Turquoise Energy Ltd. News #58
Victoria BC
Copyright 2012 Craig Carmichael - December 4th, 2012

www.TurquoiseEnergy.com = www.ElectricHubcap.com = www.ElectricWeel.com = www.MushroomOutboard.com

Features: Peltier Element Heat Pump for Home & EV Heating (see Month in Brief)

Month In (relatively) Brief (Project Summaries... and many little things not warranting separate report headings)
- 12V Peltier element Fridge is cold enuf: probe has been reading 2º too high! - Peltier Element Heat Pump for home & EV heating - 3D Printing: a 12V NiMH battery case - Battery Making - "Mini Electric Hubcap" + Bicycle Rim Motors - Mushroom Outboard: preparing for 'pot metal' propeller casting - aluminum casting for heatsinks - Torque Converter & Sprint Conversion - Greentech Exchange Solar Event, Cheap solar heater, & Crowd Funding.
In Passing (Miscellaneous topics and editorial comments)
- Banking and the Fiscal Cliff: history and beyond it
Planetary Gear Torque Converter Project
* Balanced ropes + 1 to 1 flat belt + trial with no great results
Mushroom Outboard Motor Project
* Preparing to cast the propeller
CNC Farming Machine Project
* How did I get onto another project? Don't I have enough unfinished projects?
* 18' Wooden Ladder Gantry: easy to make into a supported truss - wheels & tracks - motors - drive belts
Superinsulated 12VDC Peltier Element Fridge Project
* Is colder than measured - cold enough to use
* Commercial & Homemade Aerogel possibilities
Large Format NiMH Batteries - Take 3
* 3D printed, ventilated, stackable 12 volt battery cases minimize size and added weight.
Turquoise Battery Project
* First porous plastic electrode using my 3D printed electrode pockets is...
* An iron electrode with improved current capacity, charge rate and charge retention
* "Graphite Foil" & "Flexible Graphite" likely best solution for posode current colletors (now on order)

No Project Reports on: Electric Hubcap System, Weel motor, DSSC solar cells, LED Lighting, Pulsejet steel plate cutter, Magnetic Heat Pumping, Magnetic Motion Machine

Newsletters Index/Highlights: http://www.TurquoiseEnergy.com/news/index.html

Construction Manuals and information:
Electric Hubcap Motor - Turquoise Motor Controller
- 36 Volt Electric Fan-Heater
- Nanocrystalline glaze to enhance Solar Cell performance - Ersatz 'powder coating' home process for protecting/painting metal

Products Catalog:
 - Electric Hubcap Motor Kit
 - Sodium Sulfate - Lead-Acid battery longevity/renewal
 - NiMH Handy Battery Sticks
, Dry Cells
 - LED Light Fixtures
Motor Building Workshops

...all at:  http://www.TurquoiseEnergy.com/
(orders: e-mail craig@saers.com)

November in Brief

My September 19th, 2012 Talk at VEVA (@ BCIT, Burnaby BC)

   The Vancouver Electric Vehicles Association video of my talk at the meeting was uploaded in October. (I'm just late getting around to some of my e-mails.)

Craig Carmichael of Victoria BC presents his alternate energy inventions
Part 1 - http://youtu.be/MPaP8pnRukw
Part 2 - http://youtu.be/a5d3azR8O8Y
Part 3 - http://youtu.be/F_fx4AoLMEE

I confess... I haven't watched it yet.

Solar 12V House Wiring and Equipage

   I used the 12 volt plugs and sockets for a convenience outlet in the main panel (for an LED light to illuminate the dark closet) and for the Peltier/solar fridge. They work. I didn't cut a box into the wall near the fridge yet, so the wire is just hanging there. I keep thinking there must be some smaller box than an 1110 electrical box, to cut a smaller hole. Or a surface mount box. Since I haven't had time to make more LED lights, I bought some sample LED bulbs at Deal Extreme. One 'cool white' (6500K) one I especially liked and soon ordered a few more. According to the figures given, it's just 10 watts and 1100 lumens - fabulous efficiency. I haven't measured the power, but the brightness is there. Some reviewers at DX.com said it was equivalent to 75, 100 or even 140 watts tungsten. One of the samples burned out instantly. Later I printed out my order and found that it was a 12 volt bulb, with a standard "E27" 120 volt base. Rats, I wrecked the one for the solar power!

The main panel by month's end.
Not much progress, but the voltmeter looks nice
and the 12VDC convenience outlet for the light is the first one properly installed.
An ampmeter would be a good complement to the voltmeter, but my matching
one goes to 5 amps instead of 50. (They were bought for testing wave power.)

Fridge -- and a Peltier Element Heat Pump for Home and Electric Car Heating

   The 12V Peltier element fridge is better than I thought. The multimeter thermocouple probe was reading about 1.5 to 2º too high. When the meter read 8º, it was really 6 or 6.5º. At the floor near the ice tray, it apparently hovers around 3 or 4 and gets as low as 2. This was more in line with expectations considering the nearby ice tray and the 3" foam insulation.
   With temperatures now seeming cold enough, I started using the fridge for some day to day things. It's unexpectedly nice to use. The light 3" foam lid opens with a touch, and then you just drop it the last few inches. It closes itself really quickly, with a 'whuff' of air cushion at the end. Everything is accessible just as I'd envisioned. Since you're not reaching behind things to get other things, the time the lid is open is usually very brief, with little heat gain. And the cold air doesn't dive out the bottom when it's open like a front door fridge.
   I still need find time to make and program the smart solar control and do the drain funnel for under the ice tray.

Main hot side plate, cold side angle aluminums.
   But on the 25th, I got my nephew to cut and drill holes in the pieces I'd selected for an experimental peltier element heat pump. It has 6 peltiers at up to about 100 watts (input) each, in two rows of three. In operation at full power, each peltier cold side could be expected to cool with about 40 watts, and the hot side to heat with 100 + 40 = 140 watts. Thus the total potential was 600 watts input yielding 840 watts of heat on the hot side.
   The EV is the most interesting installation here, as the heating power has to be supplied by the same batteries that power the motor, and in winter, driving range is substantially reduced by the heater/defroster. A heater delivering any extra heat for the power is thus highly attractive.
   However, a reduction in power bills would mean a peltier heat pump is also economic for home heating versus electric baseboard heat. All else being equal (mostly assuming peltier elements have decently long service life), the power savings will rapidly eclipse a higher unit cost.

Outside heatsink fins

   In my unit, the heatsink fins go vertically. I'm hoping the outdoor cooling side can be convection reheated. (Should work on windy days, anyway. In a car, the car is moving.) The inside will have a fan. It could be window mounted (or in my case placed in an existing hole I had made long ago in the wall for a passive "Trombe Wall" solar collector, now long gone), or installed in a car. I would hedge the bets: for experimentation and initial use it would use the wall hole and run on a 50 volt, 5 amp transformer with 5 elements of the 6 in series, to draw just 5 amps for a 250 watt input, yielding 350 watts of heating power. But with minor rewiring of the elements into two banks of three, it would become a 36 volt, 300/600 watt car heater (with 420/840 watts of heating).

   On the 26th and 27th I started putting it together. The details of the hole drilling weren't very well thought out and holes didn't work out or even line up. The warm side heatsink fins were hard to fit on and then they didn't conduct the heat very well - a critical problem. I made notes for next time. (I should have had my nephew assemble it. He'd surely plan out and execute the job much more carefully next time.)
   I also began to realize that some of the "aluminum" fins were probably some sort of alloy - they weren't attracted to magnets, but they seemed much harder to bend than aluminum. It has to be pure to conduct heat well.

3D Printing - 12V NiMH D Cell Battery Holder - out of order

   On the 5th I was printing a D cell 12V battery holder. About 3 hours into the 5 hour print, things came to a sudden stop and there was smoke coming from the printer.
   There was enough printed to put in some batteries and try it out. It works. Complete cases will be stackable to create large NiMH batteries in multiples of 12V and 10AH.

