto the band in spots, too. Sticky stuff, spruce pitch! Perhaps this was part of the problem. And in June I had noticed that the guide wheels on the Woodmizer mill had slots in the rolling surface - surely to help prevent sawdust from sticking.

   So - when I finally got to it - I did likewise, cutting in some shallow grooves with the lathe on the 27th. I also cut a piece of nylon rod about 3/4" diameter (nylon is harder than UHMW polyethylene) and a piece of "U" channel steel for the front safety roller. And on the lathe I drilled a hole through the rod for a 1/4" axle. After being waylayed by an underperfroming solar panel system and also hot weather, I didn't finish until the next evening.
   After a few details I took it out on the morning of the 29th. After starting well initially, it didn't cut well at all. Again I could see that while the band guides were angled in the right direction to correct the cuts, the band twisted between them and the wood, and the cuts went bad anyway. I figured the tension must be too low and adjusted it twice. If the band forms a bow inside the wood, the middle twists the band in the wrong direction, compounding the error. The cuts got closer with each adjustment, but still weren't entirely straight. And when I continued to cut they got worse.
   The band was getting pretty hot, and gradually it dawned on me that hot metal expands, and that that would de-tension it. That might explain why it seemed to cut well for the first foot or so, both times, and then started going weird. So it needed the water cooling not only to keep it from getting so hot the teeth lost their temper, but also just to keep it from getting slack. I kept going, very slowly and stopping to let things cool, and finished the board. It was about the slowest cutting job ever. The board and the piece remaining looked awful. (And I could see the first cut had started going off - bowing - a little before it hit the sap pocket, so the sap wasn't the initial problem.)

   The next day, the 30th, I decided to try cutting another board to see better what was happening. I cut just a few inches at a time, then let the blade cool. It took much of the day, while I did other things between each little cutting. There seemed to be three unhelpful things happening. One was the band heating up. If I plucked it when it was cold, it had a certain "twang". After cutting just a few inches it was already hot, and the "twang' was almost an octave down - not much more than half the tension. When it cooled, it regained the higher pitch.

   The second one was that the blade was dull. The sawdust coming off was fine dust instead of tiny chips. Well, that was no surprise since it had cut into steel previously. (The nylon rod "stopper" seemed to prevent the band from going too far forward at least once in the latest cuttings, as it now has teeth marks in it.)

   The third was that something seemed to be out of line. The cuts - all of them - seemed to start out at the end of the cant with the guide wheels level, a straight cut at the cut line. But from there it would immediately start to cut upward, in spite of the guide wheels pointing the band ever more downward as the cut rose. If I didn't let the band get too hot it would cut through the piece without going entirely crazy, but being all along above the intended cut line with the guides aiming it down, more or less, depending on the band tension and heat, and with the cut waving up and down. The twist in the band would keep it from following the aim of the guide wheels, which were always pointed strongly downward. And inside the wood the band arched up between the two guides. The board would have a "cup" in it, greater or lesser at various points depending (no doubt) on band tension, all the way along, matched by the remaining piece underneath being domed. Where I backed up and took another run at the same section, the new cut would sometimes have less arch to it and the old cut could be seen in the finished board.
   But what was the cause of this? The guide wheels were aligned with the main wheels. The pivot pins were directly in front of and in line with the band. Could the blade be duller on one side than the other? Perhaps the next thing to do would be to sharpen a blade, since I had no more new ones. And (sigh!) order some more new ones.
   On the 31st I took out my microscope lenses and looked at the teeth on the band. It seemed the bottom teeth were duller than the top ones. Then I noticed a more serious flaw: the bottom teeth were all in-line with the band. Only the top teeth had any set to them. Well, that would certainly cause it to veer upward! It must have hit something that both bent them and dulled them at some point in the trials. With any sort of 'regular' band guides, the cuts would have just veered off and no board at all could have been cut. I examined out the other two bands and they were both better. In fact, one was almost like new. (Should I try and fix this one, or just change bands? hmm...)
   In the evening I sharpened the teeth as best I could with a 5/16" round chainsaw file. (There was nothing quite the right shape, and the file was too coarse. But the bands are hard to change and adjust the tracking and I didn't want to take the time.) Then I bent each lower tooth out a little with a small pair of pliers. I also adjusted it narrower and used the 6" guide board instead of 8". The blade didn't get hot as fast and ran more easily, with coarser sawdust flying out. I stopped every foot or 15" or so for a couple of minutes for cooling. The board wasn't perfectly straight and it still had a little bow in many places, but it was much better and had been much faster to cut.

   I (finally) made a little cooling water system with a sponge touching the band at the top of its travel on August 2nd. To thread the hole in the lexan to hold the plastic pipe fitting, I scored across the threads of a similar copper pipe fitting with a zip disk in the angle grinder, to turn it into a sharp "threading tap".
   Then I tried it out by cutting a couple of boards. Other than having to stop and refill the thin tube 2 or 3 times while cutting (13' x 6" wide boards), it worked well. The band never got more than warm. (Where did my fatter transparent tube go? I have some somewhere.)

   Again I didn't have to stop and adjust anything. But again the cutting was slow and labored, the guide wheels were always above the band aiming it down (not as badly as before) and the cuts had an upward bow in them and weren't entirely straight along the board. But at no point did the cut veer off and keep going until things jammed as is so common for band mills especially with a dull blade.
   I'm pretty sure that with a good blade everything will be faster, smoother, freer and the boards pretty flat and straight. By the end of the second board the motor (Ryobi skillsaw) started to seem pretty warm. If the cutting goes easier it probably won't get very hot even if used continuously.


   The idea works. Self adjusting band guides are a huge breakthrough for the design of bandmills of any sort. They point the band the right direction to correct the cuts, and with a half decent sharp blade/band having sufficient band tension and stiffness, the cut can only follow their aim. It should happen with so little movement up and down as to be unnoticeable, and the boards should turn out quite flat. Even a "somewhat dull" band, as long as it's symmetrical, should cut good lumber - just more slowly.
   It was probably slower than an Alaska mill with the dull, off-kilter band and the cuts weren't very satisfactory, but no gasoline was used, little sawdust was generated and some rough lumber was made. If the band is sharp the boards should turn out nicely straight with very little wood being wasted to cut lumber even to very small dimensions. Obviously the next test - now that it's set up better so it can't readily damage bands - will be to do some cutting with a sharp one.

   A big question is how wide of boards will it be practical to cut with such a skinny, flexible band using so little power. If it only does a good job up to 6 or 8 or 10" it will still be a valuable tool for cutting big slabs and blocks into lumber with little waste. If it does logs and slabs up to 18" or more, it might replace the chainsaw type of mill, entirely or at least for smaller logs, and be priceless. And of course one hopes it'll cut the thinner widths quickly and easily.

   If the best band I have left cuts really well even to quite wide cuts, I'll just order more the same. (And I'll want to find a proper way to sharpen them and set the teeth.) Otherwise I might try thicker bands, and maybe wider bands, to get reasonably perfect flat cuts. Obviously with either more tension or with heavier bands more width is possible. But then they'll need more power and pretty soon the mill will be too heavy for one person to use handheld and for a 120 VAC plug-in motor, since it's heavy enough as it is. (40 pounds) Some of the components might be lightened to allow heavier band components on the same profile. (The 3/4" shafts and pillow-block bearings are needlessly heavy. The 10" wheels could be lighter.) Well we'll see!
   But if it's needed to cut larger logs, a mill on tracks will also work better than any previous band mill. People who get one will stop wishing they had a "real" dimensional or other mill, and they won't be spending a lot of time adjusting instead of cutting.

