Turquoise Energy Ltd. News #114
covering November 2017 (posted December 2nd 2017)
Lawnhill BC Canada
by Craig Carmichael

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

Month In Brief (Project Summaries etc.)
  - Format - Feedback is welcome - Bandsaw Alaska Mill - Hybridizing the Toyota Echo - Alaska Milling - (note: Palaeos.com URL)

In Passing (Miscellaneous topics, editorial comments & opinionated rants)
 - Arthritis Cure: Borax. Still Working. - "Foambergs" (More Sea Foam from aerosol spraying?)

- Project Reports -

Electric Transport - Electric Hubcap Motor Systems
* An Off-the-Shelf Centrifugal Variable Pulley?
* Back to a centrifugal clutch? - some design considerations
* Double Set Screws: Make sure it won't come loose

Other "Green" Electric Equipment Projects
* Bandsaw Alaska Mill: Wheels and bands - making plywood wheels

Electricity Generation
* An Attic Windplant? (just an idea)

Electricity Storage - Turquoise Battery Project (NiMn, NiNi, O2-Ni), etc. (no reports)

November in Brief

Format Change

   As I'm not e-mailing the newsletter itself now but only links to it, I'm going to change the format a little. Month in Brief will have just very brief descriptions of the projects. The "detailed reports" will have the meat for those interested.


   Occasionally people write me with questions, ideas or suggestions about my projects written up in these newsletters. Perhaps others are wondering about something, or thinking I'm off base and "Yah, good luck with that!" I wish to say that I appreciate the efforts people take to write me, and that whether I use an idea or not or whether I agree or not or take advise or not, I welcome ideas and constructive criticism. And I love to chat with those who may reach this little corner of the world and visit and perhaps look things over. I make all my work public and public domain via these newsletters, and more thoughts by more people can improve the focus and perhaps together we may come up with better things than I sometimes do by going off on a tangent on my own.

Bandsaw Alaska Mill (how about "Carmichael Mill"?)

   Near the start of the month, the new "bandsaw alaska mill" idea went from "wish list" to the top of the list. It shouldn't be too long before I have a working saw, and it has excellent commercial possibilities for myself, which most of my ideas don't. So I ordered parts, and by the end of the month they had arrived and I at least had a start on making the saw. The first one is to be electric. If it works as well as I hope, there may be more models coming.

The main frame for the "bandsaw alaska mill", with 10" x 1.5" plywood wheels
and a 93" x 3/4" cutting band for up to 20" wide, 6" deep cuts.

Hybridizing the Toyota Echo

   I continued on the plan for hybridizing the Toyota Echo, switching the "how to get it moving" technique (of 3 potential techniques mentioned last month) to a large centrifugal clutch that I made almost 2 years ago (replacing variable pulley ideas, and a "too small" commercial centrifugal clutch). I also changed it from a (probably inadequate) belt drive to a chain drive from the motor to the clutch. Actual work in November was limited to ordering various needed parts as the plan changed and matured. Most of them were received by month's end except for a 28 tooth #40 chain sprocket for the motor.

The mid-shaft components sort of assembled.
This goes under the trunk and behind the right rear wheel.
Left is the drive sprocket from the motor in the trunk, through the
centrifugal clutch to the drive sprocket going to the wheel, on the right.

Alaska Milling

   My other perhaps notable activity for the month was milling up a huge spruce log into "cants" (long pieces of log much larger than dimensional lumber, eg 6" x 20"). I had started this in October, but I didn't finish dicing up the whole log until late November. Of course, this work is what led to the idea for the "Bandsaw Alaska Mill". Some of the cants are still virtually too heavy to move, and without a big forklift or something it would still be a major operation to get them out of my driveway. When I finish the bandsaw I hope to cut them into pieces of lumber and carry them off.

The remainder of the big spruce log, here with four very wide, 16 foot cuts to go to dice the last section up.
For various reasons they took over a week - including wanting help to move pieces once they were cut.
At the very least, a bandsaw would have left the bottom piece over an inch thicker by turning less wood into sawdust.
At best, it would also have made the job faster, easier and more pleasant, and left smoother surfaces.

Note: In TE News #100 for my Permian article ("all amphibians - no reptiles yet"), I referenced a website I labeled in two places as Paleos.com. Looking for it again I was for a while distressed - I couldn't find it. Had such a fine reference site simply disappeared from the web? Then I tried a web search for one of the sentences I quoted. The actual URL is Palaeos.com (which I had labeled it as in two other places). My apologies to anyone who was looking for that site and used the wrong URL.

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

Arthritis Cure: Borax. Still Working.

   In TE News #88 I mentioned having found out that arthritis and calcium buildup in the joints (not to mention osteoporosis) is caused by boron deficiency, and that little aches and pains in the joints went away when I started taking borax. Boron deficiency inhibits proper metabolism of calcium, which is why it ends up in the joints and gets depleted from the bone. Well, in case anyone thought that sounded like a silly fad that would soon be abandoned, I will say that it's still working after 3 years and that I'm still glad I found out about it!
   I seem to have gravitated to a recipe of a teaspoon of borax in a half litre (half quart) of water, and drinking a teaspoon a day, or for convenience two teaspoons every second day. One teaspoon per litre, one teaspoon a day, seemed to be not quite enough, for me. Other similar recipes and recommendations may be found on the web.
   (And don't forget to get some calcium and vitamin D to go with that. 1000mg of vitamin D per day (or 15-20 minutes of sunshine on the skin) cuts your risk of cancer literally in half. In fact, as I get older I find myself taking about 6 supplements a day. That way I expect to be healthier longer and I won't end up needing prescription medications with side effects. "An ounce of prevention is better than a pound of cure", as they say.)

"Foambergs" (More sea foam from aerosol spraying?)

