Turquoise Energy Ltd. News #127
covering December 2018 (Posted January 7th 2019)
Lawnhill BC Canada
by Craig Carmichael

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

Month In Brief (Project Summaries etc.)

In Passing (Miscellaneous topics, editorial comments & opinionated rants)
  - No Clean Air Anywhere - Hybrid Potatoes? - Yikes! More California Fire Stories? - Wonder Weather - BC Referendum on Proportional Representation Results - ESD

- Project Reports -
Electric Transport - Electric Hubcap Motor Systems (no reports)

Other "Green" Electric Equipment Projects
* Carmichael Mill Handheld Bandmill (more operating than building this time) - Bonus Tip: stack excess firewood under a tree to dry

Electricity Generation
* Solar: Storage & Going Off-Grid, CAT & HAT plugs & sockets - Real Life "Off-Grid" Tests during power failure
* Magnetic Flipping HE Ray Energy? - Dual Frequency Pulses - more tests and trials

Electricity Storage - Turquoise Battery Project (Mn-Zn, Ni-Zn or Pb-Zn in Oxalate electrolyte)
* Ugh! Zinc hydride!
* Hydro - water storage
* Problems charging oxalate - JUST calcium hydroxide as electrolyte?

December in Brief

New Chemie Batteries

   December's battery experiments disclosed several things I hadn't understood or figured out before. First, it looked like the whole oxalate idea just might have to be thrown out the window. Whatever its other properties, oxalate seemed to keep both nickel and manganese oxides from charging. (What did it do with lead oxide? I can't remember and can't be bothered to look back at that issue.)
   What might I use for an electrolyte then? I was and am very loathe to go to potassium hydroxide. After all its pH 14 solutions are what causes zinc electrodes to degrade. The pH 12 where zinc doesn't (according to the table) form a soluble ion had the promise for long life, even everlasting, zinc electrodes. It finally, after all these years, occurred to me to try just calcium hydroxide. That had seemed like a silly idea because it's so little soluble. In fact, that's why it's pH 12 instead of 14. But the result of very low electrolyte concentration is just low current capacity per square centimeter. Perhaps the best answer to that is just to have more square centimeters. That's certainly what's done with lithiums, which are made of very thin folded up sheets.
   However, I'm going to try sodium oxalate before I give up on the oxalate idea - either by itself or as an additive with the potassium oxalate. Sodium-hydrogen sulfate seems to work wonders with lead-acid batteries, so the sodium ion may act as a catalyst of sorts. By itself it's not as soluble as one might hope, but 20 times more so than calcium hydroxide.

Everything Else

   Much of the month was pretty cold or windy and rainy, which precluded working too long in the workshop or outside.

   I wanted to continue (however slowly) with the reluctance motor, but I only got as far as grinding off November's crooked weld-on of one "salient pole" on the rotor.

   I set things up better and did a couple more experiments with the magnetic flip HE ray converter, still without notable results, but I did learn a couple of things. The oscilloscope showed I had had some things wrong in my perceptions of what was happening with the pulses. Now... perhaps an L-C tuned circuit can get some better higher voltage switching happening?

   Being unable to find either of my two anemometers for months, and even finally when I wanted to measure wind speed while I was trying out the VAWT, I ordered another one. Quite a nice one this time, and I could take the measurement part and tape it to a pole to check wind higher up while I read it from the ground. (The cable, whatever each wire actually did, had a "USB" plug into the meter so I could add on a USB extension cord for more length. I have trouble believing the reader unit actually sends USB.)
   As I found, for wind power nothing is as important as finding where the most wind is.
   Where would I keep this one so I could find it when I wanted it? I decided to put it on the shelf with my various voltmeters. On the back of that shelf I found the better of the two missing ones. The same day the other one turned up in a messy drawer I had already looked in. Sigh!

And speaking of wind power I found yet another VAWT video, by "New Forest R & D" that claimed to be the world's best. It was a sort of a self-starting Darius rotor that spun faster than the wind speed. It certainly seemed to spin well. Unfortunately one couldn't make out the airfoil shape from the video. (Someone commented on the silliness of making "tilted" blades, which showed his own ignorance of how the unit is turning and has rotated farther by the time the digital camera scans from the top to the bottom. It would be easy to be fooled like that if I hadn't just seen the effect in photographing my own rotor.)

   If one could estimate the "wing" airfoils for optimum lift/thrust, this one, perhaps a two-tier version, might be more worth building than the simple "windjammer" Savonius types I've been trying out.

   On the 15th there was a big windstorm that blew down trees across the highway and across the power lines in several places. On the evening and next day while the power was out I finally installed some of the solar power equipment that was sitting around and this is written up as a project below. I was glad I had at least put the solar panels on the roof last summer. Now necessity induced me to install the charge controller and the 36 volt, 100 amp-hour NiMH batteries formerly from the Mazda RX7-EV.

   Using the 36 to 120 volt inverter wired to the Sprint EV car, I ran the fridge and the freezer toward the end as the ice cream had started to get very soft if not to melt. I was especially concerned that the well pump was 240 volts and I have no 240 volt inverter. I had to forgo a shower even while the water in the tank would still have been hot. But later in a store I found all the well pumps seemed to be 240 volts, so I guess I need to get an inverter for 240 volts. Keeping the hot water hot in an extended outage is a problem for which I see no simple solution. A water loop to the woodstove is a big plumbing job, and the extra tank needs to be placed somewhere.

   I got a fair bit of lumber milling done with my handheld "Carmichael Mill" in spite of cold and rain, and I got the automatic band sharpener working pretty well. Milling after the last sharpening seemed as good as with a new blade.

   After one more cant, I'll be able to clean up a part of my driveway and drive around the circle, for the first time since I had the trees cut down shortly after I bought the house. Wow!

   The last 1/3 of the month was a "write-off" - for energy projects - as I went to my Mom's in Comox and then to Victoria for Christmas visiting and some shopping. In that shopping I did get a 6 foot (x 1 foot) roll of .005" copper sheet/foil at Metal Supermarket for making battery electrode pockets from in case I manage to make working cells. (I hope that's not too thin. Their next choice copper sheet was way thicker - .032" or something.)

   I got back on January 2nd and didn't get going on finishing up this newsletter until about the 4th, and even then not in any hurry. (Do I have a deadline? well, I usually try!)

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

No Clean Air Anywhere

From what I've read lately it appears there is now no clean air anywhere on the planet. There is a considerable and increasing aggregate global level of combined "EO-TOX" toxins we all breathe. At this point the Earth can't clean its air of some of these without human intervention. That will be a long, slow process requiring new technologies to be developed. Even then for that to work, first we have to stop polluting. Period!!!
   The aerosol spraying program, now running evidently for around 30 years, may not have made the world cooler, but it certainly is helping to destroy it and poison the inhabitants. With the aluminum and barium particles now in the air along with all the rest of the eo-toxins being pumped into it from cars and trucks, factories, fossil fueled power plants and heating systems, fertilizer and pesticide spraying, nuclear plant "exhaust" and other sources, apparently poor air quality has become a recognized or unrecognized contributing factor directly or indirectly in maybe 1/3 of all deaths each year. There are even still radioactive particles in the air from atmospheric atomic bomb tests of the 1950s. (As a kid I remember we had to send off our baby teeth for analysis when our adult teeth grew in because the damage of these tests was being recognized. The Earth's Van Allen belts grew stronger after each test, IIRC, and only very gradually subsided.)

   Furthermore the ozone layer is evidently still getting thinner. So far it means increasing skin cancer levels, and is perhaps a factor in some of the mass die-offs occurring around the globe. At some point plants everywhere might start dying of short UV ray radiation and then the deterioration of the atmosphere could become irreversible and extinction of life would follow. I don't think we're there yet, and the collapse is coming to help bring about an end to some of our destructive ways, so, surely, it won't progress that far.

   I myself had no idea things were this bad. Why did I start reading along such subject lines? Maybe better not to know?... but then again, maybe it helps explain my occasional coughing fits in the last couple of years. After all I don't smoke... or do I? In this and every remote corner of the world they chem-spray.

   I bought an "air purifier" with a washable filter for the house when I went south for Christmas. My present woodstove smokes badly whenever the door is open - more pollution generated right in the house. (I'll be replacing it next winter.) It said "whisper quiet" but as I feared it turned out to be sort of an industrial sounding WHISPER! Noisy fans are everywhere!
   So I run it at night on low at the opposite end of the house from my bedroom. Hmm, I think it might be helping.

