I hope there was space. :)

I once had to replace an alternator on a Peugeot 308, ended up cursing and got a cold (was working in a snow storm as the car was needed next day) because it was impossible to fit my hands where I had to work, so everything was super slow.

About ten years ago, when I built my house, I decided that solar power would suffice. It did not - got a grid connection last year - but solar power is giving me the biggest part, so my grid connection is a laughable 3 x 6 amps. :)

Around 2009 or so, a fellow anarchist linked an ecological footprint calculator in a forum. I ran it, and found out that 60% of my footprint came from driving a lot (I don’t fly, except for flying drones), and most of that from buying fuel. So I built an electric moped-car. But I built it wrong and had to dismantle it, sell the metal for scrap and build another. It lasted a few years, then I was forced to switch to a temporary gas-powered microvan. When the microvan broke down, I spotted an auction: someone was selling an electric microcar which the social ministry had crashed. So I got it, managed to fix it, and finally I could produce my own fuel for driving. The road here has obviously been steep and littered with failed projects, but now I can even support other folks who drive that kind of stupidly made microcars, while I bide my time to build a proper electric moped-car.

I am working on the part of communications - not being a passive consumer but giving some and getting some in return. Found out that some folks have started experimenting with MeshCore and MeshTastic in my town, so I intend to join those nets and see if they’re usable for chatting with a friend.

I am also working on other kinds of communications (agile ultra wideband frequency hopping with SDR) but sadly, this is less solarpunk and more like miltech.

Dehydration can be done with vacuum, perhaps?

The neat thing about them:

First, the pair of habitats can be rolled by operating the cylinders as momentum wheels. If one habitat’s rotation is slightly off, the two cylinders will rotate about each other. Once the plane formed by the two axes of rotation is perpendicular in the roll axis to the orbit, then the pair of cylinders can be yawed to aim at the Sun by exerting a force between the two sunward bearings. Pushing the cylinders away from each other will cause both cylinders to gyroscopically precess, and the system will yaw in one direction, while pushing them towards each other will cause yaw in the other direction. The counter-rotating habitats have no net gyroscopic effect, and so this slight precession can continue throughout the habitat’s orbit, keeping it aimed at the Sun. This is a novel application of control moment gyroscopes.

From Wikipedia: https://en.wikipedia.org/wiki/O'Neill_cylinder

This was a interesting read. I didn’t even know that electrocaloric materials existed. :)

Having read the explanation of how it works, I think the biggest difficulty waiting for them on the curve is achieving considerable levels of power.

But they seem to firmly believe that they can get past an instrument cooler to a residential heat pump.

The news: they skipped the vanadium. It’s ferrocyanide / ferrogluconate based.

But whoever mentioned the 6000 cycle figure should kindly quote their source, because I can’t find it.

Herein, we propose a highly stable alkaline all-iron flow battery for LDES by pairing the [Fe(CN)6]3−/[Fe(CN)6]4− redox couple with the ferric/ferrous-gluconate (Gluc−) complexes redox couple, which exhibits high solubility (1.2 mol L−1), fast redox kinetics and high stability in alkaline media. The high stability of iron-gluconate complexes resulted from the stable six-coordinated iron species, enabling a stable alkaline all-iron flow battery, which can stably run for 950 cycles at a current density of 80 mA cm−2.

For anyone with an interest in chemistry, I recommend the scientific paper.

https://www.science.org/doi/10.1126/science.aec6413

Trying to think of critiques:

• the wavelength of light necessary to “charge up” pyrimidone is fairly short (300 nm, UVB ultraviolet light)

• the quantity of UVB light on Earth’s surface is limited (it is mostly absorbed by the ozone layer)

• however, one can artificially produce ultraviolet light from solar electrical power, or figure out molecules that charge with UVA or even blue light, which would be perfect

Positive aspects:

• pyrimidone looks simple, synthesis probably is not hard
• it is solid at room temperature
• it lasts long when charged
• it is compatible with a water environment when energy needs releasing

Addition of hydrochloric acid (HCl) to Dewar pyrimidone (107 mg in 0.46 ml of water) increased the solution temperature to 100°C and induced boiling within 1 s, demonstrating rapid macroscopic heat transfer to an environmentally benign medium under ambient conditions.