In some places they have icebergs.
In Hecate Strait and on the east coast of Haida Gwaii this winter we seem to have "foambergs", bobbing around in
the ocean, washing up on the shores and blowing around in the wind - even blowing across the highway in storms.

   I originally included this picture just because seemed interesting. I hadn't meant to bring up "chem spraying" again, but as I think about it, this seems to fit into a bigger picture. I can't say I remember seeing anything like it in the Pacific in decades past.
   However, in Victoria a couple of years ago it all seemed to be related to aerosol spraying, with foam and colored water appearing
even in my fish pool and rainwater barrels, while headlines noted mass die-offs of sea birds in the ocean: foam soaked birds had lost their waterproofing and got hypothermia, washing up on mysteriously foam covered beaches from California to Washington. It was thought at the time there must have been some sort of big unreported spill... but in my rain barrels too? I mentioned it in TE News #88. Mass die-offs continue daily, worldwide [TE News #101]. They only make local news now.

   Among other mentions of "chem spraying" in various issues, I did a bigger write-up of how and why aerosol spraying
appears to be altering the wind circulation patterns and hence the whole climate of the planet in TE News #109.

   The rest of the newsletter is "in depth reports" for each project. I hope these will help anyone who wants to get into a simliar project to glean ideas for how something might be done, as well as things that might have been tried or thought of... and often, 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 something of a diary and are not as 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

Wheel Drive Motor: Hybridize the 2001 Toyota Echo
* The Simplest Plan Ever *

An Off-the-Shelf Centrifugal Variable Pulley?

   After deciding the best setup would be one with a centrifugal variable pulley on the motor, I went to look for the one I had off of a mo-ped. When I found it, it seemed to be missing a couple of pieces - or perhaps just the shaft was attached to the mo-ped and I couldn't extract it. I looked back in TE News to when I had obtained the variable pulleys from Victoria Motorcycle, but I found only the output pulley, and a couple of pulleys attached to centrifugal clutches (TE News #103). Anyway there was enough to see how it worked. It didn't look as I remembered it - perhaps I was remembering a different one that I was shown at the time.
   Six small and not very heavy cylinders were the centrifugal weights. (My memory was of six ball bearings.) When the shaft was spinning these cylinders pushed out against a flat angled piece on one side and a curved angled piece on the other. The flat angled piece would have been fixed on the shaft, so the weights would have forced one side of the pulley (the other, outer, face of the curved angled piece), towards the other side of the pulley, increasing the effective diameter. It seemed like a good way to do it, but it seemed to me to be too small and light for a car. The thing to do would be to make something similar but larger and heavier, perhaps with an effective pulley diameter of 2.5" to 5". This would drive a 5" fixed pulley, yielding a reduction of between 2 to 1 and 1 to 1. The 6 to 1 chain reduction would then make the total reduction between 12 to 1 and 6 to 1, allowing sufficient torque for getting the car moving and lower RPM for higher speed driving.

   Then I went on line and found there were companies making larger variable pulleys, up to 100 HP rated. Now we're talking! One of them, "www.SpeedSelector.com", had a PDF of calculations for using variable pulley systems. It noted that there were two types, the "fixed center distance" with inversely matching variable pulleys, and the "adjustable center distance" where there was just one just as I've been proposing, and the distance between shafts changes to alter the effective pulley diameter. (I should have known it already existed!) If there was a centrifugal type, that would be the way to go, but they weren't mentioned in the document. The document also stated that the pulleys were aluminum, which carries off the heat from the belts. That's good info. But applying the formulas and figures given was a bit beyond me. I don't suppose there are a lot of models to choose from in the desired range.
   They claimed to have a complete line of pulleys and belts. I also noticed the flat-one-side variable pulley belt I bought said "www.HoffcoComet.com" on it, so it should be possible to order belts, probably just about whatever you want, from there too. Centrifugal variable pulleys are more in doubt.

   On the 6th I went back to SpeedSelector.com . They didn't have any centrifugal variable pulleys, only spring loaded ones. Their whole focus seemed to be different - industrial fans and fixed equipment.
   Then I went to HoffcoComet.com, and found they had fixed axle variable transmissions not only for 8 horsepower, but also up to even 150 horsepower. They definitely had transport applications in mind, and were an "aftermarket" or "OEM" supplier. The drive ratio ranges were better than I needed, even about 4-1/2 to 1 with a single variable pulley. I started drooling over the 40 HP model, with 7 inch pulleys and a big, fat belt. That sounded like the right sort of size, and they could recommend a bigger one if they thought it was too small. I looked for the phone number.  That would surely solve all the problems, however many hundreds of dollars it cost! Let's see, contact info... No phone, no e-mail, no "contact us" page. What? I did some more searching and found that Hoffco-Comet had gone out of business in 2009. The site I was looking at was just a legacy site, and the components Princess Auto had been selling were just leftovers! I must say I was rather crushed.

   From the HoffcoComet.com web site:

Model 94C Duster Clutch Series

This is the one to use for all V-twins and high torque applications! With a beefy 1-3/16" belt, the 94C has the rugged reliability needed when you want to spend your time riding, not making repairs. A 3.49-1 low and a .78-1 high provides a wide gear range for great low end pulling power without sacrificing top speed! Custom calibration of drive unit is quick and easy allowing you to get maximum performance out of your engine. Contact Comet Industries for recommendations.

Comet 94C Series Torque Converter
Diameter: 7-1/4" Simplicity Snowmobile
2 Cycle & 4 Cycle Fewer Moving Parts Odyssey
Housing: Stamped / Cast Easy Calibration Industrial Equipment
Bore: 30mm 1:10 Tapered 1" & 1-1/8" Performance Light Utility
HP: up to 40 Limited Maintenance
Engagement: 1600 - 4600 RPM

   Not available!?! Curses! Could I even find a belt like that one, anywhere?

