Turquoise Energy News #138
covering November 2019 (Posted December 2nd 2019)
Lawnhill BC Canada - by Craig Carmichael


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

Month In "Brief" (Project Summaries etc.)
  - 12/36 Volt DC Infrastructure - WIFI/5G Countermeasures - Ground Effect Vehicle - 100% Efficient Infinitely Variable Torque Converter - Woodstove Stirling Steam Engine Generator - New Chemie Batteries

In Passing (Miscellaneous topics, editorial comments & opinionated rants)
  - Gardening - Winter Cabbage, Sugar Beets - Hearing Improvement Further Notes - 5G Countermeasures: Biologically Harmful - Small Thots - ESD

- Detailed Project Reports -
Electric Transport - Electric Hubcap Motor Systems
* Ground Effect Vehicle (Still Prototype R/C Model... continued)
* EV Transmissions: 100% efficient infinitely variable torque converter.

Other "Green" Electric Equipment Projects
* Working with the Handheld Bandsaw Mill (& Alaska Mill)

Electricity Generation
* 5 Blade Windplant - Windplant Ducts
   - Clarkson University Experiments with Windplant Ducts
* My Solar Power System: - Monthly Solar Production log et cetera - Notes.
* Woodstove Alternator with Stirling Steam Engine?

Electricity Storage
* Turquoise Battery Project (Mn-Zn or Ni-Zn in Mixed Alkaline Salt electrolyte)




November in Brief

   November seemed to pass in something of a blur with day after day going by seemingly with little to no accomplishment. But somehow I had a lot of writing to do. And it wasn't a whole loss. I inched forward on various projects.

   I started making the DC/Solar charging system for the Sprint car, but making a HAT plug for the charge controller (to plug it into a solar panel(s)) took so long I diverted into looking for ways to make them easy to produce and didn't get it finished.
   Between me and Mike in QC we worked out to make the plug blades simply straight pieces of flat copper, and to put a ring around the inner end made of small copper tube. The ring then crimped a piece of wire to the pins.
   Somehow I had never conceived that the pipe, even 1/4 inch, could be small enough. And it doesn't make a 'clean' crimp - it squashes out. But I got on line and found some still smaller, 6mm and 5mm (and under). One of those will probably be just about right.


   The next week we tackled the sockets, studying some 120VAC electrical socket innards, and Mike came up with a simple "Z" fold, with the same copper pipe crimp at the wire end. I never think of things that would waste an extra milligram of metal myself, but these will be practical - easy to make and easy enough to use. At least they won't need soldering. And they will probably be model "A". Later will be some (bulkier ones) with screw connections - easiest to wire.
   Instead of copper, the AC sockets seemed to use brass. We figured that must be springier, and they need to grip the plug blades. I had nickel silver, which is still stronger than brass. I would however note that pure copper is by far the better conductor. Brass is likely to be better than nickel silver too - sigh, another thing I should check out before deciding what material to use in production!


Copper plug blade (ring crimped) and "Z-fold" nickel-brass socket receptacle (ring uncrimped).
They seemed to grip each other quite well.

   The next task will be to design shells to hold these pins and sockets.


   And, my fiber optic internet having come at last, I realized it had both regular 2.4 GHz WIFI and 5 GHz - the notorious "5G". Furthermore it was only a few feet from where I sit.
   Nobody mentioned the "5G" when they were putting it in. I only noticed the light later. I should have had it mounted somewhere far from where I usually am.

   We are all guinea pigs in the 5G experiment. It's a very penetrating wavelength (~6 mm) transmitted with considerable power, and it has never had the slightest testing to ensure that it's safe long-term. There are already troubling signs that regular WIFI can in fact cause people serious problems.
   No one should imagine that transmitted microwave radiation somehow goes around their body. Suing someone for criminal negligence later won't help anyone who has developed cancer or who has had a defective baby(s) in the coming years. I expect eventually "5G" will be scaled back in power of even abandoned as being too unhealthy.
   Even in front of this grounded piece of metal, the cell phone still shows four bars WIFI from it. How strong are the signals without the protection?

Ground Effect Vehicle - to open up the BC north coast and other rugged, isolated and island territories

   I modified the canard to hold the ducted fans and then mounted them. I put in a heavier (1/4" dowel) pivot since all the thrust will go through it. I redid the rear of the wing (ex "elevator") after melting the covering with the heat gun, and screwed it onto the back of the wing. I mounted the servo to move the canard up and down, and by the end of the month I tested it.
   It was starting to look like it might be finished some day.


It works!
Moving the "elevator" control on the handheld transmitter
moves the canard up and down. With the ducted fans mounted
on it, aiming it up for take-off blows the air under the wing
for more lift, a highly desirable ground effect craft feature
"incidentally" incorporated by putting them on the canard.

100% Efficient Infinitely Variable Torque Converter - for improved efficiency of road and rail transportation

   After all these years muddling around, and finally having a very clear concept of how everything should go together last summer, and then having found a piece of pipe just right for a spinnable housing for the planetary gear, I couldn't help but start working on it. This month I got and mounted on the shafts two strong double row sealed ball bearing units that seemed the ideal type for this application. (ordered in October)
   I found some aluminum disks that were usable. But I was unable to readily turn them as desired. I'll need to go to some contortions to mount them on so they fit "in the gap" of my "too small" lathe.
   Instead I focused on making a coupler to couple the motor to the planetary gear. Somehow turning a little piece of pipe to exact size was an all day job, and I still need to weld it to the end of the splined shaft stub that fits the planets gear.


Shaft coupler pipe.
It'll be "pound it onto the motor shaft and put in set screws",
and "weld the pipe onto the splined shaft stub".
I got it to pound onto the shaft aligned, so hopefully
I can weld it on still solidly held straight.

Now having all these pieces I could start trying to put things together. From the left (below):

* The shaft to connect the motor (Motor mounting off to left to be created).
* A gear to connect the assembly to the car's differential (which will hold the large gear).
* Aluminum end plate.
* Planetary gear housing (to be oil filled - the end plates seal the gear housing.)
* The other end plate.
* The big centrifugal clutch drum.

   All the parts seen here are tied together and turn together as the car moves except the motor shaft, which is connected to the planets assembly inside the planetary gear. Since they are, both end rings and the plate of the clutch drum will all be bolted through from left to right with long bolts, ensuring everything is held solid and squeezing the oil seals tight.
   Sizes and spacings conspired to insist that the drive gear had to go on the left end without touching the motor shaft. (There's more to do to mount it to the end plate. I may use sprockets and a chain instead of my perhaps dubious gears.)



Seen in the second image from the other end (below):

* The centrifugal clutch inside part (with UHMW centrifugal shoes to contact the aluminum drum.)
* An end bearing to hold it all in line. (Mounting to be created.)



   Some may ask how any mechanical mechanism could be 100% efficient. To review the operation, the ring gear/outer drum of the planetary gear is attached to the car's differential by gears or a drive chain. The motor turns the planets assembly. Since the car isn't moving, the ring is stationary and the sun gear spins. The sun gear's shaft connects to the inside element of the centrifugal clutch, which is thus turned at three times the speed of the motor.
   As the shoes fly out and engage the drum, it turns it and the car starts to move.

   As the car starts moving, there actually are losses. Even a planetary gear is only around 97% efficient, bearings turning have small losses and the centrifugal clutch is an unknown. But as it all gets up to speed and less torque is needed, the clutch shoes lock to the drum and then everything spins at once: motor, centrifugal disk/shoes and the drum, and both sides of the bearings are turning together. There is no relative motion, so nothing to cause losses. (The end bearing turns, but some shaft bearing would doubtless be needed anyway. And I suppose one might count air friction...)

Woodstove Stirling Steam Engine Generator - Reliable winter electrical generation at home

Stirling Engine Animation      
from Wikipedia            

   Owing to lack of sunlight to make electricity, and with a woodstove TEG (thermo electric generator) seeming to be not much bang for the buck, and with steam engines needing too high of a temperature, I thought of Stirling engines. I didn't understand how they worked, but I knew that like TEGs, they ran off the difference in temperature between a hot and a cold side.
   So I studied them. Youtube has many working examples. They're pretty simple, but they aren't very powerful for their size. I'm sure they can be improved.

   It seemed to me, first, that the stroke of the displacer is wasteful. It creates the most heating and cooling effect, the most propulsion, when it's at either end of the cylinder. Instead it spends too much time in the middle and never actually quite gets to the ends. That could certainly be improved by having it sit right on either end for almost 1/2 the stroke.
   Second, woodstove tops often operate just in the region where water boils - a low simmer. If a large diameter cylinder had water in the bottom, and if there was a phase change from water to steam in the heat part of the cycle, and back from steam to water in the cooling part, the volume of gas in the cylinder would change far more than just with heating and cooling air, giving the stroke much more power for the size.
   By turning an efficient alternator, perhaps as part of the required flywheel assembly, I thought a good goal would be making 100 watts of electricity - day and night for as long as the stove had a moderate fire going.

   Just what I need, another project!


New Chemie Batteries - Better, cheaper mass electricity storage

   I didn't do an awful lot on these this month. What I did do led me to I re-think a few things.
   At the start of the month I had another zinc tab corrode off from the rest of the electrode. I decided I should order some zinc that was at least said to be pure - there were no claims to purity on the "rooftop moss killer", and there was all always that black crud left over when I used up a sacrificial electrode for electroplating. How could it be pure? Would pure zinc work better? Probably!
   The experimental copper electrode with calcium oxide on it had also corroded away, but apparently radiating from one point on the tab. So I was left wondering whether that was just where it (probably) got scraped at that point when putting it into the cell, or not, and so whether the idea worked or not. But then I also thought that gold plating the copper might be a better plan. It wouldn't have to be very thick - hardly enough to see - and it would probably protect the copper nicely. (especially at manganese dioxide voltage, which is below the gold-to-gold-oxide voltage.) My old gold brush plating solution (from 2004) didn't seem to be working any more, so I ordered a bottle of "proper" gold tank electroplating solution. (Almost 200 dollars - Ouch!)
   If no metal works there's still graphite. Metal would give higher current capacity.

