Connect with us

Techniques for testing PCBs in temperature chambers

Discussion in 'PCB Layout, Design and Manufacture' started by Emam, Sep 15, 2014.

Scroll to continue with content
  1. Emam

    Emam

    58
    2
    Jul 7, 2014
    Dear all,
    I would like to know if someone has an experience with testing PCBs in temperature chambers following the production stage.
    In fact, we have to check 200-300 PCBs in 3 differents temperature : -40°C, 20°C and 60 °C. In our company, we have a temperature chamber allowing to acheive these temperatures.
    In this temperature chamber, we have a mechanical PCB fixture (based on bed of nails system) allowing to do the tests (picture attached).


    However, our bed of nail device deforme easily with variation of temperature, making the results completely incorrect....!
    in fact the contacts between the test probes and the PCB points are dependent of temperature !

    Do you know what system is used in industry, to test PCBs in different temperatures?
    How these machine can resist thermal chocks?


    Many thanks,
    Bests regards,
    P M
     

    Attached Files:

    • cpeb.jpg
      cpeb.jpg
      File size:
      29.3 KB
      Views:
      120
  2. Arouse1973

    Arouse1973 Adam

    5,164
    1,080
    Dec 18, 2013
    I have done a lot of temperature testing for my company and we do two different things. For us the problem with testing bare PCBs in the chamber is that when the PCB is in our product the thermal inertia of the enclosure the PCB lives in in normal operation can be quite large. Over temperature cycling the PCB never sees the same temperature as it would if just placed on it's own in the chamber. So for us it's not always a valid test.

    We do however for sometimes stress the PCBs and take measurements of critical points of the circuit. An example of one I have done recently was changing from a resonator to a crystal of one of our PCBs. So I wanted to know how extreme temperatures effected the microprocessors timings of a data packet. So instead of building up the whole product and trying to get that in the chamber with lots of wires hanging out of it I just used the bare PCB itself.

    I attached some wires to the points of interest and fed them out of the side of the chamber. I knew though that if it failed towards the end of the testing that I might have to do it again in the product. I took an educated guess and luckily for me it was ok. So I knew that this would pass absolutely worst case in product.

    The other issue we have found with testing at very cold temperatures is condensation formed when the item is removed from the chamber. For us this is a real life operation of our product. This can cause serious issues if the PCB does not have conformal coating on them. I think it all depends on how your product is used in real life because extreme temperature cycling can sometimes be unrealistic for real life use.

    We have found in most cases testing the product as a whole is more realistic than just the PCB on it's own. But that does depend on whether you can fit the product in the chamber :) So I have found no easier way of actually measuring real life signal other than attaching test wires to these points and measuring them outside the chamber. But remember this works for me and my products, you may have to do something a little different.

    Adam
     
    KrisBlueNZ likes this.
Ask a Question
Want to reply to this thread or ask your own question?
You'll need to choose a username for the site, which only take a couple of moments (here). After that, you can post your question and our members will help you out.
Electronics Point Logo
Continue to site
Quote of the day

-