   Just for reference here are pictures of the small (moped) variable pulley that I obtained as scrap.

When the pulley spins faster, the cylindrical weights push outward.
This forces the cast part of the pulley toward the rusty steel fixed piece,
narrowing the pulley and forcing the belt out to a greater diameter.

(I'm sure the grooves in the belt improve the efficiency considerably over a regular V-belt,
allowing it to bend around the pulleys more easily. Why don't they make them all that way?)

A Centrifugal Variable Flat Belt Pulley?

   To continue with the variable pulley theme, on the evening of the 29th I thought again about my idea for a flat belt variable pulley (TE News #101, The "Double Barrel Torque Converter?") I still like the 'almost lossless' efficiency aspect to flat belt operation, if the mechanism can handle the torque without the belt slipping. The part I was having conceptual trouble with was how the input "barrel"/flat belt pulley could be made centrifugal. It seemed like it would need a hopelessly complex mechanism. Now I see the weight of the "barrel staves" themselves as being the centrifugal weights for the pulley, and it suddenly seems more feasible.
   And instead of having a reciprocal variable pulley on the output side (complicated), I would go for a fixed diameter output pulley with varying center distance: a spring pushing the two pulleys apart to tension the belt and the centrifugal force pulling them together at higher speeds.

   I think such a mechanism would run pretty smoothly, but it just might be that the staves with spaces between them would set up a lot of vibration as the unit spins. One might make them to "mesh" between vanes like escalator stairs do. That idea might be employed to "interlock" them to solve another problem too: how to make it so the "petals" would more or less open and close together. Otherwise, the ones where there's no belt will fling wide open even at low RPMs, while the other side with the most pressure on the belt will open least, making an unsymmetrical pulley whose symmetry changes as it rotates. That might be the toughest problem. Or would it really be a problem? The design perhaps isn't far enough along to think of building one yet! And if the centrifugal clutch really works well, probably I would never bother with variable pulleys again.

Initial Idea. However, since the belt is directly on the weights, the weights on the right would fling open while others stay closed.

So a better idea would be to have them all interlock somehow.
This needs more thought.

Back to a centrifugal clutch (pictures follow text)

   Having been considering the difficulties of making a really good centrifugal pulley myself, and with this "disappearing company" disappointment, I decided to go back to the centrifugal clutch as the first method to try. This opened up an unexpected problem. I couldn't find it! Obviously I had put it somewhere after I had decided not to use it, but I couldn't recall where and on the 7th and 8th I was utterly unable to locate it again. I did however find two large centrifugal clutches I had made some time ago, and on the evening of the 8th I started thinking about them.
   I had been mildly worried about the strength and gripping power of the small diameter, rather narrow commercial clutch, and about the heat it might generate at the high RPM it would operate at, especially in stop and go driving. And if it dropped out too near the activation RPM, the flywheel wouldn't have the chance to impart its momentum to the car to start it moving. I thought about this some more and started thinking that my larger ones were probably a more suitable size for a car. Also I could adjust their operating parameters if necessary with various springs and even different weights. They were already made, and the better one might actually be the best option. They hadn't been successful as I used them when I made them, but the RPM was rather low and the load quite high with just 3 to 1 reduction to the wheels. Here the reduction after the clutch would be at least double that. And the RPM could be made still higher and the torque loading less by adjusting pulley and drive sprocket sizes.
   This time, the technique of revving up a flywheel to store up energy to start the car moving could be used, and I would configure the clutch so that once activated it wouldn't drop out until a much lower RPM. As I see it, what is needed is springs that take a fair force - a considerable RPM - to "break away" and allow the shoes to fly outward, then with the shoes out, have only a weak force trying to bring them back in. Once the weight of the shoes has been flung outward, the centrifugal force on them is greater, and (if the springs can be configured as desired) the spring tension is weaker, so that only a considerable slowing down of the drive will cause the shoes to retract again. With the car in motion, they should stay engaged. As they engage I expect there will be a considerable amount of slipping, and here the choice of slick UHMW [ultra-high molecular weight polyethylene] for the shoes and aluminum for the drum to carry heat away, will prove to work better than most possible materials or combinations.

   I eventually found the commercially made centrifugal clutch in a box I had recently put all my spare V-belt pulleys in and put away in a storage room.

Some detailed design considerations (11th)

   I decided the car wheel, even as it bounced up and down and even if it went forward and back a tiny bit, would have to be responsible for keeping itself pointed straight forward and back, and for not tilting vertically. That would mean the bar from the wheel to the mid shaft need only maintain the distance between sprockets, not shift the mid shaft's position and angle to keep the drive sprocket in alignment with the wheel sprocket in all directions. That allows for a light bar instead of a very stiff and heavy one.
   Then all the sprocket on the mid shaft has to be is aligned with the wheel, and thus in line with the wheel sprocket which is presumed to stay straight itself. I noticed on my large chainsaw that the drive sprocket was free to move in and out a bit on a splined shaft, to line up with the bar and chain. But on a car the drive sprocket is much heavier, and it would tend to get pushed from left to right a bit with the suspension and going around curves. It seemed to me it should be fixed on the shaft, carefully lined up with the wheel sprocket.
   It was tempting to mount the centrifugal clutch on the motor so its output would drive the V-belt instead of the chain. But then what about the flywheel? The input side of the clutch isn't heavy enough by itself. The more and heavier the pulleys between the motor and the clutch, the more of the desired flywheel action they would provide.
   So the clutch would have to go on the mid shaft. And then it would seem that in order that to hold the drive sprocket in excellent alignment with the wheel sprocket, there would have to be two bearings on the output shaft. And then two on the input shaft to keep the pulley in line. That was starting to sound like trouble.
   It became clear that a single shaft, an "inner" shaft, should run right through from left to right, on two bearings. A shorter outer "pipe" shaft on bushings or bearings would be free to rotate against it until the clutch engaged. This after all is how the commercial one was made. I was hoping to avoid reworking my clutch and complicating the design, but it looked like the thing to do. Then, one could toss a dice, but at first it looked easier to have the pulley and inner side of the clutch on the bushings, and the output drum and sprocket clamped onto the inner shaft.
   Then I thought that if I could bolt or weld the side of the drum to the chain sprocket, I could put a bronze bushing, 7/8" ID and 1.0" OD inside. Then I would turn the shaft down to 7/8" on that side to slide against the bushing, and use a 7/8" bearing for that end of the shaft. It seemed I had one such bushing, purchased at Princess Auto before I moved. I really needed two to get the length. (Two come in a package... where did the other one go?) If need be, I could turn down a 7/8" ID x 1-1/8" OD bushing to 1.0" OD. I had two of those. Sigh!
   Then on the 14th I went to the automotive store and ordered two suitable bearing holders. On the 17th I went to where there was internet and found Princess Auto on line, and ordered more bronze bushings, and a 1.125" tapered locking hub for the motor to hold a pulley. (along with some things for the bandsaw alaska mill)