   Finally I also decided it would probably be practical to make the cells to lie down flat instead of standing up. That'll need a new (3D printer) case design. It will also allow very large but thin individual cells that won't keep falling over.





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

Gardening - Winter Cabbage, Sugar Beets


   In the spring I had planted cabbage in the garden. By fall, just like last year, it hadn't formed heads. Somebody said cabbage had to go in the greenhouse. Certainly the ones in there seemed to grow better. On November 8th I decided to try taking a greenhouse to the cabbage. I put some walls of scrap styrene foam and glass around them, another scrap window on top plus three 12"x12", 45 watt LED grow lights. I covered over the top of the lights with a piece of plastic. It was pretty slipshod (removed for picture), but we'll see how they do - if it doesn't get too cold.

Toward the end of the month it was almost surely too cold. I started leaving the lights on full time to keep them from freezing. I'm not sure I see any growth. Perhaps if I had done it a month earlier? or two? I'll persevere for a couple more weeks, but it doesn't look promising.


   I decided to try growing carrots indoors. I hooked up the "LED indoor garden" in the corner of the kitchen. But only lettuce seems to be doing very well so far. I think the corner if not the whole kitchen is just not warm enough. I think I'll insulate around the lights and maybe even turn on the electric heat just so it heats that space. (Maybe replace one LED with a 100 watt incandescent bulb in one socket, with the small fan near it, just for the heat?)


   We've heard about the staggering global crop losses in corn, wheat, soy and even potatoes (almost nothing easier to grow at home than potatoes!), and the staggering devastation of vast herds of porkers and beefers. Now here's another whammy: as fall turns to winter the summer's stats from sugar beet farming have come in and the crops are way down: grand disaster in southern Alberta and south of the border too. People trying to source sugar simply can't get it. Rogers has apparently declared "Force Majeur" on contracts and put its usual customers on allocation: there just isn't enough to go around. So places report sugar cane crop failures, but Louisiana had a bumepr crop.

   And NASA's predictions for the next solar cycle indicate we're in for a grand solar minimum like the Dalton minimum: 30 years of cold, rainy weather - worse next year than 2019. The conditions for that grand depopulation of the planet by mid century are really starting to take form!
   So I looked up sugar beets. I've never seen them in stores in the spring, but there are some seed suppliers who have them available. I should look further to see if growing them here sounds the least bit feasible. (back to youtube!)
   Later I heard that the phenomenon is being called "farmageddon" in the USA. Farms are going bankrupt en masse (in Europe, too), there's an unprecedented wave of farmer suicides, and food companies are going bankrupt when they can't get inputs at prices they can pass on - especially dairies because the price for milk is "fixed" by the government. ("Fixed"? Don't they mean "Broken"? How about "Rigged"?)

http://www.iceagefarmer.com/2019/11/07/food-system-collapsing-actively-rapidly-urgent-warning/

https://electroverse.net/in-the-last-30-days-the-u-s-set-7112-new-all-time-low-temperature-records-vs-just-the-1605-max/

   And in the face of this, a NASA temperature map shows globally slightly warmer average temperatures for 2019 as a whole most everywhere except a big, dark "cold hole" over most of North America. Not that farming has done well anywhere. The staggering losses everywhere have different causes in different places.



Hearing Improvement Further Notes

   In TE News #136 I wrote of improving hearing by dripping vegetable oil in one's ears. It's also supposed to help reduce or eliminate tinnitus. I found this idea from more than one source. But how could it help?
   This month I read an article saying we lose hearing because of our noisy environment. It mentioned things like hair blow dryers, shavers, food processors, blenders and coffee grinders, many of which are pretty loud for short periods. In #136 I also mentioned noises that persist over long periods like refrigerators, fans and heating and air conditioning systems, traffic, and music or continuous babble - radios, TVs and so on. Since the noise is ubiquitous we pretty much stop noticing it, but I suspect that both types of noise have long term deleterious effect.

   The article went on to say that with indigenous people who don't have all these electrical noises around them all the time, old people hear as well as young people. This tends to validate the idea that QUIET is what one needs to retain hearing, or to recover it if [that's an IF in a mechanized society?] if it has been lost. And that the way the oil works is as a high-frequency selective "earplug" that works for hours, reducing the levels of noise entering the inner ear mechanism, perhaps by 10 dB.

   This is obviously a very slow acting, long term treatment - a reversal of the slow, long term loss of hearing. And obviously of course it must be done along with all other sensible hearing protection precautions one may take. And eliminating environmental noise as much as possible.



5G Countermeasures: Biologically Harmful
(Oops, I've written this up twice)

   My fiber optic internet finally came in late October. I had the fiber cable end and the interface box put in near my computer desk. It's great to have good internet again! It came with WIFI.

   When I finally bothered to look at it, I was shocked to see that it also came with the controversial 5 GHz as well as 2.4 GHz WIFI.  (now I finally know why they call it "5G".) This is getting up into a biologically destructive range and I'm sorry it's there. Even 2.4 GHz WIFI can be a damaging level of radiation if, for example, you hold a cellphone to your ear for hours per day, or there is a cell phone/WIFI station too close where you are routinely present... or sit too close to the WIFI box in your house for too many hours a day. (I have heard stories - rumours - that brain cancers have been observed from the first, and other fatal cancers from the second.) Does anybody imagine that these rays somehow go around people instead of through them? I hardly care if I have WIFI at all. Why would I want to subject myself to even more penetrating radiation? I wish I could just turn it all off. I wish I had had it installed in some far corner of the garage I rarely go into. I'd have run any length of ethernet cable in preference!
   "5G" is a very considerable radiation and there have never been tests done on it to assure that it's safe. It is down to 6 millimeters wavelength, even shorter than "microwaves" (centimetric waves) and headed toward the infra-red band. In fact we are guinea pigs in a vast experiment. I put a piece of sheet metal (cupro-nickel, but steel would work too) in front of it and connected it to a ground with a 3-prong electrical plug, only using the ground pin. That way the radiation can't come directly toward my computer desk where I sit a lot. Even so my cell phone shows four bars if I turn on its WIFI. (I had been cutting pieces from this sheet. It had a convenient "tab" that I simply bent over to hang it from the unit itself.)

   It won't help you to say, when you have terminal cancer, "Nobody told me it wasn't safe!" Anyway "they" will deny that the 5G by your desk could have had anything to do with it and you won't be absolutely 100% certain yourself. But then, "they" told you that smoking was safe, that thalidomide was safe, that glyphosate ("round up") weed killer was safe, that GMO wheat was safe (that's when widespread "gluten intolerance" started), ...



Small Thots

* I noticed a title from "Provident Preppers" on youtube, who grow their own gardens. I didn't watch it, but it said something about that they had lived for 90 days on their stored food and garden produce, without going to the grocery. Of all the crazy ideas, testing to find out how you actually make out and finding any deficiencies in your preparations!

* CDC Director Robert Redfield said people should stop referring to "a coming time" when antibiotics won't work: "It's already here." Many microscopic threats are "urgent", "serious" or "concerning".
   Along with suddenly developing food scarcity, we are setting up for the global pandemic or pandemics which will end the unsustainable "population bubble" humanity is now in. The African Swine Fever epidemic is a foretaste of how an entire species population can drop dramatically and rapidly. (I'd say globally but it hasn't hit the Americas yet. We are desperately trying to keep it out. But if some European pork tour group books a flight to vacation to Disneyworld, look out!)

* Freedom of Speech is not about being agreeable. Freedom of speech is about freedom of thought. It is about being willing to hear things and ideas that you don't want to hear. It is about letting people express themselves even if you don't agree with them - and even if you think no one agrees with them. Even if you think they are wrong and what they're saying is idiotic. It's about allowing things to be said that you fervently hope others don't believe or agree with. If what someone says really is wrong, they will gain few adherents. If it is right, you've avoided putting yourself in the wrong by arguing against it or worse, preventing people from hearing it.
   When freedom of speech is infringed by authority, it is a case of "might versus right". If we don't believe might makes right, we should want no part in suppression of free speech. Whether you like or dislike Alex Jones and what he says, when he was almost simultaneously banned from Youtube, Facebook, Twitter et al, it was a sad day and a sad precedent for free speech for all. "First they came for the Jews..." ... First they came for Alex Jones, then a whole pile of other people you won't find videos or articles from any more (even their old ones), and they're subtly going after the whole of alternative media - keeping them out of all but the most direct search results, demonetizing their videos until everybody interested has already watched them, and so on.
   Many think that the alternative media is "just a bunch of nuts" anyway. But some of the better channels provide a link to a page with the mostly credible sources of their information - which you virtually never get from the "mainstream media". From them its "experts say", "scientists say" (unspecified experts), stock or other footage pawned off as being news, even outright lies, with no provision for checking for yourself. And on Youtube and Zerohedge (at least) comments may be entered below the video, so if others have more information or disagree with the content, you can find out quickly by reading a few of them. And you can enter your own comment.
   Those who are in corporate and government and "mainstream media" power are afraid of truth. They don't want you to hear other opinions than their "narrative", and they use their bureaucratic power to stifle their airing, without being so obvious about it that there is a huge and immediate backlash that exposes their dictatorial habits.