   Anyway it seems so promising now that I'm starting to think I may get my spruce milled up into nice lumber before winter after all! (First construction project with some of the spruce lumber: a shed with racks to hold the rest of it out of the rain!)

Electricity Generation

Solar Panel Installation

   On the 14th, not only because they were going to waste and the house-shading trees were now down, but also with a view to clearing space in the shop to work on reluctance motors, I decided it was finally time to put up the six large solar panels stacked against almost the only open piece of wall in the shop. (wall without built-in shelves covering it). Four would go on the south facing roof. I wasn't quite sure about the other two. One thing I do know is that the prices keep getting cheaper, and even in this cloudy climate it's probably worth getting some more. (In lieu of getting GeV HE ray electromagnetic radiant energy working, of course.) Someone in a video said solar panels in a "utility size" array can now make electricity for 3 cents per kilowatt-hour (presumably US¢ = presently ~4¢ Canadian). And that nuclear is dead if only because the up-front costs to build are staggering. (Not that they care about Chernobyl or Fukishima.) And that wind power from ever larger commercial windplants is now getting as cheap as solar, too.
   I think 3 ¢/KWH is probably cheaper than coal. It's certainly much cheaper than what most of us are paying! (A home installation here saves about 13 ¢/KWH.) At any rate, solar and wind are certainly far cheaper for the environment than fossil fuel. The high summer temperatures and droughts we're seeing in many parts of the world are actually being fortuitously moderated by the fact that the last two solar cycles have been weaker than expected with few sunspots and hence slightly less heat from the sun. The sun's heat varies by about + or - 1%, and it must be at least at -1% now. Since the Earth is around 300 °K, -1% might translate out to average cooling of -3 °K or °C, -5 °F. Without coal and petroleum burning there would be no global warming to speak of. People are starting to die every summer in various locales from heat, even in Canada, and everywhere seems to be on fire. If it doesn't end we'll all gradually cook.

  The 3¢ figure also reminds me of BC's Peace River "Site C" hydro dam project. While I would rather see hydro in most any form than coal and petroleum burning, I would rather see distributed floating hydro, and I suspect that every year that goes by makes this dam hydro project more uneconomic and a poorer reason to flood a huge river valley environment. Anyway, the cost for me to install existing panels and equipment is 'just' my labor. (This proved more costly than expected.)

   The next question was what to do with the electricity. I had two 200 watt "Enphase" grid tie inverters and a Chinese ("Smart") 1 KW grid tie inverter. The 240 volt Enphase inverters, which I bought earlier but have never used, have to be wired into an electrical panel.The "Smart" just has a regular AC power cord to plug into any electrical socket. But some long numbers say it meets somebody's standards for grid-tie-in. Sending the power to the grid ensures that whatever power you make gets used somewhere and that it comes off your power bill. And on this island in the summer if it's dry, there isn't enough water for the one hydro plant. Then it's that much less diesel fuel being burned. OTOH, it's supposed to be inspected and approved. That seems like a lot of procedure to go to for a 1 KW peak plug-in unit.

   Then, I had the idea for a 38 volt DC power system with batteries. I could start making and trying out the 38 VDC plugs (and see if they might be improved in any way), and some 38 V LED lights (and ?? appliances). And a battery based power system will only get more practical with cheap, high capacity batteries, which, having invented them, I am presently hoping to have produced or to produce.

   Or, simplest, I could just run a 12 volt system similar to the one I had in Victoria, using the same components. And hook up all my NiMH batteries to it, since they're just going to waste right now.
   I noted that in their last price list HES had a variety of 72 cell solar PV panels as well as 60 cell. (I at first had 60 & 72 confused with 54 & 60 and wondered why - 54 are no longer made AFAIK.) The new 72 cell panels, which must be about 2 meters tall and 300 watts, are rated as "36 volts". The open circuit voltage would be about 42-45 volts. Cruelly, trying to charge 36 volts with this would be like trying to charge 18 volts with a 36 cell panel - as I found, it's just not quite high enough voltage. OTOH to charge 24 Mn-Zn cells at 1.7 volts per cell would only take 40.8 volts instead of 42-42.6 for 30 Ni-MH or around 43.2 for 12 Li-ion. The difference might be just enough. But I'd want to try charging 12 of them (18 V nominal, 20.4 V to charge) with a 36 cell panel and see before I ran out and got 72 cell panels. (First I need to have good 12 Mn-Zn cells!) So if "36 volt" panels aren't for charging 36 volt batteries, what are they for? I think it comes down to the same reason I picked it as a good standard line distribution voltage: it's the highest voltage that's safe to be around when you might touch a live connection - as in, when installing and wiring solar panels.
   As for charging, I guess an MPP DC to DC converter at the front of the voltage regulator that will be needed for an Mn-Zn battery isn't an overwhelming extra circuit to add to the works.

   Well, there's a pretty ambitious program, even without reluctance motors and HE ray energy! At least I got the four large solar PV panels up (~1 KW) on the 14th and 15th. I did a lot of climbing up and down the ladder for things left on the roof or on the ground that a bit more advance thought might have eliminated, but I spend more time getting everything ready on the ground ...especially on cleaning the glass. (Might as well have clean glass for a new installation.) And a few of the lag bolts didn't go in properly. I got 5 longer ones later and fixed it up. (no high wind came, whew!) The angle is better for summer than winter, but putting them up was a breeze compared to on my old house in Esquimalt/Victoria with its very high up, 45° angle roof. I poked my head into the attic to make sure none of the panel wires were shorted together. What to do with the electricity was another project for another day.

   Then I put the two remaining panels, the 208 watt, 54 cell "oddballs", in the dungeon (er, root cellar). And I had the thought that one of them plus a 100 watt 36 cell panel would both fit somewhere as if they were a single 100 cm by 216 cm panel and make 90 cells for 45 volts (about 55 volts open circuit) to charge 36 to 40 volt batteries for an "off grid" system without the aforementioned DC to DC converter. But the cells on the smaller panel were 160 x 100 mm while on the large ones they were 160 x 160 mm, so one wouldn't be able to get full current from the large panel, reducing the pair to about 230 watts. Still, I could do two sets for 460 watts.

   I got out the 1 KW grid tie inverter and looked at the manual. It said it was a "microinverter" designed to be tied to a single solar panel. Huh? The biggest panels are around 300 watts. Surely one should be able to parallel 3 or 4 panels to feed this inverter? Tying all four together was certainly what I had in mind. If I was going to connect just one panel - or even two - I'd have bought the 500 watt one.
   I started to think maybe the inverter wasn't really good for 1000 watts, and also that it might not limit itself to what it could handle and would fail if the panels had more than 1000 watts available. Just because the brand name was "Smart" didn't necessarily make it so. But it said it would shut off if it overheated until the temperature came down. 95% of the time (around here) four panels would be well under 1000 watts, but on those rare sunny summer days it just might max out around noon (1:48 PM PDT). I thought of connecting just three of the panels to it, maximum about 760 watts. If the "1 KW" inverter had trouble with that, I'd be pissed off.

