• State TI Thinks Resolved
  • Locked Locked
  • Replies 5 replies
  • Answers 1 answer
  • Subscribers 30 subscribers
  • Views 1748 views
  • Users 0 members are here
Support feedback
Options
Options
Related

This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

LM35: High failure rate with hand-soldered LM35 TO-92s

Christopher Deckard
Prodigy 110 points
Part Number: LM35

Hi, we've asked a CM to manufacture wire harnesses by hand-soldering the wire directly to the leads of a TO-92 LM35.  We're seeing a high failure rate.  I believe the cause is due to excessive temperature of the die (> 150deg C) during assembly.  However, when I observed the assembly process, it seemed reasonable (temperature-controlled soldering iron, dwell time was not excessive).

I can't find any reflow/soldering information for the TO-92 in the datasheet.

Has any one else had experience with this?   Or, could someone provide some specifications regarding thermal/soldering the TO-92 LM35?

Thanks,

Chris Deckard

  • 0
    TI__Expert 7855 points

    Chris,

    Let me look into this.

    -Kelvin

  • 0 in reply to Kelvin Le THS
    TI__Expert 7855 points
    Chris,

    I found this general guideline for through hole packages:

    www.ti.com/.../snoa292j.pdf
    processors.wiki.ti.com/.../Package_Reflow_Profiles

    A few questions about this CM:

    1. Have you used this CM before and they assembled similar through hole packages before. Was there issues?
    2. What kind of failure are we talking about? Please give a description of the failure.
    3. How does this failure rate change when the temperature is lowered?

    -Kelvin
  • 0 in reply to Kelvin Le THS
    Prodigy 110 points

    Hi Kelvin, thanks for your reply. 

    1)

    We use this CM for wire harnesses quite a bit.  They are adept at soldering since they use a  lot of connectors that have solder cups.  Soldering active components is not typically in their scope.  However, keep in mind that in order to solder into connectors that may have plastic inserts, one must be fairly nimble with their technique.

    They are using a weller T201.  I am confirming that it works correctly and has a 700F tip on it (assuming 60/40 solder).

    This is not a wave soldering process.  It is a manual process with a soldering iron.  Today, they received a new shipment of LM35s (Mouser).  They did a test run with 5 sensors.  2 failed the test.

    2)

    The failure modes can be grouped into 3 primary failures.  Testing is conducted with a +5vdc applied, and a DMM measuring the output pin.  Room temperature is around 18 deg C, so an output of 180mV is expected.

    mode 1) DMM reads a large DC output.  Typically, ~4vdc or around ~2vdc. 

    mode 2) DMM reads too high ~ 300mV

    mode 3) DMM reads correct outpus (~180mV), but when the sensor is handled, the output jumps around (~300, as high as 1 V).  The solder connection is well-formed, no shorting, and the leads are intact.

    3)

    I've ordered 10 or so sensors so that I can perform some more controlled tests here in my lab.

    In the mean time, any insight would be appreciated.

    Thanks,

    Chris Deckard

  • 0 in reply to Christopher Deckard
    TI__Expert 7855 points
    Chris,

    I haven't heard this type of failure due to soldering before. You should personally assemble a few parts yourself to be sure. Perhaps go to the site and use the same soldering equipment to see if you can repeat it.

    -Kelvin
  • 0 in reply to Kelvin Le THS
    Prodigy 110 points

    Hi Kelvin, I have visited the site.  They did confirm they are using a 700F tip on the Weller TC201.  I also had them use a different soldering station (a Weller with variable heat setting), and they had the same result.

    I should be receiving some sensors Monday.  I will heat them up to various temperatures to find out at what temperature they fail.  It does seem that this is the sort of testing/knowledge TI would have available, however.

    Thanks!

    Chris Deckard