Hybrid Potatoes?

   I left my potatoes from last summer in the ground because they seem to keep better there. No doubt if the ground froze very deep down, or perhaps if the soil had some composition other than very sandy, that wouldn't work so well. I'm not a big potato eater. I dug up a few on January 3rd after returning from Christmas in Comox and Victoria.
   There's white and red skinned potatoes with white flesh, yellow "yukon gold" potatoes with yellow flesh, and here we have purple "haida" potatoes with purple flesh, light or very dark purple. The last bunch I dug up were "haida". This time I just dug up 3 from a different part of the garden for supper. Whatever are these? They have white flesh.
   I thought I had grown all my potatoes from the previous year's potatoes, but with this skin they seem to be a hybrid or something. Perhaps some potatoes seeded themselves from the flowers and cross pollinated? Anyway they're quite colorful potatoes! (Hmm... somebody might actually pay extra for the colors!)

Yikes! More California Fire Stories?

   In a video (again by "aplanetruth") the author mentions the population of Paradise, California as being well over 50000 people going on more than one source. And the Weather Network said 18000 structures were destroyed. (Oops, was that in the Camp Creek Road/Paradise fire, or overall for the state?) The media said the dead were under 100, and the "unaccounted for" peaked at 1200 but dropped to just 25 after a couple of weeks. That sounds so good until it's turned around and the people accounted for are counted instead, along with knowing there were so many thousands of houses destroyed.
   The author said he went around and found around 600 people at the refugee camps. Doubtless some people would have driven off and found other accommodation - friends, relatives, motels... That might account for one or two or three thousand people. But say it was 5000... what happened to the other 45000 people? And in the surrounding area - more thousands?
   The fire came on so fast and ignited so many places at once that the main road in and out - with mountainous flames on both sides and covered in dark smoke - was quickly jammed with vehicles. There are a number of videos (dash cams?) from people who drove through this inferno and made it out. Others abandoned their gridlocked, burning cars and ran, but where could they go? The entire town was obliterated. How could most of the population possibly have got out? A tow truck driver said he counted over 200 bodies on a short trip in. But many were burned until hardly charred bones or even just ash remains.
   The author said there are plenty of residents who aren't posting to facebook and twitter any more, and he is starting a web site hoping to find out what happened to them all. One starts to get the impression of deaths on a similar scale to the nuclear bombing of Hiroshima (50000).

   In another video, a person recounted how he had been through a big California fire before many years ago (he named the fire and date) in similar dry conditions, and how in spite of the wind it took hours for it to spread from where he saw the first smoke to threaten the town. This fire was completely different, he said. Even while the fire seemed far away and there were no obvious flying embers reaching the town, pretty soon his house and the whole town was engulfed in flames. So he ran to his car and left, within two minutes of first seeing fire. He said he luckily had oxygen in the car, and that if he had taken two or three more breaths he'd probably have died of smoke inhalation. Apparently he got out before the gridlock hit, or missed it somehow. Do we suppose his neighbors all got out okay?
   There are some videos of the burned town, but evidently the military has now taken over and no one is allowed in. Is there something to hide? Where are the 50000 residents who want to see what's left of their homes? Given the seemingly impossibly small 'casualty' and 'missing' figures being spouted by the media, are they there to hide a mass genocide? And "aplanetruth" is not the only one calling it "mass murder". (The names, the names... Where would be a copy of the Paradise phone book?)
   I'm leaning toward to the conclusion that several presenters are right and that it's being done with "directed energy weapons" ("DEW"s) - infra-red lasers or something similar - from aircraft. It would also explain the "two sides of a rectangle" Fort McMurray fire perimeter quite well.
   Inventor Nicola Tesla offered the US military a "death ray" of some sort during world war two, and his papers were seized by the government when he died. And apparently they have been developing "DEW"s since about 1998 with a budget of several billion dollars, so it would be almost more surprising if they haven't come up with anything than if they have. If this is really what is happening, perhaps we should be grateful they're now using them on their own people instead of on others, which could spark a world war.

   But there are also some less bizarre aspects to these fires; the ones that leave you saying "Well, that figures." With housing so costly in and around the big cities, notoriously San Fransisco, new communities have been permitted to sprout up on "interface" border lands between the urban and the seasonally very dry woods. They are designed to lax standards with meager road access and laxly enforced building codes and fire safety regulations. In short, they're thrown out there into iffy zones previously thought to be unsuitable for housing, because of population pressure, which makes them "accidents waiting to happen." -- another result of the planet being overpopulated. Then again, what about images like this? (here, in Malibu):

Forest fires burn down houses and leave the trees?
(...But how long after the fire was the picture taken?)

   I'm definitely going to have to stop watching these disturbing California fire videos! Just as with VAWT and many other video topics, once you've watched a couple, Youtube keeps putting more and more of them in front of you as "suggestions".

   Instead, maybe let's follow the main inhabited part of Alaska, where the ground has not stopped shaking since the November 30th 7.0 quake? (but Alaska Earthquakes Center says it isn't unusual - and okay, it finally looks a little more solid with the new year.) High waves and extensive coastal flooding are in store for California this winter according to NOAA. If the worst predictions for much of the state sinking into the sea hold out, fires however started may soon be the least of their problems.
   From one video (by "Suspicious Observers", who have a good record for earthquake prediction based on solar magnetism patterns) I found out a theory for why there might be more flooding than just sea level rise would account for. It seems that (as I understand it) as the ice sheets retreat in Greenland and Alaska, the pressure on the mantle under those areas is reduced, and the crust rises up. The mantle flows (seemingly from especially the eastern and western seaboards farther south), causing coastal California as well as South Carolina and most of Florida to subside under the oceans. It is well known that Miami and New Orleans will have to be abandoned just from sea level rise, but people there are "in denial"; still erecting new buildings in Miami and trying to improve protective earthworks in New Orleans.

   Or how about the smoldering Yellowstone supervolcano, its caldera 35 by 45 miles across, where new geysers are chewing up the walkways and roads and perhaps even the monitoring equipment? Is a big eruption far off as many insist, or it could blow very soon as some think. Mount Saint Helens (its eruption in 1980 was heard from my house in Victoria BC) was an anthill compared to Yellowstone.

   Maybe some old British comedy shows would be a better viewing choice?

Wonder Weather

   But no, I had to check out some more shows, on weather, and found the nastiest hail storms ever recorded, from this last summer. In South Africa (IIRC) one broke trees and killed cattle. One in Australia only killed sheep (oops, and kangaroos). So in addition to the usual "once in 1000 years" droughts and floods, record high temperatures, record low temperatures and record breaking hurricanes, we now have nasty hail storms that rain ice balls the size of golf balls and even base balls.

Hail Storm in South Africa as seen from the outside

It did this to this piece of forest (saplings?)

and it killed cattle

On Haida Gwaii we've had no snow (or hail), but one morning I
found a strange "frost forest" covering the ground and on my car.

BC Referendum on Proportional Representation: Results

   Well, I guess there really are a lot of "sheeple" out there. Everybody complains about autocratic, dictatorial government decrees, but the proposal to reform our primitive, unfair and polarizing voting system was rejected by 61% of BC voters (of the less than 50% who bothered to mail in their post-paid ballot).
   Reading some reader comments under the news story was disheartening. I began to see the low levels of social awareness of the general population. To me, most of the "reasons" for voting against the proposal seemed to be "cop-outs" - illogical rationalizations for deciding not to take the trouble to give the matter any thought. Just be negative and nothing will change. That way it's impossible to make a worse choice than the existing one. (It was impossible to make a worse choice than that anyway.) Some thought it was just a waste of money. Some said it was "too confusing", that "too many options" had been offered, so (in spite of the option to simply vote "yes" and let others decide which option to pick) they just voted "no". Many voted against this measure to improve election of governments just because they "don't trust the government" - so "it must be some kind of trick" to get more money or take more freedom from us! Premier Horgan and others in the new BC government expressed disappointment with the result of what they had championed as a means to change our polarized politics.

   The one argument that seemed to me to have some merit was that if a "majority government" isn't elected, it's hard to govern. Decisions get bogged down in partisan bickering. An unfair electoral system is the only way to get majority governments. ("Majority government" meaning over 50% of the entire legislature are all from just one political partisan sect ("party") - so much for the ideal of a legislature being "a representative cross section of the society")
   But that just highlights the need to separate the executive branch from the legislative so that governing is separate from legislating. Ideally both improvements - electing a 'governor' separately from electing 'legislators', and picking a fairer electoral system - would be adopted simultaneously.

