Poor man's Thermal Chamber, meat packing box II

[joeqsmith]

I started out temperature testing using a cardboard box, some foam and a few resistors for a heat source.   We had some meat shipped to us that came in a nice box which I decided would work better than what I had.   

When I started testing meters, I decided to add a Peltier cooler to meat packing box, along with a camera so I could monitor what the meters were doing.  This has worked alright but the Peltier I had was fairly small and it took several hours for the box to reach 0C.    Getting below that was a gamble.   I was able to get it down to -15C with the windows open during the Winter.   You can see a few pictures of it here:

https://www.eevblog.com/forum/testgear/hear-kitty-kitty-kitty-nope-not-that-kind-of-cat/msg1371562/#msg1371562

You can also see it in several videos like this one:
https://youtu.be/uY6ZcINF7L0?list=PLZSS2ajxhiQBvWvqMVLdRQMjGofKpQUJr&t=3182

[joeqsmith]

A few days ago, I started looking at changing the box once again.  The goal is to improve the ramp times and be able to reach -10C.    To do this, I had wanted to go to a higher power Peltier along with 2-stages.    I tried stacking a couple of them on the bench and it seemed like it may work.

[joeqsmith]

Another test showing the open air loading.  The current is 13.64A and the voltage is 8.01V.  Roughly 110Watts.    The TECs can run up to 14.5V absolute max but it would dissipate roughly 400W.  The ambient side heatsink I am using would need to be much larger.     

[MosherIV]

Hey Joe.

Have you seen Dave's video on converting a fridge into a thermal chamber?
He used a fridge with Peltier type heat exchanger, so that it could work both as a cooler or heater.

Might be easier than re-inventing a thermal chamber

[joeqsmith]

Hey Joe.

Have you seen Dave's video on converting a fridge into a thermal chamber?
He used a fridge with Peltier type heat exchanger, so that it could work both as a cooler or heater.

Might be easier than re-inventing a thermal chamber

In my first post, where I provided a link showing my last attempt Dave posted about his temperature chamber.     My goals are a bit different. 
https://www.eevblog.com/forum/testgear/hear-kitty-kitty-kitty-nope-not-that-kind-of-cat/msg1371764/#msg1371764

[joeqsmith]

The adhesive is dry and I am giving it a test run.  Poor performance at best.

Some time ago, I was attempting to help a member sort out some measurements with a 74xx ring oscillator and I cobbled up this test chamber:
https://www.eevblog.com/forum/projects/waveforms-in-a-74ls04-ring-oscillator/msg646515/#msg646515

Sadly, I feel I am trying to heat a 10,000 sq foot home during the winter with a 1KW space heater.  Even with about triple the power, about all I am doing is creating more heat.   

If you look at my original thread, after I first added the peltier and prior to adding all the additional insulation, with the box empty and stabilized at 20C, it could ramp down to 5C in about 1150 seconds.   

[joeqsmith]

After adding all the insulation, I don't have a lot of data with the chamber in a steady state condition, then looking at the ramp.   

In this picture, the box is configured the way it was prior to these new mods.   There is a UNI-T UT181A inside the box and I am comparing various temps against what the box is reporting.   

I would expect that the during the ramp, we have a heat start.  With the smaller volume, it still takes 1200 seconds to ramp from 20 to 5C.  But that meter is a bit of thermal mass.   20C to 0 was 2200 seconds.   

Looking at the first part of the collection, you can see how slow the time constant is as the temperatures get lower.       

[joeqsmith]

Using the poor UT181A as a thermal mass, I let the box sit at 20C then drove it down.  It reaches 5C in about 600 seconds and 0C in 1025 seconds. 

[joeqsmith]

I decided to see what effect the blower speed would have to I increase the supply voltage from 24 to 28 volts.   I then decided to poor the coals to it and bought the power up to about double but the heatsink just couldn't remove that much heat and it started to run away.   I brought the power back down and plan to just let it set.

You can never have enough heat sink....

[joeqsmith]

Running at 14.46A at 8.361V or 120W, this was the best it could do.   A little better than what I could achieve in the winter months.   

[jbb]

It’s counterintuitive, but I have heard from a couple of sources that you may gain a bit of delta T by reducing the drive current. Have you tried it?

This is apparently because pushing more current through causes lots more loss in the Peltier (ie I squared R losses) which may overwhelm the heatsink.

[joeqsmith]

Watching the temperature of the two heatsinks, I can hold it right on the edge where it would start to run away.   It doesn't seem to gain me much running it there vs 10Watts lower.  A much larger heatsink on the ambient air side may help as I could push a lot more current through the peltiers. 

It will hold -10C with the UT181A installed, powered up with its backlight on, during the summer.     
https://youtu.be/NogCSiIKQq0

[joeqsmith]

I used a tunnel made from paper and held the blower with some wood blocks when I tested the UT181A.  Bent up some parts for it tonight. 

[joeqsmith]

My home lab is sitting around 80F which is fairly warm.   I thought I would just let the box sit as see how low it will go.   It's really struggling to do much of anything.    It made it to -12.8 and I decided to push another 5A or so through it.  Gained about 0.2 was all before the heatsink could not keep up.   I shut it down and the temperature rises fairly quickly.  I imagine most of this is the heat transfer through the peltiers.

I would guess that tripple the power with the 2-stages has gained me about 10C lower.    Not to mention is that 120 Watt space heater is doing a fine job heating the lab even more.    Much larger ambient side sink would help.  More insulation... 
   

[branadic]

I bought one of those Happy Reptile / Herp Nursery II boxes used as a poor man's thermal chamber for my experiments on voltage references and can recommend them as a cheap ready to go solution. As Dave showed in his video #101 I drilled a 40mm hole into the wall and glued come water pipe fitting to it. Furthermore I installed Smartec SMT172 temperature sensors in TO18 package to one wall together with a SMTAS04USBmini.

https://www.smartec-sensors.com/cms/pages/products/temperature-sensors.php

Attached is a profile with additional cooling packs. As can be seen temperatures below 5° are possible this way.

-branadic-

[joeqsmith]

Trying another sweep from -10 to 60C using the new setup.   This meter doesn't seem as stable as the UT181A. 

[branadic]

Please explain what you mean by "additional cooling packs".

I mean additional thermal packs for cooling put to the bottom of the thermal chamber.

-branadic-

[joeqsmith]

I assume packs like you would use in an ice cooler.    That wouldn't work very well for me.  I could get the chamber down to the pack's temperature but going beyond that would slow it down.  Just having a meter in the box slows it down a fair amount.    Not to mention slowing the heating.  Odd as you show no problems heating it with these cool packs in the chamber. 

I have a fair bit of air movement inside and the RTD is located near the center of it.  Is your sensor located next to the Peltier?  What sort of air circulation do you have inside?  Pictures may help.

[joeqsmith]

Temps started out fine but then the lab reached about 84F.  The box struggles at -10C even with the camera turned down. The camera puts out a fair amount of heat.

[joeqsmith]

Opened the windows of the lab during the evening and let it drive as low as it could go.  Used the newer BM869s  for this test. 

Once I started to heat, the fan inside the chamber froze up.   It took some time for the temp to thaw the ice.   During this time the meter was iced over, so you can't see the LCD.    Still no where near last Winter's temps.