"The truth never suffers from honest examination." - Jesus


* If humans try to go to Mars without protection against cosmic "rays", they may arrive blind if at all. Recall that the lunar astronauts reported seeing flashes of light with their eyes closed once they were outside of Earth's Van Allen Belts of protective magnetic shielding. Later some of them got cataracts early in life. That's from being just a week outside Earth's magnetic protection.
   Vegetation might grow outside the protective zone. It doesn't have delicate eyes or nervous systems. It seems like Ganymedean vegetation has spread to many spheres (even little "oids") without air, as long as there is good soil and no special harsh radiation such as Jupiter and Saturn sweep around with their rotation. (This is why Callisto and Iapetus are covered with "dark", "fluffy" complex polycyclic aromatic hydrocarbons organic matter (scientists' findings), plant life (my interpretation), but on their leading hemispheres only.)



ESD
(Eccentric Silliness Department)

* We used to have "television sets". Now with flat screens they are too thin to set anything on top of. It will fall off. Should we still call them "sets"?

* "Zero" and "seven" are single digit numbers. Why do they, alone, rate two syllables? Especially "seven" grates on my mind when counting to ten.
   Sometimes used "oh" is easily confused with the letter "O" - which also looks the same when written**. (or the word "owe", as in "I 0 1 2 thee 4 fine help.") Maybe we could shorten "zero" to "zo"? That's pretty unambiguous. Or use the British words, "naught", "aught" and "nill"?
   We could perhaps shorten "seven" to "sev"? The French word "sept" with the silent "p" - unlike "sept-ember", the 7th, er, 9th month] might be good except it is already much used as, in "television set". Okay maybe better, "sept" but sound the "p"? That has good precedent in language.

  ** I've seen a scanned book with a strange word, "loomm". It was really "100mm". And in the same book I think "ydmrn" was probably "92mm" (or 93 or 96).

* There's something funny about the flavor of wry bread.

* Lifting 100 grams is easy. Lifting 100 killergrams is hard.

* Earthly vegetation that might grow on Mars: Marzipan, Marsmallows.

* Algorithms: The daily comings and goings of Al Gore.

* Bacon supplies that may be in short supply owing to the "pig apocalypse": bacon powder, bacon soda.




   "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 just thought of and not tried... 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, consistency and completeness 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 and dead ends.





Electric Transport


Ground Effect Vehicle (first the R/C Model)

Canard & Ducted Fans Mounted on Canard

   Having decided to have a front canard to replace the rear elevator and never having seen such a thing, I found the "WISES" ground effect model video. It seemed to fly quite well with much more positive response when put to climb or descend.


The "WISES" ground effect craft - Full size model test prototype 2011. (Why does it keep
buzzing that poor guy in that boat? ...Yes, the one holding that, um... radio control... unit?)




Although it was "full size", 7 meters long, he built it lightweight and graceful,
so it would quite fly slowly. Why? Because he didn't have a high speed chase
boat. A faster version such as is planned for the manned craft would have
quickly flown out of control range. (I wonder how mine will fare?)

   Then, having decided to move the ducted fans to the canard, I looked at the "WISES" ground effect video again. I had been trying to see where the propeller was but it seemed invisible. I had thought it must be at the front of that long thin nose. Finally I noticed there was (sure enough) one on each side on the canard, faintly visible in some of the footage. Essentially the same place as my own new plan!
   So while both my reconfigurations were already made, the high-performing WISES craft provided strong reinforcement that I was probably making the best choices. At least I was in good company.

   And I believe the ducted fans are a better choice than open propellers: higher static thrust, and higher above the water, so they can be mounted lower. Around the 12th to 15th I picked away at the canard, ripping off some of the cloth, adding two ribs for mounting the fans, and filing the front and back edge spars to more streamlined inside shapes. Of course the canard was made without the fans in mind, but even so it needed to have the lengthwise strength that the famous Haida Gwaii aircraft spruce provided.
   I changed the dowels from 1/8 to 1/4 inch as they would now be taking all the propulsion thrust force along with lift. I couldn't have the canard with the motors break off and fall into the water!

   And of course, the "ESC" motor controllers had to go in near the bows beside the canard - the one part of the hulls I had covered over on top! (Won't need to get in there, oh, no!)

(23rd) I finally decided to try heat shrinking the covering, which seemed to be recommended by all the modelers for all the coverings they used. I had started to make a test piece, a short version of the canard, but then I tried on the top of the rear elevator (now to be a fixed part of the wing, but already made separate). Instead of shrinking and getting taut, the PP cloth stretched and sagged. I moved the heat gun in a little closer - not close - and a hole appeared and spread too quickly to react. I would have to sand it off and redo it.
   Okay! No heat for PP cloth/landscaping fabric! Either it'll shrink a bit and become more taut when the finish is applied, or it will stay as-is.


   I had dabbled in it just occasionally, and eventually got the ducted fans mounted in/on the canard. By the 30th I had mounted and was able to attach the servomechanism to the canard.


It works!  As the "elevator" control is moved up and
down on the handheld transmitter, the tiny receiver
picks up the position info and the servo twists,
pushing the rod, and the canard aims down and up.

   With the canard pointed up, the air from the fans is blown down under the wing. With the sides and back largely "boxed in", this somewhat "inflates" underneath like a hovercraft, allowing takeoff at lower speeds. This is an advantage built into the propeller angle of some of the radio controlled models. But in this design that steep downward air angle is obtained as part of the vertical control, and only when it's useful.

   There is one part I haven't gotten to yet: turning and maneuvering. There is no provision on the motors (or the RC control) for going in reverse. It would seem the only way is to swap two of the three power wires. For the model, at least for the first run, I think I will just have to say "if it was the real thing you could easily maneuver up to a wharf or float."
   My present plan for turning has been to turn one motor down and the other one up. But once again there is no provision for that, which would have to be a "double" throttle stick in the control transmitter. I'm thinking I need to make a special circuit board to "reinterpret" the signals from the motor speed control and the "rudder" control, and use the "rudder" control to speed one up, and slow the other one down.
   The other alternative is to put an actual rudder on the craft, and power both fans equally. As with inability to edge up to a wharf, I'm not fussy about the idea, but it may be a good stopgap for initial testing.




EV Transmissions: Off-the-Shelf Planetary Gear (for Miles Truck),
100% Efficient Infinitely Variable Torque Converter


(5th) In spite of it not being high on my priority list and trying to get out the October newsletter, having a plan for making a planetary gear enclosure and the parts to do it, I was inspired to not bother waiting for the stock gearbox and start putting together the custom one for the Sprint.
   I didn't have a bearing the right size for the motor/planets side. It was 30.5 mm/1.20". There were lots of 30 mm bearings, but none 30.5 and none 1.20 inches. I had a couple that were 1.375 inches. I found a brass plumbing piece that was a perfect fit to hammer into the bearing center, and I cut a short piece off and did so. Then I cut another piece and cut a diagonal groove across it. I widened the groove until it fit inside the other piece. So I had the bearing with two shims inside. It fit perfectly onto the flange on the planets assembly. The ring gear already had a previously made shaft on it that took a 1.0635" ID bearing, which I had some of. So now the planetary had a bearing on each side of the ring gear.

(6th) I cut off a 70 mm piece of the big (~4-1/2") pipe with a zip disk. Then I put it on the lathe and squared up the cut end. Then I used a boring bar and did some turning on the inside on one end, extended into the middle. Then I turned it around and worked on the other end, but inside only to 10 mm depth.


Starting to turn the inside of the pipe to "prfectly round" and the right size to hold the ring gear.


With the ring gear pounded in.


The motor end with the planets assembly and stub splined shaft.
Soon I realized that the "cup and cone" bearings would have leaked
oil, so I replaced them with "double row sealed ball" bearings.


The centrifugal clutch end. The long shaft on the ring gear,
 which was "ready made" from some previous attempt,
proved to be just long enough.

   Then I turned it around again and widened the inside diameter some more. After going a very tiny bit at a time (seemingly far too many times), I had a fit where I could pound the ring gear into the pipe. So it had machined ends and a solid fit. The ring gear could have gone another 1/8" in (I was just guessing the depth to cut), but its was wider side to side than the planets, so there was some room for play and it all fit. The piece of pipe was about 1/4" longer than it needed to be. (I was guessing that too, but I had left a bit extra in case my guess was short.) Here again I was in luck because the 1.0635" section on the shaft extended out an extra 1/4", so all it needed was a spacer. So everything fit first try without having to pound the ring gear out and bore deeper, and without having to cut the pipe shorter.
(7th) I noted that it was going to be very hard to get a good oil seal with cup & cone bearings at each end. Well, not much I could do about that! I could change to sealed needle bearings, but... but what? I went out and looked, and while I didn't have just the right sizes, it looked like they'd actually even be a better fit. They had side walls and were filled with grease, but I suspected a bath of oil on the inside would get in and seep through. But at least it would "seep" and not "gush" through.
   So I went online to vxb.com (bearings store in California with a good online catalog and prices listed). I ended up picking two "double row, sealed" ball bearings that I hoped would fit - and not leak, or not much. (One might be too tight, the other loose - best I could find.) I ended up 100 $C lighter. How much cheaper to make than to buy, again?

   I really still didn't have a complete plan. I took the assembly out to the car. It looked like the thing to do would be to put the chain sprocket on the motor/input end and leave the other end just for the centrifugal clutch. Other than having to make some large support with a bearing on the end to support the end of the clutch inner shaft, an extender shaft adapter/coupler for the motor and maybe cut out a recess in the sidewall for the body pipe of the planetary assembly, everything looked like it would fit in best - would actually fit in at all - that way.
   It seemed a great pity the the chain sprocket wouldn't fit around the pipe/planetary gearbox. That would have simplified things immensely. But such a large sprocket would have made for a very low reduction ratio to the differential - less than 2 to 1, when it needed 3 or 4 to 1.

   At this point the impetus ran out. On the 15th I took it to the cafe at lunch for ideas (where I was reminded that the end plates would leak without an O-ring or a gasket). I also found some aluminum plates I could use for the end plates. But basically it was now "wait for the bearings to arrive".