   We will not get another chance to make even this small improvement to our governing systems before governments as a whole start to lose cohesion and nation become largely ungovernable. And obviously we'll need improvements to be offered and promoted by local social sustainability design teams before the general public will think proposals for improvements have credibility and aren't something being foisted on them by corrupt leaders wanting more control.

(Eccentric Silliness Department)

* Ducktors are just a bunch of quacks.

* Just when did "cocoa butter" become "white chocolate"? And when did cadmium (as in screw and bolt plating) become "yellow zinc"?

* Russian anti USSA propaganda? "For 50 years, we wanted to live like you. But no longer!" - Margarita Simonyan, head of RT.com . Maybe it's because we don't have that life any more either. "The free world" seems to have shifted markedly to the East.

* How many Microns in a Macron?

* The above mention of lax building standards in California raises further questions...

 - Can you relax without first having laxed?
 - Can you relax when flying into LAX in a malfunctioning airplane?
 - Are LAX (Laxton's Progress) peas better than Green Arrow peas?
 - Do Green Arrow peas fly straighter than that airplane you're on?

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

Other "Green" Electric Equipment Projects

Carmichael Mill ("Bandsaw Alaska Mill")

[ 20 minute Video: "Carmichael Mill Update" https://www.youtube.com/watch?v=P7r6hQF3yg8 ]

   Other than modifying the automatic band sharpener to get it to work for my type of bands, which are lighter and have much finer teeth than most bandsaw mill bands (3 per inch instead of about one), there were no new developments in this project except the "Carmichael Mill Update" video posted to youtube. Basically it's a prototype that's working well and cuts good, if often somewhat wavy, lumber. (More band tension and using sharp bands minimizes waviness.)

   I cut the largest 16 foot long spruce cant into eighteen 1"x8" x16' boards in the single digit days of December. A chainsaw mill would have only made 12 or 13 boards and the rest would have been sawdust. I put them up temporarily on the trailer shelter walls with deck screws, to block the wind and rain. (I screwed on the vertical stick seen in front and slipped the right end of the top two boards in behind it as they were too heavy for one person to lift and hold in place from one end while screwing them on from on a ladder. ...Left side: wall studs on 16 foot centers?) Everything held in the big wind storms. It would be drier underneath if they covered all 24 feet of the windward end, and from bottom to top.

   I started on the cant that was underneath that one on the 12th. While the sharpener took the better part of an hour sharpening each band, and each one cut only 3 to 5 of the wide boards in the tough-milling spruce, at least I didn't have to do the tedious sharpening by hand. It allowed me to keep cutting instead of running out of sharp bands. One band, sharpened 3 times, didn't seem to stay sharp long after the third go. I switched to another one and once into the narrower cant, I got quite a few good boards with it.

   A guest watched my video and thought the mill could be mass-produced to sell for 400$. He's probably right. I've thought about making them, but I seem to be on to other things - except for cutting up my own spruce. I'm thinking I may, after all, someday get through to the end of the ones blocking the driveway and be able to drive around the circle again. 2-1/2 big cants to go (and one to sell as-is for a big beam) as of the 13th.

   One development perhaps of note: I finally got the automatic band sharpener working well.
   Earlier I put a bolt in closer to the center so the pusher would only push one tooth through at a time instead of two or three, and I moved the rod from the grinding wheel closer to the hinge so it went up and down less. Even so, this month there was another thing: I put in a piece of sheet aluminum on the turning cam so the grinding wheel travelled up and down still less for each tooth. Then it didn't dig crazily far down in the gullies between the teeth.

The next band cut like new again! The bottom of the cant took about 4 minutes per cut for all three cuts, instead of 8 minutes, 10 minutes and 12 minutes with each cut being slower than the previous one.

   I may not report on the mill again unless there's some interesting development, or if I'm especially proud of something I've done using it. It's sure nice, with so many puzzling failures and setbacks in other projects of potentially priceless value that I just never seem able to finish, to have one nice one that I can say is done (in just one year!), works great and I'm using it heavily, and which has advanced the state of the art in something. (Self-adjusting band guides seem to have never been thought of before.) It cuts using less power and turning a thinner kerf into sawdust than any other mill anywhere.

The mill (from the front) near the start of January.
The original skillsaw's gears wore out - their
grease was gone - and it had to be replaced.
(Must open this one up and check the grease occasionally!)

Electricity Generation

Solar: Storage & Going Off-Grid,
12 & 36 VDC, CAT & HAT Plugs & Sockets

   I still had the Honda Insight batteries, a total of 32 sets of 6 NiMH super-high-current cells welded together. They had proved to be a frustrating setup, yielding as I saw it a nominal 7.2, 14.4, 21.6 or 28.8 volts, which were all too far off to use for 6, 12 or 24 volts. But I should have gone one further: 5 sets in series makes 30 cells for 36 volts nominal, which is the voltage I was saying earlier is an optimum DC power voltage: 1/3 the current and wire size of 12 volts, yet low enough voltage that no one will be electrocuted from touching it. And it fills the gap right in between: at three times 12 volts it's also 1/3 of 120 volts common AC grid line voltage. (I "officially" defined it as 38 volts with +/- 15% tolerance: 33-43.7 volts, which allows for all the common "12 volt" battery types in all states of charge including being under charge. [Except high voltage pulse charging. If that is in use, the power may need to be filtered to protect appliances/equipment.])
   6 sets of 5 cell stacks, which are either 6.5 AH or 8 AH, would give 39 or 48 AH of very high current capacity storage. They would have no trouble running high load appliances, like perhaps the 36 volt to 120 volt inverter as it turns on the fridge or freezer. To that could easily be added 100 AH in the form of my three 100 AH regular NiMH D cell batteries for more storage capacity. And about another 100 AH of miscellaneous NiMH D cells if I wanted to add them in. These are all surplus since I took them out of the Mazda RX7 EV some time ago and with sufficient solar panels could tide one through a very long power outage - at least in the summer.

Real Life Off-Grid Tests

   Well, on the 15th came the opportunity - in fact the need - to test things out. There was a strong wind storm up to 120 Km/Hr or higher. I went to a cafe for lunch. The road was cluttered with small branches. The whole building shook. It was predictable: At about 4 PM the power went off. At 8 PM it came back on for a short time (10 or 15 minutes?) and then went off again. At about 4:30 PM the next day, 24-1/2 hours later, it came on and stayed on. I wondered how many trees fell across the power lines in the tempest? A couple of days later I saw a couple of pairs of them on a trip into town:

Looking south; Hecate Strait [ocean] to left

Another spot, Looking north

   The outage demonstrated that I was less prepared than I thought. One thing I did have was of course the woodstove and firewood, so there was no interruption to the heat. (And I had just noticed a video suggested by youtube from Canadian Prepper: "You won't last one off-grid winter without one of these.", with a picture of a wood stove. Those living in regions with cold winters should take note as people here and there have been without power for extended periods recently.)
   People talk of cooking on a woodstove, but my experience is that the top of any newer "high efficiency" model is never really hot enough to boil water. For coffee I put water on the stove for hours, then I brought in my propane camp stove to get it boiling. It didn't take much propane that way. In the morning I very, very gradually cooked a pancake directly on the wood stove.

   Electrically was where I wasn't so prepared. I had all the pieces, and 1000 watts of solar panels sitting on the roof, so I was surely far more prepared than most - but nothing was hooked up. It all had to be installed and wired. In the gathering gloom I couldn't find the 3 KW, 36 volt DC to 120 volt pure sine wave inverter. The just charged Honda batteries weren't all wired together for 36 volts yet. All my other big 12 volt NiMHs proved to be pretty discharged, and they were the ones with 12 volt CAT sockets wired on so you readily could plug 12 V lights in. I was hard pressed to put on a couple of low power lights for a few hours.
   I thought I'd like to try two 12 volt LED globe lights with round "power adapter" sockets on them. I found a wire with a "power adapter" plug on one end, but I had to put together a CAT plug for the other end. For that I had to find another battery that would power an inverter so I could run the soldering iron. I also had to grind the few roasted coffee beans that I had to brew a small pot. (I keep telling myself to keep some ahead in case of a power failure, but they always seem to be all gone before I roast more.) I left the inverter turned on but running nothing and within 1/2 hour its low voltage alarm started going off - it had sucked what little charge that battery had out all by itself powering nothing. Making the cord paid off because one of the globe lights in low power mode drew only 90 mA - barely over a watt, yet it was enough to see by. Surprisingly bright. I could leave that on all evening even with the low batteries. (The 15 watt 12"x12" square flat panel light was way more light, eg for reading, but drawing 1.25 amps, the batteries wouldn't have been up to it for very long.) As always I had my 6 volt bedside light with 5 D cells, which I do use and usually keep well charged. Another 12 volt light sufficed for the kitchen to make coffee and food.