Subsequently, the solution can be neutralized with an alkaline chemical, pyrimidone can be “recharged” and the cycle repeated. The summary of the article does not mention how many cycles it endures. I would be good to know that.

Not well, but it drives.

Battery capacity is small. It’s a city car all the way.

Heating is abysmal. I don’t touch it, it drains the battery. I only heat the seats (from the car 12 V system) and windscreen (with a large drone battery and two Chinese 400 W heat blowers).

Different sizes of tyres of front and back wheels are impractical. Changing headlight bulbs is a nightmare (manual says to take off the front bumper, but I deviate from the procedure and leave some screws missing, so I can take out the headlights).

The gear shifter has 2 needless pull cables (not electrical cables) which freeze in winter and cannot be bypassed easily. The motor controller borked itself (high voltage comparator error) and I had to take it apart to fix (fortunately a fix was documented).

One of the steel brake pipes rusted and leaked, and the repair shop refused to lower the battery (I have done it myself) because they didn’t feel comfortable. I had to bypass the steel pipe with a copper pipe, fortunately technical inspection did not notice.

Rear ABS sensors go faulty and start lying, producing error messages. An “original” spare part costs 200 euros, fortunately there’s a trick (installing another car’s sensor in reverse) and it costs 17 euros.

But what I can I ask, it’s a 15 year old car.

I’m in Europe, so we have 230 volts here (two times less amps needed for identical power), but… I charge my i-MIEV with 5.5 amps at night (the whole night) and it’s charged by the morning. Knowing that, I optimized my grid connection down to 3 x 6 amps (three phases, each up to 6 amps).

Of course, if I charge during daytime, I can draw power from the house inverter, so then I charge at 10…13 amps.

These guys sell those chargers for 15 euros. Not sure if the chargers work, but the plugs look compatible and they have positive reviews.

https://www.aliexpress.com/item/1005008383594311.html

To manage without an inverter and charge your bike almost directly from solar (soldering and electronics skills needed) I would consider an adjustable DC/DC converter.

1. Small step-down converter with current limit:

Input voltage range: 6-40.00V
Output voltage range: 0V-32.00V
Output current: 0-5.1A
Output power range: 0-160W

https://www.aliexpress.com/item/1005006294300036.html

2. Big step-down converter with current limit:

Input voltage: DC12-80V (can not be used for 72V battery)
Output voltage: 2.5V-50V adjustable (only buck) (maximum output voltage = input voltage * 0.8)
Output current: 20A (MAX) 25A overcurrent
Maximum power: 600W
Conversion efficiency: ≥ 93%

https://www.aliexpress.com/item/1005007391865816.html

3. With serious and silent cooling, dual displays (voltage and current), adaptive current limiting that considers temperature, input up to 120 V DC (be careful, this voltage level is deadly):

Input: DC 25-120V
Output: DC 1.2-75V adjustable
Current: 20A (MAX)
Power: 600W (Reduce current based on heat dissipation when voltage difference is high)

https://www.aliexpress.com/item/1005010659062631.html

The downside of a simple converter is that it doesn’t know about the optimal loading point of your solar panel (doesn’t do MPPT or maximum power point tracking). If you want that (it gives 30% more productivity), consider an entry level solar charge controller, but it must support 48V battery mode (I notice that your charger outputs 63V, the termination voltage of a 48V battery is usually 64V) and manual fine adjustment of the termination voltage (to bring it lower).

This might work since it appears to have a “user” (user defined) battery type:

https://www.aliexpress.com/item/1005010080097662.html

Manual here to find out if it actually would:

https://www.nikom.biz/pic_info/A656017/Homysun-MPPT-Solar-Charge-Controller-datasheet-60A.pdf

Note: dial in less than your full battery voltage and don’t push the Chinese products to their maximum advertised current. Your upmost output voltage + voltage ripple + measurement error must stay below the full battery voltage. Of course, you must have a balancer and BMS (to stop everything in an emergency) on your battery. I also recommend a timer.

If you need to step the panel voltage upwards, consider a “boost converter” or “step up converter” instead, or wire two panels in series and use a step down converter.

If you don’t mind divulging, how does that stack against timber for cost at least for the time?