   The company selling me the 5 to 1 planetary gearbox for the Miles truck was also dragging their heels. I had to phone to see what was happening and give a couple of instructions to get them to get on with it. Then on the 15th they sent a drawing of the gearbox and asked me to sign it saying it was what I wanted! Oh well, somehow I didn't seem to be getting a lot done on any projects anyway. On the 21st I finally got another e-mail... saying I should approve their shipping method (Fedex)! Do they really want to sell things?

   That same day the bearings arrived. As I expected the 30mm I.D. one wouldn't quite go onto the 30.5mm hub protruding from the planets assembly. I put the hub on the lathe and filed it down. (I was surprised I could - weren't these parts "case hardened"? Maybe not the outer shell for the planets.) As expected the other one was too loose. I started making a brass spacer. It was going to need a lot of sanding down. Somehow when I was putting a sanding disk onto the drill press, I glanced somewhere and saw some old previously made steel spacer that looked about right. Sure enough it was a perfect fit. I must have had to fit a 1-1/8" something onto a 1-1/16" shaft before - probably the same shaft in some previous attempt at a torque converter years ago.

   In the meantime I had been down to the dungeon, er, root cellar, er, storage room, and found four 3/8" thick aluminum disks I had made trying to do a flat belt to the differential years ago. The centers were too big for the bearing, but they also had some bolt holes which could be used to attach new centers. They should make good end caps for the rotating planetary gearbox.

(22nd) I went to put one of these disks on the lathe, but it was too big. I asked my neighbor if I could use his larger lathe (he had been given this beautiful modern lathe and didn't know how to use it), but the center of the 3-jawed chuck was too big to fit into my center hole, and the outer jaws were too small to go around the outside. I went home disappointed.
   There were holes in them for SDS hub bolts, and that must be how I had originally made them on my lathe, but SDS hubs aren't a perfect thing to center on. Oh well...


How to connect the motor?
Trivial in the design concept, tricky to actually do.

(24th) I left the disks for the moment. There was a question mark around how to connect the motor to the planets assembly of the planetary gear. And that connection would affect the spacing and inside size of the drive gear going to the differential, so I wanted to do it before I got too far on that. The motor had a short shaft. There was a short shaft stub that fit into the spline of the gear. It seemed to need a short hollow shaft to couple them together. On a hunch I went into a drawer full of scrap copper wire, pulled out most of the wire, and found a long forgotten 2" long, 28 mm O.D. steel pipe at the bottom. (I didn't know it was there, honest!) It looked a bit small in diameter, but I had nothing else close. It turned out that if the inside was turned out a little, it would fit over the motor shaft. I made it so I would have to bang it on, and I would add a couple of set screws. If that didn't hold, I might have to drill a hole through the shaft and put a bolt through it - worry about that if it happens.
   The other end was more of a puzzle. If I turned it out to fit the shaft stub, it would be so thin it would probably split open. There was a bit of the end of the stub shaft that was narrower, about 23 mm. I could probably turn the pipe to accept that, but it was only about 3/16" long - not much to grip. I might turn down the outside end of the stub shaft, but it might be a bad idea too. (Aren't those things "case hardened"?) Lacking other inspiration, I turned out the inside of the pipe to take the 3/16" long narrow end of the shaft. Again I made it a "pound in" fit, but still it wasn't much length to grip onto.
   Then I thought that with the two pieces tightly held in perfect alignment, I could weld them together. The idea of welding something to the shaft stub had occurred to me before, but I was sure the two parts wouldn't be in good alignment no matter how careful I was. This would be perfect because it would be solidly held in alignment during the welding.

   I fitted everything together. Somehow the images with extensive descriptions are above in "November in Brief" and I won't repeat them here.




Other "Green" Electric Equipment Projects


Handheld Bandsaw Mill (& Alaska Mill)

(7th) Along with ordering some other bearings I ordered a couple of needle thrust bearings that I thought would be more suitable for the saw than the ones I just got.

Level?

(10th) I conceive that for milling narrow or misshapen boards or when using a narrow or imperfect guideboard, it may be useful to put a level on the mill. One can then maintain the level from one end to the other as one goes and avoid cutting twisted boards.
   I thought of this as I was cutting on top of slightly convex "cupped" previous cuts without bothering to fasten down the guide board. It was too easy to tilt the mill side to side, and it is often the case one can make small adjustments by pushing down harder on one side of the mill then the other. Just a few boards after thinking of this, the guide board actually started tipping off one edge without me noticing, and my cut was more and more diagonal. With a level I should have noticed something was wrong sooner, before it was way off. Over much of the board the two faces were nothing like parallel. (I screwed the guide board down and put in chunks of bark to stabilize the net cut to get back to something like level, and that next board of course also was a wedge shape.) Anyway I got twelve 2" by 4"s from about a 24" tall cant. (The top and bottom ones, being sapwood, have a lot of worm holes.)

Ow! A Nail!

(12th) I cut another 6" slab from the log, and managed to push it aside. That went so well I decided to do another. About 4' along, and 6" inside the log, I hit a nail or something that wrecked my chain. I don't think I could have detected metal that far inside even if I had thought to try. A nasty expense for a job almost done! This is the second tree near the house that has had nails, even in this country place. It would have been over 6' up in the tree. That seemed rather high for fence nails. Perhaps it was a hook for a clothesline or humming bird feeder? The next question was, was there just one piece of metal, one hook, or were others lying in wait? If it was just one, I should be able to cut through 8 feet from the other end and just carefully cut off the last bits of wood with a small chainsaw, then pry it apart off the hook/screw/nail. It did seem like a spot a clothesline might have been hung - quite a short clothesline, but there were no other trees near the house that were more suitable for one. (It was a straight nail. I bent it to pull it out of the wood. Naturally I had been trying to rip the nail down its center. If I had just been cutting through one section, the saw might have actually got through it. ...not without lots of damage to the chain anyway.)
   But if there were more pieces of metal, or if I went a bit too far, I would just wreck another chain. Should I gamble it was just one clothesline hook? How much is that wood worth? There was still a lot of it. I could cut off 4+ feet and have a less than 8 foot log, but I still might hit another piece of metal. I could cut off 6 feet and have a 6 footer - not the most useful length boards! That would be fairly safe. I could turn the log over and start milling again on the other side?

Pulley and Link Belt Selections

(13th) I was sorting pulleys and I ran across a 2 inch V-belt pulley. Then I realized the one I thought had ordered as a "1 inch pulley" was really 1.5 inches. So it would seem that from the 5000(?) RPM skillsaw, the 1.5" pulley turning a 9" pulley (6 to 1 reduction) made the band run too slowly. Also it was too small for the V-belt to get a good grip on, so it slipped when the going got tough. I estimate the variable pulley I'm using is set to about 2.5 inches - it seems still larger than the 2 inch pulley. (I may try the 2 inch pulley - it has the right 1/2" bore so it'd be easy enough to try out.)


2 inch and 1.5 inch V-belt pulleys. The 1.5 was held on the skillsaw by that bolt.

   I also note here that when I went beck to the larger pulley, I also soon went back to the link belt. It's much better than the regular V-belt. I don't understand why they don't make regular V-belts with grooves in them like the "variable transmission" V-belts. Those would have to be better too!

Moving Along - Setting Teeth

(17th) Finally got back to it. I sharpened the band. But it didn't go well. The saw was binding in the cut. I only cut one 6" wide by 12' long board, and it was an inverted bowl across and not very straight lengthwise. I tried to cut a second, but the cut curved up badly and I gave up after a foot or so. The next day I gave myself a blister bending a 'set' into the teeth with a small pair of pliers. (That feels hard on that finger... well there's only 279 teeth, keep going and finish it up!)
   It was night and day difference! A thick stream of sawdust sprayed from the saw, which cut freely at about 6 feet a minute. I cut 3 more boards in no time. After that it wasn't quite as sharp and the cutting slowed down, but I still did 9 boards in well under an hour, finishing the whole cant.

Moral: A tool working right works WAY better than one that isn't. Also every time the band is sharpened some of the 'set' is ground off, and it needs to be re-set after 2 or 3 or 4 (?) sharpenings.

Next Cant

   Having had my chain re-ground (losing 1/2 its life) I cut in from the other end of the log until I got near the metal thing. Then the log kind of went "clunk" down on the cut and jammed it. I wedged it open and pulled the saw back out. There was still over an inch of "hinge".
   I filled the gas tank, then I cut another 6" thick slice. This one seemed to go well until about 9 inches from the end. There the saw ran out of gas. Hard to believe - really? Normally I'd have stopped it at the first sign of a sputter, but it was hard to believe it could be empty, and with such a short distance I thought it might make it. Not another inch!
   I refilled the tank, but I wore myself out trying to start it again. Typical! I hate gasoline powered appliances more and more the more I use them. I know the best way to get this saw to start again if it doesn't start with a few pulls or with removing the spark plug and giving it a few cranks is to leave it a week. Then it will usually start right up again.
I backed it off in the cut and cut to the end freehand with the electric Dolmar chainsaw. The cuts weren't very straight. Anyway I had - at long last - reached the bottom piece of the last section of the last tree cut down in 2017. The rest was bandsawing. But I didn't get there for the whole rest of November!

