Turquoise Energy News #132
covering May 2019 (Posted June 3rd 2019)
Lawnhill BC Canada
by Craig Carmichael

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

Month In Brief (Project Summaries etc.)
 - Lower Power Hot Water Tanks - Under Kitchen Sink Solar Hot Water Tank - Air Injection Iron Filter - Pipe Battery Assembly Improvement - Biggest Mistake in Solar: Not Going Solar? - Batteries

In Passing (Miscellaneous topics, editorial comments & opinionated rants)
 - More on Time Zones - The Weather and The Corn - American Armyworm Invades China - The Sudden Drop in the US Crime Rate: Many Misattributions and the Real Reason - ESD [Eccentric Silliness Dept.]

- Project Reports -
Electric Transport - Electric Hubcap Motor Systems
* Ground Effect Vehicle: Multiple Ducted Fans? - Selecting Ducted Fans for the RC model
* Seen on youtube: a very large Fully Electric Ferry and a Solar Electric Passenger Train.

Other "Green" Electric Equipment Projects
* "Off Grid" (etc) 36 volt DC Infrastructure - HAT clay socket - "DIY" Blades for High Power HAT36V-50A Plugs? - Kitchen Solar Hot Water Tank: Plumbing & Wiring (Yay, instant hot water!) - Solar Hot Water Panel(s) Versus Solar-Electric Water Heating - Badderies - Fried Charge Controller- Oops - Charge Controller Vagrancies

Electricity Generation
* [In?]Efficiency of 1000 Watt Y-Solar Microinverters - More solar trivia - Lots of cloudy days make for low power generation - Miles truck charge controller goes berserk, doubles house power usage for 10(?) days (see May 25th) - Extra Solar Panel on Lawn ups production
* Beach-Launched Tide Power: the Beach
* Off Grid Solar Installation for a neighbor

Electricity Storage - Turquoise Battery Project (Mn-Zn, Ni-Zn or Pb-Zn in Methyl Hydroxide electrolyte)
* Electrode "Pocket" with Nafion Ion-Selective Membrane Face
* "Mounting" NiMH D cell Tube Batteries

May in Brief

Me on the beach                            
   I was again frustrated in May at how little work I was able to do on any "cutting edge" green energy projects. I seem to be more and more occupied with the more mundane things - and with getting the garden in. On the first of June there were still a couple of unplanted areas, and I know I won't have enough peas to freeze some.

   Insofar as I went, I continued my push to create infrastructure for 36 volt "off-grid" and solar power systems. I figured out how to 3D print and manually smooth and shape a workable (if crude) plastic mold to shape clay for ceramic 50 amp HAT36 sockets, did it, and made the two halves for a socket. But the raw clay has to sit until I fix my long-broken mini-kiln - one more little project I didn't get to. (Hmm... I should make more than one shell before I do the firing? ...or wait and see if any changes are needed?)

3D-printed mold and red clay ceramic socket pieces formed in it.

   I spent part of the month including the last few days doing solar work for other people - the ones for the tug boats and an installation for a lady across the highway in her 70s who has lived without electricity most of her life. Even this little system with two 90 watt solar panels and a 12 volt battery for LED lights and to charge her cell phone, and to occasionally run a vacuum cleaner with an inverter, should be a big plus for her.

Lower Power Hot Water Tanks

  Hot water tanks are essentially mysterious creatures. We know they're there. We know they're using energy - more than anything else in the home except electric heat and perhaps electric car charging. But there's usually no way to see or measure it. They're hidden away in some closet or basement, and even when seen still betray nothing of their activities - whether they're doing nothing or drawing 3000 watts of electricity. How much electricity is it using? The wiring is carefully set up so even if you want to, you can't get an ampmeter on them without disassembling something. It's about the one thing that still gets directly wired in, without a plug and socket. How much electricity is being wasted when we have to run the hot water for a while each time before we get a cup of it to wash our hands? The simple fact that they take so much power when they're on, and that they're hard wired in with 240 volts (ooh!), makes us apprehensive. It says "Don't Touch!"

   Yet when I changed mine from 240 volts, 3000 watts to 120 volts, 750 watts by simply moving one wire in the house breaker box, I found no difference. Hot water is always available when I want it. But there is one difference: My 10 solar panels put out over 1500 watts in sunshine - a few hundred even in clouds. But it would take 20 or 30 panels to supply 3000 watts. So now if I use hot water when it's sunny, the grid electric meter doesn't move, instead of jumping up rapidly by the majority of what was needed to reheat the tank. So that's a keeper.

   With the kitchen sink now having almost instant hot water from its own solar tank, I prefer to go there even to wash my hands, and the big tank is now used almost entirely for showers, laundry and baths. (Why do I always want my baths at night?) 3000 watts is probably needed for a family of four or five. Likely not for a single person or a couple - which is the occupancy of the majority of homes these days.

Under Kitchen Sink Solar Hot Water Tank

   I've always said "No level of scalding hot temperature can atone for the hot water taking so long to get here." In this house, to get hot water at the kitchen sink took a minute or more, and 4 liters of hot water wasted in the pipe. The answer I have long wanted to try, a tank right by the sink, I finally did.
   I must have had the 15 liter hot water tank for the under-sink hot water for the kitchen for well over a year now, but with the new "chemical-free iron filter" for my house well water system, I wouldn't need to connect it to rain water to get un-stained dishes. So by putting it off so long I saved myself a lot of work that now wasn't needed.
The Dernord 36 volt water heating element in the 15 liter                    
under-sink hot water tank. I found 250 watts was plenty.                    
But the soft silicone(?) gasket supplied (white, under hex top) squished out            
under pressure, causing a flood in the kitchen. Get an "O-ring" or a stiffer gasket.        
   The 36 volt water heater element having arrived, until the 8th I made up pipe assemblies for it. On the 11th after a couple more trips into town for parts, I did the plumbing and installed the 36 volt heater element in the tank and the tank under the sink. Then I ran a cable across the house through the crawl space (ugh!) to connect up the 36 volts. On the 12th I hooked it up and (not without some troubles and a flood first) got the water hot. It seemed to heat surprisingly quickly even with just 265 watts of heating applied. (I wired two ~500 watt elements in series with the third disconnected.) It was hot for dishes in under an hour. I also found the small 15 liter tank seemed to cool surprisingly quickly. There was only warm water in the evening and quite cool late at night. But soon I found that it actually seems to spend most of its time shut off, so I started leaving it turned "on" at night to run on the batteries.
   I left the top of the tank with two wires sticking out that could be connected to the thermostat giving either ~260W (drawing 6.5 amps) or ~500W (almost 13 amps). The water heats so fast with 260 I don't expect to use the 500 at all.

   I still want to install a 50 amp socket and plug rather than a 15 amp, because I want to be able to unplug the heater and plug in a 36 to 120 volts inverter to run the fridge in case of a power failure. I connected the tank with "wire nuts" until I've made the socket. I wouldn't have dared put the connection down there where it might get wet (and already has) if the power was 240 or even 120 volts... or maybe even 60. 40 or less is different.

   (There are "on demand" sink water heaters that don't need to keep a tank of water hot, but they need sudden very high power that is sort of counter to the way solar wants to work. Someone on Youtube also said that they aren't very reliable and he's on his third one at his sink. But I confess they weren't even on my radar screen and I had never even thought about the idea until I saw that video, and I don't know what is available. I'm very happy with the new tank. It's great!)

   From the 13th it was a real pleasure to have almost instant HOT water (~2 seconds with the tap full on) instead of over a minute to get tepid water from the distant main tank.
   In several short sessions (my preferred way but previously impractical) I did dishes that had been piling up for 3 weeks and cleaned up the counters. After that I could do dishes right after I used them and kept the place tidy[er]. I'd always wanted to do it that way, but especially here I'd also always been loathe to run a gallon of hot water from a distant tank to get a cupful or two. Now it's just a cupful extra in the pipes. (Anywhere I can shorten the pipe or tube? I guess not unless I move the sink drain pipes. NO!)

Air Injection Iron Filter

   "The 24th of May is the Queen's birthday. [British queen Victoria 18??-1902(?)] If you don't give us a holiday we'll all run away!" (How is it the holiday Monday falls on the 17th to 23rd then?) It was on the 21st after the long weekend that I finally installed the new filter. I wanted to start in the morning on a work day so that if it didn't go well I could drive into town for parts. (The hardware store had been closed Saturday! Why would I trust them for Monday?) I had all the parts I needed, but it took most of the daytime anyway before it was finally in and working.
   On later observation it seemed that the water was just slightly orange instead of clear. But it didn't get any worse. With just the sand filter, the water started clear and then turned more and more orange and brown. And now it didn't seem to leave a stain behind it. Previously just one bath left a very visible ring in the bathtub. It was so thick and hard to scrub I had given up and just sometimes did one section of the tub, which was all brown stained again by the time I had worked my way around to the same place again. Or I would clean a few wall tiles. I had long since taken to using a squeegee after each shower to keep them from darkening so fast - not something I had ever bothered with before.

Pipe Battery Assembly Improvement

   Pipe Battery Racks for the solar, and Air Injection Iron Filter for the well water
   Putting NiMH "D" cells into long pipes of ten for 12 volts had made it practical to use them for big batteries by amalgamating 10 cells into a single physical unit. But the pipes themselves had always formed unwieldy groups, wired together but not forming a physically solid piece. The Honda hybrid batteries stacked onto little "racks". That gave me the idea. On the 16th and 17th I printed "racks" for the "D" cell pipe batteries on the 3D printer. One could make a rack of 3, 6, 9 or 12 pipes stacked 1, 2, 3 or 4 rows deep. I decided to make three sets of 6 pipe batteries to add capacity to the solar storage. (I even had to make 2 more pipes to do that. But I still had extra cells (the better part of 100 bought in... early 2016?) which might as well be put to use. And even extra pipes and ends all cut.

   Then, using the Honda batteries - for anything - had been awkward because of them being welded into tubes of 6 cells, and because there were racks only for 3 rows. I had been given the battery from one Honda Insight, but then I bought some more sets cheap on eBay, so there simply weren't enough racks. The one way to get a "standard" voltage out of them would be to pile them 5 rows deep to get 30 cells, 36 volts. So on the 18th I changed the dimensions of my previous racks and printed 12 more pieces to do two top rows on the Honda batteries. But I didn't get anything more done with them.

   The 3D printer - and I'm still using my old RepRapPro, having not yet successfully downloaded and installed the "Cura" slicer software for Ubuntu - is a wonder. Each of these 30 total pieces took nearly an hour to print, and the printer chugged away hour after hour for days, the print head and the bed whipping back and forth at rapid speeds, without anything coming loose or failing.

Biggest Mistake in Solar?

   A solar company put a video on youtube, 8 Costly Solar Mistakes to Avoid When You Design Your Solar Panel Kit. One of them was a little surprising: Mistake #4: Deciding Not To Go Solar. Here it was pointed out that solar soon pays for itself and is then a continued gain for many years, especially as electricity prices keep rising. We may not have an ideal climate in the "Pacific Northwest", usually with plenty of cloud cover in winter, but if our electricity wasn't being subsidized here on Haida Gwaii (same rates across the whole province), seeing the price of a couple of solar panels on each month's bill would surely cause solar panels to be covering rooftops at a rapid pace.