   On the 19th I met with someone else interested in converting cars and in electric powered boats. He ran rock crushing machines and had much mechanical experience. He thought that for my car design, with 8 inch V-belt pulleys, at least triple pulleys running 3 V-belts would be a good number. Ouch! Having heard that, my sense of proportion now says he's probably right, so I'd better leave extra room to allow for more belts. But it's the motor with its short shaft, and mounting bolts sticking out the same face so you can't connect in close either, that concerns me. Will it really handle a triple pulley, or will it work loose and fall off? Such a powerful, high speed motor, and so little to attach anything to! Perhaps 12" cast double pulleys might be good - a little shorter, and they would also be good flywheels. But the pulley's mounting hub sticks out well past the end of the motor shaft, and it's worse if you allow for even a thin mounting plate (or bars), bent back around the edges to clear the pulley rim - and it ought to be a good thick flat plate to hold the heavy motor. To stay out of the way, inside the mounting holes, a 6" pulley would be the maximum. Maybe I need to use a chain from the motor to the mid shaft? I guess in theory that should work just as well as a belt(s) except for needing lubrication.
   Then I looked at a 10" pulley with a flat ~1/8" plate body. It attaches to the same weld-on hub as the chain sprockets, and gives more clearance for the motor mountings. Not much weight as a flywheel, but it looked like it just might fit. The mid-shaft pulley could be the heavy one. If 10" was too small, 12" was doubtless available, and hopefully at that size, one V-belt would be enough. (Ya, 2 would be better.) I didn't have the right hub (for 1-1/8" shaft), but I had ordered one "just in case" in my 2nd Princess Auto order.

   I also note that the little commercial centrifugal clutch would look pretty scrawny beside a large 2 or 3 belt pulley. Ditto with the small variable pulley. So I think using the larger centrifugal clutch I made last year is the right decision.
   This friend also mentioned "A" and "C" series V-belts, the larger letters having wider belts. Perhaps I should be after the "specialty" larger sizes?
   I had also heard he knew where to get a centrifugal variable pulley, but when we met he pulled out a 2013 Princess Auto catalog with the same Hoffco-Comet units that I had found on line that are no longer available.

   On the 24th two bearing holders and a 7/8" needle bearing I had ordered arrived, along with some more 7/8" x 1" bronze bushings. Those should do for the part of the centrifugal clutch that doesn't turn with the shaft. (I already had a 1" needle bearing for the other end.)

   The more I checked it out and thought about it, the more a chain seemed to be a better fit than a belt. By picking relatively small sprockets, like 28 teeth, that includes to fit it physically onto the motor shaft still with room for the motor mountings. And it would most likely be more efficient - have less friction - than 2 or 3 V-belts. Why was I using a V-belt(s), again? Let's see... it was the only option for a variable pulley system. Likewise, a slipping belt clutch could hardly be done with a chain. With the centrifugal clutch, it's not needed. On the 26th I placed yet another order: another 28 tooth #40 chain sprocket and 1-1/8" hub, and some more #40 roller chain. (An incentive was a 'sale' ending that day by Princess Auto on 64 drawer storage drawers for small parts, which I need for my electronic components, which are becoming increasingly disorganized as the variety of them increases.)

   I may not be getting any assembly done as I work on other things, but the plan is improving and I'm stealthily gathering the parts I'll need when I do get at it.

Three views of the midshaft components:
* Mounting piece (attaches to car... somewhere, somehow)
* inner bearing holder & bearing,
* inner chain sprocket, from motor
* centrifugal clutch: inner part and outer drum
* outer sprocket, to wheel
* outer bearing
* outer mounting piece.

Either the inner sprocket and the center of the clutch will need to be on bronze bushings (bottom
right) to turn together and independently of the shaft, or the drum and outer sprocket will.

But the drum needs to be back a way from the outer sprocket,
in order to be behind the lower body panel and under the car trunk.
(It'll need a longer shaft for that, and still longer to accommodate a flywheel.)

Double Set Screws: Make sure it won't come loose

   (5th.) I had a hard time getting the coupling for the Suzuki Swift off the motor shaft. I loosened the two set screws, but I had to wind it off with a bearing puller with considerable force. Once I had it off, I could see I hadn't loosened the set screws enough. Huh? When I looked into that, I discovered the real problem: there were two set screws in each hole, one screwing onto the other as an additional clamp to hold it in place. Thus I had only loosened the outer set screw and left the inner one still dug in, one into the shaft and one into the shaft key. The key slid out with the coupling. The other went into a shallow hole drilled into the shaft. My pulling it off seems to have turned it into something of a slot. I had anticipated such a hole, but of course no matter how far I unscrewed the outer set screw, it hadn't helped at all.
   I guess they didn't want the coupling, transferring all the power of the motor to the transmission, to come loose on the shaft. Both the hole and the double set screws seem like very good "tricks of the trade", as they say, to use when doing similar shaft mountings. Of course disassembly is easier when you know or figure out how it was assembled.

