TI BUF634A headset driver evaluation

I have a TI BUF634A evaluation board for a headset driver circuit evaluation - I am adding a OPA1656 and using the BUF634A as an output buffer. The eval board with 1K gain/feedback resistors (non-inverting) set it for a gain of 2. Newbie question:

  1. I don’t understand the implications of gain options when testing with the QA401. I see that the output signal is so high that the QA401 goes into a constant clicking mode. I drop the input signal level to the eval board and it then handles it but I am not able to get comparative graphs that I got using an INA1620 that was closer to unity gain. I could change the gain by changing the design (adding series input resistor) - but should I or is that another approach that I should use. I can’t get my head around what I should do when QA401 signal output and the eval board output level differ.

It would be good to see a paper/video that addresses the various gain/level settings in the software and when each is used:
dBV button External Gain input/output
Max Input Level +6 +26db (20dB atten pop up on graph)
user weighting
(Gen 1/2 level setting of course)

Hi @bklein, an attenuator, especially with a headset amp, is probably the solution here. When you are testing, you are likely driving into a load that resembles a headset impedance (maybe 16 or 32 ohms).

The link below is the same problem you are facing–how to measure something much larger than the input you want to use. This is common because the most sensitive setting on the analyzer also have the best noise floor. So, you knock down the signal out of the amp before it goes into the analyzer.

If your case, if the max output of the headphone amp was, say, 8 dBV, and you wanted to use the +6 dBV input range, then you’d maybe use a 8 dB attenuator to get it where you wanted (about 6 dB below max input).

Take a look at the link below and let me know what you think

I got confused by all the gain options but in my particular case for this second headset driver option I can modify the test setup test range by 6dB and seem to get acceptable results. Or I could add a series resistor or remove the neg input resistor (haven’t seen it done with this part though) to get unity gain like the first part I tested has. Not a lot of power though so I can try the attenuator too. I was just told to work on something else though so my results will have to wait. Thanks for the help.