• State
  • Locked Locked
  • Replies 3 replies
  • Subscribers 50 subscribers
  • Views 266 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

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.

OPA2333-HT: Unity Gain at High Temp

Brandon Dyer1
Intellectual 500 points
Part Number: OPA2333-HT

Hello,

I am currently implementing the bare die version of this part as a unity gain buffer (output directly connected to inverting input terminal) to drive a single ended ADC input (ADC is the on-chip ADC of the TMS320F28335-HT part).  Between the OPA2333-HT output and the ADC input is a lowpass RC filter (1kohm and 10nF).  This part seems to be fine at 100°, 150°C, and even 180°C.  But as soon as the temperature gets to 200°C, the part begins to fail.  One notable failure I have seen, and this is from troubleshooting after failure and at room temperature, is the output no longer exhibits unity gain.  The output will be any where from 100mV to 600mV greater than the expected output, increasing as a try to measure it with a multimeter.  Looking on an oscilloscope, the output will randomly drop down close to the expected output and then increase again.  I have not had any issues at room temperature measuring the output prior to testing at high temperatures, and therefore don't believe that the scope or multimeter is causing the issues I am seeing.  Also, the signal into the op amp isn't changing, but is a constant DC value when measured on the scope (as is the 3.3V supply powering the op amp).

I guess the question I am trying to ask is whether or not you have seen this type of failure on this part before at 200°C?  Also, Is there any compensation required at high temperatures when using unity gain for this part (even if the datasheet says that it is unity gain stable)?

Thanks,

Brandon

  • 0
    TI__Guru 51971 points

    Hi Brandon,

    OPA2333-HT should be good up to 210C, though the certain performance may be degraded somewhat, see the datasheet. 

    My guess is that the RC filter is likely the issue. The capacitor 10nf will reduce its rated capacitance significantly approx. 180C or higher. The voltage divided in RC is kept changing as temperature is increasing. In other words, that RC pole is likely kept changing as the temperature is increasing significantly. If you shorted the RC filter, your issue should go away. Please check on it. 

    You also mentioned that "The output will be any where from 100mV to 600mV greater than the expected output", please let us know where the measurement point. With RC, the output voltage should decrease because of it is a pole behavior. Your output is increasing, which means that you either have a Q effect or zero or oscillation behaviors. But your circuit description is said to be open loop drive, which OPA2333-HT buffer is driving RC filter (15.9kHz) and then connected to TMS320F28335-HT part. If the actual connection is different from the above, please show us the schematic.

    Best,

    Raymond

  • 0
    TI__Guru 62290 points

    Brandon,

    I suspect the OPA2333-HT part oscillates in your configuration at temperature above 200C.  Our macro-model simulate only a typical performance at 25C but despite this fact simulating your circuit shows that the application phase margin drops down to only 35 degrees at 14kHz (see below) - the min phase should be at least 45 degrees.  Of course, all of this occur at 25C so one may assume at 200C phase margin goes to zero causing the circuit to oscillate.

    Increasing the isolation resistor, R1, to 2k improves to phase margin to minimum of 55 degrees - see below.

    Since there is no way to know how different things may be at 200C, you may have to increase R1 further to assure stability at the extreme temperature.

    For stability analysis training video, please click on following link: 

    https://training.ti.com/ti-precision-labs-op-amps-stability-introduction?context=1139747-1139745-14685-1138805-13848

  • 0
    TI__Intellectual 1985 points

    Hi Brandon,

    Adding to what my colleagues have said, at high temp, the device will not be able to swing as close to the rail as it otherwise would when sourcing the same current. We don't have claw curves for 200 degrees, but Figure 7 in the datasheet does show that the swing to the rail can quickly degrade as early as 125C. In this thread, a customer experienced issues when sourcing currents as low as 1mA above 155C. 

    That ADC is a SAR, with about 10pF of typical capacitance at RT (this could also change over temp), and so it will be periodically drawing current to recharge the internal capacitance. Depending on how quickly it is sampling, and the effective capacitance at that temp, you could be drawing enough current that the output would be pulled down - or up, if you are near the low rail. Any change in the output voltage means pulling/pushing more current from/to the filter capacitor. If the output is oscillating, then this effect gets even more pronounced, as you need to keep charging and discharging the 10nF cap in addition to the ADC.

    Between recharging the capacitance of the SAR itself, the added capacitance of the scope, and potential oscillation happening on top of it all as Marek said, there is a chance that the circuit is pulling enough current that the output is not able to achieve the voltage you would otherwise expect. More information on your expected measurement results/test conditions would help confirm/rule out this theory.

    Cheers,

    Jon