Peltier Controller / General Science and Electronics / Forums

Forums

4hv.org :: Forums :: General Science and Electronics

« Previous topic | Next topic »

Peltier Controller

previous 1 2 3 next

Move Thread

LAN_403

Bjørn

Sat Jul 21 2007, 02:58AM

Registered Member #27 Joined: Fri Feb 03 2006, 02:20AM
Location: Hyperborea
Posts: 2058

Not all peltiers are equal but the datasheets usually say 10% ripple voltage or less. Ripple can cause significantly reduced efficiency and lifetime.

Steve Conner

Sat Jul 21 2007, 08:41PM

Registered Member #30 Joined: Fri Feb 03 2006, 10:52AM
Location: Glasgow, Scotland
Posts: 6706

Peltiers are non-linear devices. They're not resistors: more like a motor, with a back EMF too. If they are driven with a current that contains an AC component, only the mean value of the current goes towards the cooling. The AC component just causes internal heating.

Therefore, driving a Peltier with anything other than pure DC makes it work against itself, thus making it work harder and shortening its life. In other words, the ideal Peltier drive waveform has a RMS-to-mean ratio of 1.

The driver I linked was a switched mode design with an H-bridge operating at 310kHz, and the output was LC filtered to make it relatively smooth. It used a digital PID controller implemented in firmware and was stable to about 0.05'C. I also designed a driver using a linear output stage and analog PID that could manage better than 0.005'C, for temperature-tuned DFB lasers that need tight control, but the details are commercially sensitive. All I will say is that I'm pleased with the digitally settable PID coefficients and current limit :^)

Firkragg: the screw pedestals were the responsibility of the mechanical design department, not me :P

Eric

Sun Jul 22 2007, 10:02AM

Registered Member #69 Joined: Thu Feb 09 2006, 07:42AM
Location:
Posts: 116

Well, I got it to work on the first try which is a bit shocking. I decided to go with a bridge configuration so that the peltier can be driven in both directions. It stabilizes at ~0.04C or better (the smallest increment I can see with my multimeter).

Fri Jul 27 2007, 08:43AM

Registered Member #809 Joined: Wed May 30 2007, 08:59AM
Location: Melbourne, victoria
Posts: 114

peltiers don't like to be driven at frequencies above 2khz, or is it below?

Steve Conner

Fri Jul 27 2007, 08:52AM

Registered Member #30 Joined: Fri Feb 03 2006, 10:52AM
Location: Glasgow, Scotland
Posts: 6706

Well, I used 310kHz and they were fine.

Fri Jul 27 2007, 09:39AM

Registered Member #809 Joined: Wed May 30 2007, 08:59AM
Location: Melbourne, victoria
Posts: 114

it must be below then, they have to much stress being pulsed, and will soon go open circuit.

Marko

Fri Jul 27 2007, 11:19AM

Registered Member #89 Joined: Thu Feb 09 2006, 02:40PM
Location: Zadar, Croatia
Posts: 3145

peltiers don't like to be driven at frequencies above 2khz, or is it below?
it must be below then, they have to much stress being pulsed, and will soon go open circuit.

No, you are missing the point. It looks like PWM regulation is always done with filter, so peltier only sees fine DC.

Peltiers are non-linear devices. They're not resistors: more like a motor, with a back EMF too. If they are driven with a current that contains an AC component, only the mean value of the current goes towards the cooling. The AC component just causes internal heating.

Steve: did you read the link I posted, the section about PWM'ing?

I understand peltier will produce seebeck voltage from it's own temperature differences. But, this vltage is tiny, and (should be

) no threat to the switch.

If peltier is driven simply in ''flyback'' regime at low frequency, there will be no AC component you are talking about since seebeck voltage never gets shorted out. (If there is AC component, seebeck current flows, and peltier then works against itself for the off time)

Or am I just completely wrong? I just wanted to understand more about those things before intuitively giving wrong advice to people again.

...

Fri Jul 27 2007, 03:37PM

Registered Member #56 Joined: Thu Feb 09 2006, 05:02AM
Location: Southern Califorina, USA
Posts: 2445

In general, you never want to PWM a peltier simply because the little buggers are notable more efficient when being run at lower powers, running them at full power for short bursts will not be as efficient as if you put a filter on the output of the bridge (particurally at low powers).

Of course you don't want to drive a pielter with anything that has an AC component, for obvious reasons... But even taking the output strait off the output of a pwm won't give you any AC unless you are right at the setpoint and it is bouncing back/forth between heat/cool... Pulsed DC isn't AC....

Fri Jul 27 2007, 10:24PM

Registered Member #809 Joined: Wed May 30 2007, 08:59AM
Location: Melbourne, victoria
Posts: 114

without geting into the whole pulsed dc and ac thing, peltiers do not like being pulsed, it causes to much stress on them, and they go open circuit very quickly. i found this out while trying to control one with pwm at a low level, and it just stopped working?

Marko

Sat Jul 28 2007, 05:11PM

Registered Member #89 Joined: Thu Feb 09 2006, 02:40PM
Location: Zadar, Croatia
Posts: 3145

So, what causes pulsed peltiers to fail after all? Anyone?

previous 1 2 3 next

Moderator(s): Chris Russell, Noelle, Alex, Tesladownunder, Dave Marshall, Dave Billington, Bjørn, Steve Conner, Wolfram, Kizmo, Mads Barnkob