Other "Green" Electric Equipment Projects

Plan for a Portable Bandsaw "Alaska Mill" for cutting lumber?
(Continued from #113)

   Reconsidering on the 4th, I started thinking that the project wasn't a difficult one, and that there should be a good market for such a saw if it worked well, with a more realistic potential profit margin than with most of my product endeavors. Instead of "someday, wish list" I decided to move it to the top of the project priority list. Making a good living for a change sounds appealing.

   I had been thinking of using 14" bandsaw wheels like my stationary bandsaw uses. On reflection, I'd like to make it as light as possible, and cutting with the chainsaw with a maximum 26" cutting width (30" bar) has convinced me wider is better. I was cutting off branch stubs and even into the wood with the electric chainsaw so the mill could get past the wider parts, and I wasn't even into the widest base section of the log. Cutting wide takes more time, but having to stop and cut out chunks that make the log too wide to cut takes even longer. I'll have to plan my cuts carefully for the base section, or use an even longer bar and chain. (I found a 36" bar with chain at a garage sale, but the chain is pretty worn out and the bar needs straightening. ...When I finally tackled the fattest section I had to make quite a few cutouts before even the 36" bar could squeeze in, doing a maximum 30" wide cut.)
   12" wheels would be lighter and the same overall width of saw will have 2" more cutting width in the middle. As long as the band will readily make the tighter bend, I'm for it. It does however limit the depth of cut, probably to about 4-1/2 to 5". I don't think I'd want to limit the depth to much less than that, so 9" wheels (the smallest common size) are out. Flexing a 3/4" band by hand, I didn't think I'd want to run it on much less than 12" wheels anyway.
   Notwithstanding the desire for more width, I may size my first model to use the same 105" bands as my regular bandsaw, which cuts almost 12" thick max. With the smaller wheels, that's likely to give about a 20" maximum cut width. That's probably also more suited to the small skill saw motor I plan to employ to power it. Also, while there's a good bit of timber on my acreage, the four spruce trees already cut down by the house were the largest, so the need for absolute widest cutting width decreases after these are done.

   I had this vision of finding some scrap bandsaw and taking all the useful parts off it - wheels and shafts, blocks, pulley and so on. Realisticly that was pretty unlikely to happen. I could order wheels off e-bay - ugh. Some people were showing making wooden bandsaw wheels and parts - even a whole wooden bandsaw mill - on Youtube. I thought I'd go with making plywood wheels if I didn't find any others by the time I was ready to start. I didn't, and the guy at the landfill said he'd never seen one show up there. (Perhaps I'll spray "flex seal" from the spray can around the outside, and not bother with tires.)

   The design continued to take form in my mind that night, and the plot thickened. Or at least, the potential cutting depth did. If the axles of the wheels were mounted under the cross bars frame instead of centered on them, the frame would be above center and the available cutting depth would be thickened with the same size wheels. Not a "C" frame but perhaps a "[" frame. For 12" wheels, perhaps 7" would be available, or it could be made deeper yet with more offset. The slotted bars sold for Alaska mills seemed ideal for the main frame of the bandsaw.

"Prior Art": Tales of a previous bandsaw alaska mill

   A neighbor said a bandsaw Alaska mill had been tried on Haida Gwaii about 20 years ago. Evidently it wasn't a hit, but it sounded to me like the problems were in the implementation, not in the idea. (Rather like some of my car transmission projects.)
   The engine apparently compared unfavorably to a chainsaw engine, so it sounded as if something like perhaps a lawnmower or rototiller engine was used. That would be heavier and more cumbersome. I plan to use an electric motor for the prototype. For a gas model, an obvious choice would be to use a chainsaw engine: small, light and flexible. Being the very same engine it would be impossible to blame it in principle for any shortcomings versus a chainsaw mill engine.
   A second objection was that one was taking one's chainsaw into the bush anyway, whereas the bandsaw was an "extra" tool having no purpose other than milling. But this is surely a matter of application. I don't really want to take an electric mill into the bush anyway. But I do want to cut my big blocks of wood into dimensional lumber without wasting so much as sawdust. And the chainsaw mill certainly burns a lot of gasoline. I want to do it much quieter and cleanly. But except for the smallest of jobs, I don't think a gas bandsaw mill in the bush would be a waste of capital, vehicle space or time - if it works better than the chainsaw mill for even some of the cutting, it's worth having as a labor saver.
   The other problem noted was that it didn't seem to cut bark well. Evidently the wood being cut was western red cedar. It's soft enough wood, but its bark rips off in long stringy strips. It was probably getting caught in the wheels and jamming things up. There are a couple of solutions to this. The first is to put in a "catcher" or "deflector" of some sort by the exit band guide blocks to deflect the bark and sawdust out of the saw. The other is to not cut red cedar. (Awful stuff to cut anyway, with long splintery sawdust that sticks in clothes and is carcinogenic.) I forsee cutting spruce and probably alder. The deflector is probably still a good idea. Perhaps the exit guide blocks could be shaped for that dual purpose.

Back to the design

   Finding a couple of wheels, it appeared that 11" wheels would give a 105" band a cut up to about 22" to 24" wide, on a frame under 36" long to hold the wheels and motor. 105" x 3/4" is also about the largest band one will readily find prepackaged in a store, and is the size my bandsaw uses. That seemed like a good size to start with if I could find or make such wheels. (I'd much rather find them than make them, but I didn't come up with anything.)

