Hacker News new | ask | show | jobs
by a-dub 3 days ago
that's a fun project. i'd be a little concerned about the servos failing to fully switch the switches or placing wear on them, maybe a dangerous failure mode where it is switched into and left in an in-between state unattended that could lead to a fire hazard.

you can buy smart outlets with the correct connector, just look for 20A versions. (honeywell, at a minimum, sells one).

https://prod-edam.honeywell.com/content/dam/honeywell-edam/h...

1 comments

The lack of feedback from the AC unit makes using a switched outlet potentially problematic. You really need to know the state of the compressor cycling before you go mucking with turning it on and off. (At least this was the case with the refrigeration compressors I am familiar with from my youth.) A current sensor could provide feedback but that goes beyond a regular smart switch.
Naw. It's not quite that weird. It's actually pretty simple. :)

The compressor can be turned off at any time. It doesn't matter where in the compression cycle it is when it becomes unpowered; it just doesn't care. It's probably ultimately going to come to rest on a compression stroke no matter what, anyway (Newton's 3rd law, etc).

While the compressor is off, the high pressure side leaks down. That's good. (Sometimes, that's a noisy process: We've probably all heard aircons and fridges and such hiss and/or gurgle a bit after they've been turned completely off in an otherwise-quiet room; that's the sound of the high and low pressure sides approaching equilibrium.)

The only trick is to wait a bit for the pressure to leak down before restarting it again; a minute or two is almost certainly more than plenty with a small loop. That wait is just to make the compressor's job of restarting a bit easier than it could be otherwise.

The risk of restarting too soon is that it's harder on things than it could be. It might be hard enough on things that the compressor stalls. Stalled compressors are bad because they use a lot of current and make a lot of heat, but there's safety cutouts present that are designed to prevent them from becoming fire just for these instances.

And anyway, it's all completely avoidable in this scenario: Set the smart plug to default to off (for when recovering from a real power outage), program a reliable restart delay into whatever the control system is (Home Assistant or whatever), and call it a day. This is functionally exactly what my Nest thermostat does.

(Some aircons already have a delay at fresh power-up as an unavoidable built-in feature, and these don't need any special handling at all.)

i don't think that's true. i still have a thinkeco ac modlet that coned sent me years ago as part of some kind of "make your ac smart" promotion.

if you turn those knobs to the off position while the compressor is running, does it actually go into some kind of orderly shutdown or does it abruptly cut power to the compressor?

The wiring diagram for this article seems to reflect it would be a hard shutdown. Most HVAC systems I have interacted with have zero problems with a sudden power loss. The only potential issue is bringing the compressor back up too quickly on the other side. You could easily control for this if you were switching the whole unit using software.
I don't think its the shutdown that's the problem but the restart. If I fiddle with the thermostat on my fridge, it turns off in one direction, but if I turn it all the opposite way it won't come back on for a few minutes. I think this is the relevant interlock. Something about letting pressure equalize.
High head pressure on startup is the problem I'm thinking of. Generally there is a delay after shutdown and a lockout on restart until the head pressure is lower. In the old refrigeration compressors I am familiar with there was an electromechanical timer to facilitate this lockout.