Solar Power for a Neighbor

   As mentioned the end days of the month were occupied with putting in solar electricity for an older neighbor who has lived without electricity for much of her adult life. Just a 12 volt system: two 90 watt panels on her roof, a "frame 27" size 'deep cycle' battery just outside, and a PWM charge controller on an inner wall.

Two 90 watt solar panels, facing south and ESE.
(The west side is shaded by a nearby tree in the afternoon.)

 Inside near the battery went a 1500 watt inverter mainly for vacuuming.
The 100 amp circuit breaker behind is for emergency shutoff.
The battery is just visible outside the window.

                          Plywood panel with charge controller,
                   solar & battery breakers, CAT12V duplex socket

   The duplex "CAT12V" receptacle will allow plugging in LED lights. (I loaned her a lamp with a multi-voltage DC light bulb for now.)

   Something was wrong and I was still troubleshooting into June. It turned out that the charge controller was "positive ground" rather than the virtually universal standard negative ground. Solar panel "minus" couldn't be grounded with all the other minuses. (The last item I've even heard of having a positive ground - much less seen myself - was some British cars of the early 1960s.)

   If the instructions had simply mentioned this important and unexpected fact, it would have save me a lot of trouble. And the battery wouldn't run the vacuum cleaner after being so badly overcharged.

Battery Development

   I made a battery pocket with nafion ion membrane covering one face to test out the idea of preventing metal ions - especially zinc - from traveling between electrodes. I haven't had a chance to test it yet. Again I am frustrated with how little time I've been able to devote to such projects with so much potential.

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

More on Time Zones

   It has evidently been noticed that people living in the west end of time zones are more often more deprived of sleep than those living in the east ends. It seems that's because the sun goes down up to an hour later. It also comes up up to an hour later, but the clock times scheduled for starting work are the same. So those in the west ends of the zones tend to go to bed at the same actual time of day by the sun, but they have to get up, by actual sun time, up to an hour earlier to go to work than those at the east ends.

   Then there's where I'm living now, Haida Gwaii, where the clock is almost a whole hour ahead of the sun... an hour and a half ahead of those at the eastern end of the Pacific time zone. And then it gets moved another hour ahead for "DST" most of the year. What insanity! Only real early birds need apply for 8 AM start jobs around here, because it's really 6 AM! Even employees in stores opening at 10 AM are actually getting up quite early to make it to work on time.
   (Even regardless of time zone position, I call DST "afternoon losing time" because that's usually my most productive outdoor working time, and when there's anything scheduled in the evening it starts an hour earlier and I have to cut the afternoon an hour short.)
   (I might also mention western France as being in a similar situation to Haida Gwaii. Apparently Germany won the war, because in spite of being a time zone and more to the west, all of France has been on central European time since 1940. But at least France doesn't go on "daylight savings" time as well.)

   My tentative conclusion is that Newfoundland has it right... the world should be divided into 48 time zones 1/2 an hour apart to limit the disparity of what "the time of day" means across the zone. And surely to God we can do away with "phony time" DST, which by the only available statistical evidence costs extra energy not withstanding that "energy savings" is often cited to justify adopting it.

The Weather & The Corn

   By early May the drenched US "midwest" was raining again. The Mississippi was flooding 8 feet higher than ever before for longer than ever before (41 days as of May 6th), with projections that there could be no crop planting in many areas even in June. Near the end of May, just 40% of US corn land had been planted. And evidently, planting so late risks getting frost before the crop is in. (And who knows these days - maybe snow?) And evidently some of it was "wet planted" - seed sown in the soggy ground or water with no hope of a crop, before June in order that "crop insurance" will pay out in the fall.
   For the latest gruesome stats on crop failures around the world see "Ice Age Farmer" and "Adapt2030". For graphic footage of all the horrible weather and other calamities from around the world from South America to Soggy Arabia to places I've never even heard of, see "Nared King" and "World of Signs". All on Youtube.

   I planted my own garden corn early, but it didn't come up. Hedging my bets, I put a bunch of seeds in a tray of potting soil inside. These sprouted. I transplanted them outside, replanted the rest of the rows outside, and planted another tray inside. But the tray was a half-hearted effort - there were nothing like enough plants to fill my little corn field. Again nothing came up outside. No frost, but probably just too cold for corn seed germination. Again the tray of seedlings inside sprouted and I relocated them outside, where the first ones were at least growing a little. I did a third tray which only sprouted gradually here and there (I think I used some older seed), and by the time some of those came up it was almost the end of May, and there was still a whole row (of four) with no corn. While things are warming up with spring proceeding, there have only been a very few nice, warm sunny days. (The solar power records under "electricity generation" below show how few sunny days.)
   There probably won't be any frost to summarily end the growing season here. Much less snow. It will probably just peter out with the days getting too cold and dull for plants to grow further. Then tomatos and things will start to go mouldy.
   Next year I will know to start enough corn indoors under LED lights for whatever cornfield I decide to grow.

   I had the theory that the deer wouldn't eat corn. The plants were too tough, and they wouldn't know to eat the cobs. If there were enough, the raccoons wouldn't get too many. So I planted them in the field outside the garden fence. But on the 29th I found that a deer had "sampled" the first four small corn plants in one row. Fooey! Do I need to fence off EVERYTHING? I bought a roll of fencing on the 31st. But my mom says to mix up an egg, vinegar, hot peppers and water and spray it on the plants. The deer evidently do have some limits to what they'll eat. (No more had been eaten in spite of doing nothing by June 3rd.)

Gratuitous picture of the expanded greenhouse.
Hopefully it will become much greener as summer progresses.

Typical of waterways along the east coast of Graham Island, "Jungle Creek" flows north
parallel to the coast for a long ways before it finally turns and heads down to the sea.
(It even ran another kilometer from here behind me before the highways department
put in big rocks and "rip-wrap" and diverted it down onto the beach at this point.)
The water and the flow is very low, especially considering that it's spring.

American Armyworm Invades China

   The American Armyworm some years back jumped continents and hit Africa, with none of its natural enemies accompanying it. It has been decimating food production ever since. It has spread across Asia and has now reached China, where it is wreaking havoc. It is so severe it has been said it threatens economic and social stability. (Adding to the African Swine Fever problem and the crop-wrecking weather in general.)
   When migrant Winter Moths hit Victoria BC in the 1970s they defoliated all the fruit trees and some other broad leaf trees every spring. Healthy trees grew a second set of leaves later, but some died and there was no tree fruit anywhere, for some years. People tried various things to kill them or keep them out of their trees, to little avail. In about 1980 biologists had located and imported the moth's enemies, a parasitic mini-wasp that injected eggs into the leaf-curling caterpillars, and a fly that laid eggs on the leaves for them to eat, which hatched out inside. The next year - in just one year - this horrible pestilence was gone and while the caterpillars and moths could be found occasionally ever after there was never any more serious trouble from them. (Next in the later 1980s came infestations of tent caterpillars, which I was familiar with as they had infested the poplar trees in the Edmonton, Alberta region in the 1960s when I was growing up there.)
   The article I read made no mention of trying to find and bring in the armyworm's enemies, only the despair and the woes they were causing. As if there was no solution. This seems very puzzling. Is no one even trying to solve this biological catastrophe?

The Sudden Drop in the US Crime Rate:
Many Misattributions and the Real Reason

from Freakonomics by Levitt and Dubner

   In the late 1980s there was a lot of violent crime in the USA and the rate was rising year after year. Everyone looked at statistics and said it was going to be a worsening, horrible blight on society. No one predicted anything else. Then rather abruptly the violent crime rate started to fall substantially. It got better and better year after year and stayed down. (For example, murders in New York 1990: 2,245. 2003: 596.) This was variously attributed by the former gloom and doom predictors to a 'roaring' economy, better education, gun control programs and innovative law enforcement tactics. Except for a quadrupling of the incarceration rate (2 million Americans living in prisons) which probably accounted for 1/3 of the improvement, decreasing profits (and hence decreasing turf wars) from selling crack cocaine (another 10%?) and an increase in the number of police (effective where it occurred) most of these actually had little effect. While all the agencies were patting themselves on the back for 'solving' the problem, economist Steven D. Levitt figured out the main reason.
   In 1973 was the case of "Roe versus Wade". By almost a century before that, abortion had been made illegal in most states. Afterward it was legal and affordable. The number of unwanted children being born into adverse family environments to (take your pick) poor, uneducated, foolish, alcoholic, drug abusing, unwed teenage mothers dropped sharply. In under two decades the crime rate followed suit. The pool of the statistically most likely potential criminals entering their criminal prime had dramatically shrunk. (At the same time, so had the pool of young girls statistically most likely to have unwanted families, so the effect was and is multi-generational.)

   Levitt connected cause and effect. Of course birth control is highly preferable and was just coming into wide availability around the same time. But as ugly as abortions may be (and we may hope it's done as early in the pregnancy as possible) here was a huge positive result of restoring personal freedom. The high crime rate was an unseen result of foolishly coercing women, not all of whom were ready to, or in a position to, or capable of raising a healthy, functional family, into having one anyway. Freedom works better than government trying to control peoples' lives.

"Know the truth, and the truth shall set you free." - Jesus

   Now imagine all children, in numbers just sufficient to maintain a stable and healthy population, being born wanted into loving, caring, well functioning families that are living the morality and ethics of the seven core values of Life, Equality, Growth, Quality of life, Empathy, Compassion and Love for humanity. In a couple of generations would the world not be a wholly different place? This is coming in the decades ahead and will become a permanent quality of our planet's civilization.

(Eccentric Silliness Department)
(with a little filosofy mixed in?)

* A website, FootprintCalculator.org , reportedly figures out how big the Earth would have to be to sustainably support the global population based on usage of materials and resources. In 1971 we evidently hit 1.03 Earths. Last year it was 1.71 Earths. Unless we become more frugal and austere, or drop the population, the Earth will need to grow quite substantially to maintain its existing population for much longer. The mounting daily mass die-offs and the horrifying numbers of species going extinct probably indicates that the oceans have been essentially fished out and poisoned with plastic, and the land is faring almost as badly.
   Isn't that just another view of what more and more have been saying when they suggest we should prepare for a lean time and that the population bubble will soon burst?

* No amount of hours spent driving in the car makes up for the missed exercise of not taking a 20 minute walk.

* Politics and Peace are opposites. Ultimately politics is separation; peace is everyone in harmony together. Politics comes about when people won't work together and compromise to find the common good and the best path forward. Bipolar politics is "my way or the highway." We won the election (51%); you lost (49%), so get lost. Your view is worthless. Peace comes when all views are taken into account and there is co-operation instead of competition and strife.

* In mid May big game hunting season opened. The big game hunters came out in swarms: carnivorous blackflies, midges ("no-see-ums") and mosquitos. Best to stay inside in the evening.

* Why do Christians hate Muslims? No doubt because mosques must surely breed mosqu[ito]es - and nobody likes mosquitoes.

* The hardware/building supply store in Queen Charlotte is called "AMS". According to the store that's short for "Aaron-Mark Services". According to most everyone else it stands for "All My Savings", as in "I took in all my savings and bought something at AMS."