   The most problematical thing of all, which had been preying on my mind ever since moving here, was that the well water pump is 240 volts. I couldn't run it without grid power, period. If I ran much water, the pressure would be gone and there would be none. I saved it for toilet flushing and short bursts of running sink faucets. The shower I'd meant to have had to be put off even though there was plenty of hot water in the tank.
   If I bought a 240 volt inverter, it would be just for the pump. And if it failed, I wouldn't have any other to use. I have several inverters for 120 volt appliances. Plus the pump was wired straight into the sub panel in the shop: even if I had the inverter, I'd have to un-wire the pump and reconnect it, with the cover off the panel. I added "120 well pump" to my shopping list. I'd wire it plug-in so it can be plugged into an inverter if the wall is dead. But in Canadian Tire on my Victoria shopping trip, I only found 240 volt well pumps, and they were 300$ and up. Perhaps a 240 volt inverter is the thing to get after all.

   The next morning I disconnected the grid tie inverters from the solar panels and put on the programmable DC to DC charge controller. I set it to 42 volts and dragged in the 'pretty discharged' 100 AH - 12 V NiMH batteries, and set them in series. Over the course of the cloudy, rainy day, the power from the 1000 watts of solar panels went from 10 watts around 11 AM (or whenever it was I'd got it all done by) up to almost 80 for a couple of hours and back to 10 by 3 PM. Short cloudy northern days with the sun at a low angle aren't optimum for solar. The controller said it had put a whopping 2.1 amp-hours into the batteries, which just hit 39 volts.

I took the two pairs of wires, each coming from two solar panels, and tied them together.
(The screw terminals on the grid tie inverters are too short to put in more than one wire,
the MPT7210A only takes one skinny wire per connection, and I didn't have a heavy
terminal strip, so the connections are hanging in the air tied together on two bolts - Ugh!)

The charge controller, set to charge to 40.5 DC volts at up to 10 amps.
I could float charge them up to 1.40 x 30 cells = 42.0 volts, but there seems no
reason to push them to the max if they'll be charged every day and not much used.
(Solar panel input is usually 32 volts open, 29 volts charging, but it was just getting dark outside.)
This unit works well, but the instructions are so inadequate for setting up its many programmable
features that people have put up youtube videos that explain how to get it to do what you want.
The solar panels are 1000 watts, but in the cloudy winter weather
with the low sun angle I don't think it's hit 100 watts charging yet.

Three 100 amp-hour, 12 volt NiMH batteries to make 36 volts.
(Beside them are some 12 V, 10 amp-hour tubes. I'm connecting and charging one each day
since I haven't got enough wires made up to connect them all together.)

   Breaking the narrative here, Jim Harrington of AGO gave me a link to the MPT7210A programmable charge controller  and I got two because they sounded like they would do just what I wanted. Now from the many youtube videos about it, it sounds like this is in fact the one to get for many purposes. It has a zillion ways it can be set up, and even multiple setups can be programmed in. The biggest complaint is the fan noise, which is terrible, and it blasts away regardless of temperature. The speed is supposed to be programmable, but doing the same thing that seems to work in a video does nothing on mine.
   The other complaint is the complexity of operation. How to use it goes from poorly explained to just not explained in the brief manual. And there's nothing intuitive or simple to setting it up. That's the main reason there are so many videos about it. There are just four buttons, but they do many different things depending whether the unit is charging or not, which one is pressed first, whether it's held down or pressed quickly, and the entire sequence.
   For example, to set it to auto-on when the sun comes up and it powers up: 1. Hit "OK" to turn off charging. 2. Hit set about 4 times until "Amps" is highlighted. 3. Press and hold "Set" again a few seconds. 4. Hit "Set" again 3 or 4 times until "OFF" is highlighted. 5. Hit "up arrow" so it says "ON". 6. Hit "Set" again. 7. Hit "OK". Then you can turn it on again by hitting "OK". The manual doesn't explain this. I had to watch part of the video, go out and do the first steps, then watch the rest of it and go out and finish it. I can't help but think it could all be made a lot more intuitive even with just the same four buttons and the same display. Meanwhile, back at the ranch...

   By early afternoon the ice cream in the fridge's freezer was melting or at least very soft, and the small chest freezer was up from its usual -22°C to -10 under the lid, notwithstanding that the room/garage it was in was probably only 7 or 8 degrees. (It was probably colder down lower in the mass of frozen food.)
   I got out the 36 VDC to 120 VAC inverter (I did eventually find it) and connected it to the 36 volt, 300 amp-hour Sprint car - half the storage of the Nissan Leaf, but accessible. I had been keeping its batteries pretty well charged, and of course lithiums don't self-discharge as fast as NiMHes. (How was it that I had to attach wires by removing nuts and hard-wiring up the inverter to the car instead of just plugging it in somewhere?) I ran a 100 foot extension cord to the house. I put in the Canadian Tire appliance power meter and plugged in the fridge. It was using about 170 watts, which gradually dropped to under 140, drawing around 5 amps DC from the batteries according to the clamp-on ampmeter. I ran it an hour or more until it stopped. It soon started again, but I unplugged it and plugged in the coffee roaster (popcorn popper) and roasted one batch, taking about 3 minutes. The 1500 watt roaster would have used in 3 minutes what the 150 watt fridge does in half an hour. After that I plugged in the chest freezer, which drew just under 100 watts. By this time it was mid to late afternoon. After an hour of that the power thankfully came back on. The car's batteries had started at 39.6 volts. They soon dropped to 39.0, but stayed there for the rest of the event. (Except when roasting coffee they dropped a little farther, but came back up to 39.)

   My conclusion is that while not much happened during a one day off-grid event (other than frantically setting everything up and wiring it, and missing a shower), a longer period would bring serious hardships as things are set up now. OTOH it was a good day to run around setting things up... there's a whole bunch of other things one can't do anyway if there's no power!

   First and foremost, I definitely need a well water pump that I can plug into a 120 VAC inverter (or else a 240 volt inverter). Then, for a period more than two or three days, I need more power generation for lights, fridge and freezer. The Sprint car, if kept charged, can run them for a while, but not more than a week or so. After that it would depend on the sun coming out to recharge the batteries. And it would take still more available power to keep a computer or two running. Milling with the electric band mill is out. Firing up the gasoline generator for fridge and freezer is a stop-gap solution for a few days. Combining the wood stove and the propane camp stove for heat should suffice for meals and coffee for a very long time. The woodstove would have to do for hot shower water, too. Ugh!
   A 20 to 50 watt thermocouple generator to put on the wood stove, that would run 24 hours a day, starts to sound like a good idea again, if that's all that can be had out of solar panels in the winter anyway. If the modern woodstoves don't get so hot on top as the old ones, perhaps it could be done with TEGs ("Thermo Electric Generators" - higher temperature Peltier modules) without any special overheat protection needed. The number of modules could perhaps be selected to send a voltage to the solar controller similar to that of the solar panels - but it would work all day and night.

   And if I'm going to have my "optimum line voltage" 36-40 VDC system, I need to start looking for some appliances that use that voltage. All I have presently for that voltage is the Sprint car and the inverter. Maybe I can put three 12 identical volt LED lights in series? It'll probably work, but it's a makeshift solution.
   And I need to finalize the design and start producing the "HAT standard" 38 volt wiring products: Plugs, Sockets, electrical box Wall Plates, and so on.

   In the meantime, I had the CAT plugs and sockets. After doing some 12 V wiring I decided after all, there was little point in changing the standard. CAT is quite good, fine as it is. There would be no RAT (Revised CAT) standard. Changing it poses two problems. One is that any little bit of traction it might have gained as an adopted standard so far will be broken. It was probably even a mistake to say I would change it some issues back. The other is that I haven't had time even to set up the 3D printer. How much time do I have to redo what's already done?
   I had planned to make the 36-40 volt HAT standard plugs and sockets about the size and shape CAT is now. What I do need to do instead is make it different enough so that you can't accidentally plug the one into the other and underpower 36 volt equipment on 12 volts or blow up 12 volt equipment on 36 volts. Perhaps the thing to do would be to make the proposed "RAT" standard, or something close to it, into the "HAT" standard. I guess nothing will be settled on HAT until I get on the 3D printer and start making something.