   A couple of days later I went up into the attic with the same box wired with fuses and diodes that I had used in Victoria and wired up all four panels. Then a nightmare began to unfold. I could find no way to get the power wire from the attic to the large closet that I wanted to use for a solar equipment room. (To skip the gory details, skip the next 3 paragraphs.) I poked a wire through the closet ceiling but couldn't find it from the attic. Then I realized it was under a different section of roof and I couldn't get there from the attic. Then, the low rafters prevented outside wall access from the attic along the whole length of the house on both sides.
   On the 21st I located an interior wall that already had some floor to ceiling wires running through it, for electric heat and an outlet. From the crawl space the bottom of the wall could be accessed; I could see and reach the back of the outlet box as well as the wires. I drilled a hole from the attic into it. It would be ideal. But how to push a wire through a wall with fiberglass insulation [why?] in it? In wandering around looking for something I found an old set of tent poles that assembled end to end with an elastic string through them. Perfect! The attic there was just tall enough that I could put the segments into the hole one by one and attach the next one, until I'd pushed it right through to the bottom. It had a solid "clunk" and I had obviously hit wood at the bottom. But I couldn't see it from the crawl space. Where could it be? Maddening! Poking up with a piece of coat hanger wire showed there was some cross piece of wood a few inches above the inside floor level. That was what it had hit against. Close but no banana. The cross piece was stopping my pole inches above where I could access. Even if I could drill a hole in it, finding that hole from the far end of a 9' pole sticking into another hole was probably problematic. Definitely all the wires in there were done before putting the gyproc on!
   Then I went to the end wall (going the long way around but at this point what the heck?) and tried to drill a hole through the top plate of the wall. That's usually two 2"x4"s stacked on each other - about 3.5" thick. At 9" I was still in solid wood. I thought perhaps I had unhappily hit a wall stud an was drilling down through it, so I tried again a few inches over. I gave up at 8" with the cordless drill motor smelling hot.
   Finally I drilled a hole through the plywood end wall into a space above the garage ceiling and then sawed a hole in that ceiling. Halejullah! - I could see my hole in the wall! But at this point I gave up and decided to run the wire down the wall inside the garage instead of inside the wall itself, and then poke it back through into the crawl space. But I had to go out and that was that - a frustrating day and I didn't get one wire run! So much for "a little before breakfast job"! Was a kilowatt of solar power really worth all that cussing and swearing? My 1879 house in Victoria was mostly simple to do wiring and plumbing in compared to this 1987 one. You could poke wires up and down through the chimney space and there was a lowered ceiling that made a crawl space between the two floors in addition to the one under the house and the tall attic space. A house 108 years newer didn't translate to "planned for utility services access"!
   The next afternoon and evening (22nd) I finished. The closet end, under the low part of the floor, was a "crawl 30 feet on your belly" [each way] job. I made it less painful by pushing pieces of cardboard ahead of me to lie on. I drilled a hole above the cement and went through two layers of wall with insulation between. I taped the wire to the long drill bit and pushed it through the hole, even right through, praying that if I couldn't find it in the house and had to pull it back out, I wouldn't lose the drill. But the wire ended up in the closet right about where I wanted it.
   Then I went up into the attic again and redid the wires to the four solar panels so that the wiring box was next to the outer wall instead of far into the attic. I pulled a bit more wire through and connected it. With the extra length of route, the #10 wire that I had thought was so long was just about the right length and I didn't cut it. Lacking cable staples, I tied it to other things in the crawlspace with a few cable ties.

   The next morning I started looking for solar equipment. Where was that Zahn DC to DC converter? Not finding it in a short search, I decided to try out the "Smart" grid tie inverter. It was sunny with some haze. The inverter worked. It drew the panels down from ~35 volts to 25 or 26. It said never to connect the AC unless the DC was connected. I wasn't impressed by the DC connection posts. They were too close together risking shorting the panels, the threaded rods were too short, and the plastic thumb nuts couldn't be tightened as much as they should be (a) without pliers and (b) without fear of stripping the short threads. It looked the "plus" wire had slipped, and when I touched the unit there was a spark from the connection. I pulled out the wall plug, turned it off, and improved the DC'+' connection. The nut still wasn't very tight, but with small pliers I bent the stiff wire around a bit farther, into more of a loop. I plugged it back in, and somewhat to my surprise after several failed inverters, it still worked. I left it there, on the entryway floor, getting very warm and with its cooling fan periodically coming on and off. It certainly seemed to be sending power to the grid.
   I had a little 'power monitor' from Canadian Tire. You plug that into the socket and then plug an appliance into it, and it tells you how much electricity it's using including present watts and amps and a running total KWH and cost (enter ¢/KWH). But it didn't seem to tell how much was going from the appliance into the wallsocket. That would be too simple! Maybe if I turned it around? But that would mean two "cheater" cords to connect it - an F to F and and M to M. Ugh! The power monitor was getting warm nonetheless while it read "zero", so I turned off the inverter and unplugged the monitor.
   Later it occurred to me the F-F could be just one of those 1 to 3 splitters that you plug into an outlet. I taped up the unused male plug since it would be live. And I sacrificed a computer power cord and put another male plug on the other end, and I hooked up the power monitor backward. That seemed to work to measure the power. But it said it was only making around 400 watts.
   The low power level suggested that at least one and more likely two panels weren't making a connection. It shows why it's a good idea to measure these things. Sigh! Back to the attic and the fiberglass dust. But not today!

   I checked it out on the 27th. There didn't seem to be anything wrong. The panels had bright sunshine. Each solar panel read ~29 volts. Both diode bridge outputs read 28.3 volts. (They were after all tied together. All together the readings said the connections were good.) Typical power monitor readings were 4.9 amps, 420 watts, 492 volt-amps and .85 power factor. The power company meter said I had used 12 KWH in a day in spite of whatever was generated. Could the power monitor somehow be wrong, being used differently than intended? And was the power company's meter not giving any credit for power generated without a permit?
   (And where was 12 KWH going in one day in July with no heat on, no laundry, no bath? That would be 500 watts average. Surely the fridge and freezer weren't that hungry. And a few LED lights at night. And some computers and printers mostly on "sleep". And the coffee maker but only for minutes. And the toaster, microwave and one stove burner, but again for just minutes. And 70 watts of LED grow lights on all day. And the well pump when I watered the garden. Hmm, I guess it all adds up!)
   I slit a 3-wire extension cord and checked the current with two different clamp-on ampmeters. While the power monitor read 4.2 amps, both clamp-ons said it was 3.5 to 3.6 amps. Even lower! It worked out pretty close if divided by the power factor. Well, I guess I'll believe the monitor. How much power was the inverter spewing out as heat, then?

Panel Connect Box in attic, with heatsink. Black wires seen from one panel on roof
go to in-line fuse and then to a yellow cable (ex extension cord). The four yellow 
cables go to the square isolation diode bridges bolted to the sides of the box with
connection wires soldered to them. (Only 2 diodes are used in each bridge.) All the
"-" wires are tied together, and the two bridge output wires ("+") are tied together,
and the "+" and "-" go to the white cable to the "equipment room" closet.     
Connections are all marrettes. (Primitive!)                      
   Then I tried the DC side. The DC clamp-on said just 17 or 18 amps, at about 25.5 volts. That's just 450 watts. (So the inverter was apparently turning less than 50 watts into heat.) From four 250 watt panels in bright sunshine? I went back into the attic with the DC clamp-on and measured the current from each panel. Sure enough, they were only doing 3.6 or 4.5 amps each -- depending on which wire and which way around the clamp-on was placed, which isn't very confidence inspiring. (Was there some way to zero it? It must have .5 amps offset... Hmm, turning it on unconnected it reads about .2 amps.) But it did say the panels just weren't putting out the promised 6 to 7 amps.
   I went back into the attic (one more time) and verified that the voltage was 28.3. It was 25.5 at the other end of the wire. 2.8 V * 17.5 A = 50 watts was being lost in that long #10 wire. I knew I wanted to keep it short!
   I also discovered to my dismay that the diode/junction box was pretty hot. .7 V * 17.5 A = 12 watts being turned into heat by the diodes. (Another place for synchronous rectification?) It needed a heatsink. I wasn't looking for yet another job! But it was hot, and screwed to a rafter. I dug out a heatsink and in the search found the Zahn DC to DC converter (18 to 40 volts converts to 14 volts, output adjustable within a range.) After putting a couple of holes in it I took the heatsink up to the attic (yet again), dimounted the box and remounted it on the heatsink. Since I didn't know the hole spacings in the box, I just made one #8 threaded hole in the heatsink. With just one bolt (and no heatsink compound), the box stayed a lot warmer than the heatsink. But it was cooler than it had been, and now the heatsink was between it and the wood.
   A while later with the sun directly in line with the house, it was doing 432 watts steady. And almost 100 watts was being lost, half of it in the long #10 cable.