   For the printer, I had visions of fried circuits, having to order a new circuit board, major rebuilding to do, and delays getting projects done.
   I had tried successively higher temperatures to get better results printing ABS, and this one was the hottest try yet, 270ºc. Fresh plastic lines sagged too much covering over holes and I wouldn't have gone quite so hot again. Probably 250º as suggested some places was about right. I was trying to insert bits of cardboard 'insulation' to minimize warping, and evidently I hit the extruder temperature sensing thermistor wire when the carriage moved towards me. (Duh - the printer can be paused!) Somehow this must have shorted something out - the thermistor blew apart and the insulation of its wire was scorched all the way along the ribbon cable. Bad temperature readings is probably what stopped the printer. It turned out that the thermistor and its wire seemed to be the only problems - otherwise, everything still seemed to work. Digikey was out of those thermistors and I had to back order. (and now they're to be further delayed!) Sigh! With the 3D printer out of order, I turned to other things. The magnetic motion machine was out too, as I wanted to print 3 more magnet arms for it.

Battery Making
A porous plastic electrode. See-through plexiglass
backing plate shows the active chemicals.
Fine powder on green face leaked through the coarse pores.
Nickel plated copper mesh current collector; RTV silicone to
seal the top (unsuccessful - it turned to goo in the electrolyte).
   I took porous ABS electrode pocket covers I'd already printed and made 3 electrodes. Two attempts to make a monel electrode with the monel mix I made in 2008 plus some graphite were failures. Only later did I remember that metallic nickel has the unusual property of not oxidizing in strong alkaline solution. Duh! It should work okay in my "moderately alkaline" electrolyte, but the point here was to make a single electrode and test it in a Changhong Ni-Fe cell with the original nickel 'posodes', where everything but the test electrode would be a known, working quantity.
   The third attempt was a successful iron 'negode'. It contained 25g of coarse ferrous oxide "sand", 10g of monel alloy, and .35g of antimony sulfide, and had a copper foil current collector. I used the coarse powder because the nano powders in the first two had been leaking through the pores in the plastic. I still put a piece of watercolor paper across the face when I installed it in case any came out. Scaled for size, it seems to work a little better current-wise than the original iron electrodes, despite being twice as thick relative to the current collector. Theoretically it's 17 amp-hours worth of iron, but it looks like it's attaining around just 2 or 3% utilization. This is probably because the coarse grit has low surface area for its volume - most of the Fe is in the interior of the grit particles where the electrolyte can't reach it. But it has the following theoretical improvements:
* The monel makes it more conductive. This leads to ability to take charge faster or more efficiently, and to drive heavier loads with lower losses. Pure copper powder should be as good or better, but monel (Ni:Cu ~~67:33% alloy) is what I had. The concentration needed to attain really good conductivity needs experiment. Furthermore, I didn't get to see if the graphite powder would have worked. From the voltages it seems it should, but I heard from one source that it bubbles hydrogen and discharges the electrode. The Sb2S3 stibnite just might make the difference to that problem if it exists. But since Mn-Mn is a wholly better chemistry, I'm not going to bother perfecting this one.
* The antimony sulfide raises the hydrogen overvoltage. This reduces bubbling of hydrogen gas, which in turn improves charging efficiency and charge retention, reducing if not curing the most negative characteristics of typical nickel-iron cells. (Hydrogen evolution may perhaps be reduced to the point where nickel-iron dry cells become practical.)
* The plastic electrode body is cheaper than metal ones. Also it can touch other electrodes without causing a short circuit, so they can be more closely packed inside a cell.

   All this has allowed me to make single electrodes to use with existing nickel ones and compare against known iron ones. It's good practice, learning, and test of a couple of theories, and has refined my cell design ideas and led to improvements such as sealing the top of the electrode pockets with wax. But the similarly long life and higher energy 2 volt manganese cells are still the main object.

   I shot video footage as I made the electrodes, and I "copied" them onto a Windows PC machine, then erased the camera to free up the space. But it turned out the computer, despite long clips taking quite a while to "transfer", had only made "aliases" - links to the files on the camera. I discovered this only when I went to edit it all into a battery making video. Only the names of the file clips remained on the computer, claiming they were the videos but with no data. Instead of editing the video and getting it online, I made another electrode and shot more footage. I didn't get around to making it into a presentable video.

   I read a new term "graphite foil" and on searching for that found rolls of "flexible graphite" on line. Was this what I've been missing all this time for positrode current collectors, for want of which I've been doing carbon (graphite) fiber and the grafpoxy mix? I ordered some and a large roll arrived December 4th just as I was finishing up this newsletter. It looks like a thinner version of "expanded graphite". (And I wish I'd ordered .010" instead of .020".) But I'll try a few things - in particular chemical surface preparations - and try it in plastic pocket electrodes, and see what I can do with it. From their cycle life graphs, it appears Aquion got good results and long life out of their "graphite foil".

Electric Hubcap Motors

   A couple of people were interested in smaller motors for converting small motorcycles and scooters to electric, as well as the motor controllers and new chemistry batteries. I had thought of a smaller diameter "Electric Hubcap" motor, which would have 6 coils and 8 magnets instead of 9 and 12. To keep things simple, they'd be made much the same way with the same components. They'd be 24 volts, ~3KW, 3000 RPM, 9" in diameter with about a 7.5" rotor, and ~22 pounds.
   In looking at some youtube videos of motors for bicycles I started to consider (tho not for the first time) a big "partial rim" motor. This is made in two separate parts. On the wheel, well out towards the rim (probably attached to the spokes) is a steel ring mounting supermagnets. The larger the wheel, the more magnets are required. The wheel becomes the motor's rotor. Attached to the frame facing the magnets is an arc of coils, a "fender" to the side of the wheel, rather than a complete circle. This drives whatever magnets happen to be across from it at the time.
   I think 3 "Electric Hubcap" coils might give a minimal ride on level pavement, 6 would give a good ride, and 9 would (probably) be overkill on a regular, lightweight bicycle. An interesting feature is that for magnets at the rim of any wheel, a small wheel will theoretically perform the same as a large one - the torque available is proportional to the torque required for the size.

Mushroom Outboard & Marine

   The fisherman who wanted to troll electrically last spring called again on the 10th, and I still haven't made an outboard. I still have the Honda outboard leg, but my 'production' Electric Hubcap motors still won't quite fit under the hood. I talked with Jim Harrington who has converted a couple of outboards to electric. An appropriate motor that would fit wasn't kicking around, so we decided to use a sealed 3 HP Leeson DC motor and controller that he had surplus at AGO Environmental, and hence available at low cost, and install it as an inboard to drive his main propeller shaft with a V-belt and pulleys, in order to get him going for next season.
   I like the direction switch on the controller: when flipped from forward or reverse, it stops at neutral. It can't be flipped from one direction to the other without stopping and releasing pressure on the switch at neutral. Where do I get these for electric cars?

3 HP motor & controller for fishing troller inboard

   Along with the battery making, I collected up supplies and equipment to cast the extra-efficient looking propeller for outboard motors. Not exactly an invention on its own, but perhaps a product in itself. And the smaller, higher RPM motors would fit in the Honda outboard leg, which could make that a good one-off product. In my first attempted casting on December 3rd, I realized in the nick of time that zinc-aluminum alloy would doubtless corrode rapidly in salt water. Melting pure aluminum on the stove worked out badly, tho it did melt and I'll try it again, and I used the mini-kiln. The first propeller blade cast had air pockets and gaps. It wasn't usable. (...fill the gaps with epoxy?) But since it was my first ever attempt at metal casting, I sanded and polished up one face a bit for a picture. I examined it with a friend with casting experience and we discussed how to get better results.

