Messages

16

Random / Re: Wacky PV stuff

« on: March 04, 2024, 08:41:10 PM »

I doubt it's going to make a huge difference though we live at the end of a dirt road and every morning our cars have a nice layer of dust and pollen. It's winter, so I expect variation even this close to the equator. Depending on what Maui Electric Company has up its sleeve I might add another row of PV and some batteries. I no longer have to worry about outages since I bought a little portable power EcoFlow thingy. In my experience, the day you add backup power of any sort is the last time you ever have an outage. But I'll add to the system and do a whole-house battery if it pencils out. This system paid for itself even though I had some prolonged outages due to microinverter problems. The lifetime generation for the system is 95.43MWh. At 43.31 cents per KWh that's $40,080, which is about what the system cost unadjusted for inflation. As a guesstimate, if I'd fixed the microinverter issue promptly it would be roughly double that. 40K for being a lazy buggah.

17

Random / Wacky PV stuff

« on: March 04, 2024, 09:59:00 AM »

It seems like everything I know needs to get upgraded every few years. For decades the gold standard for PV was installation on a slanted roof facing south, or if you had a flat roof, then racks, carefully angled to catch midday sun at a perpendicular angle. Then east/west roofs became popular. If you look at the output curve of a properly angled south-facing panel it's a single hump, with a maximum at midday tailing off to not much at all in the morning and evening. Here's what the panels at Ponohouse produce. I've never washed those buggers, I need to:

The sun angle at anything past noon is not optimal, so you don't get continuous generation. Your wiring, controllers, battery charger, inverters, etc. need to be sized for the peak, which lasts less than an hour. Conversely, panels on an east-west orientation have humps at morning and evening, depending on which side they are mounted. When panels cost bazillions, they were the expensive part of the system. Now they cost less per watt than an English muffin, so flattening out the curve makes sense. They benefit from being mounted more steeply to catch the low-angle sun, but it works like gangbusters and you don't need to upgrade the rest of the system.

Cool. But now we have Bifacial panels that are dirt cheap. They benefit from being mounted freestanding in a vertical position east/west so one panel does the work of two. They generate almost nothing at midday but do well for the rest of the day. Shading is tricky, but then it always is. They work best in a mixed installation where there are a more or less equal number of south-facing panels, depending on latitude. Shortly there will be panels optimized for vertical east/west installation that also take advantage of lower reflection, but current bifacial panels work fine. Current panels peak out at about 22% of the theoretical solar energy available regardless of installation because most of the sunlight gets reflected.

There's more. Current solar plus battery tech has crept up from 6V to 12V to 48V, driven mostly by the wire size necessary to deliver significant wattage and power expectations. Now it makes sense to use something like 400 V (the voltage of most EV batteries) because it's easy to get with ten or so PV panels in series and the wire size required is modest (current determines wire size, 1000 watts at 12 volts is 83 amps, meaning you need wire thicker than the battery cable in your car, at 400 V it's 2.5 amps). So basically, everything I learned from fiddling with Solar over the last few decades is now wrong. ....very weak Yay!

18

Prone Foiling, Surf foiling, Pump Foiling / Re: Swim Missiles

« on: March 03, 2024, 01:43:39 PM »

You were ahead of your time. Yesterday's downwind foil Maliko race was solid Kalama-style boards, AKA swim missiles.

19

Random / Re: Vision Pro

« on: March 03, 2024, 01:37:26 PM »

Or if you're just interested in something roadworthy we could just get the beast running. Not a great idea though, drivers do thoughtless enough things to bikes they can see. Even with the big flag I put on my eTrike I had people turning in front of me daily.

