AC resistivity measurement with MyDAQ on TEC

[Blokk]

Hello,

I would like to ask for advice: we would like to measure resisitivity of Peltier-elements. If we use DC method, because of the Seebeck-voltage of the element, we do not get the real resistivity value. The common method to use AC measurement in case of such devices. I would like to know, if it is possible to use a MyDAQ device for this task? As a first approach we do not need the best accuracy, this would be more as a "broken-element" test (there is usually a significant shift away from the nominal AC resistivity value stated by the manufacturer).

What would be the easiest way to carry out a simple measurement? What about if I use one analogue output of the MyDAQ lets say with 1kHz, and I drive the Peltier-element with this voltage AC source? And I connect a known resistor in series, and I measure the AC voltage drop on this. From this data, I could calculate the current, and the AC resistance of the Peltier-element? Of course I have to choose a resistor so the MyDAQ can drive the network.

what is your opinion, where to start?

Thanks!

[johnsold]

Blokk,

That should work provided you can get enough current from the AO to make a meaningful measurement.  For some TECs which I looked up the AC resistance can be a fraction of an ohm to a few ohms. The few mA which most AO lines can drive will result in millivolts or less to measure.  Although it appears that the MyDAQ has (barely) adequate resolution, the absolute accuracy is not all that good and the noise level is not specified at all.

Is the AC resistance test run at an applied DC voltage of zero or at soem active bias point? Driving a suitable AC current with a bias applied can be more complicated.

Lynn

[Blokk]

Hello, thanks for the reply!

Actually I am just going to replace 60 Peltier elements in a "home-made" 15 years old heat exchanger (I expect to find some mice too 🙂 ). This assembly has dropped performance (heat transfer power) and also I know that, a few years ago after a malfunction with the hot side cooling fans, this device was running for a certain time without forced cooling. What is quite bad in case of Peltier-elements. I have already bought 65 brand new Peltier-modules and after I get some heat conducting paste, I will replace them. And I will have 60 pieces of old Peltier modules, to play with them 🙂

Since I have this NI MyDAQ, I thought it would be fun (and would give also valuable information) to try to measure the old elements' AC resistivity. Since this task is not like I should do often, so for first, I would not like to go for more pro solutions, I will try the MyDAQ first.
 I expect that some of the modules got structural degradation during the heat shock, and this may be measurable in a shifted AC resistance.

I would do the measurement with zero applied DC voltage. These modules are Melcor PT3-12-30L (discontinued product), the nominal module resistance is about 5 Ohm.

EDIT: and would you recommend a certain circuit to try? Shell I use a known resistor for current measurement via measuring the voltage on it? Or I could use a known AC current to drive the Peltier, and I would measure the resulted AC Voltage on it?

[johnsold · Accepted solution]

Sounds like a good chance to learn about those devices. Under those conditions I would do the same thing - try it with what you have.

From a quick glance at the MyDAQ specs it appears that the AO current is limited to 2 mA. I would probably set the AO for about 2 V and use 1000 ohms in series with the Peltier device. That will keep the current within the 2 mA limit. Use two AI lines to measure the voltages on both the fixed resistor and the Peltier device.  Then you can calculate the unknown resistance.

If you were doing a production test or were making the measurement with DC bias, I would recommend using a transformer to couple the AC component into the DC line. This gets somewhat more complicated to set up but has much greater versatility.

Please post back after you have tried it and let me know what you found.

Lynn

[Blokk]

Hello Johnsold,

So far I did not replace the old Peltier elements, but I have got some Melcor CP1.4-127-06L modules to play with them...

I connected a Peltier-module in serial with the 1 kOhm resistor, and I drive this circuit with the MyDAQ AO0: 1 kHz AC +-2V peak-to-peak.

I measure the voltage drops on the resistor and the Peltier-module with AI0 and AI1 inputs.

As expected, I get voltage on the resistor a little bit under 2 Volts --> 1.994 Volts.
 So the current is limited a tiny bit under 2 mAmps.

I get signal from the Peltier just a little bit "above the detection limit", approximately 5.81 mV peak-to-peak.

From this I can calculate the resistivity of the Peltier:  2.91 Ohm .

The specification sheet tells me this module has a resistance about 2.5 Ohm, so my measurement is not far from the nominal value.

And I am not sure what the resistivity value means in the spec sheet: maybe during DC operation?

So, so far it is not bad, I can measure certainly a value. I have now 4 modules, I measured all of them, 3 of them shows exactly 2.7 Ohm, and one of them 2.9 Ohm.

I am very interested, when I get the old modules from the old heat exchanger, I will measure all of them, so I can have a histogram 🙂

Thanks for advice,

regards,

[johnsold]

Thanks for the update.

If you have an op amp available, you might try amplifying the voltage across the Peltier by 10 or 100 to get better resolution.

I do not recall which TEC manufacturer's data I looked at, but I noticed that they did not clearly specifiy how they defined or measured the resistances.

Lynn

[AnalogKid2DigitalMan]

Good work Gentlemen!

There are some good papers here, search this page for test:

http://www.tetech.com/Downloadable-Publications.html#testing

-AK2DM