Cast Aluminum Heatsinks?

   Along with casting the zinc-aluminum alloy propeller (now itself pure aluminum), I started to think it might be nice to cast straight aluminum, perhaps with fins: for heatsinks. That had the potential to speed up making several items from motor controllers to peltier element heat pumps. I was seeing various ways people were melting aluminum and got so intrigued I forgot about the obvious one. I was dubious the hotplate would be practical. On the 25th I tried melting some in a pottery crucible in the woodstove, but it didn't get hot enough where I set it. A propane torch did the trick eventually, but used a buck or two of propane. I read about making an induction heater to melt aluminum. I looked up the actual melting temperature (660ºc), and realized my electric mini kiln could do it. Duh! Problem? No problem! On the 27th in about an hour I successfully poured a small ingot of aluminum into a muffin tin hole, tho with lots of slag left over. It's doable. I couldn't clean out the pottery crucible - the stuff was stuck solid into the pores. Then I found one can melt aluminum on a kitchen stove burner in a stainless steel pot. (See the "Mushroom Outboard" project report for details.)

   On December 1st I found a source of pure aluminum drill shavings that had been just going into the garbage. I took a big but pretty light box of them. If I do much casting I'll be back!

Electric Car & Torque Converter

   I got the last piece of the new "balanced pull" slip clutch tensioning system done on the 14th. On the 15th I put the 36 volt battery on charge and since it wasn't raining or too cold, I started taking things out to the Sprint car and reinstalling them. Before, I could hardly stop the rope from holding the pulley stopped. Now it needed far too much pressure to get much drag on the pulley. Maybe I'll try some kind of noose arrangement.
   I didn't get the car to move. I think I was too optimistic and should have stuck with the 4 to 1 chain reduction drive for now. With the 1 to 1 belt drive, the converter could be putting out even twice the torque now, and still there'd be only 1/2 as much to the wheels.
   Belatedly I thought about the elementary step of measuring the torque on the shaft with the torque wrench, whether the car moves or not, or even without connecting to them. After all, I ground a hex shape into the end of the shaft for that very purpose. It proved pretty useless for measuring pulsed torque, but I should have been doing that from the beginning of the PGTC to measure results.

Greentech Exchange - Cheap solar heater

   On the 29th, after spending the day on brochures and an ad, I went to a Greentech Exchange Panel Presentation on Solar Energy. One speaker who couldn't make it in person from Seattle presented via Skype. His voice and projected face reminded me of "Hollie" from the Red Dwarf TV show, but it worked.
   His presentation was on having black metal with holes or black cloth facing the sun on a south wall and drawing solar pre-heated fresh air into the structure through it via a duct near the top. This meets HVAC fresh air requirements with less energy when the sun is out. It looked amazingly simple and cheap. I wonder how it would work with a return air hole at the bottom and solid black material? It sounds so simple I may just revive my old "Trombe Wall" from 30 years ago, but needing no glass or transparent plastic on the front. (Of course, the grape vine and other leaves would still shade it over except in early spring.)
   As invited, I set up some demos of solar 12 volt stuff and the papers on one of several tables in the lobby. It sounded like most of the panelists could really use one or more of the things I've been developing, and in the question period I said something about that and to take my card if we couldn't find time to talk.

   Rob MacGregor, the BC director of Sun Country Highways spoke with me afterwards about batteries. Sun country is building a network of EV charging stations across Canada so that electric "transportation appliance" drivers will have somewhere to go to recharge wherever they are. This addresses the biggest fear about EVs: running out of power somewhere. (Someone is presently driving coast to coast, Newfoundland to Victoria, in a Tesla. But there are many gaps still to be filled for shorter range EVs.) The second biggest complaint about EVs is the length of time it takes to charge the batteries, so a cheap, very high rate cell would be an excellent thing to have.
   I also talked with three mechanical engineering students who were about to graduate and mentioned my lagging torque converter project. It seems likely that one of them will contact me after their final exams.

Crowd Funding?

   Something I've noted repeatedly in my inventive career is that while I work on creating things to benefit everyone, almost no particular person or body feels it should devolve upon them to help foster or fund such projects when they are only one of the many who would benefit. In "R & D", Development has been the neglected little brother of Research.
   With some products now proven in principle and mainly needing setup for production, and all these markets for them, I may look into "crowd funding", which I've been hearing positive things about lately. People who wish to create new things are being funded not by rich "angel investors" or "venture capitalists", but instead by a "crowd" of ordinary people who like the idea and each contribute a small amount. Various terms and conditions for the funding are specified in advance, so the proponent and the payers are all clear what is expected.
   Thus projects to benefit the many may be funded by the many, each to a small extent. I could certainly have used something like that for previous great (IMHO) projects that impressed people and got good lip service, but went nowhere owing to lack of investment! But the Crowd Funding idea and implementation itself is a recent invention.
   But there may be other ways of funding, and there are often ways of setting up smaller scale production that cost less and reduce or even eliminate obligations incurred in order to get going.

In Passing
Incidental news, editorial opinions

Banking and the "Fiscal Cliff"

   Like most people, I've never paid much attention to the world of finance - until recently, when it started to look like catastrophe is looming and lone lunatics predicting the end of the world were joined by a rising chorus of variously qualified people predicting such a terrible financial collapse as might bring about the end of the civilization we have known for many, many generations. Even the last Globe and Mail newspaper I saw had the term "fiscal cliff" in several Business section articles. I started to think I should at least try to understand why it's happening. Youtube seemed like a place to see what "financial collapse" was all about.

   I think The best overview explanation of hyperinflation and monetary system collapse I've seen is a short two part video by Argentinian Adrian Salbuchi. (Search youtube for that name.) He says the cycle of monetary collapse is shorter in Argentina, and that they have gone through 3 hyperinflations and monetary collapses since the 1960s, whereas for us in the "first world", all we know of the subject is old pictures from the 1930s (yes, millions did die of starvation and poverty), and our "normalcy bias" keeps us from believing such a thing could happen again here. He held up an old Argentinian Peso and said it would take trillions of them to buy what one would buy in 1964.
   A lady named Ann Barnhardt said she closed her farm commodities brokerage in 2011 because after MF Global and PG Best stole all their customers' money ("and John Corzine still walks the streets a free man!") and then court decisions and new laws excused such theft, she decided that rule of law is gone and she could no longer guarantee her farmer clients wouldn't suddenly have all their money stolen. She advised them all to "get out" and keep their money in cash or precious metals. Her definition of what money is is worth hearing, too.
  One person said that precious metals were an okay temporary hedge, but that America was becoming a "third world" country and was likely to remain so for 5, 10, 15 or 20 years. He had lived in several of them and thinks that the only real riches in that situation is to have the means of producing something people need. Everybody needs food, and he had put all his money into a piece of land, a farm that that he could grow crops on.
   And there are many other points of view on the "fiscal cliff" and beyond on Youtube.

Salbuchi's proposal to have the government directly print money for specific public works, tax earnings appropriately, and then (hopefully) destroy any excess if there's too much in circulation, sounds much better to me than borrowing it from private banksters, to be repaid with interest. Where is the money for this interest to come from if the government can't print money? It has to come form money already in circulation. And if money can only enter circulation by a loan from a bank, it becomes a vicious circle of debt and more debt.
   Is it any wonder then that debts grow and grow until the Whole World owes far more money than exists, far more than can ever be paid 'back', and is now galloping full speed over the cliff of an incredible financial catastrophe?

   We must prepare with reserves of food and other needs, but ever be mindful there's a tomorrow after that collapse. The collapse will be a horrific event which may take many years to straighten out, but the banks have outscammed all the corrupt businesses through which a few families have controlled 80% of the world's economy, and hyperinflation or national bank failures will eliminate their vast monetary wealth. By the time things are re-starting, the power base of the corrupt will have been eliminated, and the world will be on a new, more sustainable and more spiritual track. Many of the fiscal cliff "prophets of doom" think the long term future will be much brighter. "The sooner the better." they say. A global change in consciousness can already be sensed.