   I managed to cross voltages without HAT plugs. I had the 36 volt battery system on the solar panels, and during the power failure I had connected one 12 volt battery of it to the wire running across the house to the closet by the living room. I wired in a CAT duplex receptacle there and ran 12 volt lights off it in the living room. Later I had been charging some other batteries and used the alligator clip jumper leads. When I reconnected to the wire, I absent-mindedly connected it across all 36 volts instead of just 12.
   When I turned on one of my 12 volt homemade LED lights, on "low", it was amazingly brilliant for about one second, then it went out. Oops, the "12" volts was 39! I moved the alligator clip wire. Later I started taking the lamp apart hoping I could repair it. Wait! There was a fuse! What a clever idea! Amazingly the 2 amp AT fuse had blown and protected the lamp. I put in a new one and it worked again.
   But if I had connected the wire with a CAT plug to a CAT socket - already present on each of the three batteries - instead of with jumpers, the incident could never have happened. This just underlines the need for proper and unique plugs and sockets for each voltage: IMHO: 12 VDC, 36 VDC, and the existing ones for 120 VAC and up.

   On the 19th I wired up my indoor garden fan, a battery, and a 10 W solar panel with CAT plugs and socket. (Since the panel is just 10 watts, I don't bother with a charge controller. I put a diode in series next to the plug so it can't draw charge back when it's dark.) Why had I been using aligator clip leads that keep breaking and falling off for this little system all this time?
   And I found the need for a couple more CAT type products. First, If the solar panel is a power source rather than an appliance, it should have a socket rather than a plug. But if it does, how do you plug it into a battery that also has a socket? Apparently a plug-plug "cheater" adapter is required. Second, I wanted to plug both the fan and the panel into the battery at the same time. I have before put two CAT sockets onto a battery, but this time I thought it would be nicer to have a splitter with a plug and two or more sockets on it, similar to those that grace so many 120 volt wall receptacles because there are never enough of them.
   Then, not to get too carried away, a very useful device for off-grid power would be an in-line power meter with a plug and a socket, to show the voltage and current. ...a low power one with an LCD display that wouldn't itself drain the battery.

   Finally after having them in use for a couple of weeks, I am reluctantly forced to conclude that the 100 AH NiMH batteries are rather worn out. One of them discharged itself from 13.5 volts to 12.7 overnight (about 17 hours) without a load. I think it was the continual float charging in the RX7-EV that gradually wore them out. Better to just charge them at a good speed and then turn it off and let them rest. And they are several years old now. Of course they're also being float charged all day on solar. For now in the dead of winter that's only about 6 hours, and I turned that float down from 1.40 volts per cell to 1.35 - total 40.5 instead of 42.0 volts. I expect the lithiums take a float charge better because they stop drawing current once they're charged.
   But it all makes me the more determined to come up with better, cheaper batteries. There's no way batteries should cost the sort of money they do. I keep thinking there can't be much more in the way of getting a better, cheap chemistry to work as expected, after all the progress I've made and all I've learned.

Magnetic Flipping HE Ray Energy?

   Okay, nothing happened in my one try at powering the device up in November.

What else could I try?

* Dragging out the oscilloscope so I could see what's going on - actual waveforms and voltages at various points.
* an antenna (Moray's earlier devices had one)
* a ground connection (Moray's devices all needed one, as I recall)
* Putting Diodes in series with the pulse source. 36 volt pulses are nice, but since the motor controller doubtless
   has rectification to capture the return energy, the big spikes across the coil will be damped out. The big spikes
   at turn-off of the pulse were a requirement. (How could I have put that out of my mind?) If I put diodes in,
   they can isolate the coil and let it fly. (When one works it out, this was a mistaken assumption. The spike's reverse voltage makes the diode forward biased anyway.)

   On the 12th I thought, what was the real problem with working on it and running tests? I started out in the summer with the car out on the lawn. Now as long as I was using the Sprint car's motor controller as the 36 volt pulse source, I was working on the floor of the garage. It's winter and it's cold out there. Plus it's cluttered. In fact, it's cluttered almost everywhere I work. Perhaps these were the things I should deal with first.
   Two things I could remedy quickly I did on the 13th: I cleaned up the corner where I had been working, and I cleaned up a bit of wood clutter by making it into a work table (long planned - just needed legs), which I then set in the corner and picked the experiment pieces up off the floor.

Dual Frequency Pulses

 Having a table to set it where it was high enough up to see the screen and where I wouldn't step on it or kick it, I brought out the oscilloscope and checked just what pulses were coming from the motor controller. It was set to 1.6 amps, the minimum, and I found it had two components: a low frequency "square wave" with the "high" part being composed of high frequency pulses. The low frequency was 8 mSec of the spikes then 8 mSec off: 16-17 mSec. (60 Hz). The narrow spikes were about 3 uSec. If I turned it up to 2 amps, then 2.6 amps, the pulse width increased to about 4 and 5.5 uSec, but the overall pulse rate remained the same - straight PWM at 16000 Hz.
   I had connected a wire to one side of the input as an "antenna". The unconnected scope probe picked up the pulses from being near it, but nothing was obtained at the DC output (which I only metered with a voltmeter) except a voltage that gradually rose from 0 to .15 volts over a couple of minutes. I also tried reversing the pulse leeds in case the pulses were the wrong polarity compared to the magnet.

Zoomed out, a bunch of high frequency pulses is followed by quiet, each 8 mSec for 16+ mSec periods (about 60 Hz)
Zooming in on the pulse areas, each one is composed of short pulses with about 62.5 uSec period (16000 Hz)
Zooming in still further, each short pulse is about 2.5 uSec.
As to amplitude, the scope probe on Channel 1 is x10, and it's driven from ~40 V lithiums.
But it's a heavy load and a long, thin wire, so we see around 30 volts (says ~3 V @ 10x scope probe).
(Channel 2 is only picking up noise from the "antenna", so the amplitude depends only on the pick-up. I should have turned it off.)
Perhaps interesting: the ground clips of both probes were unconnected - I wasn't sure where there was a ground -
and the car should have been floating, yet solid readings were attained on Channel 1.

Increasing the current setting increased the width of the pulses.
But the 16 KHz and 60 Hz overall frequencies didn't change.

   I don't know what led me to want to try the car's motor controller as a 36 volt pulse source. Somehow it just seemed like the thing to do. Now I see a potential advantage to it over any pulse generator I would have made for myself: I don't think a steel/iron "core" material will respond well to a higher frequency like 16 KHz. But it should turn on and off at 60 Hz just fine. A single frequency pulser would have to put out the same microseconds-short pulses at, eg, 60 (or perhaps 100 or 120) Hz, which wouldn't put much energy into the field.
   So this dual frequency of pulsation just might be a good key to getting such a unit to work with an ordinary iron core instead of some of the more exotic core materials others have used. (And such a setup, providing an absolute maximum 50% duty cycle with 1/120th of a second rest periods, probably would keep my motor controllers from blowing up, too! That may be why Curtis adopted it. I'll try that on the reluctance motor controller in future tests. Something else I've been missing all this time! My first motor controller was very low frequency, but I never got the idea for combining high and low frequencies. Nothing in any of the motor controller chip data sheets suggested such a thing, either.)
   Now that I understand that the pulses are of this special special nature I can make a driver to generate similar pulses and variations on them. In fact, I can use the circuit I already designed with the microcontroller - eliminate two of the three coil drivers and change the MOSFET driver to one of the better ones I bought. Then the whole thing can move indoors. I should have looked at the pulses last summer! I suppose again working on the ground made me subconsciously reluctant to bring the oscilloscope out.

[At some point here I noticed one could set the oscilloscope for "x10" probes in a menu so the numbers shown on the screen would be correct.]