Cheap Grid Tie Inverter: cheated!

   For a while I couldn't figure out why it was hardly over half of the rated amount from each panel. (I cleaned the glass! It's not fair!) Then t started to dawn on me... MPP for the panels was supposed to be (IIRC) 27.4 volts. I measured the panel voltage (all four panels) at the box as 29 volts. So really, it doesn't seem to be drawing them down far enough. That would be the "Smart" grid tie inverter's fault. In fact... as I think about it... it puts out around 420 watts for most of the day, seemingly almost regardless of the sun angle. Later in the day the voltage at the inverter was down to 24.5 volts but it was still doing 420 watts. I suspect it's really a 500 watt inverter with a "1000 Watts" label on it so they can charge more money. Why else do they say to hook it to a single panel? If you do, you'll never find out it only puts out under 420 watts regardless of what's available from the panels. I began to realize the product wasn't as advertised. Power and current don't match the specs given in the manual. There are no controls on the inverter or clues in the manual as to how to change anything. I didn't know that "Smart" meant "dishonest" and "unethical". That smarts! I can't remember how much I paid for it, only that it looked like a good deal. Perhaps it was "too good to be true." At 420 watts instead of double that, the utility meter only slows down instead of reversing. It's really only using two of the four panels most of the day.
   Oh well, at least it works. Nothing has blown up. I bought it too long ago to complain now. If I was "into" it I'd buy another one (a different one from somewhere else) and get all the panels mounted and hooked up. And I might connect those two "Enphase" 200 watt grid tie inverters to the two 54 cell, 208 watt panels. Then I'd be making well over 1000 watts to the grid... on our uncommon sunny days. And according to the utility meter, that would still be less then I'm using daily. 8 panels instead of 4, and of course fully and properly tied to the grid, should make it a surplus - on sunny summer days.

   And speaking of cheap stuff, I put another power monitor in the line. This one, "#2", read about 20-30 watts less than the first one. I put "#1" on my coffee maker and it read 950 watts where the second one had said 900. It didn't seem I had much of anything that was very confidence inspiring! The plate on the coffee maker said "900 Watts". So the "#2" monitor was probably more accurate, and the inverter was actually putting a little under 400 watts to the grid, and turning more power into heat; probably around 50 watts. Let's see... 16.5 A * 25.5 V = 420.75 watts in from the panels, and the power monitor says 380 watts out. That's 41 watts of heat. So diodes (12) + wire (50) + inverter (41) = 103 watts being lost to make 380 watts out - just over 20% losses. This is why I like to keep the number of components in the circuit to a minimum, and to charge batteries straight from the panels if possible. (And I certainly tried to find a shorter route for the cable!)

   380 watts seemed like a lot less than 420. I put power monitor #1 back on and it still only said 390. What was different from yesterday? I had changed a "computer" power cord - one from a box to the one that came with the inverter. I put the old one back on, and it went up to 405 watts! (Still not 420.) I got the fattest computer cord I could find, and 2' shorter, but it didn't seem to make any difference. What about the M to M "cheater cord"? It was the thinnest. I cut it from 5' to only 9" long. It didn't seem to make any difference either. An hour later, with the sun rising toward its zenith, the monitor said around 415 watts. Was the inverter putting out more toward midday, or was it the power monitor unit warming up and reading higher? I put #2 back on and it read about 390. Some of each?

   Well, enough of that! As it is, the system should make more power when I get it "off grid" and I'm not using this inverter - if I harness that power effectively.

   Thinking about off-grid storage, and about the Suzuki Swift fire that had almost burned the house down, I started thinking that it would be safer to be charging storage batteries away from the house. Theoretically there should never be a problem, but then there's equipment failures that might potentially deliver the full power of the panels to the batteries and cause them to overheat. So having spent all that aggravating time running the wire, I decided the battery installation should go elsewhere, outside - and the other direction was where the outbuilding clutter was. (Hmm... or maybe I could put them in a metal or gyproc box of some sort in the other garage on the cement floor, not far from the "wiring closet" and out of the weather.)

   I started looking at the power meter and noticed that even with the solar savings the house was using about 12 KWH/day - most of it when the sun wasn't shining on the panels.  (The panels, with the half-size inverter, were only saving around 3 KWH, instead of 6.) On about the 4th day, the 30th, it said the previous day had used 18 KWH. Why? I couldn't recall doing anything extra that would use all that. But after not driving anywhere for a few days, I finally remembered I had driven to Tlell and back the previous day, almost 50 Km. (It just seemed too hot to work, so I went visiting.) Recharging the car accounted quite nicely for the extra. (That dollar's worth(?) of electricity would have been around 7.50$ for gasoline around here.)

   August 1st was the first cloudy day since putting up the panels. I put the slit extension cord on and with the inverter running measured 1.6 amps going to the line. At the measured 122 volts, that was just under 200 watts. The next day was even duller. From 2.2 amps (270 W) earlier in the day the readings dropped to 1.25 A (all @ 122 V) for 150 W (and then a light, misty rain started). Obviously that was all the four panels were supplying. (probably not much more than the average of what the freezer plus the fridge were using.) So it was only on sunny days the low powered inverter was wasting capacity. Nice as July was, sunny days are a minority around here, so the loss over a year wouldn't be that great.
   But apparently if one wanted to try to offset one's own summer electricity use in this climate (never mind winter), 10 panels or more and inverters to match would be pretty minimal. If I hooked up my other two panels to make six, all the power would be in use the majority of the time. Should I get another grid tie inverter and make a real, official installation? Better to hook up the Zahn DC to DC converter and get my neglected batteries charged.

Synchronous Rectification?

   Losing 12 watts in the solar panel isolation diodes and in the process making the box in the attic hot makes me think of "synchronous rectification" whereby a power MOSFET is turned ON when the diode is to conduct, and OFF when it isn't. The voltage drop through an ON mosfet, perhaps .1 volts, is very low compared to .7 volts in a diode, and so almost no power is lost to heat. The problem is when to turn the mosfet on or off. When it's ON, the voltage is virtually the same on both sides. One would have to either periodically turn OFF the line and see if the panel voltage goes up or down compared to the load side, or else sense whether current is coming from or going into the panel with a relatively sensitive monitoring system. Or perhaps even the very slight drop across the mosfet could be used to turn a comparator on or off?
   A synchronous rectifier "panel isolation" system would virtually eliminate the diode losses. Not that that's the big loss in this system, but it probably wouldn't cost much more than diodes, which need to be installed anyway, and every bit helps. In fact, if it eliminates a heatsink, which does cost some money, it should be cheaper.

Electricity Storage

Rechargeable Battery Making
with oxalate electrolyte:

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

Electrolyte Additives Note

   I would note that although the primary component of the electrolyte is potassium oxalate, a small amount of sodium oxalate (or even some other oxalate) might improve some characteristic or other. This is worth experimentation. The chief constraint is that other than potassium, most oxalates aren't very soluble, so not much can be added.