20

Random / Re: Vision Pro

« on: March 03, 2024, 09:37:06 AM »

That's pretty cool. You might recall I built an e-trike a long while ago after I had meniscus surgery and the doc said I should ride a stationary bike. He thought if I rode outside I'd be likely to fall and ruin his work. So I built a trike. A tadpole design, like yours, but foot and e-powered. No suspension, but as usual, way too much motor and battery (More Power!!). I did a century ride over Mt. Hood with it, and collected 4 warnings for speeding in Hood River. Two from the same cop, who was losing patience with me. That thing looks like fun. It also looks like it could use a bit more power and perhaps range. We can fix that if you're so inclined. I can't seem to find the old video of me hitting 48mph at Rowena Crest, but you probably remember it. Mine's gathering dust at the shop. I was probably gonna kill myself on it, and my knee healed. I've got a spare Alfine 8-speed internal hub to replace that silly derailleur or the Nu Vinci variable speed hub that's on my trike now. The Nu Vinci is gonna be toast if I keep using it on my trike. The trike is 1500 watts and the nuVinci is rated for something like 500, though I read somewhere they uprated them to 750 without making physical changes. I've also got a two-speed pedal bracket, it looks like that's a single-speed for the hand crank. I don't mind parting out my trike a bit, I doubt it will ever see daylight again.

21

Random / Re: Vision Pro

« on: March 02, 2024, 11:26:24 AM »

Back to the original topic (sorry admin), it looks like the backlash of Apple killing the EV and the substantial number of Vision Pros being returned before the return deadline is pounding Apple's stock a bit. I'm waiting for V2 or 3. Admin convinced me some time ago to spring for the top-of-the-line Meta goggles, which are gathering dust at my shop. That isn't a dig at Admin or Meta (well, maybe a little) but just a taste of living with rampant ADHD. I'm striving mightily to resist the urge to chase the new, new, and the bright and sparkly (oh look, a butterfly!!). I'm also certainly going to cancel my Cybertruck order in favor of my new, new PV and storage battery project. Diane is cautiously pleased--mostly because she suspects I'll wind up doing both.

My real motivation for ditching the Cybertruck reservation is I'm a little tired of getting V1 and seeing V2 is way better, and I'm very happy with my Model Y, which does everything I need, including pulling trailers when I have a load of junk for the dump, building materials to carry, or want to do a remote dirt bike adventure. I also really like camping in it, I doubt camping in the bed of the cybertruck will be nicer than lounging in the back of the Model Y with streaming video and climate control on tap, and the tent I bought that erects over the back to provide covered standing room.

And I confess, I like Admin's Rivian better.

22

Random / Re: Vision Pro

« on: March 02, 2024, 10:36:27 AM »

There was a report on the news in France about solar panels and the huge increase in them recently. As Pono said they are producing a lot more than demand at the moment. The problem is a lot of them are very poor quality and end up in the rubbish pretty quickly. What's going to happen to these fields of solar panels that are popping up everywhere in 10 years time?
 I am doing some work on our house soon and have found a French company that makes solar panels that heat water as well. The heat absorbed to heat the water makes the panels a lot more efficient.

That's been a good idea, and kind of obvious, for a long time. I thought about doing it myself as a modification. It seems once an idea takes hold in opposition to something so simple that it just dies. The problem with solar hot water is that it's plumbing, which solar installers don't do, and the efficiency and utility of the early systems (we have one at Ponohouse) is marginal. The original stuff used mechanical timers and a few temperature sensors to regulate operation. The timers weren't reliable, and too often the system would pump warm water to the collectors in the dark, which cools the water. Not ideal. They use electric water heaters as the backup, but the heaters are disabled during daylight hours because otherwise the water is already heated and the temperature regulators shut off the collectors.That can lead to a lot of cold showers--I can attest to that firsthand. There are other problems, but you get the drift. Of course, any reasonable micro-controlled system instead of stupid timers could overcome the issue, but by then solar installers were convinced that a few more (now inexpensive) panels to power a conventional heater made more sense than all that plumbing. My solar hot water system is now abandoned in place, as is the huge solar pool heater we installed a few years ago that suffered from the same issues. If I wasn't such a lazy twat I'd make an intelligent controller for the swimming pool heater at least. Instead, I don't use the pool (I guess I could use it as a cold plunge) and spend my time in the 80-degree ocean.