* HOW many functions with ONE button? The LCD DC power meter I put on my solar electricity system has only one button controlling several features. But I couldn't figure them out... When the watt-hours reading passes 9,999 WH, it says 10 KWH. Eventually it goes to 11 KWH. This is a sudden 1000-fold loss of precision. Hard to track a few hundred watt-hours per day, let alone activity within the day, with that! I had to go back to AliExpress and read the instructions, a couple of times. If you click the button, it turns the LCD backlight on or off. If you hold it longer, it suddenly says "SET" and allows setting of alarm points. (What good an alarm is on something that doesn't have a beeper I don't know... but apparently the backlight flickers.) Here Patience is the key: KEEP holding it down. In a few more seconds it goes from saying "SET" to "CLR". At this point release it, and the WH/KWH reading flashes for a few seconds. While it is flashing, click the button again to set the WH/KWH reading back to zero. (If you hold it down even longer you can change the setting from "100 amp shunt" to "50 amp shunt". That would require changing the actual current measuring shunt resistor too.)

* Microwave ovens heat things up in 2 or 3 minutes. In 20 or 30 minutes... the item is completely cold again.

Headline February 7th, 2057:

Using poultry embryos and advanced Crispr25.9 gene splicing technology, Scientist Recreates a Phorusrhacid, an Extinct Giant Flightless Bird.

June 3rd, 2057:

"You called the pound, sir?"

"Yes, it's my next door neighbor. Her chicken is eating my dogs!"

   "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 and dead ends.

Electric Transport

Ground Effect Vehicle (first the R/C Model)

Multiple Ducted Fan Propulsion

   A friend, Rick Peterson, sends me various links to what are sometimes very interesting articles about green energy stuff. He sent me a link to an article about a VTOL [vertical take-off & landing] "electric jet" aircraft from Germany for "urban" use, the "Liliam". I couldn't think what an "electric jet" might be but a ducted fan such as I intended to use, and so it turned out to be. A feature of special interest was that it had 32 of them: 12 on the flaps on each wing and 6 on each side of the elevator (elevons?) of a front canard. Instead of a rudder, it used the differing thrust of the motors to steer.

   The design was perhaps not so strange for some radio controlled craft, but mind bogglingly different in many ways from any other real aircraft for carrying passengers that I've seen.
   VTOL was attained by the ability of the control surfaces to point the motors straight down. (Step aside, Harrier fighter jet!) Ducted fans doubtless gave it sufficient static thrust with less power. And they said they made it far less noisy than a regular aircraft, much of whose noise besides engine roar is propeller tip vortex noise. In 2017 Lilium[.com] got 90 million dollars funding to make the prototype 5 seater craft (the intended full size) after demonstrating their smaller two-seater could transition smoothly from hovering to forward flight. (How do other people seem to get immense levels of R & D funding? I haven't managed a penny - not even SR & ED tax credits since 2015. Even my "Fantasy Budget" for all projects combined was only 1/4 that much. Ah well...!)

   I'm not so sure an aircraft to take off and land on special pads all over town and just to cross town is practical. It might however be productively employed for hops over water, replacing slow foot passenger ferries, or in extreme terrain (swamps, jungle, canyons...) where distances are too far or it's impractical, or the volume of travelers is too small, for a bridge.

   However, such a high speed service over water could also (and I trust better) be provided by a ground effect vehicle using much less energy per kilometer, greatly increasing the range or the carrying capacity of a given size craft/vessel.  The multiple ducted fans is certainly a fine new idea and a timely one as I was just about to start making the wing, which will hold the motor(s). Multiple fans should be able to solve at least two design issues that have been in my mind.

1. Having the thrust above the wing tends to push the nose downward. Yet if the propeller is large, to put it lower puts it into the water. Land aircraft have long wheel struts to keep their low propeller(s) above the ground. Thus "amphibian" aircraft with floating bodies (including the electric one shown in last month's issue) have the propeller back by the tail to minimize the nose-down effect. But for the ground effect vehicle, to blow some air under the wing for additional lift means putting the propeller at the front of the wing, as low as possible. Multiple ducted fans of smaller diameter would allow the thrust to be down lower, still with water clearance. Additionally, the vertical aim of the thrust might be adjusted with pivoting fans.

2. I have seen 'catamaran' shaped R/C models flying diagonally for a considerable distance before straightening out when the rudder is employed to make a turn. Keeps going straight on when turned... the collision stats will start mounting up! I was going to have a large central "tail" ("dorsal?") fin sticking up for lateral stability especially when turning. That would be fine on the model. On the piloted version it would have to be made of transparent material or something to prevent a serious blind spot in the view to one side that would make "taxiing", maneuvering and docking extra challenging.
   Multiple fans, especially if they could be aimed left to right, might take the place of the fin (or at least reduce its size).

Selecting Ducted Fans

   Previously I was wondering if the 5" ducted fan was large and powerful enough. Then I started thinking of 4 (or even 6?) smaller ones, perhaps 3 inch diameter. I'm thinking of mounting them under the front of the wing, where they would all contribute to "blowing up the bubble" under the craft that will help it lift off as well as providing forward thrust.
   So at this point, the next thing to do was to look on line and see what's available in the 3" size. Surely there were some better ones than the 5" one I have, which is the largest available but by no means aerodynamically optimized. Then I'd need a motor and controller for each one. More minimally, just getting another 5" fan and having two would fulfill the essential requirements, at least for the model.

   On the 25th I sat down to look. I already had a browser window open to AliEpress, and I just did a search there. Sure enough there were many pages of results. It wasn't that I didn't spend a little time looking over a couple of pages of other choices. But at the top left hand corner, the first result, was the biggest size, a 90mm ducted fan with 12 blades. 3-1/2 inches. It looked very good. It looked powerful. "for model jets". You could buy it with a recommended motor already installed. I wasn't prepared for the prices, thinking of 30 or 40 $. With motor it was 130 $C. But considering the alignment problems I've had with the 5" one, getting them already assembled sounded like a good plan.
   Then they needed BLDC motor controllers, "ESC"s ("electronic speed controls") in hobby parlance. 120 amps was recommended. The store didn't have one! I searched again and found a store that did, a "Sky Flight" controller. Another 40$ each!

   I sat a few moments looking at my on-line "shopping cart". Two of each, left and right. That was the minimum. I wasn't going to go for four at those prices, and it sounded like two would have plenty of power. Was I really going to do this? I decided that if I was serious about the ground effect craft, I needed to. It has such potential! I clicked "purchase". There went yet another 350$ I didn't have. I get Canada Pension starting next year. I wonder when money will start hyperinflating? Not something to count on when incurring debts! (I got the solar panel system done for the off-grid lady across the highway by the 31st... oops, June 3rd. That will pay for them.)

   Let's see... 120 amps at 22 volts is 2640 watts: 3-1/2 horsepower! Each! And in total 16 pounds of static thrust!  Put it nose up and it could probably lift off vertically! It starts to dawn on me that having not just "sufficient" but "impressive" or even "excessive", power will make for not just working but impressive demos that can create the sort of enthusiasm needed to attract funding for more development or perhaps even commercialization.
   Then, what to use for the full size manned craft? Twenty of the very same ducted fans? Those should actually fit across along the front under the wing. That would be 160 pounds of thrust. This would seem to be the approach of the VTOL "electric jet". But I suspect bigger ones will be needed.

   What part to build next if I wasn't sure how the fans would mount on/in/under the wing seemed to be an academic question since I didn't have time for building much of anything. (It's early June and there's still areas of the greenhouse that aren't planted yet!) I guess I'll wait for the parts to arrive. Before the end of May however I decided to just make the whole wing and mount the motors on top, in "amidships" on each side. It should have lots of take-off power without blowing the air under the wing, and the small diameter ducted fans wouldn't do much forward tipping.

From the 90mm 12 blades Ducted Fan Order Page:

It looks better made than the 125mm one I bought before.
It has a bit of a front rounded 'scoop' to bring in more air.
12 blades. (and a spare fan.)
It has a center streamlining piece at the back.
It comes with recommended motor, assembled and aligned.


Brand Name: FMS
Item Name: 90mm 12 blades Ducted Fan
Motor: 3546 KV1900(the diameter of the shaft: 5mm, the length of the shaft: 27mm, the max power:3000W , banana plug: 4mm)
Weight: 327g
Usage: Suitable for RC Airplane

Recommended setting:
6s 22.2v 5500mAh 35C Lipo battery (not include)
ESC: 120A    (not include)

- All composite materials processing and injection molding processing, light weight, power system 327 g, has ultra high rigidity and toughness.
- First-class mold design,ensure the blades perfect dynamic balance,the blade cooperates the duct in place,sound effect and efficiency value are comparable to all metal duct on the market.
- FMS pioneering thermal design,in does not affect the efficiency of the case,ensure the motor's normal working,no need to worry that the motor heating will affect the life and performance.
- Reasonable structure design,you can arbitrarily choose different manufacturers motor,this item can e assembled to all the 90 ducted airplane,no need to make too much manual changes to the aircraft.

Electric Ferry Danmark to Sweden

   I've noted a couple of small electric ferries operating in Norway. Here's a very large one plying a 4 Km route between Danmark (Hey, that's how the Danes spell it) and Sweden. This one was converted from diesel electric to plug-in electric by adding sheds full of batteries on the top deck. Evidently the conversion is a plus in every way.


Sweden to Denmark

Batts: 800 V 4,200 KWH
4 propulsion Motors 1.5 MW each.
46 crossings per day - every 15 minutes
4 Km
10 KV charging in 10 minutes / 6 minutes.

Inside one of the battery containers on the top deck

An automatic arm comes out from a tower to recharge the ferry when it docks.
No human intervention - they need to charge fast during
the short docking while cars are unloading and reloading.

   There is one aspect to all this that somewhat puzzles me: I thought there was a bridge between Danmark and Sweden. Why do they need a ferry? (But I suppose if I looked into it I would find the routes are quite separate.)

Solar Train

   Also from the "Fully Charged" Youtube channel that did the ferry video... In someplace called Byron Bay in Australia, a short railroad between the town and a resort was converted to solar electric. It's a level track so it doesn't take much power. The train, fully restored and converted "historic" units, has some solar panels on the roof (flexible ones to match the original curved roof), but mostly it recharges at each station. (through the rails, perhaps?)

Other "Green" Electric Equipment Projects

36 V DC "Off Grid" Infrastructure

HAT36V-50A Ceramic Sockets

   I printed the "bottomless" mold on about the 7th and tried it on the 8th. The clay didn't come out very well, but it was much better than with the previous mold. So I went about fine tuning this one. Someone named Eric who I've never met said the walls should be sloped 3 degrees to make it easier to get the clay out. That seemed reasonable. Trying to program that in with OpenSCAD would be a daunting task, I'm sure. I needed also to smooth the side walls, so on the 9th I filed them and put a slope into them that way. After I was reasonably satisfied that I had smoothed it off as best I might without going to exacting lengths (both halves), I then coated it with the liquified wax.

The mold with smoothed and slightly angled edges, and the clay pieces made.

   That night I stuffed fairly stiff clay into one side. Pushing on the bottom sections carefully, one area at a time, the clay suddenly fell out. It looked workable! So I did the other half. Then I did some edge trimming, then dug out the bolt holes with a small piece of brass tube. After it dried I would need to fix my mini-kiln to fire it. Somehow that didn't happen during all the rest of May.

"DIY" Blades for High Power HAT36V-50A Plugs?

