Post Go back to editing

LTC4332 Temperature problem

Category: Hardware
Product Number: LTC4332

We have tested two different batches for now. First one from Farnell, the product was LTC4332I. Second time from mouser we tested LTC4332I and LTC4332H devices.

With Farnell part the IC stops forwarding valid data if the ambient drops under 0 degrees C.

With mouser parts the threshold is around -8C, although the part is rated to -40C.

The sad thing is that the marking on the part itself does not specify the temperature class so we first taught it might be a mixup by Farnell, but now it seems unlikely.

We added a hater resistor on top of the IC to locally heat it, and that fixed the problem, so now we know that the problem lies with the LTC part. 

Any ideas/comments?

It seems to be unusable in an industrial product!

  • Hi Tarmogr,

       These parts should operate at the lower temperatures specified by their grade in the datasheet.   While lower temperatures should not cause the parts to misbehave, it is possible for some of their characteristics to change with temperature and this could cause a marginally operating system at room temperature to fail at other temperatures.

       Note that the successful operation of the LTC4332 depends on the characteristics of the link between the two LTC4332 devices.  Factors such as:

    • Configured Link speed
    • Length and type/construction of the link cable
    • termination network specifics (if present)
    • protective network specifics (if present)

    can have an impact and so knowing more details about the link might help point out where to investigate further.

    It also would potentially be helpful to understand exactly what is meant by "stops forwarding valid data".  Waveforms when working and not could be useful.   The LTC4332 has a nLink pin which reports the status of the link.  Does this does this pin do as the temperature varies?


    Eric