• State Resolved
  • Locked Locked
  • Replies 11 replies
  • Subscribers 47 subscribers
  • Views 1405 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.

TL082B: offset change after aging test

Howard Zou
Genius 12365 points
Part Number: TL082B
Other Parts Discussed in Thread: TL082, LM50,

Hi,

my customer is using TL082 to sense the output current of their DCDC power supply.

Before the aging test, they will measure the output of TL082 by applying 2.6mV to Isamp, and will get a value like 100mV at pin 1 of TL082 at 25C temperature.

Then they will test the whole equipment in a chamber for 24hour, during which the board temperature(measured by LM50) is up to 100C.

After the aging test, they will measure the output of TL082 with the same input, now they will get a value like 14mV at pin1 at 25C temperature.

So the offset of TL082 changes after the aging test.

My question is:

Why would we get such big error? Is it what should be expected?

100C board temperature should not cause the damage of the device TL082BCDR since the absolutely maximum junction temperature is 150C.

And we are measuring the difference at both 25C temperature.

  • 0
    Guru 140200 points

    Hi Howard,

    the operating temperature range of TL082BCDR is 0...70°C!

    Kai

  • 0 in reply to kai klaas69
    TI__Genius 12365 points

    I know, but operating temperature is the range that we guarantee the parameters, I think that even at 100C the device is not damaged. And by going back to 25C, the parameters should keep the same as original.

    Besides, I want to know if the offset of an OPA should vary or not after temperature cycle, are there anything like hysteresis on offset?

  • Former Member
    0 Former Member in reply to Howard Zou

    Hi Howard,

    Higher temperatures can prematurely age the part causing parameters to shift.  When the board went into the oven, was the power to the board and op amp on or off? 

    You should consider the fact that the offset of the amplifier is very large relative to the input voltage.  So, even small changes in the offset voltage will cause substantial changes in the output voltage.  You can see this by running a DC sweep of Kai's circuit and seeing how sweeping the DC offset voltage can change the response.

    Also, is this the only thing on the board or are there other parts of the circuit that can be the issue?

    Regards,

    Daniel Miller

  • 0 in reply to Former Member
    TI__Genius 12365 points

    Daniel,

    then I would like to know if we use a Q grade version TL082 which could work -40-125C.

    Can we guarantee that the offset will keep almost the same at the same temperature within one socket?

    My main concern is: Is offset shift a normal thing, or it's implicating that the device is broken?

    When the board went into the oven, was the power to the board and op amp on or off?  The power was on.

    Also, is this the only thing on the board or are there other parts of the circuit that can be the issue? This is the only thing.

  • 0 in reply to Howard Zou
    Guru 140200 points

    Hi Howard,

    as Daniel already mentioned, the offset voltage of TL082B is amplified by a factor of 28. So an output voltage change from 100mV to 24mV translates to a offset voltage change of 2.7mV. This is still well within the specified offset voltage range of -5mV...+5mV of TL082B.

    When heating up a populated PCB many different effects can cause a shift of offset voltage.

    1. If the populated PCB is contaminated with flux and dirt, heating up can change the contamination which can result in a change of leakage currents and by this the offset voltage after cooling down.

    2. Heating up a soldered PCB, like it is done in classical burn-in procedures, can relax mechanical tensions in the circuit and within the chip. This can change the offset voltage and help to overcome the "infant" change of long term drift process. Consequently, again heating up the PCB will result in a much lower shift of offset voltage.

    3. Heating up a PCB made of FR4 to 100°C and above is critical because this may exceed the maximum continuous operation temperature of FR4. I have seen many boards heated up to 100°C which looked severely twisted after cooling down. Twisting the PCB will also twist the components and OPAmps resulting in a change of offset voltage. Because of this, we never "burn-in" and cure a populated PCB at temperatures above +85°C!

    4. Many ovens show a considerable "tidal range" (change of temperature during temperatur regulation). This aggravate the issues discussed in 3. It can also cause an exceedance of maximum operation temperature of certain components containing thermoplastics. Another reason not to increase the "burn-in" temperature of a standard populated PCB ober +85°C.

    Kai

  • Former Member
    0 Former Member in reply to Howard Zou

    Hi Howard,

    If the device was exposed to 100C while still on then that may very well have damaged the part.  Moving to a Q1 part that can go up to 125C will certainly help.  We cannot guarantee that the offset voltage will not change.  In fact, over time it will.  This is called "input offset shift."  However, it should be more controlled as explained by Marek Lis in the post I quoted above.

    I think it would be best for the customer to consider an amplifier with a lower offset and lower offset drift across temperature.  The signal they are trying to read is pretty small and the offset voltage is relatively large in comparison.

    Please let me know your thoughts or if you need help selecting a new part.

    Regards,

    Daniel

  • Former Member
    0 Former Member in reply to kai klaas69

    Hi Kai,

    Thanks for sharing this information.  I do not have experience with heating PCBs myself.  So, this is all quite useful to learn about.

    Regards,

    Daniel

  • 0 in reply to Former Member
    TI__Genius 12365 points

    Daniel,

    I prefer to consider it's broken. Because we have tested hundreds of boards, only about 1% board will have big offset change after aging test, most board offset change is neglectable. So although Kai's post list a lot of factors that could lead to the change of offset, I consider the change very small for "good" part.

  • 0 in reply to Howard Zou
    Guru 140200 points

    Hi Howard,

    but please abandon the idea of barbecuing the populated FR4 PCB in the oven at 100°C...

    Kai

  • Former Member
    0 Former Member in reply to Howard Zou

    Hi Howard,

    I am not sure if I have understood you correctly, but here is my response.  Please let me know if I have misunderstood.

    Yes, it could be broken.  However, it has probably been damaged by the heat beyond the absolute maximum ratings.  Operation beyond this region means that the performance can no longer be assured and the part may be damaged.  If the customer moves to production and exposes their op amp to these conditions, then any return will be turned away because the absolute maximum ratings table has been violated.  It would also be a good idea to check with the board manufacturer whether the customer may be damaging the board.

    Regards,

    Daniel