Retesting the lead-zinc cells

   On the 10th my visitors left, and my [presumably more pure] calcium hydroxide arrived. After sitting a couple of weeks or more, one lead-zinc cell read about .8 volts, the other about .25. Both the lead and the zinc plates seemed to be well stuck to the separator paper. So it wasn't possible to remove them for inspection without wrecking the separators. I did a first battery experiment: squeezing out the old electrolyte and diluting it in water, and putting in new in the .8 volt lead-zinc cell with the three zinc sheet plates. But first I used one of the nitrate-nitrite test strips in the old electrolyte. There didn't seem to be any. Was there some mystery contaminant in the pottery supply calcium oxide that caused discharge, or was I wrong [again, still] and the cause lay elsewhere?
   The lead-zinc cell I had been testing had started doing funny things before I had quit testing it last time, and I found the "+" terminal lead (wire) had fallen off the lead (Pb). I soldered it back on again.
   The voltage seemed to drop pretty quickly. I charged it longer, 3 or 4 hours. That helped. Then I remembered the other cause of self discharge, oxygen getting in. I covered over the top as best I could with modeling clay. Charge current wasn't high and the short circuit current was under 1/2 an amp. Perhaps the concentration of electrolyte was too low?
   It didn't look like I had got far on elimination of remnant self discharge. was still down to 1.700 volts in about 20 minutes, and losing almost 2 mV/minute. Now, even assuming it had anything like a seal, how much oxygen was in the air space under the lid? Could left over oxygen, and poor sealing, be the real culprit? Or worse, if the zinc bubbled hydrogen and made pressure while charging (once it was mostly or fully charged), might it suck more air in later, each time or continually? Or if the lead was fully charged, it would bubble oxygen inside the cell, leaving it there to gradually discharge the zinc electrode. Commercial manganese-zinc dry cells with their very low self discharge were, after all, well sealed. (And also they didn't get recharged.) Perhaps what I mostly needed was a proper case, and perhaps care not to overcharge?
   I had just dumped some K2C2O4 and CaO into the empty cell, put the electrodes back in, and then added some water. I pried open part of the top and dumped in some more K2C2O4. Charge current didn't seem to go up, but when I shorted it it went up to .7 amps. But wasn't this the same cell that put out 7 amps short circuit current when I first made it? Why now just 1/10th of the current? Instead of 1.3 volts into a 1 ohm load, it just managed .38. A little later I took it off charge and shorted it again. The current started at over .8 amps and instead of dropping clawed its way up to a full amp over 10 seconds or so. With a one ohm load it started near .28 volts and again rose in several seconds to .382. Perhaps it was just short of oxalate, and what I had added was taking time to mix with the pure water that I'd soaked the workings in to try to eliminate the old electrolyte. It continued to improve.

   By the 12th the cell was acting funny again. The voltage readings were jumping all over. I pulled pretty hard on the "+" wire and the voltage reading rose and became steady while I held it. The obvious culprit was the soldered connection, which was underneath the modeling clay cover. It's amazing how fast things go bad inside a battery if it's not right. No figures were given in the solubility table for tin oxide or oxalate, but basically tin with a positive charge was going to dissolve or at least disintegrate in most anything and leave the wire with a high resistance connection to the cell. ...and could its reactions have anything to do with the gradual self discharge? (Can you solder with lead and no tin? Where did my flux pen go so I could try that? I wanted it last time but couldn't find it. I finally did, much later.)

   At that point I was tired of the Pb-Zn cells and working on a new cell.

Lead-Acid Battery Capacity Note

   In the last issue I estimated that the small 'motorcycle' battery I bought was about 4 amp hours or less. It weighed 1.67 Kg. Now I've found another old lead battery of similar size, 1.52 Kg, which says on it "4.0 amp-hours". So the estimate was probably pretty close.

Battery Type Energy Storage Comparisons

   Again I can't help remark on the contrast: an alkaline "D" cell is 12-18 amp-hours and I weighed a couple at 143 grams each. Both types, Pb-Pb and Mn-Zn, yield their maximum amp-hours at low discharge rates so we compare the 4 with the 18. It takes 8 Mn-Zn cells for 12 volts, so 1.14 Kg for over four times as much energy - from a cheaper chemistry that can now be made highly rechargeable.
   Furthermore, compared to a Ni-MH "D" cell, with 1.5 volts instead of 1.2, and with 18 amp-hours instead of 10, it's 1.5/1.2*18/10=2.25 times as much energy storage. And by weight, 163 g (Ni-MH) / 143 g (Mn-Zn) makes it 2.56 times the energy by weight. Since Ni-MH more typically puts out 1.25 volts and Mn-Zn may be somewhat under 1.5, the comparison may be somewhat overstated, but we may fairly say a Mn-Zn "D" cell has twice the energy storage capacity of a Ni-MH "D" cell and well over twice by weight. And very low self discharge.

NEXT CELL: A New Larger Cell in a New Case

   On the 11th I decided to make a Mn-Zn cell with bigger electrodes, 4" x 3". (=100 mm * 75 mm = 75 sq.cm faces... One manganese dioxide electrode with a zinc electrode on each face should give a few amps.) I would cut flat pieces of ABS plastic and 'glue' them together with methylene chloride for a custom size case.

Manganese Dioxide Pocket Electrode(s)

   I put the 'from scratch' Mn-Zn electrodes from last month into water to dissolve out any electrolyte contamination. After a day there seemed to be a bluish color in the liquid. I wasn't sure if it was dissolved or precipitate. It didn't look uniformly distributed. I decanted some water off the top and it looked clear. Best guess was copper oxide powder. Maybe the copper mesh wasn't going to last forever, even if the electrode itself would? (Then again, the electrode had been disturbed more than once and was falling apart in general.)

   I wanted to try out cupro-nickel for the new current collector for a MnO2 electrode instead of just copper. But I only had sheet, not mesh. And it was heavy and stiff. (Why did I get such a thick sheet?) I couldn't see how the powder could be induced to stick to that, so I decided to make a "pocket" electrode with the MnO2/graphite powder held inside a thin "pocket" of perforated sheet metal. I took a piece of the old #28 gauge nickel-brass from 2008. That at least seemed more suitable in thickness. I tried punching holes in it with the pin frog in the press, but I stopped at 3-1/2 tons and there were no holes, only slight indentations. The trouble with etching holes into it is that I don't have a silkscreen or PCB print ready to do it with, and it seemes like a fair project to make something decent along those lines.

Pocket Electrodes: Perforating Sheet Metal with a "Bed of Nails"

   Somewhat frustrated I took a roofing nail and a hammer and punched a few holes. That at least seemed to work, and somehow it suggested the idea to drill a whole bunch of holes in a row in a piece of wood to slip roofing nails through. They could be pounded into the sheet metal one at a time without needing a huge force. I didn't put the holes far enough apart and the heads overlapped. I found some other small nails still with flat heads that wouldn't go through the holes, and made four rows of twelve holes for these nails more closely spaced. I used the milling machine and cranked the handle 4 turns per nail to get the holes evenly spaced. (A CNC drilling setup would have been ideal for this. And next time maybe I'd make enough rows to do the entire electrode.) I set it on the piece of sheet with a softwood 2"x4" under it and hammered on the nails until they had all gone in a bit. The metal was perforated at every nail. I then did my best to put another set of holes in the spaces between the first set. It was hard to tell about the alignment and they weren't even (and if you turn it over to look closer the nails fall out), but it did seem workable and I did both faces. To my surprise the nails had little tendency to want to stick in the holes they made in the metal and the metal pulled away easily. I bent some of the sides partly up to make the "pocket" into which the powder could be placed and compacted by squeezing it between the faces. Then the bending could be completed with a hammer or pliers.