A focused bit of American Financial History (an inevitably slanted view)

   There's nothing new in financial collapses, except for the global extent and the extraordinary degree of control the main banksters have attained over governments everywhere and over everyones' lives and fortunes - their ability to extract the wealth from most everyone, everywhere, by fraud. They foist worthless "credit derivatives", "credit default swaps" and "collateralized debt obligations" - weapons of financial terrorism rated "AAA investment" by the corrupt ratings agencies, who are under their control - onto other banks, pension plan funds, municipalities and other unsuspecting saps, running to trillions of dollars. Where in the early 1930's 1600 US bankers went to jail for their roles in causing the great depression, banksters and the ruling "cleptocracy" caused the 2008 crash with impunity and continue merrily on to bigger and worser things.

   But the finance problem was understood centuries ago. Britain grew its empire to cover 1/4 of the globe by a central bank which provided vast sums of credit to build a navy and army through fractional reserve financing, which then enforced British rule at the expense of the colonized people (including the Americans... and not to mention the common British people) to pay off the loans. The following is attributed to Thomas Jefferson. It seems it's not a real or single quote, but rather contains some of Jefferson's thoughts and words from 1802 to 1816, liberally appended to to sharpen and modernize the points he made. It does well describe America today, especially with millions homeless while more dollars worth of foreclosed real estate is in the hands of the banks than is owned by homeowners (albeit the bank holdings are doubtless 2007 inflated "real estate bubble" figures):

   "If the American people ever allow private banks to control the issue of their currency, first by inflation, then by deflation, the banks and corporations that will grow up around them will deprive the people of all property until their children wake up homeless on the continent their Fathers conquered...I believe that banking institutions are more dangerous to our liberties than standing armies... The issuing power should be taken from the banks and restored to the people, to whom it properly belongs."

 The indictments of Andrew Jackson (US president 1829–1837) sum up the problems so well they could almost have been written this year (Wikipedia):

As President, Jackson worked to rescind the bank's federal charter. In Jackson's veto message, the bank needed to be abolished because: Following Jefferson, Jackson supported an "agricultural republic" and felt the Bank improved the fortunes of an "elite circle" of commercial and industrial entrepreneurs at the expense of farmers and laborers. After a titanic struggle, Jackson succeeded in destroying the Bank by vetoing its 1832 re-charter by Congress and by withdrawing U.S. funds in 1833.

   Like today, the bankers of 1832 threatened collapse of the economy if they didn't get their way. Jackson replied that "You are a nest of vipers" and to do their worst - he would see them off. (cartoon is from that time) His veto of their charter renewal solved the problem, AFAIK until 1913 when the "Federal Reserve" corporation was created and usurped the US treasury as the printer of US dollars.
   Later (1834?), someone attempted the first assassination of a US president, saying Jackson's banking policies were keeping him poor. (His two pistols both misfired but later both worked reliably.) He was pronounced insane, but a disgruntled banker or two may well have put the ideas that Jackson was to blame for his ills into his weak mind - and then given him the pistols. It would seem likely that there's also nothing new about bankers and other vested interests trying to eliminate leaders that don't knuckle under to them.

   I've mentioned that I think America might dissolve into chaos and civil war after the collapse. Now hundreds of thousands from many different states have "requested that [their state] peacefully secede" via a federal government web site. Texas soon had over 100,000 who think they want it to be the lone star state again. (In fact it went from 21,000 to 70,000 petitioners in a couple of days after RT[.com] did a news article on it. Do so many Americans get their news from RT?) Doubtless it's largely people who didn't like the federal election results, but the numbers and the Texas "Secede!" bumper stickers indicate considerable disquiet. This movement could gather momentum rapidly, as the central government seems to be doing so little of real value for the wellbeing of its citizens and seems so out of control.
   The USA is next door from here, has had great influence over the world, and is well documented. What of the rest of the planet? Every country and region is different and has different conditions, but the more I see, the more it looks like the financial collapse will leave no land untouched. For example, China's industrial economy was partly built on selling goods to the west, which is now broke, so suddenly vast unemployment and abandoned construction works are spreading there too. Hyperinflation and bank runs too keep rearing their ugly heads in surprising places. Water is starting to gush through leaks all across the dam.

   Over long time periods geography is the biggest determiner of administrative regions. Might a looser North American union of the entire continent eventually come into being? with the component regions having more independence, being better governed, and providing people more freedom, than exists today? Of course, all intermediate regions including continents are of relative value only insofar as they improve peoples' lives. The whole planet will eventually be unified in some way as best seen fit at the time - it's the final way to end wars and have true, complete human brotherhood. But I think I've said such things before in Fundamental Principles of Democratic Government - Towards Utopian Systems of Governance.

Planetary Gear Torque Converter (PGTC) Project

   I left the project for a while, with the last piece being some way to attach the clutch/tensioning rope so it could tension around the big pulley with balanced forces. I was trying to straighten a scrap piece that looked about right for one side of some sort of bracket but was bent 90º, and it attained a bit of an "S" shape. I thought that might be useful, so I took it and I eyed everything up together. The bend put the end in about the right position relative to the pulley. "Some sort of bracket" was thus reduced to this simple bar of steel attached at one end with two bolts (instead of one and a separate diagonal piece to keep it from twisting) with an eye bolt sticking out to attach the rope to. If it bends in use, I'll use a still stiffer piece.
   At the other end of the pulley, I ended the friction rope ends pretty short, clamping them to two thin ropes which go over two smooth cylindrical 'slides' and then clamp together into one that goes through two small pulleys to the gearshift/tension cable.

   Originally, I could hardly stop the rope from locking up the clutch-pulley. Now when I tried it out, I couldn't get enough tension on the ropes to put much load on the motor - the pulley just spun. I took a small grinding wheel and (at risk of considerably shortening the life of the rope) roughened up the inner surface of the pulley as best I could. This helped some, but somewhere it needs more leverage. Maybe some kind of noose. A mechanical student I talked to later asked about steel "bands" that have been used with planetary gears, which I've heard referenced before. Perhaps I should look into that.
   I finally ended up grabbing the clamps on the ends of the rope (with my fingers just too close to the pulley) to tighten it. The car didn't move regardless of heavily loading down the motor. After some fiddling around and trying variations, I gave up.
   Evidently I was too optimistic about the torque it would produce. I still don't see why it shouldn't be hundreds to one. But with the one to one belt drive, the converter now needed 4 times as much torque as before. Even if it now had twice as much as before - surely it has more than it did - it would now be 1/2 as much at the wheels and might not move the car.

   I took it all off the car again to disassemble and re-install the 4 to 1 reduction chain drive. Better to have limited top speed than none at all. If it moves, it can be tweaked. If it doesn't, it's too hard to see what's happening.

...Come to think of it, I should be able to measure the torque on the output shaft with the torque wrench regardless of how it attaches to the drive shaft, or even if it's not attached at all. That's why I ground a hex nut shape onto the end of that shaft. Duh! That would be a good step. I wish I'd thought of it while everything was assembled, and even more, for each trial from the beginning.
   Another thing I should do is make a bigger 36 volt battery than 20 amp-hours. This one will only supply 50 or 60 amps without losing voltage, placing a limit on the torque of the motor. (Let's see... 36V/30AH is: 3 * 3 = 9 battery cases at 5 hours each is 45 hours of 3D printing. Up it to 70 AH and the Sprint should be able to run around Esquimalt.)

   But with the weather being mostly cold and wet, and having lots of indoor things to do, that's about all I got done on the car or torque converter.