Worse yet, bifacial solar panels in ground mounts deliver power from the front and back from reflection--those designs are the darlings of the industry, sparking new approaches like vertical mounting in east-west orientation (I know, sounds stupid, but it works).

I've heard the rap about junky Chinese panels, it's like the rap about everything Chinese being substandard. It's propaganda (I'm old enough to remember "Cheap Japanese Junk"). I bought a pallet of Chinese panels before the tariffs made them expensive, and they are beautifully made. They produce more power than they should and are clearly well-constructed--they look better than the Japanese, Canadian, and US-made panels I've experimented with. I have two of them on a container for testing and six on top of my stupid RV project, where they have been performing faultlessly for years. I expect they'll be fine twenty years from now. The lifespan of solar panels is a wacky issue. No one really knows what it will be. The big commercial installations replace their panels frequently--not because they aren't working just fine, but because better and cheaper panels are available every year. If you have 10,000 250-watt panels in a huge field and you can double your system output by replacing them in the same mounts and the same wiring with 500-watt panels for a third of what the 250-watt panels originally cost what would you do? There's a thriving market for used 250w panels in the US. I can buy them for $30 each from a vendor (Santan Solar) or sometimes more or less free if you connect to the source. When 750 watt panels come out I expect to see lots of bifacial 500 watt panels for 30 bucks. Even at 208 bucks for 550 watt bifacial, PV is now $.37 per watt. Back when we could build 1000 megawatt nuke plants for 1.5 billion (we can't anymore) that was more than a buck per watt. Yes, it could run all night and only got refueled once a year (for millions of dollars, probably billions today), but it's a lot more than $.37. And we thought nuclear power would be too cheap to meter.

23

Random / Re: Vision Pro

« on: March 01, 2024, 05:58:50 PM »

I expect Oil is going to get more important, and more expensive as we quit burning it, though if geological hydrogen doesn't turn out to be the next cold fusion, the use of oil to make ammonia will collapse faster than coal for electricity--it's super easy and much cheaper to make ammonia (NH3) from hydrogen. You could do it in your kitchen (I don't recommend it), still using the Haber-Bosch high-temperature method without the petrochemical middleman.

When I commented that oil is dead, it's as a source of electrical generation. It gets lumped with "other" now. The fact that spurred my WTF moment is the current price for 550 watt bifacial PV panels--I can buy them for 208 bucks in modest quantity. When I bought 30 Sanyo 250W panels for my house in Maui 12 years ago I paid $40K for them--in 2010 dollars. Roughly three times the output (depending on how cleverly I mount them) for 15% of the price.

And a minor correction: When you consider the energy content of a barrel of oil you need to at least allow for Carnot efficiency, and Carnot efficiency is the theoretical best case for a heat engine. The actual efficiency, disregarding energy required to obtain, transport, and refine oil, is about 28% for anything other than direct heating. It should be feasible to calculate the total energy value of a barrel of oil, but it would depend on how and where it was used. In any sort of actual engine, it would probably be about 20% of your man-hour number. Still impressive, oil has been very, very good to us.

24

Random / Re: Vision Pro

« on: March 01, 2024, 09:16:49 AM »

Oil is dead, coal is dying fast (could it be any faster??), Nuclear is flatlined and just got surpassed by renewables, the only growth is natural gas and renewables, and natural gas looks like it might be a spike because the cost of even combined cycle gas turbines is rising while PV cost is plummeting. How weird is this?? Who knew?? All the dipshits saying "your EV is powered by coal" are talking out of their ass even more than I thought. This isn't government mandate-driven, it's cost-driven. And it's not cheap Chinese PV because those are tariffed out. Though it's a big deal on a global scale because china produced 50 percent more PV this year than the market wants. Africa and any part of the third world that China wants to influence is gonna get paved with panels. We're going to start being able to see PV installations from space.