First try at a perforated nickel-brass pocket electrode.

   Then I started thinking about electrode size if I made more. If I was using sheets 6" x 12" as all the nickel-brass sheets were, the optimum use of the metal would probably be to make 5" x 2.5" electrode pockets from a single piece of metal. They would be cut to 3" x 12", and a 1.75" square would be cut out of one end, leaving a 1.75" long x 1.25" wide connection tab sticking up. After punching the holes, the edges would be bent up 1/4" to form closed edges that would be crimped over so the powder couldn't fall out. The bottom 1/4" and then the bottom 5.25" would then be folded up, closing off the top, bottom and sides, with the MnO2 & graphite powder contained. (It might need something to hold the middle areas from bulging. Rivets across? or just push everything into the cell and let separator pads press against the pocket?) A key feature is that it would all be one single piece of sheet metal for the most solid possible connection throughout and hence highest current capacity. The rolls of zinc metal being 3" wide as well (oops, 2-5/8"), zinc pocket electrodes could be done pretty much the same way.
   Since the nails didn't tend to "stick" to the metal, I thought I'd do a new "bed of nails" block with fine, short finishing nails - perhaps 5/8" in a 1/2" plastic block since they don't need to stick out very far. With the small heads they could be very closely spaced. I would do them in an "X" pattern with alternating rows at 45° angles for closest spacing. I might just make it big enough to do an entire electrode in one stamping without moving it around - several hundred nails. Or 2.5" x 2.5". If it's not a mass production tool, and if it doesn't make the ideal pattern of very fine, very closely spaced holes, it's at least an adequate, simple and replicable way of perforating metal sheets.

   Almost since I started trying to make batteries ten years ago I've wanted to make pocket electrodes, but I kept running up on the rock of not finding any satisfactory way to perforate a piece of sheet metal - or even graphite or plastic. After all the unusual and unsatisfactory things I've tried... Here it Is!

   On the 15th I tried making a better "bed of nails" block with 3/4" finishing nails. I measured one nail as 1.13 mm diameter, and I used a 1/16" (1.58 mm) drill bit. Instead of turning the crank 4 turns I went 1.5 turns. I drilled 5 rows of 31 holes and broke 3 drill bits. It was just as well I stopped there because it wasn't very good and it didn't work very well. For all I was using a 'precision' milling machine the holes weren't even at the top, and the drill wandered and they came out in wild positions at the bottom. Trying to use it, it was too even a pressure over too small an area. It felt like I was pounding on a solid instead of nail points, and the sheet metal formed a dish under the nails instead of just being punctured where each nail was.
   I took out the finishing nails and put in cigar box nails in every second hole in the first, third and fifth rows. That seemed to work adequately. I know the commercially made ones have far more tinier, closely spaced holes. Doubtless that's optimum, but I expect there's a lot of room for variation without losing a lot of current capacity.

Pick a Size

   I cut a 3" x 12" piece of nickel-brass for a shell and I went over the whole electrode perforating a patch at a time. Then I bent it up to make the outer shell and cut off the part that wasn't the tab at the top. (101 elephant jokes... Q:How do you carve an elephant? A: Get a big block of marble and chisel away everything that doesn't look like an elephant.) BTW you leave the jagged edges of the holes inside. It's more current collector surface for more of the powder to contact to.
    I weighed the piece: 60.1 grams. That of course is dead loss as far as adding amp-hours is concerned. But if it was much thinner it wouldn't be very strong. So the more powder I can stuff in there, the less the negative impact of the weight of the sheet metal. (After all, you don't want negative impact in a positive electrode!)
   Dimensions excluding the terminal tab were 2.625" x 5.0" x .3" or 67 x 128 x 7 mm. Hopefully the electrode inside will be 125 x 64 x 6 mm. That's 160 sq.cm counting both sides. At 25 mA/sq.cm that would be 4 amps. It might put out 10 if shorted. Or it might (hopefully) do substantially better. Making it would tell.

Folding it up after perforating

Loaded with 25+ amp-hours of MnO2

   For the other side, would simple zinc metal plates suffice, or should they be double layers or zinc powder filled pockets? Before going to too much trouble with zinc electrodes, I decided to fill the MnO2 pocket and see how much electrode substance it had. After packing in some MnO2 paste (which now included some Veegum and a small squirt of Sunlight dishsoap) and then compacting it and squashing out liquid at 5 tons pressure (over the whole electrode) it weighed 180 grams, so 120 g of paste. I dented the sides of the electrode with a hammer and screwdriver while it was in the press. It seemed to stay shut, but it really needs something more positive to hold it. Perhaps if I hit the edges with a nail or an awl that would 'rivet' the two edge pieces together.
   I took apart a couple of "F" cells. The "+" powder weighed about 80-90 grams and one rather corroded zinc can was 15, the other 17. It probably started out well over 20 grams, but it still shows how little zinc metal it takes to match a positive electrode oxide powder... at least if it's in chloride that gets dissolved away as it discharges. 25 grams of zinc is 20.5 amp-hours if it's fully utilized. So it might be a guess that the cells are actually under 20 AH.
   Taking 85 as the median, the electrode had 120/85=1.41 "F" cells worth of MnO2 mix. If we take an "F" cell to be 20 amp-hours, that would be 28 amp-hours. (And let's not forget the potential of another 14 amp-hours at 1.2 volts instead of 1.5 volts for 42 amp-hours... and another potential 14 at about 1 volt, but that's getting pretty far down... and assuming sufficient zinc.) So with two such Mn electrodes in the cell, that would be 56 amp-hours with another 28 in reserve at lower voltage.

More zinc electrode ideas

   To match 56 amp-hours requires 68 grams of zinc, or matching the 84 amp-hours that's available down to 1.2 volts, it would be just over 100 grams -- if the zinc was 100% utilized. But if it's only 10% utilized, that's 680 grams or 1 Kg - more than all the rest of the battery.

   Last month I thought of folding up zinc sheets to give them more surface area. This month it occurs to me to also put zinc powder into the folds. The powder (if not too fine) should add further zinc surface area while taking nothing away from the mechanical properties or electrolyte penetration of the metal sheets.
   I also wonder how much surface area could be added to the sheets by some deep gouging or perhaps coarse sanding? That might be an alternative to texturing in the rolling mill. It should even be better - if sufficient surface convolutedness can be attained.

   I tried folding the zinc sheet by primitive methods rather than in a sheet metal break. It was thin and soft enough to fold over any sharp corner, so I started thinking of how some good folds such as those in the cross section drawings of last month might be accomplished by hand for a prototype.
   I put a spare 2.5" square zinc electrode in a vise with two thin pieces of hard metal. I bent it around 180° each time and then reversed it and moved it up for the next fold. It ended up 2.5" x 1", less than 1/2 its original face size. But it was pretty irregular. And I would rather somewhat stretch out the zinc for the extra surface rather than simply fold it. (Thinner plus more surface area.)
   I started thinking also of making aluminum rollers - for the rolling mill or to drive with a motor - with the requisite surfaces on them to reliably and uniformly create the fan-folds, or with slots to hold harder metal to bite into the zinc. (Or could I find two splined shafts, or wide gears with fine teeth, that meshed "just right"?)