   The spec for the blades is 2mm x 8mm x 20mm. There are no common solid wire sizes that will flatten to 2x8mm.  It's not even a very common thickness of copper sheet to find. However, 1/2" copper plumbing pipe has about 1mm thick walls, and 1" pipe has about 2mm. So one could slit a 1" pipe, pry it open and flatten it down to get 2mm thick sheet copper. Alternatively, one could slit open a 1/2" pipe, flatten, and cut it 8mm wide by double the desired length. Then fold it in half. The closed end is the outer end, to be inserted into the socket. Perhaps most ideally, the other end can be shaped to hold the wire between the two pieces. Be sure it fits in the shell before soldering.

Kitchen Hot Water Tank - Plumbing & Wiring

   I've always said "No level of scalding temperature atones for the hot water having taken so long to get here." The answer I have long wanted to try, a tank right by the sink, I finally did. It's great!

   Near the start of the month I put together a rather complex set of pipes and fittings to connect the hot water tank under the sink. Then I remembered that since I had done any plumbing decades ago, people had switched to using flex tubes with threaded fittings on both ends to connect faucets to the rest of the plumbing. I supposed that the fittings on faucets were some special size, but when I took the one off the hot water tap it proved to be the most common size, "1/2 inch NPT" - the same size as the hot water tank. So the tubes would be too. When the flex tubes were created, they were quickly adopted and everything else was cast aside. They are not only flexible, but the nuts are loose and can be screwed on without turning the tube, giving threaded connections without the need for "pipe unions" having a loose nut.
   I would still need the rather complex little assembly with two shutoffs I had mostly already put togther to split the hot water off from the cold and then reconnect the cold, but the connection from there to the tank, as well as much of I had meant to use from the tank to the hot water tap, could be replaced with two flex tubes. Pfew! But I couldn't find one long enough so I had to make an "L" piece so the tube would reach.
   Having finally done everything, I found the old valve I'd used leaked to the outside world. The "O" ring rubber was perished. I didn't have any the right size. So on the 9th I had to go into town, for the fourth day in a row, and get a new one. Since I would have to shut the water off, I wanted to start the job in the morning just in case something didn't work out and I had to run out yet again and buy some other part before I could turn it on again. The 10th was out. That meant waiting until the 11th. Rats! How long was this going to drag out?
   On the 11th I finally got the main pipe assembly in. It took a couple of hours and two tries. It looked like a solder join leaked... then it didn't. I had to cut the original plastic cold water pipe. I really hated to do things that were irreversible in case something didn't fit right in the new assembly. Plastic "pex" pipe had replaced the plastic pipe type used when building the house in 1988 and there was almost nothing that fit it. The selection of plumbing parts on this island is pretty limited if I needed something else. It could mean no water... for how long? But there was no choice. Happily the parts I thought would fit did.

   I found the Dernord element was (after all) the right size for the threaded hole in the tank and I put it in. I disconnected two of the three 400 watt elements, leaving 400. Then it occurred to me I could put in a switch and switch it to 400 or 800, or even two switches and have 400-800-1200.
   The original thermostat went into a special tube in the original element assembly, so I couldn't use it. Another hot water tank thermostat I've had kicking around for decades turned out to have at least two problems. It was junk. (And I threw out several others when I moved - rats!)
   So I made a hack job of putting the original thermostat up against the side of the tank, inside the insulation. It might not respond as fast or as well - and it sure wasn't proper - but it should suffice.

   Knowing I couldn't scrounge any very long wire, I had recently broken down and purchased a 75 meter roll of AWG #14-3 house wire cable "for solar projects". (146$ - I knew was saving money by scrounging/saving wire!) I used about 20 meters to go from the solar to the tank at the far end of the house. With the bare ground wire, "#14-3" actually has four #14 wires. The black and bare, and the white and red, can each be doubled to make the equivalent of two #11 AWG wires, which would be rated for 30 amps. 1200 watts at 36 volts is actually 33 amps. First, that doesn't seem like serious overload. Second, I figured I'd probably never use all 1200 watts anyway. I'll start with 400 (11 amps) and see how that works out, and then go to 800 (22 amps) if that heats the water too slowly. (It turned out 2 in series for 250 watts was plenty.)
   I tried my best to arrange it so I'd only have to crawl though each section of the dusty crawl space once, but such plans rarely work out. The worst part is a long section only 16" high, crawling on my belly on the concrete. I found a solid wall at the end of that with nowhere to put the cable through. I called someone to help me the next day, but he wasn't on the island. I decided by gosh I would finish it that same day after all. I went back in with the drill. In spite of good luck pushing the wire through that wall, I had to go in to the other low section with a sand floor three times to pull the wire up under the sink with a coathanger wire and tape - the cable came off the first time. The next time I folded both the coathanger and a copper wire around each other before I taped them together so I could pull harder. I got out alive, if sore and tired and coughing up dust for a while. An interesting thought is that while the water heater was under the sink, if it had been 120 or 240 volts, I don't think I'd have wanted the power connection to be under the sink where it can get wet.
   Everything was ready for the hookup except that I didn't have the first ceramic HAT36V-50A socket fired yet. Ah well... "wire nuts"! Definitely time to call it quits. Plumbing, 36V heater element installed in water tank, water tank installed, then a long wire run across the whole house in the crawl space. That was lots of progress for one day!

   Next day (12th) I hung up the wire with some wire staples in in the easier access places, hooked up the water heater, and hooked up the other end of the wire in the breaker panel. I was going to use the new surface mount breakers, but there was no place to connect ground (B-) wires except inside the box. It would seem there's a need for a surface mount grounding block for external breakers.

   First problem: When I turned it on, it was only 275 watts instead of 400. So I took the water heater cover off (not so easy being all plumbed in now - it would have been impossible without the flex tubes!) and checked it. I ended up hooking the power wires directly to one element, and it was 460 watts. Bad connection in the thermostat? (Later I noted that's the power with two elements in series. No doubt I had it wired that way by accident.) I put it back with two wires sticking out that could connect to the thermostat either one element (which turned out to be about 500 W with full batteries) or two in series (265W). Power seemed good. After a while the water was getting warm, then hot. I did a few dishes.

   Next problem: The MPT7210A was putting under 4 amps into the batteries, while the tank on 225 W was drawing 6 amps. The MPT7210A is supposed to be 10 amps, but now that it was actually being asked to put out 6, it wouldn't! It wasn't drawing much from the solar panels either - they were almost at open circuit voltage. The unit was running cold, definitely not limiting itself because of overheating. It just wasn't even trying. It was putting out 135 watts while the water heater was drawing 250. The batteries were getting lower.
   It turned out that if it's turned OFF and then ON again, it wakes up and delivers the current, but I didn't know that at the time. (And manual intervention certainly shouldn't be required!)

   I decided to connect the PowMr 60 amp charge controller. That meant some rewiring, which took quite a while. The PowMr needs over 50 volts in to put out 40 volts, which meant taking two panels away from the grid tie and putting them in series with the other two. That's well over 60 volts open circuit - my one beef with "buck" controllers. I would rather they "boost"ed from 30 up to 40. (Don't touch that, no! put the nut in the nutdriver and don't touch it while putting it on! Only touch the handle of the nutdriver, not the metal shaft!) At this point I did get to use a couple of the surface mount breakers to disconnect the solar and the batteries. The PowMr and the Y Solar grid tie inverter at this point seemed to fight over the two panels still connected to the grid tie. Soon I gave up and turned off the grid tie. The PowMr was however much more satisfying than the MPT7210A: it was putting out 500 watts instead of 135, and recharging the (now lower) batteries even with 235 W going out to the water heater.

   Next problem: I went into the kitchen and found the floor flooding. The soft plastic (silicone?) gasket that came with the Dernord heater element was surprisingly soft. I had wondered about it. Now it had squished out from the join as the water got hot and the tank was spewing water - by now cold again. I made a new gasket from a piece of hard rubber but it leaked. I looked at the original element. It had a simple rubber "O" ring. I put that on the Dernord and put it back on. Everything was so wet I couldn't tell if there was some minor leak, but it seemed okay. I put it back together (still minus the outer cover), connected the 400 watt element and turned it back on. With the batteries at full charge that was doing 512 watts. There was still almost an hour before the tree shadows would shade the collectors. Could I get some hot water again? In 45 minutes it certainly felt hot coming out the tap. That didn't mean the whole tank was hot. It was still heating at 500 watts.
   And the PowMr was still putting out a similar amount and keeping the batteries up. It certainly seemed to perform well. If set to 40.5 volts, when the tank went on within seconds it would be putting out as much power as was being used and even a bit more to keep it at 40.5 volts.

   Next problem: The water got too hot. I don't like scalding. I discovered that one set of contacts on the thermostat was just "pass through", not a switch. I moved the wires to the other side. There it switched on and off properly. While I was at it, if it was going to heat that fast, I connected the wire with two elements in series instead of the single element. The series pair drew about 265 watts with fully charged batteries. (137 W each. They should never burn out!) I'm pretty sure I'll never want the full 1200 watts or even 800. Probably not even 500.

   The water was seemingly heating more than twice as fast as in the experiment a couple of months back with 100-140 watts. I'm putting that down to the element sticking into the tank from the top instead of from the side near the bottom, and the Dernord element being considerably shorter than the original. So it's probably not heating the whole tank very quickly, just the top half. That is actually fine with me if I don't run out of hot water doing dishes.

Solar Hot Water Panel(s) Versus Solar-Electric Water Heating

   The other thing I would note is that I never got anything like this sort of hot temperature when I was doing solar hot water with a hot water collector and pipes around 1980. Notwithstanding all the work installing it, this one has far less plumbing and was much easier than putting in a hot water solar panel and a solar water preheat tank - to feed an electric tank which would use grid electricity to finish heating it. This was much simpler and more satisfactory. And once the water is hot, the solar power is available for other things. I think except perhaps for swimming pool heating, electric is the way to go even for solar hot water.
   For December when there probably won't be enough solar power many days, I think I might just add on a battery charger that plugs into the wall to keep up the 36 volts, and stick with using the DC system. Of course, only four of my ten solar panels are on that so far.

Heat Loss

   Aside from having the cover off the small tank, the pipes coming out the top including the pressure/temperature relief valve, were hot from the water. I had used some hot water, yes, but just four hours after shutting it off, the water was hardly tepid. The pipes seemed like major radiant heat leaks. Some pipe insulation was surely needed. On a shelf I had a very fat (3" O.D.) piece that I cut and put on before bed, and a couple of thinner bits that I put over the flex hoses.
   To skip ahead a day, the water was pretty cool again just 2 or 3 hours after I shut the power off as the sun went behind the trees at 4:30 (PST). I guess with such a small tank (15 L) it's probably inevitable it will cycle on and off frequently, and one just can't expect it to keep the water hot for 8 hours.

   The next morning (13th) was cloudy. I turned the water heater breaker back on at 9 AM. It was using 265 watts. The PowMr managed to put out 200, so it was usings batteries for the other 65. But the solar power crept up as the sun rose - and then the clouds lightened a bit - and soon it managed to equal the drain, and then to recharge. By 9:45 the water coming out the tap was pretty hot and by 10 quite hot.

Hot Water! Clean Dishes!