Mushroom Outboard (Outboard Motor from Scratch) Project

   For the moment, this seems to have become "The More Efficient Propeller Casting Project". I started looking into casting the plastic propeller blades in metal. It seems "sand casting", as one might suspect, involves more than just sand, which would obviously crumble. "regular" sand is mixed with very fine sand so the pores are smaller. The simplest binder to hold the sand grains together is bentonite clay. This is thoroughly mixed and water is added.
   I thought a zinc-aluminum alloy would be easy to do because it can be easily melted on a stove burner. I could get a good hotplate and do it all outdoors. A web page [www.baldwinmanufacturing.com/zinc_AL_alloy.html] indicated this alloy would be a great choice for more reasons than I suspected. Another web page I mentioned before talked about the zinc-aluminum "stovetop metalcasting". [www.gizmology.net/stovetop.htm] On that site is the(?) formula for ZA-27. I presume those are weight %'s rather than volume %'s. He said:

I melted a handful of post-1982 (copper coated zinc) pennies. Once the zinc melted, I dissolved a bit of aluminum into it, to produce an alloy similar to the various ZA alloys. ZA-27 (71% Zinc, 27% Aluminum, and 2.2% Copper) is on par with cast iron for strength, but melts at a much lower temperature and is only two-thirds as dense.

Gray Cast Iron ZA-27 ZA-12 ZA-8
Compressive Yield Strength 65,000 psi 42,400 psi 33,400 psi 36,500 psi
Ultimate Tensile Strength 40,000 psi 60,900 psi 39,900 psi 54,200 psi
Density 0.258 lb/in³ 0.181 lb/in³ 0.218 lb/in³ 0.228 lb/in³
Melting Point 2795°F 709-903°F 711-810°F 707-759°F

I say the alloy I made is only similar to ZA-whatever because I didn't measure anything - this run was just to see if it was possible at all. To get a near perfect ZA-27 alloy would require $1.99 in post-1982 pennies, one pre-1982 penny, and 100 grams of aluminum. This assumes that the copper on the pennies won't oxidize, which is a stretch - to make sure there's enough copper, I'd add two pre-1982 pennies.

The pennies took about 15 minutes to melt from a cold start, and the aluminum took perhaps another 15 minutes to dissolve. Then I turned the burner off and let the blob cool.

   I hoped I could stick the propeller blades in a sand casting mold and simply pull them out, leaving the cavity behind to pour molten metal into. Then I'd add a cylinder on top as an attachment piece, and maybe a bit of a funnel to pour into above that.
   I learned more about "cope" and "drag" sections of a mold, slightly improving the odds it'd work and learning vaguely known procedures. I bought sand, fine sand and bentonite clay, the essentials for 'sand' casting. On the 21st I made a small wooden box to pack the mix into. On the 22nd I bought a 120V hotplate.
   Then at a recycling place I got some old corroded pieces of ship anode for a few pounds of zinc. The surface looked awful, but inside was bright, shiny zinc. The corroded surface should just become excess slag on top.
   I could also see wanting to melt pure aluminum for casting heatsinks, which could be easily molded. But to melt aluminum evidently required a lot of propane. I found a site with what look like good instructions for making an induction heater: http://inductionheatertutorial.com/
   Then I remembered my mini-kiln could melt it. I tried it out. It took an hour - 15¢ of electricity - and I poured a little ingot from a pottery crucible into a muffin tin. Gosh, liquid aluminum - it actually works!
   I tried to melt some in a pot on the hotplate, then on the kitchen stove, but the thick bottomed pot turned out to be laminated not solid. On the stove, the base layer detached and warped, holding the rest of the pot off the burner. Notwithstanding that the bottom of the pot glowed dully red, none of the aluminum in it melted.
   Later I watched a youtube video of someone doing his own first aluminum casting, and learned a few things to do and not to do, improving my odds for success. (My wooden mold box had to be remade.)

   When I went to cast the prop on December 3rd, at the last moment I decided it should be pure aluminum - wasn't zinc used as sacrificial anode? Why would I think it would last in salt water? I pulled the zinc out of the thin bottom stainless steel pot on the large stove burner before any melted, and put in only aluminum - drill shavings and some fat chunks. It did eventually melt the aluminum. But between the thin shavings of aluminum (I had thought they'd disappear into the zinc) and I kept taking the lid off the pot to check it, it ended up with more oxidized slag than aluminum. Some bright aluminum was melted and would run across the bottom of the pot under the slag if it was tipped. This demonstrated that It Can Be Done! Then I made the fatal mistake of stirring it with the tongs. This mixed the oxide into the liquid aluminum and everything turned into mush that wouldn't run or pour. (Oxides melt at much higher temperatures than pure metal.)
   I'll try melting aluminum on an electric stove again. Next time I'll pay attention to the following points:

1. Use a big stove burner rather than a small one.
2. Use a flat bottom stainless steel pot with a lid that fits well. The lid is all the more important as the aluminum is spread in a shallow layer over a broad area, and thus is quite subject to oxidation.
3. Don't open the lid any more than necessary. It lets in oxygen as well as letting heat escape. It'll probably take at least 1/2 or 3/4 of an hour to melt say 200 grams of aluminum, so frequent checks are pointless. (The 1500 watt mini-kiln takes an hour or more.)
4. Use big chunks of aluminum (eg, 1/4" or thicker pieces) rather than drill shavings or thin sheet metal. The thin pieces end up mostly as oxidized slag.
5. Don't put in borax flux. It's commonly used, but it seemed just to be trouble both on the stove and in the kiln.
6. Put the mold close by on the stovetop, perhaps on another burner. Lay out metal - perhaps cake pans or cookie sheets, anywhere the pot of melted aluminum passes over. Be very careful. Always wear safety glasses. A full face shield and leather gloves are better when you lift the pot. Watch out your pot and lid handles aren't getting burned or melted - especially, you wouldn't want to drop the pot. I'd much rather do all this outdoors, but that's not where the stove is. The hotplate I bought for that purpose doesn't get hot enough. I don't think it has the 1000 watts it says on it - seems more like 600 or 700. It was good for burning off some smoky oils (or ?) on the aluminum outdoors before bringing the pot into the kitchen, but that's all.

L: Removed top "cope" part of mold. (The little vent hole to the right
didn't flood with aluminum no matter how much I poured on.)
R: Breaking away the sand/clay mix

Prop freed of sand.

   For the mold I used 10 pounds of damp sand and 1 pound of bentonite. With a bit more water, it made quite a nice solid casting mix that held its shape. I could almost call it "sandy clay" in spite of the small proportion of clay. I found that the cope and drag didn't have to separate on a straight line - I could pack more sand in the bottom part in one area to fill under the curving top of the prop blade, and put a bit of a dip on the convex side. so configured, the prop blade pulled out freely enough, as hoped, from both the top and the bottom, and the mold was made.

   I poured in aluminum (melted in the kiln), to the extent of making a big puddle of it on top. None came up out of the tiny air vent where I expected it to. I had an experienced friend over, who arrived just in time for the pouring, who had a couple of misgivings about my mold and technique. The bottom, the outer end of the blade, filled nicely but seemed rather rough, and the tip (bottom) was quite rough. The upper zone didn't fill completely and had some hollows and gaps. It didn't appear to be appreciably smaller than the original part.

