Thermal testing setups

RE: Thermal testing setups

Gough Lui — Thu, 03 Feb 2022 11:35:23 GMT

I certainly don't make enough to justify that kind of cost for a proper thermal chamber, nor do I really have that much of a need for it. It's more the curiosity factor for me - i.e. what happens at the extremes. But whether we like it or not, all our testing is influenced by temperature in some way - whether that be component drift, changes in behaviour or even changes in test equipment accuracy.

So for my own informal battery testing where 20/21/23/25 C +/- 1 C is the demand, I have resorted to my modified Peltier-based car fridge to "trim" the room temperature appropriately to keep the battery under test at the right temperature, assuming the broader room temperature stays within the limits of accuracy for the test instrument (usually something like 23 C +/- 5 C). My unit has gone through a few design improvements and now I have two which are both in-use at this exact time, although the Peltier module recently failed on the second and was replaced. The advantages for testing Li-Ion batteries is very evident in the meausred capacity graphs which have a strong correlation with temperature - controlling for temperature is necessary to get an "accurate" value.

The puny little units I have won't really go past about 18 C to 60 C, but on the upside, cost me only about AU$60 each in total, which I can afford for a pure hobby use. Your setup definitely has me in awe - I'm glad it earns its keep!

- Gough

RE: Thermal testing setups

bhfletcher — Fri, 28 Jan 2022 16:41:23 GMT

Hey [mention:296c89b25d2c4d2ea7d35085b61af3e8:e9ed411860ed4f2ba0265705b8793d05] -- what was your thought process behind purchasing vs. renting the thermal chamber from time to time?

RE: Thermal testing setups

bhfletcher — Fri, 28 Jan 2022 16:39:47 GMT

I also work from a home office, which is awesome but does create some challenges. I remember several years ago working through an issue on some of the original PicoZeds where the eMMC wasn't readable at cold temps. It was intermittent from board to board, but you could make any board fail by getting it cold enough. I found a board that would consistently fail at somewhat cold temperatures, and then moved a mini-fridge into my office. That was enough to allow me to reproduce the problem and coordinate with engineering in determining a fix -- we either had to change the operating mode with a slower frequency or we had to remove a mux. We did an update for PicoZed Rev E that removed the mux and operates at high speed even at cold temperatures.

A mini fridge is not ideal, and not very controlled, but it did the job for about ~$100.

RE: Thermal testing setups

wolfgangfriedrich — Fri, 28 Jan 2022 01:51:19 GMT

Living in eastern Canada, I can usually test from -20degC to +30degC with humidity over the course of 12 months by going outside . Unfortunately, the setup is very un-controlled. [emoticon:4191f5ee34e248a29fa0dbe8d975f74a]

RE: Thermal testing setups

dougw — Thu, 27 Jan 2022 22:03:12 GMT

At work we have a couple of hot/cold thermal chambers, including one big enough to walk into. At home I don't do anything beyond hit it with a heat gun. If I really wanted to do something more accurate for hot soaking I might employ one of my 3D printers with an insulated cover over the temperature-controlled bed. I actually used this method the get a 3D printed object to a precise temperature where it could be bent without breaking or self warping.