QHY9 Cooler Being Funny?

[E_ri_k]

The cooler on my QHY9 was behaving a little strange the other night. First night of imaging it was fine, -35 at 60%-70% power. Next night it was as if it was struggling? Sat at 100% power but only managed to reach -25 or so, and it was freezing outside!

After a little fiddling around, I managed to get around it by setting the cooler to say -6, and watch the cooler power drop quickly to say 50%, while the sensor temp remained quite cold, then set it back to -35 pretty fast. Like fooling it into working?

Tried both DC 201 power supplies and same problem? Any ideas?
Erik

[alpal]

Do a test inside & measure the voltage on the 12V power supply
as you adjust the temperature lower & lower.

Even a poorly connected fuse could cause that.

[Meru]

Hi Erik,

I had a similar issue a while back with my QHY9, but mine would either constantly show 100% and actually force the cooling to 100% (so regardless of my set point it would cool all the way down) or it would show my CCD sensor at something like 90C and force cooling to 100%.

I took my camera back to Theo a few times, different power sources and the TEC controller, eventually we had to replace the mainboard from memory. Never had an issue since, so maybe give Theo a shout? He's pretty good with fixing things up

[E_ri_k]

Cool, thanks, I'll give that a go!
Erik


Thanks. I'll check the voltage, if it looks normal I'll speak to Theo.
Thanks guys :)

[E_ri_k]

Never got around to testing anything, but it's come good again by itself:screwy: I'm wondering if it could be a software issue?

[Garbz]

No.

This is a feature of imperfect TECs called thermal runaway.

The idea is that there's a limit of how fast heat can be dissipated from the hot side, and there's a max delta-T between the hot and cold face of a TEC. Once that delta T is reached (and it'd dependent on the TEC surface temperature at the heatsink NOT AMBIENT TEMP), heat will flow from the hot side to the cold side. The controller compensates by making driving the TEC harder which makes it worse and worse and worse.

On a cold day things are better. The heat of the chip will warm the cold side and the cold ambient temperature will cool the warm side more efficiently. But on a warm day you may find that after you get past a certain point things break down quite quickly. I typically avoid going over 80% on my TEC, but most of the time I don't have problem getting to -25degC, especially in winter where I can do it at around 50% power. Also this problem is more likely to happen at startup where the controller will overdrive the TEC to get it to cool as quickly as possible. Warm ambient + warm cold side at startup + lots of power into the TEC is a recipe for thermal runaway.


If you suffer thermal runaway, drop the controller into manual and back the power off to something like 30%. You'll see the temperature rise and then at some point start dipping again. When it settles out you can try going back into auto.

The only engineering solution to this is a massively over-designed heatsink, or a clever control scheme to detect it and back off when it happens.

[E_ri_k]

Thanks for the info Chris! I understand what you are saying, last night after posting here, it played up again:rolleyes:

It was a warmish night, and it struggled a little, 40% to get to -11! Eventually got down to -33 or so, but at 100% I manually raised the set point to back off the power, and it stabilised at -34, with 70% power.

Maybe I will try gradually dropping the temperature in 5 degree intervals?

Erik

[E_ri_k]

Just tried cooling in 10 degree increments, and it worked fine. Sitting on -35, at 70%. Fingers crossed! I'll try this method from now on and see how it holds up.
Erik