   Before doing that, the next thing to try was to block the suppression of the flyback at turn-off, by putting a high voltage, fairly high current diode in series with the pulse source (instead of across the coil to damp it when it shut off). I tried on the 18th. I found I only had lower voltage diodes, but I had an IRF840 Mosfet good for up to 500 volts. Its body diode was good for a fair number of amps, so I soldered the gate to the source so it would never turn on, and soldered two wires to it.
   I checked it out on the oscilloscope and found there didn't seem to be a very notable spike. It seemed to me the coil should have far more turns, of finer wire. (Channel B goes straight to the motor controller. Channel A goes through the 'diode'. I thought it was thus free to spike as far negaitve as it wants, but it only did a few volts.) One of the problems with energy device documentation is that it seldom seems to give the size, characteristics and number of turns of the coils.

   Let's see... there are presently (were) 16.5 turns on each half of the coil. If that little turn-off spike is maybe 9 volts, and we want 100 volts, we need 11 times as many turns, or 183 turns (give or take a couple of dozen). Let's see... unwinding one, it's less than 4 feet of wire. So make it 45 feet to be safe. That's probably more in keeping with typical wiring for that sort of voltage... but much harder to thread through a toroid, and to keep count while doing so! I found some #19 AWG wire. That was about the thinnest I had much of without going really fine. I wound it, but I'm not sure why I continued after a certain point. It was obviously too bulky. Even #22 would have been a dubious fit - 24 or 26 would have been better.

   So I went to look for the fine magnet wire that Jim at AGO, having had no use for it for quite some time, had virtually thrust upon me. I didn't think I could possibly use it, winding lower voltage motors as I was at the time, but this is the use! There was a spool of #28 AWG and one of #30. I used the #28. Thanks Jim!
   And I was thinking, that if the pulse widths were wider, the spike voltage would be higher. So I decided to wind just 100 turns on each side. If the voltage was too low I could widen the pulses to put more energy into the coils. I don't know what a good voltage is anyway, just that it's a lot higher than single digit volts.
   The outer diameter of my pipe "toroid" was 45 mm, the wall thickness was 3 mm, and the height (pipe length) was 19 mm. It took over 14 feet of wire to do each side. (according to my calculator.)

   On the 20th I wired it up, took it out to the garage, and connected the motor controller and the 40 watt lamp. This time, the pulses looked nothing like the previous ones. There was no 60 Hz, just continuous 16 KHz pulses, and instead of being a very low duty cycle, they were around 90% ON. Apparently the controller was unable to put the set 1.6 amps into the (effectively) 50 turn coil of #28 wire, so it was putting out all the voltage it could, trying.
   I tried taking out my two 22 ohm, 10 watt resistors, to provide a load. In parallel, (11 ohms) the controller kept saying "field coil output shorted". With 44 ohms it worked sometimes, but it was still long pulses with no 60 Hz 'off' times. And the resistors got very hot pretty fast. With 22 ohms, it mostly said "shorted" but worked one time and still had full pulse width and no 60 Hz. Apparently the car's pulse generator was expecting motor field coils for a load and it wasn't going to give me what was wanted now that my coil was probably close to what was needed. Unless I managed to load it down just about the same way it was before, and connect the unit in parallel with that.

Thinking of that, I got out the original coil from last summer and did that. It worked - at least, sometimes. (The other times, the motor controller says "shorted", which has been happening all along.) When it did it gave the desired pulse forms: short pulses at 16 KHz going on and off at 60 Hz.
   Probe "A" was mistakenly on the "+" side of the drive, before the diode. "B" was after the diode, on the anode. Instead of a spike at turn-off, the big spike was immediately at turn-on. Why? It was around 95 to 125 volts from peak to peak as it went negative in less than a microsecond.
   This is probably about the right range. The only problem was that there were still no results. No output.

   As it got dark I went inside and drew it out on paper. There were so many things that could be one way or another - so many binary "1" and "0" choices... Diode polarity. Magnet polarity. Drive polarity. And there was another one: polarity of the two coils, one on each side of the 'toroid'. I had connected them opposite since they were on opposite sides of the toroid. But that was wrong, wasn't it? With the drawing there I realized that instead of both sides pushing the magnetism from the magnetically connected end toward the gap end, one was pushing one way and one the other, around in a circle. Net result out into the iron horseshoe: zero. I swapped the two ends of one of the coils.

   The next day I went out again and figured out some more. The spike at turn-on instead of turn-off was because the diode was the wrong way around. The rise was simply the turn-off time of the diode, and the fall was because the diode stopped conducting, so it was like "turn-off". But it was just stray transients. There hadn't been much time for energy to accumulate in the coil, so there would be no notable magnetic output. After the spike, the voltage settled at about +20 volts - the same as the negative side of the drive.
   But if I had the diode the right way around, my blithe assumption that it would allow the coil to spike negative was wrong. Just as the "+" drive rising to 40 volts pulled the coil to +40 volts minus the diode forward drop, the "+" drive at zero wouldn't allow the coil to go more negative than zero minus the diode drop, around minus one volt.

The drive from the motor controller as seen at the dummy load coil, "-" side and "+" side.

The "+" side of the drive (a) from the motor controller and (b) following
the diode at the unit's coil, with the diode the right way around:
Essentially the same less a diode drop (< 1 V)

The "-" drive, and at the coil on the "+" side following the diode, which was in backward.
That's the only time a coil spike was seen, but it was at the wrong time and with little energy.
(I'm not sure why a timing offset is seen. It seemes to be just in this one trace.)

   Okay, for someone trying to do groundbreaking work, I sure made a lot of stupid mistakes. All a learning curve! I always take comfort in what Roger the plumber said to me in about 1995: "The only people who never make mistakes are the ones who never do anything."

   So. Even with isolation diodes, was it possible to drive the coil so that the turn-off spike wouldn't be damped off? It didn't look like it. Maybe with a capacitor in series? Nothing like a tuned circuit to maximize an oscillation! Time to refresh my memory of that aspect of circuit design from BCIT in 1975. But it would have to wait until after Christmas.

The circuit... what circuit?

   I had already been thinking about and working on a pulse generator.

   On the 18th and 19th I spent some hours doing the revised circuit board for a microcontroller based pulse generator. I printed out the finished board deign, then I started thinking that the whole thing was pretty simple. While an LCD display showing all sorts of info would be cool, why not just use a couple of simple chips and forget the microcontroller? That would make it much easier for others to duplicate and eliminate the need for programming The programming requirement would add a level of complexity beyond the grasp of many DIY builders. OTOH, for the prototype, the easier it was to see what was going on, and the more the adjustments and operating parameter possibilities it had, the more the things that could be tried out simply by changing the programming instead of having to modify circuits.
   So the prototype should be run by microcontorller, and doing the circuit hadn't been a waste of time. (Whew!) Once optimum performance was obtained, a dedicated circuit could be created.

   How to do a dedicated hardware circuit? Feedback would be the filtered DC output voltage. That, after all, is the desired result. One could modulate the pulse widths manually easily enough using a 555 timer and a potentiometer. But how to do it from voltage feedback? No doubt it could be done.
   But perhaps even that is a needless level of complexity. It could be even simpler: If the voltage was below the setpoint, there would be pulses. If it was above, there wouldn't. So: a 556 dual timer providing the 60 Hz and narrow high frequency pulses, and an LM339 quad voltage comparitor turning that on or off. But how to gate the pulses? Use a second and third comparitor on the LM339 as switches to gate the signals from the 556? Or add a CMOS logic triple 3 input AND gate: Low frequency is high AND high frequency pluse is high AND voltage is below setpoint will send the pulse to the coil driver. Or something like that.

   I decided to continue using the car motor controller until and unless I got some results. The idea of a tuned resonant circuit, if it worked, or perhaps some other idea, might necessitate more changes to the design, so I would wait.

Electricity Storage

Rechargeable Battery Making
... NOT with oxalate electrolyte?

   As the battery performance deteriorated, I started thinking it must be the zinc electrode that was the problem. But how could that be? I looked at the old one. The connection tab seemed almost ready to break off where it came out of the case. With a little touching, it did. This was something I has also noted with the nickel-manganese cell negative electrodes with zinc current collectors: where they were not in contact with the Mn powder, they deteriorated and I ended up switching to graphite current collectors. I didn't think it would happen since I wasn't pushing it up to the -Mn (metallic) voltage. But there it was.
   I suspected it had something to do with hydrogen gas bubbling off, but there are no electrochemical reactions shown to combine zinc with hydrogen. Some reading on line on the 10th disclosed that in flooded nickel-iron batteries, both the nickel hydroxide and the iron absorb a large amount of hydrogen over the life of the battery. In fact, the nickel hydroxide absorbed up to 22 wt% - four times as much as a 2015 "target to aim for" for hydride storage. The problem with using them as hydrides is they don't release the H atoms under normal operating conditions. Surely then zinc electrodes will also absorb hydrogen - and apparently in doing so become brittle and break.
   A big part of the problem is doubtless that I put in charged (metallic) zinc and probably quite discharged MnO2 (Mn2O3, MnOOH...). So as the MnO2 side charges (or is it to Mn(C2O4)2 in oxalate?), the zinc side merely bubbles hydrogen. I didn't think it could be absorbing it because there were no electrochemical reactions shown for it, but turning it into a hydride is what these relatively new papers seemed to suggest. The second electrode wasn't (yet) as brittle as the first one.