   It was a real pleasure to have hot water in the kitchen, in 2 seconds, instead of tepid water after 70 seconds! I could wash dishes as I used them, instead of letting them pile up for weeks then spending an hour and a half at it, and still the "hot" rinse water got cooler and cooler because I wasn't running it fast enough. This is how I've always wanted to do it. Having a central hot water tank at a distance has always been an obstacle, and this one was an extreme example.
   I cleared up much of the kitchen sink clutter by 10:15. I filled a small metal bowl rather than the whole sink, but I mostly just squirted a bit of dishsoap into whatever I was cleaning and ran hot water into it. Before I was done the water wasn't as hot as it was at first. But it had only been powered on 75 minutes at that point, at 265 watts. I picked away at the mess and wiped parts of the counter a few times as I passed through, and before the end of the day, the kitchen started looking almost habitable. Hot water any time you want it is great!

   Last time I had noticed the energy meter on the DC panel, it read somewhere around 1000 watt-hours after some months of use mainly just for the night light at 2.4 watts. But after a couple of days of heating hot water in the daytime, I noticed it had now hit 3709. I was finally making some substantial use of the DC power system!
   The next day (14th) I turned it on and in a couple of hours(?) it said 4333 WH. So 624 watt-hours to heat up the tank from room temperature. I wasn't using much hot water in the kitchen (3 or 4 quick rinsings), and by 4:30 PM it said 4483. That meant that it only came on very occasionally and had used just 150 watt-hours or so to keep the tank hot all day, 8 hours. Maybe it wasn't such an energy hog as I had been thinking after all! At that rate, all I had to do to save the batteries yet have 24 hour hot water would be to not use very much hot water in the kitchen at night. Even if I did some dishes after dark they shouldn't be down too far by morning.
   So I tried NOT shutting it off for the night. (What?!? Actually USE the batteries? For more than a night light?) At 6:30 PM it had gone up to 4578 - another 95 WH. At 7:30 it was 4641. Apparently it uses around 50-70 WH for each "top-up" reheat?

May 14th, 15th, 16th
Time (PST) / Watt-Hours Reading / WH Used by water heater / Battery Voltage
09:00 3709   0   40.5  Turned on circuit breaker - water heated from room temperature to scalding.
11:00? 4333 [40.5 - being charged]
16:30 4483 150 [40.5] - (I turned the thermostat down a little at 11 AM so it may not have come on for a while.)
18:30 4578   95  V??      (Didn't record battery voltage. It was well over 39 volts.)
22:20 4721 143 38.24
24:00 4768   47 37.61    (Voltage drop doesn't look good! [but maybe it had just shut off?])
  6:30 4982 224 38.6      (Too late to see minimum voltage - it's already charging 3.2 amps)
  8:30 5048   66 40.5      (Total 1339 WH - includes initial heat-up of 624 WH. Only a few rinses of hot water use.)
14:30 5248 200 40.5
17:30 5311   63 40.5
18:30 5385   74 40.0      (I turned voltage down to 40.0 - go easy on the poor old batteries. Ouch! Water is hot! 817 WH in 24 hours.)
20:00 5452   67 40.0      (Sun almost down - behind the trees - still getting 35 W of solar!)
21:00 5518   66 39.2      (Caught it in the act - 235 W. Voltage dropped to 37.?? while it was running, but recovered to over 39.)
24:00 5649 131 37.6      (again... 215 W and barely 36 volts while running. 891 WH over 24 hours.)
  1:30 5718   69 ~37      (why am I still up?)
  4:30 5787   69 36.9      (Did I really get up to do this? Batteries: 12.7, 12.6, 11.2 -- AHA! It's just ONE battery that's the problem!)
  9:30 5919 142 39.9      (~880 WH in 25 hours (= 845 in 24).)

   Reading the voltages was a bit of a pain. If the meter on the PowMr would say 14.2, then the one on the breaker panel would say 14.57. One voltmeter would say 14.3 and another 14.4. I finally started using the one on the breaker panel at night. It was obviously a bit optimistic, but the display always showed voltage, and it had a fourth digit to decrease round-off error.

   Next improvement would be to install the "chemical free iron filter" so the well water doesn't stain everything brown. Then it'd be perfect!

   Oh yes, and then to install the 240 volt plug and socket so I can unplug the well pump from the wall and into the 36 volt DC to 240 volt AC inverter instead, and so still have water if the grid power is off. Then it'll be perfect. [Still not done June 3rd]

(Then can I move on to some more exciting project?)

   On the 23rd I tried to chart how fast the temperature rose, with just a 250 watt (or thereabouts, depending on voltage) heating element in the 15 liter tank. Considering how long it took to heat up with around 120 watts in the experiment a couple of months ago, the performance seems surprising. But again, it was probably heating the top half of the tank, which only gradually conducted down to the bottom half. The water was quite warm to start with, having been hot the previous evening before I shut off the breaker (owing to battery/charger problems) . (I didn't record the initial temperature or the time I flipped the breaker on.)

~11:20? Start
11:30 37.2°
11:40 40.1°
11:50 42.0°
12:00 42.5° ...dropping battery (bad connection - fixed)
12:10 42.8° Huh? I think my measurement is using too much hot water.
12:20 43.1

   Owing to there being almost no solar and the batteries dropping, and so the power going down instead of staying constant, I ended the test and shut off the breaker. Time to get to repairing the main battery I had out! One problem with the test was that I was actually running hot water to measure it at the faucet, and that the temperature sensor took a long time to heat up, so I was actually using a fair bit in order to measure it.

"Universal" Power Control?

   Shutting off the breaker to the hot water for the night a couple of time brought me back to the need for a voltage control. I checked out the "output" circuit on the PowMr. Although I could set the battery charging to various levels (thank goodness) the output shutoff defaulted to 11.2 volts and could only be programmed up to 12.5 (*3=37.5V). I would pick at least 39 volts and probably 40. This arbitrary maximum level would mean the NiMH batteries would be over half way to flat before the load would shut off. I only wanted the water to heat when they were fairly well topped up, ie, when the sun was keeping everything charged. For the lack of imagination by limiting its shutoff to this arbitrary low voltage, it couldn't be used for that purpose.

   At this point I'm thinking once again of designing my own control. Now I'm thinking of using a microcontroller and a small LCD display and making it highly programmable. Even with allowance for voltage drop in the line. In fact, it might have 3 or 4 outputs to control that many loads. One or two could be N-channel MOSFETs for simple resistive DC loads, including high power loads. Another might be a relay for anything. Another output or two might be a simple digital output to control (eg) a solid state relay. Thus it could be a controller for all solar and intermittent power purposes, installed at the main equipment panel. Also it could have a temperature sensor input(s) to control heating or cooling of any sort, in which case it would probably run only one thing, installed at the equipment it would control.

   Another thing it could do in a central setting would be to switch on another load "load B" when "load A" was satisfied. For example, it could turn on a baseboard heater once the hot water tank was hot.

   That could be an actual valuable, salable product! (When am I going to make it and program it, again?)

Arcing Thermostat

   A reason cited by some for not using DC water heating was that the thermostat might arc. That seemed to me to be a pretty paltry reason. Surely an appropriate thermostat can be had! 120/240 volt AC circuit breakers, when they bother to also give them a DC rating, seem to be rated for about 40 volts DC. I figured a 240 volt water heater thermostat should likewise be okay at 40 volts DC. I had ordered a couple of high rated ones on line, but the one I had proving to be defective, I presently had only the original 120 volt one from the tank.
   At 10:50 (on the 12th), it having had another 1/2 hour to reheat, I moved the dial to see how close it was to shutting off. When it clicked, there was a green glow from within and a hissing sound. It was arcing, and not shutting off! I did that a couple more times with the same result. After a little bit I tried again and it shut off normally. There was just a quick green flash. But it was certainly disquieting! I hope the "240 volt" ones arrive soon! Meanwhile I thought I'd better keep it shut off at the breaker when I wasn't home. This would be a good place for the "universal" control above, set up to measure the voltage and the temperature, and to switch the water heater via a power MOSFET - solid state; no mechanical contacts arcing.
   But I turned it on and off a couple more times, and left it on while I was home, and it didn't do more than the momentary green flash. I know it went on and off by itself at least once - probably 2 or 3 times. I decided it must be all right... for now. It continued to behave itself into June.


   One fly in the ointment for the kitchen hot water was that I didn't want the tank to run at night and drain the batteries. That would mean manually turning the breaker on in the morning and off again before the sun goes down. That increased my desire to make a "universal power control" that will control most anything solar or even multiple things, via flexible control schemes to be worked out by the user/purchaser himself for various circumstances.
   But after a couple of days my apprehension was relieved when viewing the watt-hours meter on the solar breaker panel indicated the tank (the only daytime load) only came on occasionally and would use under half a kilowatt-hour overnight if I wasn't running hot water. I shouldn't fill the sink at night, but quick rinses/washes with the tap running slowly wouldn't make much difference.

   The NiMH "D" cell battery bank however, while not completely useless, appeared to be quite disappointing in actual use after all the repair and replaced cells. They lost a lot more voltage than they theoretically should have. At first I thought well, most of them are 5 years old now and were used hard and perhaps abused in charging in the converted Mazda RX7-EV. Still a 4.7 KWH battery bank shouldn't drop from 40.5 volts (1.35 V/cell) to maybe 36(? - 1.2 V/cell) from supplying just 0.5 KWH of night time electricity. (How can I ever find the time to finish my 'everlasting' battery experiments and then get them into production?)

   Then I got up before dawn with the hater heater turned on, and checked on them. They were down to 36.8. That was getting toward "flat" - not very good for 4600 kilowatt-hours of batteries having been drawn down by less than 1/2 a kilowatt-hour. As soon as my head hit the pillow it occurred to me I should check the individual 12 volt sections. I got up again and found: 12.7, 12.6, 11.2. AHA! It was the one battery that I hadn't done any work on that was 'bad'. I had thought it was okay. Evidently I should take that one apart and (no doubt) replace a few cells, too.
   The next day (16th) was heavily overcast with raindrops and the solar was only charging at 75 watts. I did some dishes, then at 10:30 went out and looked. The water heater was running. The batteries were at: 13.0, 13.0 (steady) and 10.3 and dropping. Definitely some repair was needed! Ugh! More work! But when I broke it into parts, none seemed to discharge especially fast. It did appear however that a fat jumper between the two boxes of 50 cells each, somehow had a high resistance connection. And the 'extra' 30 amp-hours in three tubes was attached to the wrong one. Could it be that the lowest battery was operating on just 50 amp-hours instead of 130? That might explain the problem and it needed to be checked out. Too bad it was on the bottom, underneath the other two box batteries and some tube ones.

   During the day on the 17th I checked: 13.5, 13.8, 12.9. The lower box was 80 mV higher than the upper one if the water heater was on, with the difference being at the negative terminals. despite a fat #6 or so wire between them.
   I was also surprised that the middle battery was up to 13.8 volts. It's acceptable, but I hadn't meant to charge them that high, only to 13.5. It shows how having batteries out of balance can wreck the less leaky ones that hold charge better. If I had set them to the highest steady charge level for NiMH, 14.0 (total 42.0 volts), the middle one would have been overcharging.
   That night before bed, I went out and checked: 13.07, 13.14, 11.60. I turned the hot water breaker off (as I had the previous night too).

   It wasn't until the 24th I got the bottom battery fixed. I lost count, but I replaced about 7 "D" cells in one box and 6 in the other. give or take a cell. No wonder it got low so fast! Once again I note that it was mainly but not invariably batteries with the "+" button facing up that were bad. This was the oldest battery from the RX7 (2013?) and there was some rust around a few of the solder joints.

