cold plate

Sep 09, 2015 59 Replies

We've been doing some component temperature testing lately, in our huge blue temperature chamber. That's overkill for testing 0603 resistors, and settling time is mediocre, more suited to cooling a keg overnight. We were thinking it would be neat to have a cold plate, a peltier heater/cooler with a small work surface, a few inches square maybe, that we could attach little parts to for testing. It would have to be closed-loop controlled with a USB or serial interface, so we can automate temperature sweeps.



Anybody doing this? Any suggestions?


Den onsdag den 9. september 2015 kl. 20.38.41 UTC+2 skrev John Larkin:

any temperature interface you can find a bit of code and a serialport/ethernet to a bench supply?

-Lasse

I guess we could talk to an interfaced power supply and some interfaced thermocouple instrument, and close the loop in software, but I'd rather buy something if I can. We have too many engineering projects already. Manual control would be nice too, without firing up a computer to make a quick non-automated test.

Den onsdag den 9. september 2015 kl. 21.02.53 UTC+2 skrev John Larkin:

something like this maybe?

formatting link

just need a current source that can be controlled by one of the outputs

-Lasse

On Wed, 09 Sep 2015 11:38:24 -0700, John Larkin Gave us:

If you mean "Blue", that's a brand name and is capitalized.

Otherwise it would be merely a paint color choice.

On Wed, 09 Sep 2015 11:38:24 -0700, John Larkin Gave us:

Use a Metcal soldering station drive unit with the peltier in place of a soldering wand. Modify the heat/positive thermal feedback loop into a cold positive thermal feedback loop, and voila! (or monitor the hot side of the peltier in said loop)

On Wed, 09 Sep 2015 12:02:43 -0700, John Larkin Gave us:

Use this, an SSR, and your peltier of choice.

You could use an IR device for faster response but that would be overkill.

Pretty cheap these days too. Hard to find the old, larger form PID factor... damnit.

On Wed, 9 Sep 2015 12:12:49 -0700 (PDT), Lasse Langwadt Christensen Gave us:

Yep... that's the animal. Some even come with a couple SSRs from Amazon.

Yes, I have one of those controlling a small domestic benchtop fan oven for quick temperature testing. I had to do a small amount of wiring to connect the heater element via a solid state relay, and it works far better than I'd anticipated, around +/- 1K.

I kept the oven thermostat in circuit to act as an upper limit in case something goes wrong.

Cheers

Syd

Den onsdag den 9. september 2015 kl. 21.31.39 UTC+2 skrev DecadentLinuxUserNumeroUno:

peltiers are not very efficient when cooling and if you drive with pwm it gets worse

-Lasse

The Metcals pump RF into the iron, and use a curie-point sort of mechanism within the tip to control the temperature. There is no electronic control loop.

Yes, get a keg that fits the apparatus :-)

Many years ago I designed a multi-channel TEC control thingie but you can always buy the PID units as complete modules if you need only one. The Laser diodes were mounted to a Peltier carrier and with the PID set aggressively this thing could swing in a matter of a second. Not sure what temperature range you need. This company is IMHO more on the expensive side but they often have a good solution for on-off sitations:

formatting link

Regards, Joerg http://www.analogconsultants.com/

On Wed, 9 Sep 2015 12:55:09 -0700 (PDT), Lasse Langwadt Christensen Gave us:

The PID is not pwm, it would be merely for turning on and off the power supply driving the peltier and that duty cycle hardly qualifies as pwm. Sheesh.

On Wed, 09 Sep 2015 13:02:06 -0700, John Larkin Gave us:

Yeah, they now have stands that turn them off by simulating full temp when one places the wand in the stand. I just got several.

But I already knew that. So I must have been sub-consciously testing you.

I an *so* relieved that I passed that test.

I want something like this, maybe smaller,

formatting link

but I don't want to pay $7K for it.

I used one of these for checking temp tracking on RF power amp bias circuits. Management decided that it was cheaper than wasting CO2 to cool the big test chambers: I don't know enough about the product or pricing to offer a specific recommendation. I suggest that you also buy or build a styrofoam ice chest for insulation.

Jeff Liebermann jeffl@cruzio.com 150 Felker St #D http://www.LearnByDestroying.com Santa Cruz CA 95060 http://802.11junk.com Skype: JeffLiebermann AE6KS 831-336-2558

Den onsdag den 9. september 2015 kl. 22.59.09 UTC+2 skrev John Larkin:

what would you pay for one? how much power do you need?

-Lasse

Wrong. There are quite a few environmental chamber companies that paint their enclosures blue.

You might be thinking of "Blue M" chambers: which today prefers white paint to blue, probably to differentiate themselves from the rest of the blue horde.

Jeff Liebermann jeffl@cruzio.com 150 Felker St #D http://www.LearnByDestroying.com Santa Cruz CA 95060 http://802.11junk.com Skype: JeffLiebermann AE6KS 831-336-2558

Would $50 including shipping do? One catch, no temperature control. Kinda looks like just two peltier junction plates and a big power supply.

Jeff Liebermann jeffl@cruzio.com 150 Felker St #D http://www.LearnByDestroying.com Santa Cruz CA 95060 http://802.11junk.com Skype: JeffLiebermann AE6KS 831-336-2558

Join the Discussion

Have something to add? Share your thoughts — no account required.

Didn't find your answer?

Ask the community — no account required