Voltage limits of TECs

[Pack34]

I'm in a bit of a bind.

On one design we have a TEC that has a bit of a lead time and was hoping to substitute it in case we run out. Dimensions, delta-T and everything else are the same but the voltage ratings are slightly different. The old part has a maximum voltage rating of 4.3V, while the new one is 3.8V

Now, the TEC driver does have a 5V supply but it only peaks up that high very briefly at initial boot, then it's just temperature stabilizing. It's regulating the temperature of a small piece of metal. Steady state operation is <1V.

I would say that current 4.3V TEC is a bit under-speced and it would be ideal if the driver was running off of 3.3V.

How bad is it for the TEC to receive a voltage larger than it's rating? With motors you can overdrive them a bit as long as it isn't prolonged. Am I playing with fire by checking off that the 3.8V TEC is a suitable substitute in a pinch? It looks like it's driving fine, but could the voltage rating cause serious issues down the line?

[Kleinstein]

Usually the current is the more important factor, not the voltage. If you use more than the so called optimum current (which is about the same as the specified maximum), the cooling effect will go down again. Otherwise it should be no problem to use a higher current / voltage as long cooling of the hot side is sufficient and the temperature is not close to the upper limit. It is more like a practical limit and than due to possible thermal effects. So short pulses should not be that bad. If cooling of the hot side is not perfect, the optimum current will be reduced - so the turn over to lower cooling usually comes earlier than the optimum current specified in the DS.

For regulation one should avoid going higher, as the curve will change slope, so simple regulators will not work in the range well above the maximum. If you need to work close to the maximum current one should take into account the nonlinear (about an parabola) relationship of current to cooling power.

A lower supply or TEC with a higher rated voltage could ease on the inductor for smoothing.

[Pack34]

Thanks!