The solar power system NiMH "D" dry cell batteries, now in "racks" - mostly 5 or more years old.
Bad cells have been replaced in the three 100 amp-hour boxes and in the three 60 amp-hour
tube sets theoretically providing 160 amp-hours at 36-40 volts - 6000 watt-hours of storage.
A single tube with an unscrewable end at the front is charging 10 spare cells.

Also seen is the new "Air Injection Iron Filter" for the house water from the well.
Water quality is much better without the water gradually or quickly turning
brown and staining everything after it comes out of the faucets.
(Its backwash cycle is 75 minutes(!) instead of 10-15 for the old filter.)

Hey! I thought I had cleaned up those loose pieces of wire before I took pictures!

   After I fixed the third box battery and put it back in, when I went out in the morning the MPT7210A was only putting out 150mA to keep the batteries up - much the lowest I'd seen. Before fixing most of them it had been, IIRC, over half an amp to keep them up. Before this last battery and the tubes, it was at best 250-350mA.

   Now that I've made the "racks" for the tube batteries I think I like them better. There is at least no disassembly and unsoldering to isolate the tubes and see which one(s) may be causing a problem. OTOH, for each tube that has a problem, the end has to be cracked open, and reglued with methylene chloride to close it again. But again no soldering to replace cells.

   Having now done all the boxes I started checking out each tube. Not just a few but most of them seemed to have low voltage once disconnected and separated. When I opened them each one typically had one bad cell. I attribute it to the hard use in the RX7-EV (excessive current drain) with the continuous charging at the highest float voltage, 1.38-1.40 volts per cell.
   By the 28th I seemed to have got them all. Now there was 160 amp-hours in all three 12 volt sections, around 6000 watt-hours. If the water heater was on at night, the voltage held pretty well. And I had two odd tube sets of 30 amp-hours each - the "quintos" and the former car starter. One more set of three and it could have 190 amp-hours, about 7000 watt-hours. I still have lots of cells to do that. Going to 200 AH I'd be getting into a few dicey ones. (The exact watt-hours figure depends on what one decides to call the voltage. Mostly it's over 38 volts. I wouldn't want to draw them down below 36.)
   I am certainly hoping they will provide long and reliable service with a float voltage of just 1.33-1.35 volts per cell and very moderate load currents. 6.5 amps (water heater) spread over 16 sets of cells in parallel is a lot less than 70 to 200 amps from a car motor spread over 10 sets. If it's really supplying the whole house, steady-state currents should still be under 15 amps (fridge+freezer+water heater+some lights...) most of the time.
   I left the kitchen solar water heater on on the night of the 29th, and at 6 AM on the 30th it had come on 3 times using over 200 WH. The batteries, now all 160 AH and supposedly repaired, read 12.85, 12.8, and 12.65. When fully charged they seemed to be around 13.3, 13.2, 13.0.
   Might there still one bad, or just more leaky, cell in the top (formerly bottom) newly repaired box battery or the 6 tubes in parallel with it? I thought I should probably separate them into box top and bottom halves and the tube set, and see if one of them dropped more then the others. Or was it just natural variability with old, abused cells, perhaps combined with the fact that I wasn't really "topping them up"? Did I really want to get further into this?
   Out of sequence... on the night of June 1st I went out late at night and found the water heater on - and the battery voltage down notably. Now to my surprise, while two batteries seemed fine one had in fact dropped to 11.5 volts. But it was the supposedly "better" bottom one, not the "lower" top one. I turned the heater off and it bounced back to over 12.5 volts. It was disquieting that it should drop so far with just a 250 watt load, but it wasn't really "85%" discharged as 11.5 volts open circuit would indicate. I swapped the 60 AH tube sets between the lower two batteries hoping to get better balance. It seemed to help.

   I don't think I would recommend to anyone to use NiMH dry cells for big storage unless they were getting a really good deal on them. I do again lament the killing of the big flooded NiMH cells as used in the EV-1, RAV4-EV, et al. And I am trying again to get MnZn or NiZn cheap, 'forever' cycle life battery chemistry going with an ion selective membrane, pH 11 electrolyte, and jelled electrodes.

Fried Charge Controller- Oops

        The PowMr charge controller when it was wired right and working
   On the 22nd I dug out the 100 AH box battery from the bottom of the heap for inspection and repair. When I went to hook everything back up with other batteries, the ground wire to the PowMr was too short by about 2 inches. I got another wire and hooked it up. but to the wrong terminal. When I turned the breaker on there was a big "Zap!" and it blew. The PowMr had two terminals for two wires to each connection, and I had connected the new ground to the +36 volt battery terminal. I'm not sure why that blew the controller, but electronics are delicate. I have the impression the microcontroller blew rather than the heavy electronics. Perhaps it just zapped the flash memory and lost the program. Anyway I can't reprogram it.

   Obviously the operator is the the one responsible. That's who put the wire on the wrong terminal. But without trying to blame others for my mistake, I like to analyze contributing factors. One could say the driver of a car is at fault in an accident, but obviously if he hadn't been on the road it wouldn't have happened. The fact that he was is a very obvious contributing factor.

   We might itemize the following contributing factors.

1) The labeling of the terminals was unclear. It was just small white figures silkscreened onto the circuit board, in the deep shadow under the terminal block itself. (On the MPT7210A the terminal markings are larger and printed right on the front of the case.)
2) There was insufficient light in the area, contributing to the above. This is typical of places where equipment is installed. I couldn't read the markings but since I thought I knew which terminal was which, I wired it anyway.
3) The "hot" +36 volt wire was white instead of red. Because in AC wiring black perversely is hot and white is neutral (which stupid color scheme I complained of just last issue of TE News), putting a ground wire right beside a white wire triggered no mental alarms. (Apparently neither did the yellow tape.) It's not that I shouldn't have known - it's just there's a confusion factor from there being two incompatible wire coloring schemes.
4) It's hard to get red wire. That's why it was white. White and black are everywhere. Black is ground. White is completely ambiguous. That one can't readily get the right color wire is a problem in doing DC wiring.

   If any of these contributing factors hadn't been there, probably the accident would never have happened in spite of my usual lamentable glossing over of "the fine details" in this sort of job and working "on autopilot" without much presence of mind, a habit I've never fully overcome. (Maybe a few more hundred dollars of blown equipment will do it. Nah, probably not!)

"The only people that never make mistakes are the ones who never do anything." - Roger the plumber, heating system specialist.

   It was the last of a bunch of connections I made to connect up a temporary set of batteries while one of the three "main" ones was out of service. There were of course things I could have and obviously should have done. In the first place I should have wrapped some red tape around the white wire when I put the "B+" wire from the breaker to the PowMr in place weeks ago. In the second place, being unable to read the tiny letters in the dark shadow in the dim garage, I should have gone and got a flashlight and shone it to read them and verify the connections.

   I decided to try the MPT7210A again.

Charge Controller Vagrancies

   The MPT7210A is a rare boost controller that raises a lower solar panel voltage to a higher system voltage. Most are buck controllers requiring  the power input to be higher voltage than the desired system voltage. For a 36 volt system this means the solar panels must be put in series in pairs, putting out over 50 up to 80 volts - into dangerous territory if touched.
   But it does appear to have something of a flaw. When a load is drawn from the the system, it seems absurdly slow and reluctant to respond. When the water heater comes on, the MPT7210A takes minutes to gradually, stingily even start bringing up its output current much. In the meantime, power is being drawn from the batteries while the solar panels are almost idle. Why after several minutes should the batteries be down 2 volts* from the setpoint, supplying 6 amps, and yet the controller has only increased its output from half an amp to 2 amps, while the solar panel voltage is way above the MPP and they're hardly being drawn from?
   But if the MPT7210A is turned off and then on again, it "wakes up" and quickly ramps up to supply the required 6 amps and more. Obviously manual intervention shouldn't be required. Of course without intervention, when the water heater shuts off the voltage will gradually recover, but why take it out of the batteries in the first place when there's sun and the controller can supply 10 amps?

   In contrast, when the 60 amp PowMr charge controller is installed and the water heater comes on, within seconds it is supplying as much power as is being drawn to keep the battery voltage at the setpoint, so the power essentially goes straight from the solar panels to the load. This pretty much eliminates any cycling and draw-down of the batteries. Isn't that how it's supposed to work?

(*Down by 2 volts in just minutes? Yes, there was still a weak battery on the 25th, readings: 13+, 13+, 11+ ~=38 volts. Dropped in 10(?) minutes from setpoint 40.5. Shortly after this I put the third 100 AH box battery, repaired the previous day, back in.)

Electricity Generation

My Solar Power System

[In?]Efficiency of 1000 Watt Y-Solar Microinverters

   On the 26th I hooked a third 305 watt panel to the grid-tie inverter already having two the same on it. On the 27th I noticed that it didn't seem to have 1.5 times as much output. So I disconnected one, then two and got the following readings:

1 Panel  : 235 W
2 Panels: 445 W
3 Panels: 635 W
4 panels: 815 W (Trailer roof with four of the same 305 W panels)

   From one panel we would expect that four would make 940 watts, but only 815 was coming out of the trailer according to the meter. So each panel was adding less power than the previous:

#1    235 W (This is about what the 75% "realism factor" says you can expect from these panels in real world conditions.)
#2 + 210 W
#3 + 190 W
#4 + 180 W

   The sun was pretty high but not yet at its zenith. The only panel pointing straight at the sun was the one I had placed on the lawn, panel #3, which didn't seem any better than the others. The conclusion is that the Y-Solar microinverters, while being rated for 1000 watts peak and 900 continuous, are more efficient when more lightly loaded. So I'll go back to a previous idea: that they are cheap enough that it's best to have more of them running with lighter loads. Probably three panels on one inverter is plenty. Two is even better.

One grid tie inverter on the wall, one in the (turned off) freezer.
Hide and Seek

   If I put two inverters in the trailer each with two panels, instead of one with all four panels, at noon I should get over 900 watts instead of ~830-850. But I looked everywhere for the fourth grid-tie except where it was and couldn't find it. It must be around here somewhere... but then I've failed to find things before because they were on edge instead of lying flat, were a different color than I thought, were out instead of in a box, or even were hidden under a sheet of paper. It would surely help if I could simply remember where I put it! Surely it should be in the electronics shop near the other inverters? I looked one more time and found it at the far end against the wall with a just-sufficient piece of opaque white foam plastic draped over it to hide the shape and view of the on-end box.

   I put this fourth grid-tie in the trailer and connected 2 of the 4 panels to each one. I thought I'd see how much power there was at noon, but it soon clouded over. At one point in the afternoon it lightened up (not completely cleared), and the output briefly hit 899 watts. The previous high was about 860 in full sun, so it must have been working better. The next day was sunny and at some point I found it was making around 885 watts where it usually was around 830 to 850 at the most.
   It makes a little more electricity. Is it worth it? For simplicity call it 50 watts extra, so 20 hours - two sunny summer days - to make an extra kilowatt hour (worth 13¢ at the current subsidized price, the same throughout BC). 20 days for 1.30$, 200 for 13$ (call it a year), and somewhat under 2000 hours - 10 years - to pass the purchase price of somewhat over 100$. If OTOH we take the real price of diesel generated electricity here of 50¢/KWH, that's 1/2 a dollar in 2 days, 5$ in 20, 50$ in 200 days (a year) for about a 2 year payback.

