31 replies to this topic

[M_Ants]

Anyone every try making an ant hibernation chamber? https://hackaday.com...wn-mini-fridge/

Mini fridges are expensive to run and it's not cold outside always. Regular fridges aren't always convenient either so I might makes something like this with temp. control.

Edited by M_Ants, June 4 2020 - 12:17 PM.

[Ants_Dakota]

I actually might try this if it is not too pricey. Thanks!

[M_Ants]

A reply!   Not sure how great it works or if it's worth the time but I thought it looked cool.

[SleepyAsianAnter]

I just looked around on craigslist until I found a cheap used wine cooler

[B_rad0806]

Just use your fridge. Every ant that I hibernated in there lived and did really well coming out of hibernation.

[Ants_Dakota]

if you are allowed to use the fridge.

[KitsAntVa]

Honestly ant companies like THA should make these and sell them.

[Ants_Dakota]

Honestly ant companies like THA should make these and sell them.

maybe i can with AntsDakota. will think about it. tell me if you are interested!

[KitsAntVa]

I’m interested actually

[Ants_Dakota]

well, i would not be able to create and sell them, as i would need to do something with getting a permit, but i could sell all the parts packaged together and have instructions. how does that sound?

[KitsAntVa]

that does sound interesting ill think about it, but the instructions better be easy because i'm already to lazy to read them  

[Ants_Dakota]

that does sound interesting ill think about it, but the instructions better be easy because i'm already to lazy to read them  

if i do this, i will try to make them as easy as possible.

[Domagoj]

So I started to cobble up something, just to see what's the ballpark for temperatures, power etc. This is the busy part, the rest is just an off the shelf thermostat and an insulated box, which I'll try to put together in the following days.

Anyway, for those of you playing along at home, there are two Peltier elements in parallel, hooked up to 12V, mounted on a 68 000 mm^2 aluminium heatsink (black) on cold side, and a 10 mm alu plate for cold side. The cooler on top is a CPU cooler (Cooler Master Hyper TX3 Evo).

I still haven't put the thermal paste, since that thing is messy and I'll still need to disassemble everything a few times, but here are the numbers after about 10 minutes of run time in open air (no insulation), just sitting on my desk:

Total power: 80W (78 W Peltiers, 2W fan)

Ambient temp: 24.0 °C

Hot Side (alu plate): 44.7 °C

Cold SIde: 16.1 °C (note the condensation on the left side of the black heatsink)

 

I consider this quite promising, and worth further exploration. I have extruded polystyrene sheets 5 cm thick, which I think will be sufficient to keep the heat out.

I might squeeze some additional performance if I replaced the screws that connect hot and cold side (they are regular steel at the moment, stainless steel would provide lower heat transfer) and the ones holding the radiator to hot side (again regular steel atm, replacing with brass would offer better heat transfer). Of course, thermal paste is a must.

[Ants_Dakota]

keep me updated!

[Domagoj]

I didn't have much time today, but managed to apply thermal paste and hook up the thermostat.

Still no proper insulated box, but I wrapped the cold end with some shop rags and ran a test.

The thermostat has a safety feature where it will not turn on the chiller less than 1 minute after turning it off so in order to extend the cycle times, I put a 1-2-3 steel block (about 500 grams) under the cold side to add some thermal capacity. Sensing probe is stuck inside the block, not touching the aluminium heatsink.

Settings as follows:

Ambient temp: 24 °C (75 °F)

Relative humidity: 65%

Chiller turns on above 10 °C (50 °F) and turns off at 8,5 °C (47,3 °F).

After a few cycles to allow for the system to achieve equilibrium, it takes about 1 minute 30 seconds for the temperature to rise from 8,5 °C to 10 °C where the chiller is turned on again. There is some hysteresis, so the temperature continues to rise and reaches 10,5 °C before starting to drop down again. It takes about 15 minutes to reach 8,5 °C.

I've also noticed the total power dropped to 70W (down from 80W on my tests yesterday). I'm not sure what is the reason for this, but it may be due to mechanical damage to the Peltier elements if I overtightened the screws. I'm not sure I did that, though.

Anyway, this comes to about 90% duty cycle, which is a bit more than I expected. In this configuration the device would consume about 1,5 kWh per day, or about $6 worth of electricity per month. 

I suppose the situation will significantly improve once I put the cold side in a properly insulated box, since the rags I used were noticeably colder than ambient temperature, and there was a fair bit of condensation going on (that is a huge power hog). Putting the entire setup in a colder room would also significantly drop the energy requirements.

 

That's it for now, I'll try to get going on the box building part over the weekend.

Edited by Domagoj, September 25 2020 - 12:38 PM.

[ANTdrew]

This is like Back to the Future ant-keeping. Love it.

[Domagoj]

I made the insulated box and assembled all the bits. Unfortunately, it now seems certain that there is either damage to the Peltier elements or that they were faulty from the beginning (always a chance with ebay stuff) and just needed an excuse to fail completely. In my last test, total power dropped to 40W and after a couple of hours of run time, the chamber temperature dropped only to about 16 °C (24 °C ambient).

Whatever the case may be, I'm redesigning the functional bits. I'm adding additional two elements (four in total). This requires that the cold side be upgraded to a bigger heatsink and will receive a low power fan (sub 1W).

I'm also dropping the CPU cooler on the hot side since it has a small footprint and would not mesh well with the large cold side heatsink. It is being replaced by a regular (albeit on the larger side) conventional alu heatsink (no heat pipes) with some fans on it.

Since I'm doubling the number of Peltier elements, my previous power supply (12,5 A) is no longer adequate, so I'm replacing that with a 400W computer PSU.

I'm also building in a safety feature I think may help prevent damage to the elements - the heatsink-elements-heatsink assembly is now being held with spring loaded bolts, instead of directly by bolts. This will hopefully provide some play room for thermal expansion and prevent elements from crushing themselves.

 

More to come.

[Ants_Dakota]

yes! maybe you could sell this on our website! try to make it as inexpensive as possible, though

[AleeGuy]

yes! maybe you could sell this on our website! try to make it as inexpensive as possible, though

If it will cost more than a 70 then I think there is no point of selling that because there is better wine coolers on Craigslist that cost that much(maybe even cheaper)

[Domagoj]

It would certainly cost more than $70, since I'd charge for my time alone more than that, but for me personally and my use, the idea is not to build it down to price, but up to spec. I am willing to absorb that price myself, but there is no way a one off production or even a small scale production run can ever compete with mass manufacturing in terms of cost.

A used wine cooler found on garage sale with a decent thermostat built in will always be a cheaper option, but like I said, I can't find what I'm looking for locally and buying off of ebay or amazon comes with shipping price comparable to that of shipping an unherded herd of rabid goats half past the edge of the world.

 

For those looking for budget solutions, I'd recommend getting a cheap mini fridge (optionally putting it in a cold room (sub 18 °C / 65 °F) if it can't reach desired temperature) and hooking it up to a thermostat so it doesn't freeze the ants if it happens to be too powerful).