It’s too expensive but practically indestructible, except in salt water. Unlike wood, it does not offer any favourable thermal properties, so you can only do unheated sheds with it. I came across a pile of it on a bankruptsy auction, which made it possible to afford its use.

I recommend to find some assistance. Even one other person can help immensely at certain points.

I have built a 2-storey shed on a metal frame alone, and it involved highly inconvenient and more than moderately dangerous stages. I would not do it again.

I managed only because I relied heavily on industrial aluminum profiles for machine building (engineer’s Lego beams).

Apparently yes, and they’re a bit late, as the Ecoflow & clones train has already left the station.

If you ask a random Ukrainian drone pilot “how can I support you”, one of the more likely answers would be “help our unit get another EcoFlow, or a good clone from the short list”.

In their bunkers and base stations, they need lots of portable power without a thermal signature. Best if silent too, but most imporantly, it has to be cool. Generators are kept far at the back in really safe places, because they shine quite brightly in thermal vision. They also need oxygen and output poisonous exhaust, so you can’t take them to the same bunker as people.

All I can say is that I read the stories, enjoyed them, and gave them away in a freeshop for others to read. :)

Heyva Sor is real and if you can, I am sure they welcome any help. :)

https://en.wikipedia.org/wiki/Kurdish_Red_Crescent

I have a few solar panels installed in 2008.

They are mounted in an extremely bad location: on a south facing wall, without proper airflow on the back side.

They continue to work, even if output is degraded. Newer planels installed in the same location overheated and their elements cracked, indicating inferior manufacturing quality, but the oldest batch is not showing this symptom.

If you don’t mind buying from Guangdong and being hit with import and customs fees, then on Linux, this gadget presents a standard V4L (Video4Linux) USB camera.

https://www.aliexpress.com/item/1005007938478574.html

The cable is delicate, you’d have to find a way to secure it against damage (e.g. tape). For the 256 x 192 pixel sensor, focal length 4 mm will get you an ordinary field of view (e.g. 45 degrees), while 9 mm will get you a very narrow field of view (e.g. 20 degrees). Grayscale image, very fast frame rates, automatic calibration against a shutter (occasional sub-second stops in video).

Note: never ever point at the Sun. It has poor solar protection, I almost ruined it when my code erred and scanned across the Sun, leaving a massive black track which took several hours to fade. The shutter closes when it sees the Sun, but not fast enough.

Coincidentally, it is cold here too. Walked around the house with a USB drone camera. Window edges are radiating like crazy. There was no difference between a window with shades down vs. a window with shades up (I hoped to find a difference, but was disappointed). However, edges need considerable extra foam.

What makes me worry is the size of the “reflector”. I wonder what symbol rate it is possible to get if the “reflector” is literally everywhere above you - the time delay between start of reflection and end of reflection might be considerable.

I found a resource with practical advise about using NVIS, including using meteorological data (ionograms) to determine the best frequency, antenna diagrams and such. (His drawings are in Spanish, but I think most English-speakers can decode Spanish antenna diagrams with a few clues, since meter is metro, and frequency is frecuencia.)

https://www.ipellejero.es/hf/NVIS/english.php

Some additional interesting antenna designs (vertical yagi on ropes, spiral antenna, etc) and propagation shemes, can be found in this research paper.

They have a passage which addresses data rates, and the data rates are poor…

In [110] two alternative channel access technologies are tested and compared in wireless mesh networks: Automatic Link Establishment (ALE) and fixed-frequency MAC protocols. These are also described in the book of Johnson et al. [111]. In [112] we find measurements on a MIL-STD-188-110C [113] link over distances of up to 160 km, providing the users with bit rates up to 9.6 kbps in 6–9 kHz RF bandwidth. The standard gives the option of selecting bandwidth and modulation, and includes wideband modulation with up to 24 kHz bandwidth with elaborate coding and interleaving.

So, some military radios use a narrow slice of 6…9 KHz for each channel and push 9 kilobits per second. Widening the slice used will give more speed. Enough to send SMS, I would say. A text of 1000 characters could be 8 kilobits, and would require initial air time for 1 second (+ exponential back-off resends, most likely). Let’s assume a total of 3 transmission events per message. An hour contains 3600 seconds, divide by 3 --> a channel at 100% capacity could accommodate 1200 messages per hour.