Skilsaw Chain Beam Saw

   Someone sent me a picture of a chain attachment for a handheld circular saw, for "beam cutting". I'm not sure how practical it looks, but as it's related, here it is.
   One interesting feature is that like my bandsaw mill, it uses a skillsaw as a motor to power something else.

http://www.praziusa.com/12-beam-cutter-model-pr-2700/





Electricity Generation


5 Blade Windplant - Windplant Ducts

   Until the morning of the 6th no wind had blown to turn the windplant since October 23rd. Then it made a few watts for a little while before stopping again. In the evening it picked up and the next day there was a fair blast for a while, creating figures of 45 to 65 watts. It was dark overcast and that was almost as much as the DC solar power was making.
   In fact, while typically things were quiet, there were quite a few days with wind in November, and I would often see the MPT7210A charge controller putting out 3, 15, 20 or 50 watts in the wind. Occasionally it would hit 60s to 70 and beyond. On the 13th it was blowing all day, with 40 to 65 watts being the most common readings, and it went just over 100 for three brief readings. And it would do it at times of day when there was little or no solar collection. So it is by no means a useless appendage to the energy system.
   I don't notice the noise much. I do think it's quieter than a 3-blade windplant. When the wind is blowing the waves are crashing in and the level of noise is pretty high anyway. (This place, for all its virtues, is a bit close to the sound of the ocean - a roar on windy days - for my taste.)
   On the 13th it all seemed worthwhile. Maybe I should do some sort of venturi surround for it after all and get some higher mileage out of it? But there wasn't much wind for the last 10 days of the month.

Clarkson University Experiments with Windplant Ducts

   Of course I'm not the only one to think of improving the performance of a propeller type windplant. At Clarkson university they have been experimenting and found they could literally double the performance simply with a flaring exit to the fan, similar to the one I drew last month as the exit to the venturi duct.


"Ducted" wind turbine (2017): just a short flared exit "duct" doubles power -
without doubling the price. What would a whole venturi duct do?
(On the left wall louvered fans blow controlled air at the unit under
test. Many of us only dream of having such test facilities!)


Other windplant duct experiments by Clarkson.
Putting together the flared exit of the previous photo
and the "back end of a venturi" configuration of the
right side unit, they'd probably hit around 2/3 of the
effect of a complete front-to-back venturi duct.
(Or, they just might be doing better than I think
with those clever vane arrangements.)

https://www.youtube.com/watch?v=EXxA-RkwuRY



My Solar Power System

Water Heater, Sprint Car & Lithium Batteries

   I noted last month that I had put a switch on the water heater, hoping I could leave it on on dull days. Opening the switch puts the third heating element in series with the two already in series, bringing the watts down from ~250 to ~160. When I did that the weather doubled down and got duller and cloudier so that not even 100 watts was being produced at midday. With the very weak NiMH dry cells battery, turning on the water heater in either mode quickly brought down the voltage.
   By the 7th I was tired of nothing more than luke-warm water and a sink full of dishes.

   Last month I had made a DC/solar charging circuit to charge lithiums without danger of any one cell hitting too high a voltage, and connected to some 40 amp-hour blue colored lithium batteries I had bought used about 4 years ago. I had always tried to keep them charged up. Now at the end of October I tried to power the solar system with them, and discovered that 3 of the 14 cells were useless. They had voltage until one tried to use them, then they instantly went to zero. A fourth cell was very weak. So there weren't enough good ones to make 36 volts. This was unexpected. Perhaps they were bad when I bought them? I might contrive to use some completely different lithium cells in series to make up the last two, but for now it seemed impractical to try and use them.

   Other than the inaccessible 24 KWH Nissan Leaf battery, that left the ~11400 rated watt-hours of lithiums in the Sprint car, at 36 volts. I thought I had made everything so I could connect the car to the solar power with 70 amp Anderson connectors except the final "extension cord" to connect them together. Connector in car... Connector to panel... presently plugged into the NiMH batteries.

   I now made that cord, unplugged the increasingly unsatisfactory NiMHes and plugged in the car/lithiums, and turned on the water heater with the full 260 watts setting. The lithiums quickly dropped from just over 40 volts to 39-1/2, but only gradually continued down to 39.14. The next time I checked, the water heater had shut off and it was back up to 39.68 (eventually to 39.81 and overnight about 40 volts). Time to wash dishes!

(8th) Even with high capacity batteries that actually work one can't go on too long without recharging them. For charging from 120 VAC I had done the first two 6-volt chargers each on 2 series cells under the hood last month. (putting out a constant 7.0-7.2 volts) Since I hadn't yet done the rear, I connected my two Canadian Tire 'trickle' chargers on the two 12 volt sections of the 36 volt total in the rear of the car, ran a 120 VAC extension cord from the front to the back (under the trim and next to the doors), and put an AC power bar in the back. I plugged in all the chargers and put an AC power meter on the main cord.
   The two old chargers in the rear drew just 15 watts each - hardly over 1 amp at 14 volts. The new front ones started out at 135 watts, indicating charging at almost 10 amps, and gradually dropped from there. That's faster than the water heater discharges them. It's hardly a guess which 12 volt section will be charged first! After a while (well over an hour), the front was down to 45 watts while the rear two were still at 15 each.
   At first I still turned the water heater off at night, and I unplugged the chargers to the front of the car. I note that these are not very well regulated - the voltage changes over time probably with temperature. Furthermore the knob is much too easy to turn accidentally an hard to get just where you want. Adapters with screwdriver 10 turn trim pots - and better regulated - would be much better if they can be found. A plus of these ones is that the AC input can be 100 to 240 volts. That means they could be plugged into a 240 volt J1772 car charging station - although there is no need or use for plugging them into more than a regular 120 volt outlet. (With 5 amp chargers at 36 volts it won't be a fast charging car by any means. 10 or 20 amp power adapters would be better if I could find them.)

Sprint Car & Solar Lithium Battery Chargers

(Now, should this be under batteries, solar power, or electric transport?)

   At this point, I had done nothing to recharge the lithiums from the solar. I decided I should put an MPT7210A right into the Sprint and connect it straight from the solar panels. (a second, lighter extension cord.) That would be compatible with my DC lithium charging circuit, and when I put the long-planned solar panel on the car roof, it would be all ready to hook up. And since I had determined that slow-charging 12 volt sections from solar was safe, I would use just 3 DC to DC converters for the 36 volts. As I thought about it, I would use two of them and adjust the 7210 to the exact voltage for the uppermost section, 40.2 volts, from which the middle and bottom sections would be subtracted leaving 13.4 volts per section.
   How to connect? I wasn't going to hard-wire any part of a car to a stationary wire. The solar panels were about 36 volts... How about my 36 volt HAT connectors? I decided a 15 amp one should be adequate. I had a couple of the 3D printed shells.

   I took an absurd amount of time to wire up the plug - a couple of hours. I had to make the pins by flattening a piece of #10 AWG wire and notching and drilling them per TE News 129, and then soldering the wires to them. It didn't go smoothly. I pulled out the wrong size pins in the drawer but didn't realize it. Redo. For some reason the first drill I tried simply wouldn't drill through copper. Then the holes were too small for the wires. I soldered a wire on the wrong side of a pin and it mattered: redo. I dropped a box of 100 tiny screws on the carpet. They wouldn't go through the holes in the plastic shell so I had to drill those out bigger.
   I'm convinced now that the thing to do is to do injection molds for CAT and HAT plugs and molds for putting clay in to make porcelain sockets, as well as molds for wall plates and everything else needed or useful, and to make stamps that will stamp out the various pins and receptacles... and to manufacture and sell them to all the places that sell solar and low voltage equipment. Maybe Amazon? That way when anyone - including me - wants to do 12 or 36 volt wiring it'll be simple and straightforward. Somehow no one else seems to have done anything like it yet. (If they did and used my pin size and spacing specs, I could just buy them and my problem would be solved.)

   Next I thought about the steering diodes. The "car alternator" diodes lost over .8 Vf. Recently I had uncovered some 60 volt, 30 amp Schottky rectifier diodes I had lost track of. (And I now put all my diodes into some plastic part drawers all in one place.) These had actual part numbers so I could look up the specs! They only lost .49 V max at 15 amps - only about 60% as much wasted power - and even less since the MPT71210A only puts out max 10 amps.
(15th) I soldered four of them to little pieces of copper to screw to battery terminals at the 12, 24 and 36 volt levels, with the anode inputs from the DC to DC converter outputs. I mounted the MPT7210A in the back of the car and ran a ground wire from under the hood to the back.

   That was about as far as I got. Owing to the time it took to make a plug, I started thinking about making them easier to make, and about how to produce them for sale. I enlisted Mike's help and we at least came up with easier to make and use pins for the plugs and almost equally simple "Z Fold" receptacles for the sockets. The small writeup on that process is in "November in Brief".

Month of November Log of Solar Power Generated [and grid power consumed]

(All times are in PST: clock 48 minutes ahead of sun, not PDT which is an hour and 48 minutes ahead. DC power output readings - mostly the kitchen hot water heater - are reset to zero daily, while the others are cumulative.)

October 31st    2.98+.24, 888.72 => 2.78 KWH [[email protected]:30] mostly cloudy, a bit of sun in the afternoon.