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

(All times are in PST: clock 48 minutes ahead of sun.)

Date  House solar KWH, Trailer Roof solar KWH - day total KWH made [power co. meter readings] weather, usage...

April 30th 69.91, 245.80 - 15.46 [65940@9:30, 942@19:30] Sun. Few chemtrails. Finished charging car (5 Hrs at 1500 W from 9:30)
May 1st 77.40, 252.05- 13.74 [65954@11:30, 959@21:00 ] very light overcast/chemtrails. Sunny later PM. Drove 55, started charging car
  2nd 82.13, 255.88  - 8.56 [65965@9:30, 975@20:00] clouds with sunny breaks. Finished charging car but not much solar.
  3rd 86.79, 259.41  -  8.19 [65987@13:00, 991@19:30] Clouds.
  4th 90.24, 262.32  -  6.36 [66000@10:00, 009@19:30] clouds. again. 55 Km & 2/3 charge car.
  5th 93.80, 265.58  -  6.82 [66022@9:00, 030@21:00] clouds. A couple of short sunny breaks. rest of charge car.
  6th 96.80, 267.89  -  5.31 [66040@10:00, 055@21:00] heavy cloud. 55 Km.Charged car, laundry x2. Not much help from solar!
  7th 99.52, 269.94  -  4.77 [66082@19:30] Is it my imagination or is each day duller than the last? 60Km & fast charged
  8th 106.90,275.60 - 13.04 [66101@22:30] sun but dense chemtrails, later PM clouds, 55 Km & slow charged until 6PM.
  9th 115.66,282.36 - 15.52 [66108@9:00, 115@19:00] sunny! yay! 4th daily drive to town 55 Km & slow charging with solar help from ?2PM.
10th 123.46,288.56 - 14.00 [66125@19:00] sun but chemtrails. Drove to Masset in Echo to pick up Air Injection Iron Filter for DW.
11th 132.38,295.38 - 15.74 [66131@9:00, 132@18:00] sun, clear. Bath AM. With HWH turned down to 120V/750W, solar reheated the tank over some hours - house power meter didn't move. Now if only I could get the car charging down to that sort of level, real energy independence would be feasible on sunny days - at least from a power use perspective, or if tying all the inverters together with the grid down works. (Even if you could only drive to town one day out of three.) Of course the other alternative is always... More Solar Panels!
12th 137.86, 301.50 - 11.60 [66139@9:30, 141@20:30] Put in kitchen hot water tank. Had to disconnect the 1 KW section of 4 panels and hook up PowMr charge controller. It didn't seem to get on well with the grid tie inverter, so 2 panels "lost" to grid tie by being set in series to higher voltage, and 2 I have to turn inverter off if there's a heavy load (ie the water heater) on the batteries.
13th 142.10, 306.62 -  9.36 [66150@9:00, 151@19:30] cloudy, bit of sun. No bath, no laundry, no drive, only big power draw was night bedroom heat.
14th 144.78, 310.09 -   6.24 [66157@9:30, 166@19:30] cloudy, drizzle later. Bath AM. Drive 55 Km but only part charge at 1500 W.
15th 149.65, 315.96 - 10.74 [66179@20:00] some sun. Finished car chj.
16th 151.01, 317.98 -   3.38 [66194@14:30 - oops I left BR heat on - BR heater was trying to heat house, 199@20:00] Clouds, bit of rain.
17th 154.74, 322.77 -   8.52 [66209@9:30, 230@19:30] clouds, bit of sun, bit of rain. Fast chj after 85 Km drv. Laundry x2. Hi energy day!
18th 157.35, 326.06 -   5.90 [66248@11:30, 261@20:30] more of the same. Fast chj after 55Km. Trailer heat on all day (250W = 6 KWH).
19th 161.05, 330.41 -   8.15 [66274@12:30, 282@20:30] Thin clouds/dull sun.
20th 163.46, 333.47 -   5.47 [66302@20:00] like last few days. 35Km drv.+slow chj. 3 to 8 PM.xxx
21st 165.48, 336.33 -   4.88 [66323@22:00] nuthr dul dae (and just why should the garden grow with no sun... and no rain? Thank goodness for the garden hose! But how about some real sun?)
22nd 170.91,342.67 - 11.77 [66335@10:00, 350@25:30] Sun! - rats, morning only. PM light clouds.
23rd 173.04, 344.75 -  4.21 [66362@13:00, 373@22:30] Heavy overcast - almost no solar. (power failure at night reset house solar meter)
24th    2.94, 349.21 -   7.40 [66401@20:00] cloudy with sunny breaks - 85Km, fast chj.car ~16 KWH? (power fail evening reset house solar meter)
25th    8.67, 356.18 - 15.64 [66427@22:00] Genuine SUN! bath, car chj.(on low 1500W) in sun...
Where on earth was all that grid power being used??? This time it looked really odd... the bedroom heat was for once off, I hardly did anything all day - I didn't even water the garden (and hence run the well pump)! The Miles Truck?!? I had plugged the truck charger in maybe 10(?) days ago. But why would it use any appreciable power for a float charge? I went into the garage and checked. No heat was on but the whole usually chilly garage was warm! Inside the truck's cargo space above the batteries it was even warmer. The "Delta-Q - Quik-Q" 1100 watt charger was blasting away, boiling off the water in the batteries - I could hear it boiling out! That was certainly what had been draining the house power. Day and night, using most or all of the solar plus the grid power. That's the same brand of charger that caught the Suzuki Swift on fire. I hate them more and more! And apparently I haven't learned my lesson about charging electric vehicles inside a garage. I hope it hasn't torched the batteries. It certainly can't have done them any good. I'll have to check and refill all the cells. And I plugged it in "just to keep the batteries up"! The only "good news" was that nothing caught fire! and the truck still moved. I didn't however drive it far inside the garage.
   I wonder when I would have thought about the truck again if I hadn't been recording power usage and been puzzled that it was so much higher than expected? There wasn't a day under 20 KWH from the 16th to the 25th, so perhaps it was the 16th when I plugged in the truck.

26th 16.08, 363.11  - 14.38 [66434@10:30, 437@22:00] Sun. Light chemtrails PM gone before evening. I set a 305W solar panel on the lawn and connected it to the grid tie with two the same on the house roof, total 11 panels instead of 10. With 1200W at the house and 830W from the trailer, at last it was exceeding 2000W. In the sun. In late May. And with the truck charger off, grid consumption was 10 KWH instead of 20+, in spite of doing a load of laundry with hot water and ~3 KWH going to the clothes dryer.
   Perhaps I'll put the panel as planned on the Sprint car and arrange it so I can plug it into the grid tie if the car doesn't itself need charging. I'll make it so it will swing up to at least 60°, and one can park and orient the car to get the best advantage - even moving it around during the day if desperate in December. Maybe I'll even arrange a few more panels that can be rotated or moved around by the house, for more in the morning, more in late afternoon, and especially, more in winter, the roof panels being at such a flat angle they lose up to 50% of what little sun there is in December.

27th 25.27, 370.13 - 16.21 [66442@20:00] Sun. With 11th panel makes new daily record. BR Heat, Bath AM. No drive. Use from grid just 5 KWH!
28th 31.75, 376.03 - 12.40 [66447@9:30; 451@21:00] Sun earlier AM then clouded over. Later 55 Km Drv. & part chj. @1500W. Had to disconnect 11th panel to close garage door - didn't hook it up, had 4 panels doing DC/water heater only for part of the day. Also on this day I split the 4 panels in the trailer between two "Y-Solar" 1000 W grid tie inverters. Trailer hit 899 watts in a short sunny break.
29th 41.14, 383.60 - 16.96 [66457@10:00, 66460@20:00] Sunny. Hooked up #11 today; Another record! Bath AM & finished charge car. The trailer was generating around 885 watts with the 2 inverters instead of maybe 830-850 with a single one. If we call it an extra 45 watts, that's about 5% more from more lightly loading the inverters. The 'extra' inverter was another ~100+$. That makes for a long payback time at 13¢/KWH - over 5 years.
30th 50.37, 390.74 - 16.37 [66466@21:00] sun w. a few kemtrails.
31st 53.10, 393.27 -   5.26 [66469@9:00, 485@20:30] Lt.Rain. (no BR heat) 85 Km drv. fast charged.
June 1st 56.94, 397.18 - 7.75 [66502@20:30] cloudy, occasional sunny spots in PM. 60 Km drv.Fast Chj.
2nd 68.56, 399.19 -    3.63 [66517@19:30] RAIN, more RAIN! (We REALLY needed rain!) 49 Km Drv.- chjd.after meter reading. The 4 older poly panels went to the DC system only, to keep kitchen hot water hot, so real total might be ~4.4 KWH (with a little power wasted from the 4 panels).

   From the 11th the DC system started using some real power - for the 36 volt, 250 watt kitchen under-sink water heater. I was often turning it off at night because of a bad battery (or "just because"), in which case it probably used .65 KWH per day. Otherwise it seemed to be around .8 to .9 KWH/Day, probably depending how much hot water I ran and how hot I set the thermostat. I didn't run very much hot water because it only took a couple of seconds to get it, and then I was usually just washing or rinsing a few things - a few more seconds running slowly. The total amount drawn would have hardly started the water on its long journey from the main water tank to the kitchen, so the new tank is doubtless saving energy. Where was I going with this? Oh yes: add .8 to the "total daily KWH generated" (total to the grid), especially from the 22nd on. (There's still power going to waste from the DC tied panels - depending which charge controller I have connected and whether their grid-tie inverter is on at all.)

   My power bill for March 13th to May 10th arrived. Little graphs on it showed that last year in the same time period I had used about 37 KWH/day, and this year only 20. Counting the times I must have been sending a little free power to the grid, it would seem the system generated around half the power I used. Unless I did more driving last year (probably). or it was colder (not that I recall). The answer to reducing the other half would be of course more solar panels and maybe a VAWT windplant... or to get the HE ray energy device working. (At the rate I'm going, will that ever happen?)

   For the month of May itself, the house meter says I've used just 17.5 KWH per day from the grid. Of course a lot of that is for charging the Nissan Leaf after driving, too... not to mention the rampaging Miles truck "Delta Q" charger.

   Something becoming quite clear is that the power used from the grid is less when the sun is shining. This is due to it being warmer (needing less heat - including letting the woodstove go out), as well as to the extra energy input directly from the solar panel system. The bedroom still used a bit of electric heat at night most nights to keep me at my comfort level. I used to say Victoria (BC) wasn't cold, but it had a 9 month heating season. This seems to be even more true up here. (And then finally May gave way to a chilly start of Juneuary.)

Low Light Performance

   It has been said that monocrystalline solar panels have better low-light performance than polycrystalline. I decided to try a few readings to see what happened. First I simply looked at the output to the grid from the house, with the DC charging turned off. That includes the four polycrystalline panels totaling about 1000 watts on one grid tie inverter, and two monocrystalline panels totalling 610 watts on the other. Then I turned off the inverter on the monocrystalline pair and read again, giving just the polycrystalline panels. I did it twice a couple of hours apart to get a better average. The monocrystalline is of course the difference, so there's no need to turn the others off and the monos on.