   On the 10th I decided to shrink the pulleys a bit more yet to 10". I don't think that's beyond the "radius of elasticity" for typical 3/4" cutting bands, and the smaller the wheels, the smaller the saw can be. And with the axles now being below the frame "backbone" instead of in-line, a 6" cut depth should still be possible. But a special motivation for 10" wheels was that's the largest size I could turn on my machine lathe if I had to make them, as it appeared I would.
   10" wheels would give the 105" band a 27" cutting width, but the axles would be over 36" apart, which would make construction with the 36" bars tricky. The next band size down I think is 93". I have one because it's what my bandsaw took before I added a 6" riser block so I could cut guitar backs, up to about 11" thick. (93"+6"+6"=105"). It appeared the 93" band would reduce the cut by (surprise, suprise) 6", to 21". But then I could easily mount the wheels shafts on "pillow block" bearings which a store here could order for me, without their mounting holes exceeding the 36" frame length. Of course, frames longer than 36" can be had, and with a little silver soldering a band of any desired length can be made. But perhaps for a first attempt - with a 120 V motor of just 2 HP or so - I should be happy with a maximum 20" wide, 6" deep cut (or thereabouts) in a 36" frame with an off-the-shelf band, and consider that it can still be very valuable for cutting smaller logs and for dicing up the cants from larger ones into lumber, rather than to deal with splitting the largest logs itself. If it works well enough, it could potentially replace my 7.5 HP electric sawmill on tracks, and it could undoubtedly be made for a lower cost than any bandsaw sawmill on a track, which opens up manufacturing possibilities.

   Later I could try upping the power and dimensions with a gas model using a chainsaw engine. I visualize attaching the chainsaw motor driving a V-belt pulley instead of a bar and chain, and using its two bar bolts and nuts to attach it to the bandsaw frame. I could see a saw like that that ripping through big logs like a demon! A potential advantage from a manufacturing point of view is that it could be sold for lower cost without an engine, leaving the buyer to attach his own chainsaw power. I hope the trigger isn't too awkward - it'll be in an even worse orientation than on a chainsaw Alaska mill. But people manage those okay. Then again, if it was placed at the right hand side instead of the left, it would probably be quite okay. The operator would then walk down the right side of the log instead of the left.

Note: In checking out wheel sizes and band lengths, I noticed that the teeth on the band were pointed the wrong way for the envisioned cutting direction. I twisted the band around and found it wasn't too difficult to turn it inside out, which reversed the teeth. Problem? No problem!

Geering Up and Starting to Build

   I bought a pair of 36" extruded metal Alaska mill rails, which happened to look like a very good choice to be the bandsaw's main backbone, on the 7th. From a local automotive store I ordered four "pillow block" bearings for 3/4" shafts on the 14th, and on the 17th from Princess Auto, I ordered "H" taper-lock hubs. (On the way home after placing the order, I remembered I had no 3/4" shaft and should have ordered some, and that I should have also got a couple of 7/8" and 1-1/8" shaft weld-on hubs in case I might need them for the car project. Shopping is definitely easier living in a major center, and with real internet access at home.)
   That day I also dug out a piece of 3/4" birch plywood and cut out four 10" discs [on the bandsaw, of course!] and glued them in pairs to make two fatter discs (1.5") to use as bandsaw wheels with the taper-lock hubs. I had the thought that for production, molded PP-epoxy wheels could be very, very light, and strong. (That would mean getting the new CNC router working to make the molds. Should I really have traded the old, working CNC router for a new model that now needs to be set up, both in hardware and software? But the old heavy one took up a lot of space, and it was a lot of weight that I didn't have to move up to my new home.)

   The next day (18th) it was raining and almost freezing, so I decided to work on making the bandsaw mill wheels and pass on trying to roll the big log 1/4 of a turn and then Alaska mill another big slab off it. I sanded the outside rims as smooth and round as I could get them by eye on my stationary [belt &] disk sander.
   I didn't have the 3/4" shaft "H" hubs yet, but the 1" ones were the same except for the center hole, so I could use one to get the wheels done. I marked and drilled out the center holes: a tapered "H" hub center hole and 4 mounting bolt holes around it. I wondered how on Earth I was going to mount the wheels on the lathe to shape the center holes, to 1.6" at one end tapering to 1.5" at the other. But the holey gods were kind that day. I drilled the center holes with a 1.5" flat blade drill bit in a hand drill, as my drill press won't take a 5" wide item. What with boring through the thick pieces and angling the drill back and forth a bit to get it to cut, the hole proved to be already almost the ideal size and taper, and the hub inserted into it without doing anything more. (A few file strokes could get it fitting in closer if desired.) Then I looked in the lathe paraphernalia box and found a holder I had made for making small gears. One end had a #2 Morse taper to fit into the lathe spindle and the other end was a 1" shaft - perfect to hold the 1" "H" hub! Everything was so close to maximum possible size that I had to assemble it all on the lathe, and disassemble it in order to remove it, but it just fit - the 1.5" thick, 10" diameter wheels just fit on, both ways! I contrived to use the tool holder with a steel milling tool as a rest for the wood chisels, and that too was just barely able to reach the right position in front of the wheel rim. On the lathe I trued up each outer rim to one diameter all around, and then honed the edges to give it a little "wooden barrel" convexity to keep the band running centered. By suppertime they were finished. I trust that was one of the bigger jobs for making the mill. (Let's see, what else is there?... making and assembling the frame parts, mounting and connecting the motor, making entry and exit blade guide blocks and holders for them, making the cut thickness adjuster, making the water drip blade cooling system, making a sheet metal(?) body and guards to make it safe... Okay, it won't be done in a couple of days!)