We discussed the probable reasons for this and what to do about them:

1. The roughness at the bottom was probably from pulling the prop out essentially sideways to the surface. This may have roughened the sand surface, and bits of sand probably fell into the bottom, explaining some bottom voids. And I may not have tamped down the sand sufficiently to press it well enough. (I did want to be able to pull out the blade!) Other than to change the orientation of the mold entirely, I could try oiling or greasing the surface of the prop so it (hopefully) slips out with less disturbance, and tamping harder.
2. This is guessing... With the tiny air vent, made by strategically poking in a wire, the aluminum quickly froze around the vent and prevented air from coming out. This caused the aluminum to build up air pressure and stop filling the upper area, causing the voids.
3. The aluminum may not have been hot enough, causing premature freezing. But I'm not convinced this was the problem since I poured a good puddle on top, and the aluminum in the crucible was still liquid as I set it down, and after a few moments, thinking the pottery crucible might crack from temperature strain (not to mention the wet grass, soon burning), picked it up again with the tongs and put it back into the kiln.
4. (I thought of this afterwards) These factors didn't seem to completely explain it. It wouldn't do to pour too slowly and have aluminum touching the wet clay freezing up faster than the pouring, but think I poured too fast, and aluminum slopped up into the air vent (and froze) before it should have reached that height. I was surprised how fast the mold was full. In fact, the cast blade weighed 76 grams and the puddle on top was 111 grams. The crucible full of melted chunks could easily have cast all three blades for a prop.
5. I should have built up more of a "funnel" in the sand at the top while forming the mold, so that the puddle of aluminum would have more height to exert more pressure in the top part of the mold. This is indeed the part that was full of gaps.
6. Once the prop (and a plastic funnel) were removed prior to casting, I should have avoided disturbing the sand, as grains would tumble into the mold. I might even turn the bottom half upside down to dump out any loose grains.

   These improvements I'll soon be trying and I'm optimistic I can make a good propeller(s). I've thought of casting the entire prop in one piece, but I'm not very confident about getting good results from such a complex shape, so I think I'll stick to casting separate blades and hub. As a bonus, the pitch will be adjustable.
   I'll make a hub with a 14mm hole and shear pin slot to fit my Honda outboard. When I get some motor in that, I can test times, amperages and speeds between the shallow pitch prop, the steeper pitch prop, and this new propeller, perhaps on a couple of different size boats. (Or maybe I can give it to someone else to try out while I make molds to make the "mini-electric hubcap" motor and do one to put in the Honda. Or make a whole "Mushroom Outboard". Hmm!)

Notwithstanding that it doesn't look usable, I roughly sanded and polished up my first ever metal casting.

CNC Farming Machine Project

   A first thought for the faming machine was to involve someone else, who has more experience than me in both CNC and farming. He seemed less enthused than I expected and had other ideas, citing the high cost of mechanical parts for such an undertaking. But we might still accomplish some things together.
   The second thought was that I should do a small unit and test things out on my own garden. And then, that an 18' wooden extension ladder section might make an admirable 'bridge' or gantry to go over the garden: Build a strong double "roof truss" shape out of more pieces of wood. Screw it all to the inside of the rails, leaving the rails and outside clear for the wheels of the carriage (made of 3D printed plastic?).

   This makes for a carriage that can roll from one side to the other, with wheels on top and underneath the track so it can't come off or tip. That should be sturdy enough to hold a rototiller and run it back and forth through the soil. I have two cordless lawnmowers someone gave me, and in fact three such 24 volt motors. These and optical position sensing devices might be used in place of stepper motors for motive power. Other details are more vague.

   For rails, I considered that one could screw angle iron to 4x4 wooden beams as rails, similar to my sawmill rails, which use angle iron and aluminum 3" x 3" tubes. (or even use my sawmill rails... hmm!) I bought 2 wheels that run on angle iron edges. (The other side gets plain wheels that run on the wood: only one side guides. Otherwise the slightest misalignments between rails would cause trouble.)
   This is another project for which I think flat drive belts (with barrel shaped pulleys) would be ideal. They would have to be made in custom lengths far longer than typical pre-made belts from a store. This is really a traditional situation where flat belts would be chosen.
   I'm trying to decide whether the gantry should operate at a fixed height with the tools moving up and down, or should itself be able to raise and lower. A low profile would be best for the stresses of plowing, rototilling or raking, but passing over tall corn, or a few other things, needs a lot of clearance... Okay, I think I just decided the gantry needs the "Z" axis. A stepper motor at each end driving an identical screw (like the Reprap 3D printer's Z axis) is probably a simpler arrangement than linking two raise/lower screws mechanically over such a long distance. (My sawmill does 4 screws with bicycle gears and chain.) No lawnmower motor for "Z"!

12 Volt Solid State Refrigerator Project

>It's Colder than I Thought!

   The Peltier element fridge cools better than it seemed. Belatedly comparing the convenient multimeter probe I've been using all along, I find its readings are too high. In a typical location it said 7º, but two regular thermometers in the same place said 4º and 4-1/2º. Near the top, it said 9 and the thermometers both said 6. Duh!
   At the floor near the ice tray, it apparently hovered around 1 or 2. This seemed to be more in line with expectations considering the nearby ice tray and the 3" foam insulation. Then I got some more thermometers and started dipping them in the tray of ice and water to calibrate them. It turned out the truth was somewhere in the middle, about 1.5 or 2 degrees colder than the meter's reading. But it's the difference between being nervous that things might spoil and reasonable confidence.
   It just goes to show you shouldn't believe everything you read without verifying it, even if it didn't come from mainstream media. The floor near the ice tray is typically 2.5 or 3º to 4º.


   I'm finding the fridge needs to run about 11 or 12 hours a day. Less and the ice completely melts and temperatures start to rise. Otherwise, since the cooling is accomplished by an ice tray that's always about 0ºc, the temperature is pretty constant, varying only by a degree or so.
   I still need to make the smart solar control and the drain funnel for under the ice tray. I now have what's needed and just need time to get to these details.


   With temperatures inside now seeming to me to be just cold enough, I started using the fridge for some day to day things. I put things most likely to spoil near the ice tray end where it's coldest. Lettuce at the other end. It's unexpectedly nice to use. The light 3" foam lid opens with a touch, and you can lean it open against the wall if desired. After you lower it part way you just drop it. It closes itself with a 'poof' of air cushion just before it hits. Everything is very accessible just as I'd envisioned. Since you're not reaching behind things to get other things, the time the lid is open is usually very brief, with little heat gain. And the cold air doesn't dive out the bottom when it is open like with a front door fridge.
   I think it's just as well I went for the simple single hinged lid and didn't bother with multiple sections or compartment drawers. They'd really have been superfluous as well as a nuisance.

Commercial Aerogel - Home Made Aerogel?

   Aerogel at R10 is the best thermal insulation besides a vacuum. But it hasn't been available. Now it comes in flexible matt rolls and in small strips for facing 2x4s in house walls, to reduce heat loss through the wood itself. The structure of these products is to have some sort of solid or flexible web for strength (think perhaps of green nylon pot scouring pads), and then fill in the space with highly insulating but flimsy aerogel. The strips are costly; I didn't enquire about the rolls and have apparently lost both URLs.
   Then I got yet another link, to how to make aerogel: http://www.aerogel.org/?p=990 .

   I don't think I'm about to follow up on aerogel insulation myself at this point, but it sounds exciting. An R20 fridge with 2" walls would be more compact yet better insulated than R15 with 3" polystyrene foam. Or larger inside with the same exterior dimensions, eg 5.7 cubic feet instead of 4.125. A 4 or 5 amp peltier element could replace my ~7.2 amp one or it could run fewer hours during the day, reducing electrical consumption.

Large NiMH Batteries - Take 3

   The first basic type of large NiMH battery, of necessity built up from small dry cells owing to suppression of the large sizes by corrupt interests, was soldered together cells placed in plastic food containers. In cars, the solder joins tended to eventually vibrate apart. The second design was to put them end to end in PVC 1.25" (for D cells) irrigation system pipes. These work very well, but are needlessly bulky and the sizes are often inconvenient. The third major variation is to 3D print custom ABS plastic cases to fit 10 D cells (12V) in minimal space. This of course is only possible now that I have a 3D printer.

