Both of you.
It seems silly to solicit feedback from a source where I know nobody goes… While my target demographic is exceptionally narrow, it’s also the only people who have an opinion I consider valuable or useful.
I’m trying to design an off-grid modification for a chest freezer, to make it into a refrigerator. A super-efficient, top-opening refrigerator that works way better than a typical dumbass refrigerator on a ton less juice.
You might think this is a really easy thing to do. All you gotta do is change the thermostat to 33 or 34 degrees, right?
No, not at all, here’s why:
The power inverter burns juice when it’s on, supplying power to other stuff. Usually it eats about 400 Watts on it’s own. That’s not a small thing in an off-grid system.
What’s that got to do with the freezer then, eh?
Duh, when the freezer is just sitting there, it’s using a tiny trickle of power, but it keeps the inverter from going into sleep mode. So, even though the freezer is only pulling microamps, it keeps the inverter powered up and sucking 400 Watts. Even when the freezer compressor turns on, it burns less than 200 watts as it pushes phase change…
So, my trick is to make a battery powered microcontroller with a thermal sensor. Stuff the thermal sensor up the drain hole, since this won’t be a freezer anymore and there won’t be ice. Still, spills, but whatever. Using the drain hole for sensor access. This part is no problem. PICAXE + Stainless encapsulated DS18B20 on a waterproof wire. The mass of the stainless alone helps attenuate hysteresis, so I can get pretty tight on the temp range I choose. A relay physically disconnects mains when target temp is reached. The system then pulls micro power as it checks the temp every 10 seconds until high temp triggers, pulls in the relay, and two things happen. First, the obvious running of the compressor. Two, this now connects the battery charger and power is replenished for the time it was running unplugged. I also throw in a bit of a safety measure in the form of a 2 minute time delay for heat soak/pressure surge. Don’t want it kicking off and then right back on again. Since I’m dodging the factory countermeasures, need to implant my own.
Well, there’s this thing called inductive start… The compressor, while running, is remarkably efficient. But at the moment this thing trips the relay, the contacts have to endure a huge inductive surge, even though operating current is quite low. It’s a worst case scenario for a mechanical relay, and it happens every single time… The relay will die a very premature, glorious, flaming death.
Then you say, well, of course, just use a Solid State Relay. It’s what they use on the huge compressors in your air conditioner unit, and that’s just a bigger version of exactly the same thing, so why the hell not use this obvious solution?
Because of the very same problem I mentioned at the top. Solid State Relays have current leakage. Even when off, a teensy bit still flows, and this would keep the inverter in a constant on state sucking 400 Watts, defeating the whole purpose. It’s like the AC version of a MOSFET. You never really get absolute 0 current.
So, instead of either/or, why not put the SSR in series with a standard mechanical relay? Have the PICAXE pop the mechanical relay on first. Which does nothing except close the contacts. THEN pop the SSR. That way, the mechanical relay can make the full circuit disconnect to avoid the leakage current, and the SSR doesn’t allow the inductive punch to land until AFTER the contacts are already solid, thus protecting the mechanical relay from an early demise. Shut down in reverse for same reasons.
Could even use a much cheaper mechanical relay since it won’t be taking so much abuse. Already sourced a 25A SSR (way overkill to avoid a need for heatsinking) for under $5. Have a handy 10A relay board for under $3. This is the Golden Age of tinkering with electronics…
Sound like a plan?