Sizing things up. The tool on the left is
the one that allowed me to mount the
wheels on the lathe.
   After doing a sample fitting it seemed to me that 6" from front to back seemed excessively wide, but various parts thicknesses added length to the shaft. I reflected that 1.5" thick wheels were somewhat overkill for .75" wide saw bands. On the other hand, they would handle a wider band if I could find one (from somebody's bandsaw mill?) in case that seemed to cut better, so that made them better for a prototype. What I could have done was make one full 10" plywood wheel piece and cut out the center of the other one so it was just a rim. It would have the same outside shape but the center would be thinner and the pulley center could overlap it - it would be thinner, and lighter. (I'm hoping to keep it well under the weight of the 40 pound chainsaw with mill, which I can barely handle. The lighter the better! The wheels weigh 915 g each - about 4 pounds total. Their hubs add another pound. That's 5 pounds out of... 30?)

I decided to total up some weights to get an estimate for the total weight:

ITEM (qty)
WEIGHT (grams)
TOTAL (grams)
Wheels (2)
Hubs (2)
Axles (2)
Aluminum 6" pulley (if 6" is a good size - my other pulleys are heavier.
But an 8" or 10" pulley might be better: slower saw band with more torque.)
Backbone (2 pcs)
Pillow Block Bearings (4)
500 (?)
3/4" x 96", 2 teeth per inch, saw band
Motor (Ryobi Skill Saw)
 quite close to 5000

That wasn't an all-inclusive list, but it came to 12152 grams or 27 pounds. By the time the case and miscellaneous parts were added, the saw might be around 35 pounds - hopefully not more than 40 with the cutting depth/slide rails ("Alaska mill" parts) included. So much for my 30 pound pipe dream!

   A lot depends on the motor's weight. Now, where are those OEM lawnmower motors to be found? I'm sure they're the best: 120 VDC motors (with a diode bridge to run on AC) and permanent magnet stators, available up to at least 12 amps. No electricity wasted in field coils, so maximum mechanical output power per amp. (120 V * 12 A = 1440 W = 1.93 HP.) And not so much over 100$. I think those are the motors for a production unit.
   OTOH Ryobi skill saws complete with motor are around just 60$. And to mount it, one might just bolt down their base plate to the unit. Then the cutting depth adjustment becomes the belt tightening adjustment.
   But after I run the saw with the Ryobi, I could temporarily remove the 12 amp motor from my own electric mower and try it out for comparison.

   I digress with various thoughts and no actual progress because I was waiting for the ordered parts to arrive. The pillow block bearings were supposed to arrive on the 19th, but they still weren't there on the 21st. I needed them and the 3/4 inch shaft to start sizing up the overall assembly.

Milling with Chainsaw Mill (might a bandsaw do better?)

   In the meantime I continued milling with the chainsaw. I rolled the big log 1/4 turn with a hydraulic trolley/floor jack, a bit at a time, wedging pieces of wood under it so it wouldn't roll back. On the 20th I finally had the flat board mounted on top and I cut the rough first slice off the top - and then into firewood pieces. On the 21st I cut two 6" thick slabs to make 16 foot 2" x 6"es out of. The very first slab cut earlier before rolling the log was mostly 30" wide, that being the maximum width the 36" bar attached to the mill could cut. (Several protruding chunks and some of the butt end had to be cut from the log to narrow it so that the mill would fit on.) That cut took around 14 minutes of actual cutting, excluding time spent putting in wedges and refilling the gas and oil part way through, and for me a few very brief rest breaks out of the exhaust and noise. As always it excludes the time spent sharpening the chainsaw and refilling the gas and oil - with every slice. And the time starting the saw, which sometimes went well and sometimes was exhausting in itself, when 10 or 15 pulls did nothing and it required removing the spark plug and giving it a dozen pulls for no apparent reason (wasn't flooded) before it would fire when reassembled. After rolling the log, the top cut at right angles to the first ones took about 7 minutes for a roughly 18" cut. On the 21st I cut two slabs, one in the morning and one in the afternoon, each taking 10 minutes of actual cutting, yielding two 6" thick by roughly 24" wide slabs with one bark edge, 16 feet long.

   After being diverted for some days, on the morning of the 27th I went to cut the log into the last two pieces. The saw wouldn't start. I spent more time trying to start it than I would have spent milling. I finally threw it in the car and drove into town with it to the dealer - another hour driving, tho not entirely for just this one thing. The dealer changed the spark plug and it started with one pull. I bought a spare spark plug. Late in the afternoon I got back to the log. Again it started with one pull. Wow! Almost all the way through, the saw ran out of gas. I guess it only holds enough for about 12 minutes of cutting. After refilling it it took a number of inconvenient sideways pulls to start again, from being warm and turned sideways in the log, but it did start and I finished the cut. I made this one 8" thick instead of 6", to cut some 1" x 8"s or 2" x 8"s from.
   The remaining bottom piece was also about 8" thick, but one edge would be bark and irregular, so the boards will be narrower - 6" if standard lumber dimensions are adhered to. It would have been thicker - over 9" - except that three of the chainsaw cuts turned over an inch thick of good wood into sawdust. A bandsaw would have saved most of that and probably allowed some 8" boards. (The top cut on the log doesn't count as wasting good wood because the sawdust theoretically comes out of the top scrap piece, so I only count 3 cuts [9/8"] instead of all 4 [12/8"].)

   This last cut took about 13 minutes of actual cutting, in spite of being a somewhat thinner cut than the previous two. That puzzled me at first. I thought it was cutting fine. Did I not sharpen the chain as well as usual?
   Then I remembered I had seen a couple of sparks as it passed a protrusion where a limb had been, near the start of the cut. It would seem there were a few chunks of gravel embedded in the bark. That side of the tree had been lying downward, for months, and the rocks must have got pressed in. I don't suppose bandsaw bands will like sand and rocks any better than the chainsaw chain does. (For city trees, carbide teeth that would cut through embedded nails, hooks and clotheslines was one important advantage of my pivoting circular blade mill, small and slow as it is. I suppose it would do loose bits of rock without much problem too. Hmm!)