November

Solar: House+DC, Trailer  => total KWH [grid power meter reading(s)@time] Sky conditions
  1st    4.00+.19, 889.33 => 1.82 KWH [[email protected]:00] light overcast.
  2nd   5.41+.56, 890.50 => 3.14 [55Km.chj.car; [email protected]:00] cloudy AM, sunny PM.
  3rd    7.20+.42, 891.87 => 3.58 [[email protected]:00] Mostly sunny.
  4th    8.08+.39, 892.65 - 2.05 [[email protected]:30] cloudy.
  5th    8.79+.11, 893.28 - 1.45 [[email protected]:00] cloudy.
  6th    9.11+.19, 893.54 - 0.77 [[email protected]:00] Clouds rain and light winds (2-5 watts from windplant)
  7th    9.82+.72, 893.84 - 1.73 [[email protected]:30] Dull, windy (up to 45-65 W at one point from windplant)
  8th  10.86+.50, 894.76 - 2.46 [85 Km,car charging; [email protected]:00] Mostly overcast.
  9th  12.52+.05, 895.75 - 2.70 [60 Km,charging car; [email protected]:30] Largely Overcast, a little sun.
10th  13.65,(.61*),896.58- 1.96 [[email protected]:00] Mostly overcast. (*Recharging lithiums on 36 volts system from wall not solar, so it doesn't count for solar collected.)
11th  14.95,(.95*),897.41- 2.13 [[email protected]:00]
12th  16.40+.33, 898.37 - 2.74 [[email protected]:00]
13th  16.77+.35, 898.87 - 1.22 [[email protected]:30] (there's some wind power in the DC .35, .62 KWH figures)
14th  18.00+.62, 899.69 - 2.77 [55Km,charging car; [email protected]:30] Some sun.
15th  19.80+.07, 901.07 - 3.25 [85Km,ChjdCar; [email protected]:30] Quite a bit of sun. (But quite short days now!)
16th  20.16+.15, 901.33 - 0.77 [55Km,Chj; [email protected]:30] Clouds & rain.
17th  21.37+.42, 902.33 - 2.63 [[email protected]:00] some clouds, rain, sun.
18th  01.83+.37, 903.58 - 3.45 [[email protected]:00] mostly sunny (power fail during night)
19th  03.47+.30, 905.28 - 3.64 [[email protected]:30] Sunny!
20th  04.74,(1.28),906.14-2.13 [[email protected]:00] Some sun. I ran out all the hot (hardly warm) water in the sink because it stank, until it ran completely cold. Then I turned on the heater and later with water that didn't stink I did a good load of dishes. Thus the high usage - which came mostly from the car lithiums, from the power grid. I turned down the temperature a couple of times, but if the water is in the tank a couple of days, (ie if I don't use most of it, daily) it starts to stink. (sulfur?) I know the well water is mineral rich, but why does the kitchen hot water smell, but the water in the main hot water tank tank never does? And I con't remember any smells last summer when I put the tank in.
21st  05.64+.35, 906.99 - 2.10 [[email protected]:30] Mainly sunny, rained a bit anyway (sun from way south, rain from above.)
22nd 05.68+.01, 907.01 - 0.06 [85Km&Chjd; [email protected]:00] Clouds & Rain. Holy Nothings! Could a day's collection get any closer to 'gooseegg' than that?!? From a full 15 KWH in summer that's ~.04%. One could do 100x better in space way out near Jupiter! PS: One can see I didn't turn the kitchen water heater on this day - just lights on the DC power.
23rd 06.51+.01, 907.52 - 1.35 [55Km&Chjd; [email protected]:30] Mor cloudz and rain. Left KHW off again.
24th 08.21,(1.04),908.71-2.89 [[email protected]:00] Sunny, rain before dark. Left KHW heater On, on the car/lithiums.
25th 09.77,(.60),909.88 - 2.73 [[email protected]:00] Mostly sunny.
26th 10.58,(.62),910.81 - 1.74 [[email protected]:00] Snow, then sunny.
27th 13.71,(.92),912.35 - 4.67 [[email protected]:00; 55Km,ChgCar] Sunny. (Extra sunny - No jet trails)
28th 15.33,(.14),913.18 - 2.45 [[email protected]:30] Mainly sunny.
29th 17.33,(.18),914.60 - 3.42 [90Km,Charging; [email protected]:00] Sunny. (Sky full of jet trails by sunset)
30th 18.12,(1.11),915.01-1.30 [55Km,Charging; [email protected]:00; 50Km,Chj.] Not sunny.

December
01st  18.84, (.71), 915.41 - 1.12 [[email protected]:30] Even more not sunny.
02nd 19.58, (.63), 915.84 - 1.17 [55Km,Chjd; [email protected]:00] Clouds and Rain.

Daily-
KWH-  # of Days (Nov)
Made
0.xx  - 3
1.xx  - 8
2.xx  - 12
3.xx  - 6
4.xx  - 1
5.xx  -
6.xx  -
7.xx  -
8.xx  -
9.xx  -
10.xx-
11.xx-
12.xx-
13.xx-
14.xx-
15.xx-
16.xx-

Monthly Tallies: Solar Generated KWH [Power used from grid KWH]

March 1-31: 116.19 + ------ + 105.93 = 222.12 KWH - solar [786 KWH - used from grid]
April - 1-30: 136.87 + ------ + 121.97 = 258.84 KWH [608 KWH]
May  - 1-31: 156.23 + ------ + 147.47 = 303.70 KWH [543 KWH] (11th solar panel connected on lawn on 26th)
June - 1-30: 146.63 + 15.65 + 115.26 = 277.54 KWH [374 KWH] (36V, 250W Hot Water Heater installed on 7th)
July  - 1-31: 134.06 + 19.06 + 120.86 = 273.98 KWH [342 KWH]
August 1-31:127.47 + 11.44+91.82+(8/10)*96.29 = 307.76 KWH [334 KWH] (12th panel connected on lawn Aug. 1)
Sept.- 1-30: 110.72 + 15.30 + 84.91 = 210.93 KWH   [408 KWH] (solar includes 2/10 of 96.29)
Oct.  - 1-31:  55.67 + 13.03 + 51.82 = 120.52 KWH - solar [635 KWH - from grid]
Nov. - 1-30:  36.51 +   6.31 + 26.29 =   69.11 KWH - solar [653 KWM - from grid]

9  month total March 1 to November 30: 2044.50 KWH made; [4683 KWH consumed from grid]

Things Noted

* At some point in October I gave up trying to charge the car at 1500 watts during the day in order to use power from the solar, and just used the 3800 W charging station. There just wasn't enough of it for it to help. What solar power there was would have been taken by other loads around the house anyway - easily by the 250 W travel trailer heater and the 175 W(?) LED grow lights for the indoor winter garden. And the sunny hours were too short to bother even if there had been enough power during the day.

* On the 10th to 13th (and various other days) I didn't drive anywhere. This is reflected in markedly lower electrical consumption on those days.

* Owing to the low winter collection I looked up Stirling engines (see article below). I think I can make a better one for the woodstove, "sort of" a low pressure steam engine! A generator would be attached. 100 Watts?



Woodstove Alternator with Stirling Steam Engine?

(apologies for poorly written rambling article. I want to get this newsletter out tonight!)

(22nd) In the winter weather and low sunlight, and with the windplant only very occasionally firing up, I thought again occasionally about using the woodstove for generating electricity, and how the seemingly promising TEGs (thermoelectric generators - sort of a special Peltier module - had in the end seemed like "not much bang for the buck" and I didn't build it. I thought that boiling water on the stove and running a steam turbine could give a heck of a lot more power. But realisticly the top of the woodstove is rarely hot enough - unless the steam was in some sort of closed loop in a partial vacuum to reduce the temperature water would boil at. That would make it a very complicated project.

I have also thought off and on about Stirling engines, also called "external combustion engines", invented by Robert Stirling in about 1816, after Guillaume Amontons' "Fire Mill" hot air engine in 1699. (over a century earlier!) They run off any amount of heat, in fact off the temperature difference between one end and the other. I didn't understand how they worked, and owing to the mechanical difficulties of constructing a piston sliding in a cylinder they didn't seem like anything I wanted to get involved with. (I was having more than enough mechanical trouble with building "too flimsy" or "insufficient" torque converters.)

   I wandered off to youtube and found that "MyfordBoy" had made a couple of very elementary (not to say "Mickey Mouse") "Beta-type" Stirling engines. They spun madly, admittedly with no load. To do that with just a candle for power, they must have very low friction and be quite efficient. An old CD was the flywheel. Furthermore, his videos showed just how to make them from start to finish. Surely even I could make one!
   The most notable feature was that the power piston was replaced by a diaphragm. The material was latex from single use gloves, or balloon rubber. This seemed to me to be the key: no piston with rings to make friction. That was why they worked well - not to say why they worked at all. Technically the round piece of wood in the center of the diaphragm with the rod connected was still a piston. It just had the diaphragm for a seal instead of rings sliding up and down in the cylinder. It made a perfect seal that was also simple to make with no high precision parts required. Cool!

Stirling Engine With Diaphragm Instead of Piston:
https://www.youtube.com/watch?v=klbBn_3ufV8

   Could this clever design be made in a larger version with some real power, say with a car tire inner tube for a larger diaphragm? With an aluminum base to sit on and take heat from the woodstove, and an aluminum heatsink radiator up at the top, similar to the animation from Wikipedia? This seems to me to have great potential to turn a generator.
   I figure the target is around 50 to 100 watts with a fair fire in the stove. (I hope that's not too optimistic.) This is much better than a 100 watt solar panel because it would run all day and night: eg, 2.4 KWH of power per day from 100 watts. (That's more than I'm getting many days in midwinter from a dozen solar panels total rated 3.4 KW.) It should also be sufficient to charge a battery to power a few LED lights and run an inverter to power a freezer or fridge in a power outage.

   So I conceive of something akin to a big "coffee percolator" (anybody remember those?) sitting on the stove, with fins around the top and a substantial flywheel spinning above it. The bigger the thick aluminum base, the more heat it would collect from the stove. The bigger the heatsink fins, the better it would dissipate the heat at the top end and together the more power it could make. The moving parts had to be large and robust enough to make good use of the temperatures thus provided.
   MyfordBoy's Stirling engines, like the others I looked at, had to be started with a spin by hand. The flywheel then carried them through the non-powered parts of the stroke. Self-starting would be a plus, but since one has to feed wood into the stove anyway, giving it a spin would be an extra little thing to do after lighting a fire.

   Then the generator could either be a separate unit, or more simply the flywheel itself could have magnets on it and there could be some coils installed in something around the outside. With the materials I already have, I would just make the flywheel an axial flux magnet rotor and make it a highly efficient Pigott type alternator.

   I could actually see this being a viable commercial product! Why does no one seem to have done it this simple way before? I guess the way to find out is to build one and try it, and see how well it works!

(29th) I occurred to me that the way to make the big cylinder was a flat aluminum plate welded to an aluminum ring. Then it occurred to me that sounded like the bottom of a pan. I looked around and found an old aluminum pressure cooker. I'd never used it but still it seemed like a waste to cut it up. I decided to try the thrift shop the next day.
   But then it occurred to me that it would be even better if the heat from the bottom wasn't being conducted through to the top. What would one use as the edge material if not metal? And how would it be bonded to the bottom plate?