25

Random / Re: Vision Pro

« on: March 01, 2024, 12:25:25 AM »

The apple car will be absolutely amazing IF it ever arrives.  That's the thing.  Apple would never launch it unless it is - that's core to the engineering culture there.  They've been screwing around with the project for 10 years now and I bet we'll never see it.  But if we do, I'm confident it will be jaw-dropping

Apple killed its EV project. I suspect they realized, as I recently did, that there are good reasons to expect small, sub 200-mile range EV cars to be extremely cheap soon. It's going to be extremely hard for late movers to make money, and even Tesla might actually be fucked depending on how the politics shake out of keeping Asian cars (like BYD) out of the US and Europe. It's also getting extremely cheap to do alternate energy projects. Fairly weird for a nuclear geek like me to contemplate. The weirdest part is that there are good reasons to build large PV arrays with small-ish storage batteries. Completely counterintuitive, but the math is clear. I've been planning the opposite project for my shop--150KW of battery (two Tesla Model Y salvaged batteries) and 30KW of panels, but when I did the calculations I'm better off with 30 to 70KW of battery and 80 KW (or more) of PV. Even in the cloudy Pacific Northwest it works out.

All this stuff happened while I was distracted with other stuff--and I pay attention to this kind of thing. While no one was watching the US became energy independent and renewable energy went from negligible to exponential growth. WTF??

26

SUP General / Re: Broke the Leverlock paddle's handle. Retire or replace part?

« on: February 26, 2024, 01:21:01 PM »

Any hollow shaft, including a glued one, can take on water. The paddle heats up in the sun, increasing pressure inside so the air escapes from even a pinhole, and then you stick it in water which cools the gas inside and creates a low pressure that sucks water in. You have to periodically empty it unless it's been perfectly glued. One of the many reasons we used hot glue to assemble Ke Nalu paddles. Unfortunately, the parts need to be designed for hot gluing for it to work optimally, but even if it's not perfect, it's better than epoxy.

27

SUP General / Re: Broke the Leverlock paddle's handle. Retire or replace part?

« on: February 26, 2024, 09:19:03 AM »

I don't know what the diameter of the leverlock shaft is, but it looks thick--you might be able to epoxy a replacement handle in place--the ones on Amazon look OK. You'd need to take the adjustment stuff out of the remainder of the extension shaft. If you want to retain the adjustability you could hot glue it in place and adjust with a heat gun. There's a lot of crappy-looking paddles for sale on Amazon. I'm glad to be out of that business. The market for premium paddles has to be grim now.

28

Random / Re: GPT 5, Sam Altman, What's coming

« on: February 25, 2024, 07:36:36 PM »

The investment in building and processing ever larger and more fine-grained language models is going to get crazy fast. ROI is hard to find, investment dollars are not. The biggest opportunity is fraud--companies pretending to build something new, taking the money, and "pivoting". And here I thought only crypto exchanges could effectively pivot into things that make even less sense. The biggest opportunity in Crypto used to be the greater fool scenario, but they seem to have briefly either run out of fools or they just aren't great enough. Is that feasible for AI?

29

Random / Re: GPT 5, Sam Altman, What's coming

« on: February 22, 2024, 07:34:01 PM »

My brother has just started a business in Australia doing VR training and using AI for simulations etc. I don't really understand it to be honest. He has done some presentations to pretty big companies. They quite often just look at him with a blank look at the end of it. " Is that actually possible"? Half of the people in the room then realise they just became redundant and fight pretty hard against it.
 It's definitely here and developing at a rate no one expected. Because it's AI!

They're right. Any senior exec meeting being shown a presentation of what AI can do--including most CEOs, are staring into the abyss.

30

Random / Re: GPT 5, Sam Altman, What's coming

« on: February 22, 2024, 07:32:14 PM »

There are zillion Frankie AI covers. The ones that have dynamic range are funny--they mirror the way Frank used to drop an octave or two to handle high notes. The lady gaga stuff is unmanageable.

I want to hear Ima Sumac do something by Mariah Carey.

and of course now that I've looked at the available tools it's trivial to do that. We're SO fucked.