   But I started to think this was mostly ways to get the same zinc sheet into a smaller area. Except for stretching it a bit to get somewhat more surface per volume, it didn't address the low utilization of the sheets. Surely powder should do that? And surely the best way to do that would be to make pocket electrodes - same as the "plus" side only considerably thinner. The volume of zinc powder to make up the required grams of zinc wouldn't be much.
   The zinc metal for the pocket was just half the weight (32 g) of the nickel-brass one (60 g), and it would contribute to the amp-hours.

  That would determine how much zinc (getting hopefully, say, 20% of theoretical capacity) would be needed to match the amp-hours.
   And I found I had a jar of somewhat coarser zinc powder to try out for pocket electrodes.

   Ugh! I had thought the zinc strips were 3" wide, but on the 15th I measured the roll as 2-11/16" wide. The new roll was 2-9/16". So much for all the metal being a neat 3" width! Time to recalculate potential sizes. Better finding out now than after making more "odd size" electrodes that won't match. But plain zinc sheets should match the width of the folded cupro-nickel pockets quite nicely.

Bending up the edges of the zinc electrode


   Making pocket electrodes and zinc sheet (or pocket) electrodes gives the idea that with solid pieces of metal facing each other in a flooded cell, simple plastic spacer pieces to keep the separator sheets from touching are often used rather than paper or fabric sheets. It's simple and inert. And in experiments, electrodes in a cell can be swapped around without being "stuck" to a separator paper.

ABS Plastic Case

   On the 16th I cut out a case. Inside it was 2.75" wide, 5.25" tall, and 36 mm thick. I used 1/16" ABS for the outer faces, and 1/4" ABS for the top, bottom and sides, so it had to be 3.25" wide, 5.75" tall and about 40 mm thick. The top had an extra piece of 1/16" of the outer dimensions to keep the fat piece from falling in. It was black except I made one edge white to be the "+" side, and the top face white as well. It was all "glued" together with methylene chloride. (surface dissolves in a bit of methylene chloride squirted on at the joins through a very fine "syringe" tube, and then re-hardens as the methylene chloride evaporates.) It weighed 138 grams.
   I filled it with water and it leaked part way up one side. Fixed with a little more M.C. I was glad I thought to try that first. Unfortunately it wasn't the only leak.

   How did it look next to a 3.4 Kg, 3.2 volt, 100 AH lithium? 2/3 as thick, 1/2 as wide, 2/3 as tall.  Maybe 1/5 the overall size. But it would take 2 to get the voltage and 1-1/2 to get 100 amp-hours (actually 84 AH - at rated voltage). That's 3, total 3/5 the size and maybe 2/3 the weight. But then if you really needed the range (and assuming sufficient zinc), you would still be able to drive maybe 35% farther at lower power as the Mn2O3 converted to Mn3O4. That would of course all assume these were ideal capacity cells and that you could draw 100+ amps from them as you can from the lithiums. Hah! Fat chance of that. But if my first cells don't meet expectations, that's still the potential.

Cell assembly and Tests

The cell: Case, zinc pocket electrode, MnO2 pocket electrode, electrode separator, lid
(slots for electrode connection tabs were cut into lid later)

   I designed the cell to hold 2 "+" electrodes and 3 "-" with 1/16" separators between. But for the first tests I just wanted one of each. What was the current capacity for one pair of faces? Would the amp-hours double (along with the amps) when a second zinc electrode was added? On the 18th I filled in much of the the empty space with pieces of plastic. I used the zinc pocket electrode with no powder in it, just the zinc sheet. Having added no calcium oxide yet, I wet a brush and painted it on the outside of the MnO2 electrode. If it was going to suffuse into the electrode, it would have to do so through the perforations. Then I mixed up 500 cc of water with 180 grams of potassium oxalate. It didn't all seem to dissolve (not quickly, anyway) so I added another 100 cc of water. Each time after it settled, there was still a K2C2O4 layer caked on the bottom. Admittedly the water was from the fridge, so the 36 grams per 100 cc maximum figure may not have applied until it warmed up. A hydrometer test showed 1.15 g/cc.

   I got the cell filled with electrolyte (~100 cc) and connected in the evening (18th). It initially read .699 volts and only put out .22 amps when shorted. Of course the zinc started as 'charged' metal, so the old MnO2 mix from the dry cells with the corroded zinc must have been pretty much discharged to Mn(OH)2. To charge that the zinc electrode would then have been bubbling H2. I didn't see any bubbles, but the level had gone down to where it was harder to see - and it looked like in spite of my precautions, the cell leaked.

With the power supply giving a 1.7 volt charge it initially drew 140 mA, which dropped to 80 by 1/2 an hour. This didn't seem like much current going into such a big battery. And it would only put out an amp for a moment when shorted, which quickly dropped to 3/4 amp. This seemed pretty disappointing from such a big cell. There were a number of possibilities. First, the separator plastic was 1/16" ABS. A piece of paper would get them closer together, and that seems to make a difference. Second, there were far fewer holes, spaced much farther apart, than in commercial pocket electrodes. That might well be limiting the rate at which the electrolyte could carry ions back and forth. I decided the second "+" electrode would have to have at least double the density of holes and see if that helped. The maximum instant short circuit current rose with charging, even to two amps, but whatever it was it dropped off to well under an amp in a second or two.

   I put a 2 ohm load on it and it delivered .25 amps for about 6 minutes. Then I found a couple of 1.5 ohm resistors (arg, WHERE have all my low ohms, 5 watt resistors gone? There's a big bag of them somewhere.) and changed to 3 ohms. It put out over .21 amps (.63 volts) for a further 10 minutes. After that I stopped it. In a few minutes it had recovered to 1.2 volts and then I resumed charging, which started at .21 amps but dropped to .12 in a few minutes and .07 a while later.
   Soon I put all four resistors in series for 5 ohms. It started at about 1 volt, but dropped fairly rapidly to .82 volts - 165 mA, and carried on a much slower descent from there. After 7 minutes it was down to .75 volts/150 mA and by ten minutes, .72/144 mA. It recovered to 1.23 volts and then I put on the 11 ohm load without a recharge. It started at 1 volt or so and after a couple of minutes was down to .9 V/82 mA; 5 minutes .876 V; 10 minutes .862; 15 m, .851 V; 20 m, .840 V/76 mA.

   These tests seemed to show the cell has some good storage capacity, but is unable to receive or deliver it at desirable rates of speed. Even at low rates it delivers with serious voltage drops. I'm thinking an insufficient number of holes in the sheet metal are drastically limiting the effective interface area over what it should be.

   I charged it overnight, but by morning too much electrolyte had leaked out and I disconnected everything. In the evening I fixed the leak (I hoped) and put it back together with fresh electrolyte. It started out at .84 volts, so it hadn't kept much charge. This time it started charging at 1/3 of an amp but it dropped to 80 mA pretty quickly. At the risk of potentially generating some permanganate I raised the charge voltage to 1.8 and current went up to 120 mA. Then I thought that maybe generating some permanganate for a limited period might actually be helpful: dissolved MnO4- could reform into MnO2 in contact with the electrode, but only in contact with it, so connection could be improved. Perhaps even better would be if I put some KMnO4 into the electrode mix when I made it. Something to try in the next electrode. For this one I raised the charge voltage to 2 volts for a while. It started charging at 200 mA. At least that was getting a little current into it!