   So my next endeavor was to put 12 grams of yellowish ZnO powder, which needs to be charged to Zn metal, into the (2nd) zinc electrode. Another consideration was that my ultra-fine zinc (metal) powder, at least the "flakes" jar, seemed to be so densely packed that it kept electrolyte out and didn't work well. By putting in ZnO, when it charged to Zn, there should be some spaces left over between the zinc atoms for the water. At least, with little knowledge of crystallography or how atoms behave, that was my theory.
   I sprinkled in the powder and smoothed it off with a teaspoon.

   Then I pressed the electrode closed, then I put in ordinary staples. It was tough going, but they went in. If I had used the heavier staple gun, say with cardboard behind, I'd have had to bend all the staple ends over. Finally I took a hammer and pounded the staples flat anyway, which pretty much put some momentary pressure all across the electrode to compact the powder.

 12 grams of zinc oxide is 820 AH/Kg * (65.4 / (65.4+16.0)) = 7.9 AH theoretical capacity. Add to that the capacity of the degraded sheet zinc itself? Surely over actual 10 amp-hours? A second electrode would probably be needed to get it up to 20 or 25 to match the plus side.

   It was soon charging at 120 mA. Same sort of low currents as before. I decided it was time to try some sodium oxalate to see if that would help boost things. I combined some oxalic acid with some sodium hydroxide. (I left it a little acidic in preference to having free sodium hydroxide.)
   It still had low currents, and it still had the same sort of high self discharge as all my cells of any chemistry have always had. Nothing I try seems to fix it. Why can any commercially made cell sit there and hold its charge, but all mine bleed off? I have some theoretically great stuff, all going to waste because I can't actually say to anyone "Here's a great battery that will work well." What, what am I doing wrong?

"In theory, there's no difference between theory and practice. In practice, there's a great difference." - (I forget who said that).

Manganese-Metal Hydride cells?

   A thought occured to me... Sometimes my cells have quite a few amp-hours but usually at voltages below a volt. Could that be from hydrogen storage rather than from the chemical reactions I was expecting? MnO2 at pH 12 is only +.25 volts. Hydride at pH 14 is -.833 volts, and less at lower pHes. Found a Pourbaix diagram for just water: it's around -.72 volts at pH 12. So  +.25 - -.72 = .97 volts. Hmm!!! A manganese-metal hydride cell, with zinc as the hydride? How much hydrogen would zinc store? Or should I say, how much hydrogen would zinc oxide store, since zinc would convert to its oxide form (or oxalate) at such a low voltage and stay there. If the corrosion of the zinc sheet metal is an indication, it either holds a lot, or is easily subject to weakening by hydrogen infusion.

   Then I looked up, specifically, "zinc hydride". Unmentioned in the electrochemical charts, zinc will form a white powdery compound, ZnH2, in the presence of hydrogen gas. As I say, I had previously looked to make sure it didn't, and it wasn't shown the electrochemical charts. They didn't include gas. This is probably what I've been missing, or at least is something I've missed seeing all this time.  I mistook the white powder for zinc oxide - and vaguely wondered why it didn't seem to all charge to metal with all the charging.
   Of course, if it was so easy to make a hydride that would work well in a battery with something as common as zinc, it probably (but never say never!) would have been found years if not decades ago. Of course it would be a problem that it needed H2 gas to make the hydride rather than splitting water into H+ : OH- ions and absorbing the H+. And it said ZnH2 "gradually" decomposed into Zn metal and H2 gas - or very quickly in acid. It might not come out "on demand" when one wanted electricity, and especially not by turning OH- ions into H2O. (But it means all I have to do is wait to decontaminate the electrodes. How long? Also the H2 comes out "quickly" at over 90°C. I set it on the woodstove for some hours.)

   Single use cells don't have the hydride problem because the hydrogen is only formed during charging. And in a rechargeable cell, if the zinc isn't overcharged, the gas won't form and the problem won't occur. In trying to charge the MnO2, I've been overcharging the zinc and it bubbles hydrogen. Apparently I need to charge the MnO2 electrode before it's put into the cell, so that both electrodes are already charged when starting. Then, it needs to be positive limited so that the positive side bubbles oxygen before the negative side bubbles hydrogen. When that happens, as in NiMH and NiCd dry cells, the oxygen finds its way to the negative electrode and causes discharge, so that hydrogen is never formed. The reason given is that hydrogen gas pressure would burst the dry cell. Nobody ever said anything about hydrogen contaminating the electrode!

New zinc electrodes... Works!

   It seemed like a good chance that the hydride accounted for the self discharge. So the next thing to do was to make a new zinc electrode, put it in, and not charge it. The MnO2/Mn(C2O4)2 should have been long since charged unless the hydrogen affected it, too, and the metallic zinc is already the charged form. So the cell should be full voltage, should not self discharge, and should work well to run a load - at least until it's charged.

  Before disassembling the cell, I let it sit for the better part of a day. The voltage dropped to about 1.15 volts, rapidly at first then more and more gradually. Whether it was still dropping very very slowly or not I don't know because that's when I took it apart, but that particular voltage acquires significance below.

   I made 2 zinc electrodes on the 14th, just single sheets of zinc, sanded to roughen the surface and etched for micro roughness. (Then cleaned in solvent to get rid of any chloride remaining from the ferric chloride.) I put one on either side of the MnO2 electrode and put it together. I added a little (more) sodium oxalate. (And I finally remembered to check the pH. It was about 12, as expected.) The cell read 1.4 volts and rose over 10 minutes to 1.45. That much seemed promising. Then I shorted it and the currents were no better than usual. And it took a while just to come back to 1.2 volts. So I wasn't expecting much.
   But I connected the 11 ohm load. As expected, the voltage went under a volt, to about .97 volts. Unexpectedly it started to rise, and was soon up to 1.018 volts in a minute. It hit 1.019 and held over 1.018 for over 5 minutes. But it didn't start dropping like on most trials. Supplying ~91 mA, it stayed up there. It started dropping by about only 1/2 a millivolt per minute instead of tens of millivolts per minute. Well, yay! Being pushed down to a volt to drive just 90 mA didn't say much for current capacity, but for once, finally, something was behaving like a battery should!
   After a full hour, having delivered 90 mA-hours of current it was down to .994 volts, just 25 mV below the highest point. It started at under 1.3 volts before the load was connected, and when disconnected, it only rose up to 1.152 volts after 10 minutes, so it couldn't be said it that sourcing 90 mA had dragged it down from a 1.5 volt level.

   But why should the voltage be so low? First we can apparently say it was dragged down about .2(?) volts by the "heavy" current being drawn. So we might call the essential open circuit voltage "1.2 volts".
   This was fully charged metallic zinc. Either it was ~ -1.2 volts or it wasn't. It must have been. Let us look to the manganese side, then to see about intermediate voltages.

Reading between graph lines as best I can here are the theoretical voltages:

MnO2 (+.28 v) - Zn (-1.17 v) = 1.45 V
MnOOH (-.02 v) - Zn = 1.15 V
Mn3O4 (-.27 v) -Zn = .90 V

Mn(OH)2 to Mn(metalic) reaction is more negative than zinc, so Mn(OH)2 wouldn't make a voltage.

   If just a few Mn's were charged to MnO2 (Mn valence 4), we might get the initial 1.45 V, but as soon as any current was drawn, they would be discharged leaving MnOOH (or Mn2O3; or Mn2(C2O4)3) (Mn valence 3). 1.15 volts was in fact exactly what was left after the discharge test (after recovery).
   I think it's a pretty safe conclusion then from the voltages seen that the manganese, in spite of the countless hours of charging it had with the other zinc electrode, was in the state MnOOH (or Mn2O3; or Mn2(C2O4)3) rather than MnO2 or Mn(C2O4)2 (ie, Mn valence 3 instead of 4), and that for some reason that's the state it wanted to stay in rather than charge further. Putting in a higher voltage forced some of it to charge to MnO2 but it didn't want to stay there, so it self-discharged to MnOOH and the voltage wouldn't stay up long at the ~ 1.45 V level.
   In fact, in the next discharge test, it continued down from where it left off after the first one, as if no charging had taken place.


