Getting thermostat accuracy in small space with drafts

[Sugarkryptonite]

Hi guys

I have a small space (say around 100 sqft) that I’m air conditioning with a 5000 BTU window unit.

Right now I “hacked” the temperature control of the unit and am using a standard STC-1000 to control the room temp.

I have one central “exhaust” duct from the AC that is facing straight up, centrally on one side of the room.

The problem I’m having is that no matter where I put the temperature probe from the 1000, it seems to be getting air drafts from the AC exhaust. Because of this, the unit runs fine until the temperature is met, unit turns off. At this point I’m watching the temperature reading on the 1000, and it goes up pretty quickly, like 1.5 deg C in about 3 mins, enough to kick the unit back on, which doesn’t seem right. The duty cycle is just too high vs what it should be.

There are minimal heat sources in the room, say maybe 100w. I only use the unit at night so no sun to compete with.

I’m thinking of maybe putting the temperature probe in a plastic box (like a plastic outlet box with a cover) with a couple small holes drilled in it, so the probe doesn’t get any air drafts from the AC exhaust.

The reason why I know it’s off is I’ve also tested the unit using an on off repeat timer, whereas for example I set the unit to run for 10 mins, then be off for 20-30 mins and repeat (basically no thermostat involved), and the room temp is better controlled. Problem is I have no temperature control using this method, which I need.

Anyone have any suggestions for my temperature probe quandary? Is putting it in some protective box a decent idea? Just trying to brainstorm before I go buying more parts and waste more time

 

[mac_1103]

Where is the probe now? Have you tried setting a 20-30 minute compressor delay on the STC-1000?

 

[Sugarkryptonite]

Where is the probe now? Have you tried setting a 20-30 minute compressor delay on the STC-1000?

I’ve tried moving it around the space in various locations that would be out of the drafts. The only place that made a significant difference was placed under a small storage box, but still not ideal.

Also no I have not tried to set the compressor delay, but maybe that would be useful to stop it coming on so fast.

 

[day_trippr]

So far, nothing is addressing this:

1.5 deg C in about 3 mins

The AC thermostat is just doing its job. Is there a way to lower the heat gain?

Cheers!

 

[Sugarkryptonite]

So far, nothing is addressing this:


The AC thermostat is just doing its job. Is there a way to lower the heat gain?

Cheers!

As I mentioned. Heat gain is minimal (100w). I believe the probe is reading incorrectly low due to drafts

 

[mac_1103]

I believe the probe is reading incorrectly low due to drafts

I'm guessing that what you mean by this is that the probe is reading the temperature of air coming out of the AC unit rather than the average temperature of the room. So maybe what you really need is more drafts - to "homogenize" the temperature in the room. I would stick a fan in there and see what happens.

 

[Sugarkryptonite]

I'm guessing that what you mean by this is that the probe is reading the temperature of air coming out of the AC unit rather than the average temperature of the room. So maybe what you really need is more drafts - to "homogenize" the temperature in the room. I would stick a fan in there and see what happens.

That is a possibility. I think part of the problem is that the air coming out of the AC is so much colder, that even mixing it better, the thermostat gets satisfied, then when the unit turns off the air suddenly gets much warmer. Not sure though I’ll have to test it out more. I did have a couple fans installed already to try to homogenize the air but didn’t work great

 

[Bobby_M]

Just put the probe inside a small jar of water. The mass of the water will act as a buffer.

 

[Whisky River]

Have you tried submerging the probe in a gallon or two of water? That’s usually a good way to even out erratic air temperature readings and if you plan to use the space for fermentation or conditioning it does a better job of approximating what’s going on inside a batch of stored liquid.
Edit: Too slow. Listen to Bobby

 

[Sugarkryptonite]

Just put the probe inside a small jar of water. The mass of the water will act as a buffer.

Interesting thanks

 

[Sugarkryptonite]

Have you tried submerging the probe in a gallon or two of water? That’s usually a good way to even out erratic air temperature readings and if you plan to use the space for fermentation or conditioning it does a better job of approximating what’s going on inside a batch of stored liquid.
Edit: Too slow. Listen to Bobby

No but I’ll give that a whirl thanks for the reply and info