   But it gradually became apparent that it still leaked! The case business seemed to be trickier than it appeared. Cutting the edge pieces had to be precision width and the ends absolutely square. I decided it was time to work on something else for a while. The first time I managed to get my thumbnail in at a seamor two . Naturally I figured that was where the leak(s) was, and I 'glued' liberally. On the 23rd there was nothing so obvious - it was much tighter everywhere. But I had a look with the 10x microscope lens. There were great gaping chasms everywhere around the edges, and plastic with bubbles in it. It looked awful. Leaks could be anywhere. But there was probably only one, somewhere.
   I put a small piece of ABS in a jar and dripped in some M.C. That softened it, so I could smear it into anything that looked like any sort of gap. I did have to put it in the jar again and after just a smear or two, and soon find another little piece, but soon I couldn't see how it could possibly leak. But I had become more suspicious and tried filling it with water and setting it on a dry paper towel. Any wetness should become evident, and it was soon there... a corner that looked fine was dripping. Again, that was probably from the bottom piece being just a hair wider than the edge piece, leaving a hair gap. And then aother one from the opposite corner. Hmm, wrong corner, or was it both corners on this side? Then I filled it to the brim. Another leak, 2/3 of the way up one seam! This one could only be blamed on rough cutting, or virtually no pressure to push the faces together. I refilled it and left it a while... and there was still a damp spot. Water was seeping out of a corner I had already "fixed". After that it seemed to hold.
   This case had more leaks than the US government! What a headache! I must get the sides absolutely precision, all pieces identical width and with absolutely square ends, and very smooth cuts. Then the sequence... First glue the bottom to one face, then the edges to that face and the bottom, all square and lined up. Put a heavy weight on it to hold it and leave it a while. Then glue the other face on and press it again. Hopefully such improved procedure should solve the problems. Otherwise I might have to get into plastic welding - ugh.
   23rd: The cell finally held water and I put the battery back together, sans lid, and on charge.
   24th: I could see occasional bubbles of hydrogen coming off the zinc as the MnO2 charged. But if the cell really had 25 AH of MnO2, and if it would only charge at 160 mA, it would take a week to go from "empty" to "full". And since the charge current at 1.8 volts would gradually drop off as it charged, it could be considerably longer. That might explain the slow changes over a day of charging. After about two days, the voltage started staying higher longer and the short circuit current started going up. There were momentary currents over 4 amps, but they quickly dropped off to 1/2 an amp.
   I tried sticking in an extra piece of plastic to squeeze the electrodes, and in particular to better press the MnO2 mix powder against the perforated metal, but that made it worse instead of better. More holes, more holes! I suspect twice as many would double the continual current and the charge rate, and four times would quadruple it, etc. Even more would be needed before the law of diminishing returns set in. The thing to do would be to make the second "+" electrode and put way more holes in it.

   I'll get back to it. I think it's a great size cell to produce multiples of. But for the moment I moved on to other things. Is there an easy and better way to do the perforations?

   Also I saw a friend with a table saw and we cut some new edge pieces from my 1/4" ABS. They were smoother than my cuts, and the pieces were all precisely the same width as close as I could tell. I have some hopes for the next case to be watertight on the first try.

Better, Faster Perforating: Ideas for Rollers with "Nails"

   Even after at last having something that more or less worked, I kept thinking about how to perforate sheets of metal better - to make more holes better and faster. Making such big plates for large cells, perforating each one was pretty tedious, and the more the hole density, the more laborious. I kept coming back to the rollers idea; one roller with short "nails" sticking out and the other with matching holes for them to punch into. But everything would have to perfectly align, with the rollers geared together.

   Finally I had a bit of an idea. Instead of holes, the other roller could have "slits" going around for the nails to pass into. Then they only had to line up side to side - much simpler. One roller could be driven and the other would simply follow it, and if it slipped a bit it wouldn't matter.
   Another thought would be to have the piece slide across a single roller of nails with a piece of soft wood sliding across it. The nails would then simply go through into the wood.

   For a second idea, I could use the milling machine and the rotary table to mount the roller to be perforated, in order to line up the holes to insert the "nails" into, both axially and radially. (I'm assuming a tiny drill bit can be coaxed to go in straight and not break if I get it right! There's probably some way to ensure the drill depth is consistent, too.)
   For a third idea, the roller could do 1/4 density. The piece could be fed through twice with the holes punching in different places. Then too, the holes could set be slightly off center so that if the 3.0" wide piece was turned around, the rows of holes would align in between the first rows.

   At first I keep thinking too that the easiest thing for me would be to make the rollers to fit my jeweler's rolling mill. At the cost of making custom rollers that wouldn't fit anything else, that would give them a solid mounting, a thickness/spacing adjustment and a crank handle.
   OTOH rollers could simply be made from 1" or larger shafts, and then they'd fit on 1" needle bearings or bronze bushings. Then one could either attach a crank handle or a much speed reduced motor. They'd need solid steel sides to hold the bearings and keep the rollers at the right distance apart.
   I wonder if the nails could go into an aluminum cylinder instead of steel. That should be a heck of a lot easier to drill the array of holes into than hard axle steel. I might even say "possible" versus "it'll never work!" I found an aluminum cylinder in my scraps, about 1.5" O.D. and (?)24 mm I.D., and 4" long. That might be just about ideal, other than having to turn a 1" axle down to the I.D. to fit it. Or maybe I could use a boring bar and increase the I.D. to one inch?

Lead-Zinc Cell From Scratch With Rolled-up Sheet Metal Electrodes

   On the 24th my 1' x 2' x 1/32" sheet of lead arrived in the mail. It was heavy, 4 # @ 2 #/sq.ft. The material was 15 $US and the shipping (IIRC) was 25. I'd have got more for a volume discount, but the shipping just went up rapidly.

   Someone on a list had mentioned the idea of just rolling up two strips of metal. It was for off-grid storage and they didn't care about weight and bulk. In 1865 Gaston Planté did just that to invent the lead-acid battery. He rolled up two strips of lead with separators and filled the container with sulfuric acid. The lead-zinc cell should be able to be done the same way, of course with one zinc strip and the oxalate/lime electrolyte. And of course this is what I ordered the lead strip, just to try it out. I decided to cut 2 foot strips of lead and zinc, about 2-5/8" tall (the width of thinner roll of zinc).
   Perhaps the cleverest thing I did was to start cutting at 1" instead of at the end, and then to cut a tab on the far end. Thus the first two rolled electrodes (if there would be another one) would each be one piece with a terminal tab that was part of the lead sheet, without wasting any of the sheet. No corroding tin solder in this one! Then I rolled it in the jewellers rolling mill both to stretch it and to put a pattern in the surface, both to give it more surface area for its volume. It ended up expanding from 23" to 30.75". The thickness went down from .9 mm to .6 -.7 mm. So the dimensions were about 69 mm x 781 mm x .6 mm, and it weighed 435 grams. (Wow, almost 1/2 a kilogram. Of course everything depends on what percentage of theoretical it will actually use. It won't be 100 amp hours, but if it uses 10% it might be around 10 amp hours.) The interface area, counting both sides of both sheets, is 1078 sq.cm. Even at just 10 mA/sq.cm that should be 11 amps, and it's more likely to be 50 amps momentary. (If it gets over 20 I'll have to use a clamp-on ampmeter rather than a DVOM!)

   I put the zinc in the mill without cutting it off the roll and textured/stretched it. Then I cut it at 31". I couldn't do the tab the same way since there was no more width to cut from, but I cut the end 3/4 of the way and then folded the tab piece up. It seemed to be shorter than the lead, so I ran it through the rolling mill a couple more times, which helped a bit but seemed more just to put kinks in it.

The remaining original sheet of lead (Pb) with the
piece I cut off and stretched in the rolling mill

The two stretched sheets, zinc and lead, to be rolled up together
with separators between as a 'rolled electrodes' Pb-Zn cell.
These will go in a soup can or a plastic pipe section.

   Easy as it should have been to finish it, I turned to other neglected projects for the rest of July.

Haida Gwaii, BC Canada