   The point to writing here of these chainsaw milling activities is to illustrate what it is hoped that bandsaw mills can replace. If it's easier to push and cut with, or if it cuts faster for similar effort, I for one could do more cutting. And the cuts should be smoother. So far I haven't managed more than two cuts in a day. I've averaged at best one per two or three days from: saw won't start, taking the saw in, brushing off the rust and straightening my new (used) bar, and a couple of new chains (after wrecking one... all this 'prep' stuff is done now, I hope!), and from cutting off protrusions and carefully setting up the top board on fresh log surface for a straight, flat cut, which I've now done 6 times for the three sections of log. (The bandsaw alaska mill won't fix that headache. But I just might come up with a better technique!)
   With a maximum 20" wide cut this first electric mill won't break down the biggest logs that the chainsaw with the 36" bar (30" cut) does. But it should be able to do the smaller ones, and it certainly should be able to cut the big cants made by the chainsaw into useful lumber sizes, without making much sawdust.
   I've saved a couple of smaller logs, upper ends of the big trees, to try the first "Carmichael Mill" out on once it's ready and has reliably sliced a few boards off of cants.

Building Continued

   On the 24th the bearings and the 3/4" hubs arrived on the same day. I was still missing the 3/4" shafts to really start putting things together, but I did a little test fitting that evening. It appeared that the maximum cutting depth, by cutting right down at the backbone, would be about 6-1/2". But this could be extended to anything short of the 10" wheel diameter by mounting the wheels' "pillow block" bearing holders on extension blocks. The possibility for cutting at least an 8" thick slab off a log to then slice off boards or planks of that width was attractive. That would need at least 1.5" standoff blocks; preferably 2". Then again, if the blocks were 3" thick, the wheels wouldn't protrude above the backbone bars at all, and a flat top cover could be used instead of a shaped one. But the thicker they are, the more solid and well supported they and everything will need to be.

   On the 28th the 3/4" shaft finally arrived. I cut off a 6" piece for the end with the pulley and 5.25" for the other end. Fitting everything together so it all turned smoothly turned out to be one of those simple sounding jobs that take several hours. The "H" bushings wouldn't tighten sufficiently against the shaft in the wooden wheels to lock them on, and it needed four 3/4" ID washers on the shafts to keep the pulley bolts from hitting the bearing mounts, but amazingly I had just that many. By evening I had the basic assembly but still with a couple of things to change that would require disassembly again. One wheel wobbled badly and the washers I had used for the tightening/adjustment bolts were too small and were just digging into the wood. (Aren't washers supposed to spread out the load? These days they seem to make them too small and so thin they bend into a cup when tightened. What's the point?) I disassembled and fixed it later.
   It turned out that with the 93" cutting band the bearing holders were almost exactly right at the ends of the 36" extruded tubes. Can't ask for better than that! I mounted the pulley on a protruding axle behind the frame instead of between the two pieces, in order that it wouldn't have to be disassembled to change the V-belt. The cutting band fits around both wheels, so it can also be easily changed.

The essential frame of the saw, upside down. 10" x 1.5" plywood wheels, lightweight 6" V-belt pulley, 93" cutting band.
With the band on and no case I had to set it on a thin board - one end of my radial arm saw table.

Right way up. If nothing sticks down from the frame
in the cutting area, the maximum cut depth is over 6".

   I spun it around with the pulley (already watch yore fingers on those moving teeth!) and reflected that there's still much to be done before it's ready to slice into some wood.

Electricity Generation

An Attic Windplant?

   On about the 28th I was lounging in the bathtub and hearing some whirring vents this house has on the roof. Perhaps this idea is unique?: Whenever a wind is blowing past a house, there's a high pressure windward side and a low pressure lee side. If one put one-way vents into the soffits (and end walls?) that would allow air to enter but not to exit, so pressure would be built up inside the attic by wind from any direction. Any wind would cause air to flow upward through a vertical air pipe extending like a chimney through (or near) the peak. The wind turbine is placed in the pipe. That's it in a nutshell.

   All the moving parts would be in the attic, out of sight. The air intakes would be covered by a screen. Even some floppy plastic that lays over the screen, held along one edge, might suffice for "one way vents". One could embellish this by having a swivel pipe on top of the chimney, with a vane so its exit always points downwind, to maximize the pressure drop through the chimney/pipe. If it's noisy, perhaps some sort of muffler could be made for the air outlet.
   I haven't tried to figure out what sort of power could be made, or what airspeed might be attained by what wind. The airspeed near the ground is lowered by obstacles such as houses, and it's generally considered "the higher the better" for windplants. But extensive soffit vents could present many square feet of entry to rapidly admit many cubic feet air per second into the attic. So the air velocity in the pipe (and hence the available energy) might be higher than one might expect. The pipe should be pretty short and a fair diameter so as to not be a bottleneck.
   Consider a "regular" windplant with a 2 meter diameter propeller - a pretty large home unit. That would have 3.14 square meters of air across the blades at the full speed of the wind. Think of 'a sheet of plywood' to visualize this area. Now consider 3.14 square meters of soffits or end wall allowing air into the attic. If the chimney was .314 square meters (.632 meters or ~2 feet diameter), whatever speed the air came in at would be multiplied by 10 on its way out the pipe. That would be a pretty big pipe but a pretty small turbine, easier to do.

  Having thought of it, I don't think I'll try building it. The results would obviously be very dependent on the building and the wind conditions there. The effects of the air pressure in the attic might have to be taken into account. All rather unpredictable. My place might actually work out well, but I'd rather press on with the lambda ray energy converter... after the bandsaw alaska mill... after hybridizing the car...

Haida Gwaii, BC Canada