(Total - Poly only = Mono only)
11th - full sunlight at 11:30 PST: 1120 - 675 = 445. 445/675= 65.9%.
                               13:20 PST: 1190 - 703 = 487. 487/703 = 69.3%. (sun more straight on to panels)

So if the Hanwha "Q cell" monocrystallines were actually better in low light, then on a cloudy day their percentage contribution should be above 70% or so. Next: wait for a cloudy day. On the 12th there were instead high altitude "chemtrail" clouds.

12th full sun but chemtrails: 11:00: 898 - 405 = 493. 405/493 = 82% (turned off the four 1KW panels' inverter instead)
Oops DC is on, add 100 W to the four: 998 - 405 = 593. 405/593 = 68.3%
later on... 965 - 563 = 402. 402/563 = 71%

   Tentatively the monocrystalline panels seemed to perform slightly better than the polycrystalline in percentage terms when there was some cloud cover than in full sun. But the DC production from the polys wasn't consistently accounted for so this wasn't a very reliable test. There were lots of cloudy days afterward, but I got busy putting the solar water heater in and I forgot about doing the low light comparison readings... and then the water heater started taking away much of the polys' grid-tie output.

Beach-Launched Tide Power

   Notwithstanding that I decided not to attempt the 100 KW (or much smaller without looking for funding) floating tide power unit, when I drove to Masset to pick up my "air injection iron filter" water filter, on the way back I stopped right about where I would want to deploy such a unit in Masset Channel and took a look at the beach there. It was just what was wanted. A nice wide beach at low tide, flat, solid (not muck) and not too rocky, and the highway and power lines were just up from it - easy access. The channel was relatively narrow here, so flows would be faster. There was lots of room to lay everything out and then let the rising tide lift the vessel. Its self-steering rudder would then take it out into the channel where it sensed the fastest current. (And with all the other control stuff and tactics I mentioned in previous issues.)
   The anchor cable and the power line would come to the anchor which would remain on the lower part of the beach. The electric cable might be buried from there up to the power poles across the highway. The vessel would stay toward the near side of the channel, leaving lots of room for vessels and log booms to go around.

40 Km Long Masset Channel (AKA Masset Sound), where the tide
journeys back and forth from the ocean to empty and refill huge
Masset Inlet, virtually a small sea in the center of the island twice a day.
Looking southeast to east, at about 6-8 Km from the north end.

Looking north toward Masset and Old Masset with
the ocean just to the left in the farthest distance.

Off-Grid Solar Power Installation for a Neighbor

                      Plywood panel with charge controller,
                solar & battery breakers, CAT12V duplex socket
   As mentioned the end days of May were occupied with putting in solar electricity for an older neighbor who has lived without electricity for much of her adult life. I put two 90 watt panels on her roof, a "frame 27" size 'deep cycle' battery just outside, and a PWM charge controller on an inner wall. Inside near the battery went a 1500 watt inverter so she could vacuum. A duplex "CAT12V" receptacle will allow plugging in LED lights.

   But things weren't working right and I was troubleshooting into June. The charge controller was absolutely frying the battery. I had carefully grounded all the "-" terminals safely in a box.
   Investigation with a meter finally disclosed that the charge controller was "positive ground". To connect the solar panel minus side to the battery minus side was to bypass the controller and put the full power of the panels straight to the battery. Ditto with the switchable load output to shut the outlets off if the battery gets too low. Instead, it was all the "plus" sides that were tied together internally. Nothing was mentioned about this critical detail in the instructions.

   The last item I've even heard of having a positive ground was some British cars in the early 1960s, so to say this was a surprise was an understatement. I can understand why they might PWM modulate the solar's negative side and switch the load negative side instead of the positive side, as n-channel mosfets are better or cheaper than p-channel, but THEY MIGHT WARN THE PURCHASER! They include a card of several warnings, but that's not on it.

   The rewiring wasn't too bad. I had to get up on a ladder and change several connections in a junction box in the ceiling and on the main panel. Then I had to go back yet again after I later realized the load/outlet sockets negative was the same way. But I hadn't tied frame ground to solar minus in the solar panels, so I didn't have to go up and take one off the roof (fyooh!)

   But the battery seemed reduced in capacity if not fried by a few hours excessive charge that first morning - at least for now it would no longer run her heavy 1000 watt "shop vac" vacuum cleaner, instead dropping in voltage in a couple of seconds until the alarm came on and the inverter shut off or the 100 amp breaker blew. (I only put the breaker in as a manual shutoff in case the inverter was going up in smoke or otherwise needed disconnecting... you can actually trip a 100 amp breaker?!?) And it was all because of one vital piece of missing information that should have been supplied with the charge controller.

   I took over a lamp with a 12 volt bulb to plug into the DC system. (the first electric light besides flashlights that's ever been on in that house!) At this latitude the days are so long at this time of year she won't use it much, but she'll appreciate having electric light come fall and winter. After all, lights, cell phone charging and vacuum cleaning were the stated objectives. I didn't try to "upsell" her to a 36 volt system with four+ big solar panels and a hot water tank. But I did suggest that she go in for a couple on a bulk purchase of golf cart batteries that someone else intends to make soon.

Electricity Storage (Batteries)

Electrode "Pocket" with Nafion Ion-Selective Membrane Face

   Obviously I wasn't going to get anywhere until I had "glue" for the nafion. On the 10th I went down to the cellar and looked in a box to see if I had anything the least bit like the recommended "barium metasilicate" glue. It seemed I had barium sulfate and barium carbonate, from the pottery supply.
   Looking things up on Wikipedia it seemed both the sulfate and the carbonate were insoluble. It also seemed that barium in contact with glass would itself bond with the glass and "set" as barium metasilicate. I thought the carbonate was probably the better bet to "set". Would that work with ABS? It didn't sound very likely, but I decided to try it.

   (If I could get the nafion to glue solidly to an ABS plastic frame, I would also try Osmium doped cellophane film later.)

   On the 19th I made a little box with one face and 3 edges. The top stays open for inserting and removing electrodes at will, and will be kept above the level of the electrolyte in the cell so nothing gets around the nafion. I opened the nafion package and cut a little piece to glue on to cover the front. I was surprised that the nafion seemed like a pretty strong, thick sheet of flexible plastic. Somehow I had expected it to be thin and brittle.

    On the 20th I read up on barium sulfate and barium carbonate. Son of a gun - barium carbonate is used as a cement in mortar and concrete. On the 27th (still having had no time to look at it) I thought that if I put some in a test tube with some water... perhaps there was how to make some barium metasilicate.

   Instead of etching the zinc in ferric chloride I put it in a small puddle of hydrogen peroxide. The zinc wasn't very flat, and the peroxide only seemed to hit a few spots so it wasn't very effective. And then it evaporated. (The here unseen back side was however much better than the front.)

   Regardless of theory, on June 1st I (finally) put 4cc of barium carbonate into a test tube, added 3cc of water and stirred, making a stiff paste. I smeared this all around the edges and then slapped on the nafion. I put it face down on the flat countertop with a weight on it. Would it glue it on just like that? On June 2nd it had dried or set, and looked like it just might work. I'll find out... hopefully very soon.


From a Website   http://barium.atomistry.com/barium_silicates.html

Barium silicate occurs in nature under the form of various silico-aluminates, such as harmotome, brewsterite, and barytafelspar.

A barium orthosilicate, Ba2SiO4, melting at 1750° C., has been mentioned, but it is probably not stable in contact with water.

Barium metasilicate, BaSiO3, may be formed by fusing a mixture of barium chloride and carbonate with silica. It may be obtained as the hydrate, BaSiO3.H2O, by precipitation of a soluble barium salt with an alkali silicate. A hexahydrate has also been described, prepared by the action of barium hydroxide solution on sodium silicate or colloidal silica. The heat of reaction of dissolved barium hydroxide with colloidal silica is 8 Cal. It is probably decomposed by water, but the decomposition is prevented by the presence of barium hydroxide in solution. Barium silicate is formed on the walls of glass vessels containing barium hydroxide solution.

The anhydrous metasilicate forms small crystals showing a weak double refraction. The density is 4.435 and melting-point 1604° C. The heat of formation from barium oxide and silica is 26.3 Cal.

The metasilicate is not isomorphous with calcium or strontium metasilicates. With α-calcium metasilicate it forms a double compound, 2CaO.BaO.3SiO2.

By treatment of hydrated silicic acid, or powdered quartz, with barium hydroxide solution, or by boiling barium metasilicate with water for some time, a number of hydrated acid silicates may be obtained. Crystals of barium disilicate, BaSi2O5, have been observed to separate in optical glass in the form of hexagonal plates.

A compound, 2BaO.3SiO2, which forms a complete series of solid solutions with the disilicates is also known.

The use of barium silicate as a constituent of glass has already been mentioned. Its employment was first suggested by Dobereiner in 1829, and its effect upon the properties of the glass is much the same as that of lead. It gives a higher refractive index and greater brilliancy than lime alone, and it has the additional advantage over lead of being unaffected by the products of combustion in the furnace.

With barium aluminate a cement may be formed of composition 2(SiO2.2BaO).Al2O3.2BaO. It sets hard with water, but, after some time, it splits, becomes soft, and deposits crystals of barium hydroxide.

"Racking" NiMH D cell Tube Batteries

   It was great putting the NiMH "D" cells in sets of ten in "PVC 1-1/4 inch irrigation pipes" with ends drilled from 3/8" ABS with a 1-5/8" hole saw. All those unwieldy little batteries were made into a larger pipe assembly as a 10 amp-hour, 12 volt battery that was easy to handle. However the individual tube batteries were then unwieldy in groups - individual loose tubes connected only by wires, to be carefully picked up together to move and set down carefully on a flat surface. or occasionally taped together, but still not really a solid package. In the Mazda EV they had been set inside a wooden box. Now I had seen what I wanted to do from the Honda Hybrid batteries: make little racks that they would sit on, each row above another and then all bolted together, and that would make them into a 'solid' unit.
   Now I wanted to temporarily swap out the bottom box battery to check it and fix any loose connections or bad cells and it was awkward to replace it with a bunch of loose pipes. And essentially that was what was missing to assemble and connect the Honda batteries into the system, too.

   I set about making some "racks" on the 16th. To make holes for the tubes in plywood I had too sizes of hole drills: too small and too large. I finally decided to do them on the 3D printer in plastic. Of course then they were just right and no splinters. I was glad I did them that way. The only problem was that each little rack took almost an hour to print and I needed 18 of them - six for each 'rack' of six batteries. The printer was doing it, not me. But I had to be there the moment each print was finished to turn the bed heat back on (it turns it off when any print is finished), or else I had to wait around for another 1/3 hour for it to warm up again. It cooled much faster than it heated, and it never quite reached the set temperature. (It worked fine for printing PLA plastic, where the bed temperature was 55°C. For ABS the set temperature was 110°C and the prints slid off the glass below about 96.)
   On the afternoon of the 17th I started the printing again. I had intended to just print a few. But somehow I was always there just before or just as it shut off, or within a minute or two, so each time I started the next print. Before midnight all 12 were done.

   On the 18th I shrank the dimensions a little and did 8 "interior" rack pieces for the Honda batteries. I couldn't do them wide enough so the interior racks were in pairs. Their shapes held the rows in place, and the entire 36 volt, 6.5*6 amp-hour battery was very heavy, so for the outer strips I decided to simply use wood - 1"x2"s for the bottom were not "overkill" for strength. But I didn't get any farther than printing the racks in May.

Haida Gwaii, BC Canada