Wikipedia Image Caption:

Beta-type Stirling engine. There is only one cylinder, hot at one end and cold at the other. A loose-fitting displacer shunts the air between the hot and cold ends of the cylinder. A power piston at the open end of the cylinder drives the flywheel.


How it works
------------
Referring to the animation, there is the power piston (upper moving part), and the displacer (lower).

0 degrees rotation: The displacer rises. This means most of the air in the cylinder is at the bottom, where it is warmed. Warming air must either expand or the pressure will rise. First the pressure rises.

90 degrees: The increased air pressure pushes the power piston (or diaphragm) upward and the warmed air expands, losing pressure and returning to 'medium' temperature as it does so.

180 degrees: The displacer now moves down to the bottom. Most of the warmed air is now above the displacer and exposed to the cooling fins. As it cools it must either contract or the pressure will fall below the outside air pressure. Initially the pressure reduces.

270 degrees: The relative vacuum of the contracting air pulls the power piston downward. The air volume shrinks and the air temperature goes back to 'medium'.

The two arms and the rotating wheel keep everything in sync. The wheel is also a flywheel, keeping everything in motion through the unpowered parts of the cycle.

Design features suggest themselves. The larger the hot bottom area, the faster the air will heat and the greater the power will be. It probably doesn't need to be very tall, so a cylinder and displacer much more short and squat than the animation - just as MyfordBoy's are configured. The displacer should be as lightweight as possible, perhaps a hollow "tuna tin" sort of shape only bigger. The "power piston" diaphragm and the cylinder it's in(?) needn't be anything like as big as the lower part. The smaller it is, the faster and farther the rising and falling air pressure will move it and the higher the RPM. Larger will be lower speed and more torque - but the same theoretical power, since the power comes from the heat from the stove and the cold from the radiating fins.

A late thought came to me at 2:30 AM on the 25th (or is that an early thought?): The top surface of a woodstove often operates right around the boiling point of water or a little over it. If a little water was placed in the cylinder, then while the displacer was up, the water might boil (a little or a lot) in the bottom. This phase change would cause much more pressure than just expanding air. Then when the displacer was down, the steam would condense on the radiator and fall back down as water drops, causing a lot of contraction and vacuum. It would have much more power for its size. Maybe that would make it some sort of steam engine after all! Some details have to be thought through...

   Trust me to start out with the idea of just copying something, and to soon be wondering if and thinking how significant improvements might be made to the whole idea! What's the real point of taking on another project if it's not going to be a "world's best" or "world's first" or something? Other than to have a reliable source of winter electricity, of course.

   By the 27th my idea was that it would be a very shallow cylinder with a displacer like a pancake. The whole bottom would be sitting on the woodstove, and much of the top would have cooling fins. There would be a slot in the mechanism to give it a little slack. The idea of that would be that it would hit the bottom of the short cylinder, pushing all water out of the way, and sit there for part of the stroke. Thus the water would be out of contact with the woodstove surface for a certain period, say 1/4 of the cycle, and would be cooling and condensing, sucking the power "piston" down.
   When the displacer lifts off the bottom, the water rushes in and starts to boil, The expanding steam and air pushes the power piston out again. With a spring the displacer could be made to rest at the top of the cylinder for 1/4 of the cycle, blocking access by the steam to the cooling fins.

So the cycle is:

1. The displacer reaches the bottom and sits there, covering over the heat plate from the woodstove, and no more water or air is heated. The steam and air cools on the surface inside from the cooling fins on top and the steam condenses, dripping down onto the displacer.
2. The cooling air/steam vacuum causes the power piston/diaphragm to pull down while the displacer still rests on the bottom.
3. The falling piston turns the flywheel until the displacer starts to rise. It goes to the top as the piston crosses the bottom of its stroke.
4. With the displacer at the top, the steam has no access to the top surface just inside the cooling fins, but the water at the bottom rushes in and boils, heated by the hot surface sitting on the woodstove.
5. The boiling water/steam/air expands and pushes the power piston up.
6. The rising piston turns the flywheel until the displacer starts to descend. It goes to the bottom as the piston crosses the top of its stroke. Then the cycle repeats.

(28th) In fact, why have the displacer be driven by the flywheel at all? Ideally it's either at the top or at the bottom all the time. It could be lifted by activating a solenoid, with the switch on the flywheel. Or else, it should go quickly to the top or bottom at the half way point on the flywheel and sit there for almost 180 degrees of flywheel/power piston travel. If mechanically driven by a cam on the flywheel, a slot instead of a hole would allow it to rest on the bottom, but a spring would have to be used to allow it to rest on the top as the cam continued upward. Actually for both those reasons, the displacer and its mechanism should be as light as possible consistent with being able to keep up to whatever RPM is attained, since lifting it, and compressing the spring toward the top of the stroke, both require work from the piston.

   The displacement cylinder with its bath of water need only hold a small volume. The displacer would be a thin pancake and the cylinder would be a thicker hollow pancake that holds it. Most of the steam would be forced to move into and out of a smaller cyclinder above it with the power piston or diaphragm in it. It should also be largely a heat insulator rather than a heat conductor. Using a piece of 1/4" plywood in the prototype suggests itself.

   As I think about it, any Stirling engine would work better if its displacer sat at the bottom or top through most of the stroke. A down side might be more noise from the rapid (even sudden?) stops and starts of the displacer instead of a gentle sine wave motion?

   Just what I needed, another project! I talked about it with my next door neighbor. He's a great project doer, and he might be willing to do the work and make one for himself if I show him what to do. And there's another person in Victoria for whom this sort of project would be just up his alley. If I do a few experiments and most of the designing... and probably make the actual generators... these people might like the fabrication project. But it seems worthwhile. Millions of people who burn wood could make sufficient free power for their fridge, LED lights, charging devices and batteries and so on in winter if they're off-grid, or during power outages.

   Before sending this newsletter I had collected an aluminum pressure cooker from my own storage as a test cylinder that would hold pressure, and a 10 inch aluminum frying pan with straight sides - a perfect shallow cylinder - from the thrift shop, and some "happy birthday" balloons.





Electricity Storage (Batteries)

Turquoise Battery Project: Long lasting, low cost, high energy batteries

   I did a few little things, but mostly I gave battery making a rest for November. I was getting a bit burnt out on them and I needed to do some other things - finish milling my spruce and do a bit on the other neglected projects.

Zinc Protection

   Near the start of the month I ran a few more tests on the nickel-zinc cell, and added some salt. It didn't have very much capacity or current and the salt didn't do much for it. Then after a few days, it quit working entirely. It seemed familiar... I could only think of one thing that would cause such a sudden failure.

(17th) I [finally] pried the cell open. As anticipated, the zinc terminal tab had turned to brittle zinc hydride up at the surface of the liquid where the hydrogen bubbles would collect, and finally broken off. In fact, 3/4 of an inch of it was missing, with a chunk lying loose inside. Of course: the zinc tab had no jell and no zircon to raise the hydrogen overvoltage to prevent gassing.
   I had painted the tab with polyurethane spray paint to protect it when I made the electrode, but I noticed when I opened the first cell it was in that the paint was disintegrating in the alkali and loose chunks of it were drifting around. (Just a couple of flakes of red remain.) It would seem protecting the zinc is indeed necessary, but that PU isn't the thing to do it. Perhaps the tab needs to be electroplated with some other metal - maybe lead? Tin? Copper? Or dip it in melted plastic? In epoxy? I'll try a couple of other paints first.
   (What was it my brother said on the phone again? He knows a lot more chemistry than I do. But not specifically electrochemistry - more organic chem. I remember when I tried to make a transparent acrylic battery case and it swelled and cracked. To him it made sense that polyacrylic acid (PAA) would become a gel in alkali.)
   Now that I'll be making a new electrode and a new cell, maybe I'll try that Minwax Polycrylic that I bought for the ground effect craft. Acrylic is something different than polyurethane. Maybe it'll form a protective gel?

   But there may be another factor at work here... I've been using "roof moss killer" zinc because it was the only sheet zinc I could find in Victoria. I've wondered how pure it is, but I've managed to keep that thought out of my mind for a long time now. But why is there black crap when I etch it, and black crap left over from a sacrificial zinc electrode when I'm electroplating? Whatever it is, it says it can't be pure zinc.

(27th) I got on line and found some "high purity zinc sheet" from 3 different stores on Aliexpress. First 20 pieces that were 100x100x .2mm. Then I thought it would be nice to have some bigger ones to make larger cells, and ordered another five sheets, 140x140x .2mm. That was as big as they came. Then I decided as long as I was experimenting to try some that was even thinner. It came in a roll 100x1000x .1mm. That's probably too thin, but I'll see if it gets holes in it or has any other problem. It all came to almost 50$c. I hope there is a cheaper wholesale price somewhere for manufacturing. It's likely to take the better part of a month to arrive. But it just *might* solve the problem... or maybe even 2 or 3 problems.

Case Sleeve

   One thing about gluing cases closed: it made it hard to get in and see what was happening. Busy as I've been, if a tab had broken off in an open cell I'd have seen the problem right away. In fact I'd have seen it getting brittle beforehand. I got the idea to make a close fitting ABS box that the whole cell slid into. Not only would it not matter if the case leaked, but maybe I wouldn't glue the cover shut.
   I cut the plastic pieces, but I didn't get it glued together. Just as well as another idea came to me. - read "Layout" farther down.

Positive Electrode

   The nickel oxides 'trode hadn't performed very well, in spite of being an old and well known chemistry. It shows that it is my technique that must be in some way faulty. It did however work, equally in straight alkali (KOH) and with salt (KCl) added. That was to be expected from the pourbaix diagram, but after seeing that lead oxide didn't behave anything like I expected in alkaline pH, it was an experiment that needed to be run. And it now seemed to be verified.