   I decided to design the NiMH 12V D cells battery case while I waited for another hour long printout. Hah! With all the various intricate little desired features, it took the whole day. Then it took the better part of an hour for Pronterface to make the g-code file. Then it took 5 hours to print just one.
   In fact, the base layer alone took almost an hour. Printing directly on the glass is done very slowly in order that traces don't curl up as they print. Since for this object the base layers covering the entire area only make for a solid side wall and aren't needed structurally, it would seem reasonable to use a solid piece of plastic and glue the complex part of the frame to it, to cut an hour off the printing time. Furthermore, the corners warped up as it printed. If there was no bottom, might I hope for less of that?
   But it looked nice as one solid part. The top will be flat plastic. It seemed obvious that it should be simply cut from a flat piece. But to get the holes for the screws lined up by hand looked tricky. It would be slow, but easier, to print even the flat tops! ...or maybe just make one flat top to use as a drill template.

   I misjudged or mismeasured the first one - it was 3mm too short to hold two cells in series. The other dimensions and sizes were perfect. Nevertheless, I found and scribbled out a half page of small changes and fixes. When I printed the second one, I tried stuffing some pieces of cardboard around the partly printed object, hoping if it was kept warmer with insulation it would warp less. A hot wire touched another when I touched a wrong place with the cardboard. The printer instantly stopped and started smoking. I pulled the plug. Enough of the holder was printed to try it out and even to use it, as seen here. It seemed only the thermistor sensing the extruder temperature was fried, and a wire going to it, but the printer was out of action until another could be found.
   Then I read the the way to reduce or prevent warping is to cool the bed temperature. But I've found that the white ABS plastic, specifically, has a tendency to break loose from the glass except at the highest bed temperatures. Next time I'll use another color for anything large. Or PLA plastic. I ordered 6Kg of PLA plastic.

   I'll try such battery holder design variations as mentioned when the printer is running again.

Turquoise Battery Project

An Iron Electrode - Ni-Fe

   I decided to try and make an electrode this month after having printed several porous plastic electrode pockets on the new 3D printer last month. I bonded a porous plastic electrode plate cover to a solid ABS back piece to form a plate with 4 porous ABS plastic electrode pockets, about 57 x 15 x 2.4 mm, porous only on one side. I cut out a flat copper foil current collector and fitted it in.
   Nickel-iron batteries keep coming up in discussions, and I decided to make an iron electrode. I thought of SAFT's making of highly conductive sintered copper-iron 'negodes' in the 1950s, as mentioned in Alkaline Storage Batteries. I didn't have any fine copper powder, but I had monel powder: nickel and copper alloy. Both elements are compatible with an iron electrode, having lower reaction voltages.
   Then again, I thought, why not make a nickel negode with the monel mix that I started with in 2008? To 30 grams of the monel mix I added 15g of graphite powder to make it more conductive (15-20 ohm readings), and I stuffed it into the electrode, ramming it in with a little piece of plastic to compact it. When it was full I glued a printed plastic cover over the top. There was a bit of a crack over each pocket, but I didn't think much about it.
   The plan was simply to use it with a couple of Changhong nickel hydroxide 'posode' plates in their NiFe cell case. It should be 1.21 volts open circuit. (= +.49 - -.72) I put it all together and filled it. It started out at .6 volts, but charged steadily. However, it didn't pause long around 1.2 volts but continued up to 1.5 and started bubbling audibly. Performance was poor and short lived. When I disassembled the cell, I found that much of the substance of the electrode had been percolated out of the little cracks at the top by the bubbles. And a little powder had apparently also made it through the perforations in the face of the pocket. Maybe I should have used a little coarse iron oxide powder and made an iron electrode, after all.
   Being stubborn, I tried again with the monel mix. This time, I used (nickel plated) expanded copper mesh for the current collector and acryllic plastic for the back of the pocket plate. The powder inside could now be seen through the back. I filled the pockets with powder, ramming it in as best I could, and it looked good through the plexi.

   This time, I got out some silicone RTV cement and caulked the top shut. Now the only way out for the powder was through the perforations or to burst something. In assembling the cell, I put a piece of Arches 90# watercolor paper in front of the porous side - between it and the metal positive plate facing it.
   To make a long story short, it was another failure. And on disassembly, I found that the silicone had turned to gooey mush in the potassium hydroxide solution.

   Only later did I remember that metallic nickel is the one metal that simply won't oxidize in pH 14 alkaline solution, so the monel was never going to work at that pH. Duh!

Installing with paper and plastic spacer,
with the Changhong nickel 'posodes'.

Third time lucky?

   This time I went with 25g of the grainy or 'sandy' Fe3O4 'black iron oxide' powder [pottery supply], 10g of pure monel powder (not the mix) [micronmetals.com] - this time just to improve the conductivity of the iron - and .35g of Sb2S3 [american pyro supply]. Adding a bit of water I mixed and pressed these together in a press, dried the flat (but not very coherent if touched) 'cake', then I torched it with a small propane torch for about 10 seconds to sinter the powders together. (On the next one I torched it too long, and there was a large patch of reddishness that would have been iron oxidized to Fe2O3 - rust. This won't charge because it's an electrical insulator.) Then I scraped the material (loose powder again with a few crumbly clumps) into a small plastic container. Resistance measurements were in the megohms. That didn't seem very encouraging, but it wasn't open circuit and I decided to try it anyway. (In fact, it worked better than the original Fe 'trodes.) I wondered if the hydrogen bubbling in the last two was caused by the graphite and didn't want to add any this time.
   I mixed in a few drops of sunlight dishsoap and water and mixed again, then used it to refill the first electrode (which had been removed and easily emptied by running water through it). It took virtually all the material with hardly a gram left over.
   This time, I sealed the top with sticky casting wax. Later inspection showed this to be intact. Again I used the art paper to mask the pocket perforations.
   I calculated that 25g of Fe2O3 should have about 17 amp-hours. It turned out to be more like .2 amp-hours - 1% utilization. This would be owing to the coarse particle size having far less surface area for the volume than a nano-powder.
   After an hour to soak I started charging slowly at 40mA. Since .04 * 24 = .96, that meant that the number of days it took to charge was virtually the number of amp-hours it had. It didn't take a day.
   Yet it was soon evident that this electrode was working better than the previous attempts, as the voltage rise was gradual. At first, it only wanted to put 1/2V, 1/2A into a 1 ohm load. After a couple of hours, that was up to 3/4 and it dropped much more slowly with the discharge. Charging voltage rose from about 1.30V to 1.33V in that time, and the rise or drop as the charge was put on or removed decreased from 50mV to 25mV: the resistance was dropping as the monel, somewhat oxidized by the torching, charged back to metallic form.
   But when the voltage hit 1.5 before bedtime (after 5 hours), I got nervous and took it off charge for the night. In the morning it was 1.25 volts. I tried a load test. It put out over a volt into a 10 ohm load (>100mA) for nearly 1/2 an hour and looked like it would continue for hours at very gradually decreasing voltage just under a volt. But I wanted to charge it further, not discharge it. Probably 1.5 volts was a fine charging voltage since the theoretical Ni-Fe is 1.43. The antimony sulfide should prevent hydrogen gassing below maybe 1.7 volts. So I put it back on charge, this time at 90mA until it got back up to 1.5V.
   Towards the end of the day after 16 hours of charging, the charge voltage was up to 1.6, and it would deliver over .9 amps for a bit into 1 ohm, and almost 3 amps if momentarily short circuited. With the effective electrode surface/interface area being about 5.2 x 5.5 cm or 28.6 sq.cm, that was 31 mA/sq.cm current density for the 1 ohm load or around 100 for the short. Scaled for 1/6 the interface surface area, the short circuit figure is better and the resistive load similar to an original Changhong Ni-Fe cell. In evauating this, we must take into acoount that the electrodes are about 2mm thick instead of 1, with a current collector of smooth, solid copper foil at the back rather than a mesh or with holes or coarseness to increase its surface area contact with the electrode substance, and that it has quite low amp-hours capacity. Thus its current performance seems considerably better than the original iron electrodes.
   I left it on charge overnight and in the morning I ran a load test. For 1.5 hours it had a 10 ohm load. It ran most of that time from about 1.1 volts down to .9 volts. I decreased the load to 20 ohms and ran it another hour, after which it was again under .9 volts. Doubtless it would have run somewhat longer at still lower voltages or lighter loads. It was working, but it didn't seem notably good or practical. But late in the evening, after hours of recharging, I tried again (10 Ω) and it ran over 1/2 hour above 1.3 volts/130mA. This was into real, practical battery territory, and much better than the first discharge where it sank under 1.3 volts before 2 minutes were up. It probably probably shows why flooded batteries get cycled a few times when they're first made.