   Could that have been an effect of the hydrogen? As a gas it would have entered the positive electrode as much as the negative. Perhaps at the voltage of MnO2, the H2 combines spontaneously with the Os to make Mn(OH)2, an overdischarged form that needs to be charged again. So that's my present theory. If that is so, once again, it will only happen if the negative electrode is overcharged and bubbling hydrogen gas.

   Another possibility is that it has something to do with oxalate. If so maybe there's no point trying to charge it higher than 1.15 volts? One good thing is that if it can't charge to valence 4, it will never charge to permanganate, valence 7, which would eventually dissolve. OTOH it would be poor if it only moves one electron instead of two, halfing the amp-hours per gram - and the cell is only theoretical ~1.15 volts instead of ~1.45. That's a lot less energy.

   On the negative side, the zinc would turn into ZnC2O4. Could there be some problem reversing this reaction too? Or was it just the plus side?


   So... how do we get the Mn side initially charged up to MnO2 (or to Mn(C2O4)2?) to match the charged Zn 'trode? I tried looking it up but found no useful information. For converting nickel hydroxide into nickel oxyhydroxide one uses bleach (sodium hypochlorite, NaClO) and I suspected that would work with manganese too, so I took the cell apart and put the electrode in bleach solution to try to oxidize it to MnO2 non-electronically. A bit of the powder came out: the gas probably pushed it out the too large perforations and the cracks around the edge. Then I saw tiny bubbles coming out for a while. It also started turning the connection tab blue-green - copper oxide? I had planned to leave it in overnight, but I didn't want to lose the connection tab, so I pulled it out, and then rinsed it by putting it in water and changing the water a few times. (Please, no chloride in the cell!) Hopefully enough Mn2O3 would have become MnO2 to make the cell a solid 1.45 volts. Then the question was whether that would stay or if the self discharge would appear and the cell would drop back to 1.15 by itself.
   I put the cell back together. It never went above 1.15 volts. Let's see... nickel hydroxide wouldn't charge to oxyhydroxide. Manganese wouldn't charge to dioxide level (really to Mn(C2O4)2). Maybe it won't charge at all? I was reluctantly forced to the conclusion that oxalate was somehow forcing the voltage down to where substances won't charge. For all the seeming advantages, it just wasn't working.

   But I could try sodium oxalate again - ether by itself as the main electrolyte, or else as a "catalyst" for the potassium oxalate, noting how sodium (as sodium-hydrogen sulfate) helps get lead-acid batteries to charge better by getting the sulfate off the plates. Might it help get oxalate out of the zinc electrodes? Or might it help get oxalate into the Mn electrode? It's not very soluble, but it's 20 times more so than just calcium hydroxide, so I won't give up on oxalate without a bit more experimentation. (I ordered some on January 5th as I edited this. That way I know I can trust the quality. I had been wanting to for some time, but hoped to put the order off until I wanted another chemical or two as well. I still hadn't thought of any, but I finally broke down and ordered a microscope too. It was the cheapest one and I hope it's half decent. By "Coulter Optical" - the same outfit that made my low cost 10" reflector telescope about 3 decades ago. That had very good optics.)

   A final thought occurs to me: It's eay to convert between HOH and OH- ions in KOH solution. A reaction can use or create a bit of water, which is balanced overall, and perhaps in initial charging some hydrogen or oxygen bubbles off to attain the balance. With K2C2O4, if a C2O4-- ion is absorbed into an electrode with none released by the other electrode, it leaves two free K+ ions, which will no doubt convert with water to 2 KOH and release hydrogen as H2. Maybe I have to convert the MnO2 into Mn(C2O4)2 initially before it's put into the cell? Otherwise, what about the extra potassium when its C2O4-- is absorbed into the plus electrode during charging? Is that why it won't charge?
   If I'm using Ca(OH)2, what about using CaC2O4? so any left over Ca can become more Ca(OH)2? Oh ya... CaC2O4 isn't soluble.

   Insoluble? Egads... does that mean one could use CaC2O4 <==> Ca(C2O4)2 as a positive electrode substance?... or is the voltage (+1.55 V) too high?... What about in ethaline DES?... or is Ca(C2O4)2 an insulator?

(Wild possibilities are multiplying... Egads, how did I get started doing chemistry?... oh ya... for better batteries!)

JUST Calcium Hydroxide Electrolyte?

   Something occurred to me that I should probably have tried long, long ago. Just calcium hydroxide for electrolyte. It seemed like a silly idea because it's so insoluble compared to any other electrolyte. Here's a table of solubility for various potential electrolyte substances (grams of substance per 100 mL of water):

NH4Cl 37 Std. dry cell - dissolves the zinc
KCl 34
alternative, but also dissolves
also dissolves metals
some aspect of this doesn't
seem to be working??? (so far)
electrolyte, or catalyst for above?
Used in all previous alkaline cells:
OH- ions, pH 14 - very caustic
Ca(OH)2 0.173
OH- ions... but not very soluble!
BUT it makes for pH 12 solution
instead of 14 - much less caustic

     Other "alkaline earth"
       metal hydroxides:
Mg(OH)2 0.000963
A.K.A. insoluble
Sr(OH)2 1.77
no doubt too soluble - pH 13+
Ba(OH)2 3.89
too soluble - pH 14

   What does insolubility do? It limits the current capacity. My currents have been so low anyway, what's the difference? Maybe the oxalate does nothing useful at all and that's already why the currents are so low? Other than that it has real and unique advantages:

1) Zinc, nickel and manganese won't dissolve in it
2) It makes the solution pH 12 instead of pH 14, so zinc doesn't form that soluble ion that plagues it at pH 14
3) It is virtually bound to work - just like potassium hydroxide - because all the ions and reactions are only hydroxide.

Low current batteries that WORK would be much better than high current batteries that don't work. I would be satisfied with batteries that need a much larger surface interface area because of slow electrolyte as long as they actually work, last a very long time, and have very high energy storage. They would be (as I have been trying to create all along) an improvement over everything out there.

   I suppose a similarly very, very weak solution of potassium or sodium hydroxide would do much the same thing. But with calcium a reserve can be included in the cell to replace any that becomes calcium carbonate. No such reserve is possible with potassium or sodium. And calcium carbonate (rock) isn't soluble, so it exits the solution (lodging in an electrode somewhere, no doubt). Unless the top is wide open it should take an age to convert all the spare Ca(OH)2 to CaCO3. Potassium and sodium carbonate are soluble. The only answer for them would be to change the electrolyte, as is in fact occasionally done with flooded Ni-Fe (etc) alkaline cells.

   Now... did I need to make new electrodes because the old ones were contaminated with oxalate, or could they be cleaned out? Perhaps it could be dissolved out in... hmm, gasoline? That was the first thing to try (17th, AM)

   I suspect that with just Ca(OH)2 electrolyte, I could go back to using higher voltage nickel for the plus side. Ni-Zn would be highest energy voltage and energy (cars); Mn-Zn would surely be cheapest overall (on-grid or off-grid energy storage). (And then there's Ni-Mn or Mn-Mn?)

   Later I checked out the other "alkaline earth" elements: magnesium, strontium, barium and radium hydroxides. (No figure was given for radium hydroxide in Wikipedia's table. I wouldn't have wanted to use it anyway.) No others were close to calcium. Mg(OH)2 was insoluble and the others apparently at least somewhat too soluble. They'd have made it pH 14 or very close to it, and so seemed to have no advantage over KOH. Ca(OH)2 was the one suitable one.

On-Grid Energy Storage?

   It's one thing to speak of storage by water reservoir above a hydro generator to provide electricity for weeks and months. It's quite another to speak of a few hours storage just to smooth out intermittent natural energy sources and daily load fluctuations to avoid turning (eg) a diesel generator on and off as loads and intermittent renewable power sources come on and off line.

   Pumping water to a storage tank on a hill to run a hydro generator might work well for that. But could a horde of inexpensive Mn-Zn oxalate batteries accomplish that job too, probably more efficiently and for a lower price? This is a major potential use if they can be made to work.

Haida Gwaii, BC Canada