   For the poor performance, poor conductivity comes to mind. It only had 5 wt% conductive carbon black additive. Doubling it seems like a good thing to try.


My NiOOH electrode at 200x.
The open side surface is a bit lumpy,
and a few fine carbon fibers protrude.
The side against the current collector was smoother.
There is a bit more light color than in the com-
mercial electrode, which I attribute to the gel.


The smoother NiOOH electrode from a Tenergy NiMH dry cell at 200x.
The fine grains looked pretty similar.
(Rats, how did both photos of it turn out so fuzzy?)


The Tenergy metal hydride electrode at 200x for contrast.
The surface was against the fibrous separator sheet.

Calcium-Copper Oxide

   And then there's the calcium-copper oxide current collector idea. Copper sheet is a much better conductor than graphite sheet, and even with an oxide film on it it might be a substantial conductivity improvement. The other (MnO2-Zn) cell had sat idle for a month. The voltage of MnO2 was high enough to oxidize the copper sheet behind it. Now I took it apart too, to see what it looked like. (Again if it wasn't glued shut I'd have done so a while ago.)
   The copper tab was corroded off. But interestingly, it was mainly where it went through the slot into the upper part of the cell. Possibly all the calcium was scraped off it at that point during assembly, leaving unprotected copper. And then oxidation had eaten in from the bad edges thus created. With the MnO2 electrode brick removed, there were also a couple of areas, "lines", of black copper oxide next to the edges near the bottom of the copper foil. But on the whole the main surface looked uncorroded, front and back. That seemed very promising.

   So it looks like the idea might work and make a higher conductivity cell, but it needs further study. Perhaps there is a method to fully and completely coat the copper sheet with calcium before inserting it into the cell?


Or... Gold Plated?

   If that doesn't work, there's something that has a very good chance: gold electroplate the copper. The reaction voltage of gold to gold hydroxide (Au(OH)3, ~+.6V @ pH 13) is higher than Mn2O3 to MnO2 (+.25V). So a gold plating wouldn't oxidize. That's assuming I'm using nickel manganates because they should recharge properly, and that their reaction voltage is that of manganese dioxide. (Or else that I've found a way to properly recharge to MnO2.)
   The reaction voltage of gold to hydroxide is about the same as nickel hydroxide to oxyhydroxide, so it would probably oxidize in a NiOOH electrode. But gold is such a heavy element this oxide might well remain solid and bonded as a one molecule thick layer protecting everything underneath - just like with titanium or aluminum (at neutral pH), and nickel at pH 14.
   Gold costs money, but it doesn't have to more than angstroms thick. (What's an angstrom again?) (If it was cheap one might have a gold-oxide/metallic gold powder-paste electrode and the cell would be a solid 1.7 volts or so with amazing current density. I don't even want to try that out - what would I do if it worked really well? It is known to work well with copper-oxide/copper, gold's column 11 relative on the periodic table, but with copper the reaction voltage is so low that even with zinc for the negative, the cells are less than one volt.)
   Now... where was my old bottle of Caswell Plating gold electroplating solution? A fairly diligent search didn't uncover its hiding place, but the next day (27th) I had another look and unearthed it in with cans of spray paint.
   Now... would it still work, or did I need to get something new? There were some almost microscopic bits of gold around the rim and on the bottom. With a stainless steel anode the brush plating didn't seem to work. (28th) I found a tiny piece of gold, soldered it to a piece of copper wire as a rod, and tried with that. No difference. But some very few little streaks of dark color had appeared with both.
   One last thing... I tried heating the solution while stirring, to maybe 70°C. Then I tried again witht he gold anode. A few more dark streaks. I poured the 30 cc of liquid into a beaker and tried plating in this bath, again with the gold anode, at a very low current. This produced a more uniform dark coating.
   I finally decided that the dark coating was probably the gold, but that it was so thin I was also seeing the copper underneath, and they were subtracting colors. Even when the solution was new I had had a hard time getting a thick enough coat to make it really look yellow. "A few angstroms thick", I said. Was it enough to protect the copper underneath? Or did it mean there was some gold and some bare copper? Once there was any hole in the coating, the copper would gradually dissolve in the cell.
   Or maybe the combination of gold oxide (if it formed) and calcium hydroxide would react together and form the desired non-corroding surface on the copper?

   I left the piece in the plating solution overnight, but it didn't change color. I finally decided the coating probably wasn't really gold but some impurity, and that I should get some real tank immersion gold plating solution. Let's get enough gold on there to make it at least vaguely yellow! I ordered it on the 29th, again from Caswell Canada.


That purple-black color don't look much like gold to me.

   I thought shipping said "Fedex". (There was no other option. What have people got against the post office? I'm sure it's much better than it used to be. I haven't lost anything in the mail in about 13 years.) The receipt however said "UPS / Purolator 1-4 days depending on location". If they sent it UPS I may not get it in 2019!

   The alternative to finding a sheet metal or coating/plating that won't oxidize of course is to use sheets of graphite. That has higher surface resistance and hence lower current capacity than a metal. But it does work, and it's not entirely incompatible with the flat cells idea if the corner tab bends up gradually as a ramp instead of making a sharp fold straight up.

Layout

   I've been making skinny batteries that always want to fall over.
   It has occurred to me I that if the current collectors are metal, I can have them lie flat. The connection tabs can be bent upward to come through the case at the top. In fact, the corner can be slit and a tab bent up without needing any extension to the square or rectangular sheet of metal. If the bottom electrode was slit and bent one way and the other the other way, that would just leave a little gap with no electrode in the corner where both terminals are.
   And that could form a little reservoir in that corner, where one could open a top plate or cap on to check the level and add a little water into. And as long as the cells aren't to hold pressure, there's essentially no size limit.
   They could for example be 12 by 12 inches. That would give about 850 square centimeters. If one could get 200 mA/sq.cm, that would be around 170 maximum amps. That's for a single layer cell less than 1/4 inch tall inside, which might also hold around 30 or more amp-hours depending how thick the electrodes are. A stack of 8 cells for 12 volts might be 3 or 4 inches high.
   (It's tempting to think of sticking the electrode plates themselves to an edge frame so they form the top and bottom of the cell, instead of having a plastic top and bottom covering them. That would be great if and only if one could keep them from leaking around the edges. I think I won't try it.)

(27th) I trimmed my idea down to about 7 x 7 inches for now, realizing that that's about as big as the 3D printer can handle. Still somewhere around 50 maximum amps and 10 amp-hours of storage.
(28th) On ordering pure zinc I reduced that yet again for now - the biggest sheets I could find were 140x140mm, or 5.5x5.5 inches. 100x100mm (3.94") were cheaper and I ordered some of them too. So much for 12x12 inches! 5.5x5.5" and 4x4" are only 21% and 11% as large. 7x7" would have at least been 1/3. Then again, perhaps 100x50mm or even 50x50mm is a good size for testing.



Lithium Battery Charging (120 VAC Power)


Note: Where to put it? I have moved lithium charging with DC power to fall under the heading of My Solar Power System under Electricity Generation. Hopefully I'll have finished writing about the AC charging soon.

   I connected the Sprint car to the 36 volt solar DC supply. While connected it took over heating water in the kitchen and running the DC light at night. The car is "36 volts" (really closer to 40) with 300 amp-hours composed of three 100 amp-hour lithium cells in parallel at every voltage level. Call it 38 V * 300 AH = 11,400 watt-hours.

   Having installed the two 120 VAC source chargers in the lowest 12 volt section, set to provide 7.0 volts, I measured the voltages on each cell shortly after charging with the two 6 volt chargers. At the (nominal) 0 volt, 6 volt and 12 volt levels cells A, B and C were connected together, so the only places they could be different were at the 3 and 9 volt levels. The notorious lithium imbalances between cells are plainly evident in each pair of cells. (Are the ones at 3.28 volts actually fully charged? The AC line current supplying the chargers dropped way down, so presumably they must be. What happens if both cells in a pair refuse to rise above 3.28-3.32 volts? I guess they just keep drawing current?)

Nominal
Voltage
Level
Battery "A"
Cell Voltages
Battery "B"
Cell Voltages
Battery "C"
Cell Voltages
9-12 V
3.59
3.57
3.44
Upper 6 V
6-9 V
3.35
3.36
3.51
Charger
(Cells connect-
ed together at 6
volt level.)
3-6 V
3.28
3.61
3.28
Lower 6 V
0-3 V
3.66
3.32
3.66
Charger


   I will remark that I'm quite disappointed in these particular adjustable power adapters. First, multi-turn screwdriver trimpots would be far better than the single turn pot with a knob on it. First they are incredibly touchy, hard to set the voltage on. Then the slightest touch, accidentally brushing the knob, changes the voltage. Unless it's being used as a cheap lab power supply, whatever you're using it for you want to set the voltage to some value and have it stay there.
   My second complaint is the same problem: the voltage doesn't stay where you put it anyway. The load and temperature regulation are terrible. In some applications going up or down a volt (or considerable portion thereof) may not matter, but in others it does - it sure does in battery charging.
   Nor am I fussy about them drawing current from the batteries when unplugged. The "car alternator" diodes I put in to isolate them drop over .8 volts - wasting over 10% of the charging power. (I should probably look for some Schottky diodes with a lower drop.) And when the batteries are charged and current is reduced, the drop decreases, raising the voltage to the cells just when it's not wanted any more.
   I can't see these as being reliable for unattended lithium car battery charging. I'm sorry I bought a dozen. Apparently I should have started with two and checked them out. Next time I'm shopping on line I'll be on the lookout for something better. OTOH these were the best I could find when I was looking - I may just have to make do.




http://www.TurquoiseEnergy.com
Haida Gwaii, BC Canada