   I consider it's probably worth trying a straight monel nickel negode again in a lower pH electrolyte just to prove it works. Of course, the voltage will only be about -.65 volts at pH 12. If the one I still have still doesn't work, I may try one with the pure stuff and no graphite, and torched to oxidize some and to fuse particles into larger agglomerations. (With maybe a little nanocrystalline nickel oxide or 'fluffy' nickel hydroxide thrown in to up the surface area per volume.) Nickel-iron batteries are known for their long cycle life, yet Fe3O4 gradually gets converted to Fe2O3 and the capacity gradually decreases once there's less iron capacity than NiOOH capacity. Nickel may be .2 volts lower than iron, but it's more conductive - especially as monel with its copper - so it loses less voltage under load. So the difference may be less in practice than in theory. Plus the reactions are reversable even if overdischarged, so it has indefinite life span without loss of capacity. Still further, it's more dense than the Fe3O4 and so a lot of monel fits into a small electrode, making lots of amp-hours in a small space.

   All this wasn't getting my Mn-Mn chemie going, but it was a simple way to assemble and test the plastic electrode construction on single electrodes, while trying out the copper and stibnite improvements for Ni-Fe that nobody else was doing, just to prove the point.
   I learned the need to seal the entire pocket plate very carefully and figured out to seal the top with wax but not with RTV silicone. In fact, I started to think about new electrode and cell construction ideas and techniques.
   I also used it as an opportunity to shoot some video about battery making. Doing a video is something of a project in iteslf, but it should really bring it to life for a lot of people who want to make batteries but have no idea where to start or what to do - who seem to be about where I was when I started in 2008. Unfortunately when I "copied" the files from the camera to the PC, all it did was make an alias/link to the camera files. But it cleverly took a long time, longer for longer videos, to trick me into thinking the files were being saved. Of course I didn't discover this until I went to make the video. When I erased them off the camera to free up space, they were gone, and I had to make another electrode and take videos all over again. I had also planned to grab a few stills from the videos for this newsletter, so there are few pictures here. But they're representative.

   I took another look at the Fe Pourbaix diagram and noticed that contrary to my previous belief, it in fact looks compatible with a lower pH electrolyte. At pH 12 it's about -.82 volts. Since nickel oxyhydroxide goes up to about +.65 at that pH, cell voltage for Ni-Fe/pH 12 should be a little bit higher than in KOH/pH 14.

Next Steps?

   The next 'negode' with Fe3O4 nano powders wasn't successful. I tried painting the face with egg albumin, which I then boiled, but the KOH electrolyte dissolved it, allowing powder to leak out. Furthermore, the dissolved albumin caused rapid self discharge even after briefly rinsing the electrodes, putting paper in front of the iron one to catch leaking powder, and replacing the electrolyte.
   Either I need to get a 3D printer that does finer traces and smaller pores, or stick with grainy powders, or find some other way to make electrodes. Camosun's printer makes considerably finer pores, but I certainly can't afford their prices. I've decided to stick with porous plastic pocket electrodes for now since so many other constructions I've tried haven't worked, and with the grainy Fe powder for now, but to crush it further with a hammer. It still wouldn't be nano and shouldn't leak out much. Presently the pottery supply has only nano "black iron oxide", but "granular magnetite" is close enough the same thing. I hope.

   With the one working Fe 'negode' to test with, next is a 'posode' with a graphite current collector to work in the lower pH electrolyte. Once both sides are working properly, I can work on getting best amp-hours and performance, and then try out manganese electrodes both '+' and '-'.

Posode Current Collectors: "Flexible Graphite Foil"!

   I saw a youtube video on "Aquion" batteries and looked up the site. To my surprise they had a paper to download that described how their batteries worked. Their chemistry (manganese, salt) and philosophy has quite a bit in common with mine, tho their route was somewhat different. For me the chemistry was mostly of passing interest, but a term used perked my interest: Graphite Foil current collectors. It turned out to be something the "Graphite Store" website, the place I looked a couple of years ago for a carbon based current collector solution, simply didn't have. But it's the stuff people evidently have been using in various batteries (including the very first design I ever looked at), that I could never find. As is often the case, I just didn't know the name to look it up or ask for it.
   And someone may well sell it for some unrelated purpose (eg as "gasket material") locally, but again I don't know where to look or who to ask. But I went on line and found it seems to be quite common. I ordered a roll of .020" x 5" x 108' of "flexible graphite" 'gasket material' from www.minsealorder.com - about 50$ list for 324 electrode plates worth.
   Unless it has some problem (like the "expanded graphite" gradually swelling up and decreasing in surface conductivity), it'll be much better. Since this is what others seem to be using it is likely to work fine, in which case all my work with carbon rods from dry cells, carbon fiber, 'grafpoxy' and other graphitic materials may have been a dead end, and added a year or two to the Turquoise Battery project as a whole.

"Practical" Battery Production? (meaning: ability to make enough for an EV or solar installation in a reasonable time, instead of hours to make one cell)

Finding my 3D printer's resolution is rather coarse for what I bought it for - porous battery electrode pockets - I'm starting to think again of other means of electrode and battery production. Squashing an electrode flat from a face takes many tons of pressure. If the electrode is very big at all, a 12 ton hydraulic press is inadequate.
   Earlier efforts to compact the material from one end resulted in mashing the mesh current collectors. Yet it seemed to work reasonably well in a recent pocket electrode with copper mesh. Plus, a sheet copper current collector seems to work well enough (so other sheets should too, including 'flexible graphite' sheets). (Potentially, small holes could also be punched in such sheets to increase contact area.)
   So then, a tall, narrow, well polished steel box as an 'electrode die', with a funnel attached to hold the material, and a matching thin (~3mm) punch could use a much smaller press such as the book press to compact a solid electrode, in several plunges with more material dropping in each time. The (bottomless) die would then be raised and the electrode extruded down out of it. Any desired aspects of this could be automated. The briquette can then be dried and torched whole, instead of having to torch it and then crumble it and feed it into a plastic electrode pocket.
   Next, a piece of plastic would be 3D printed that is exactly sized to the width of the cell cavity, perhaps a slightly widened section with slots in the cell walls to accommodate it. Tho thin, this is porous and it has channels to allow bubbles to rise to the surface. And it has a lip around the edge. When a piece of paper is pushed against it, the lip ensures there are no gaps around the edges. (With a finer resolution of printer to make finer pores, perhaps the paper might be dispensed with.)
   The cell is constructed: 1, 2. metal current collector, pressed with negode, 3. a sheet of watercolor paper, 4. the plastic divider, 5. a sheet of watercolor paper, and 6, 7. the posode pressed with the graphite current collector. Since the thickness of each piece can be carefully controlled, this whole assembly is made to fit exactly into the cell cavity, and it is put together and pushed in from the top. Wax is dripped in to cover the tops of the electrodes but not the center vents.
   It's a liquid filled battery, but there's very little liquid in it except whatever reservoir is added above the tops of the electrodes.

Victoria BC