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.

THS3491: Min/Max values or normal distribution for HD2 and HD3.

Part Number: THS3491
Other Parts Discussed in Thread: OPA2677,


The Typ values for HD2 and HD3 are listed on page 6 of the data sheet.

Could you tell me the Min/Max value of this value?

Or could you tell me the sigma value of the normal distribution because I want to know the variation between HD2 and HD3?

Best Regards,


  • Hi Nishie,

    The way we get Min/Max values is based on a gaussian distribution up to +/-4 sigma and is typically applied to DC specs. Prior to distribution, we typically only test the part's DC performance, and we use the Min/Max to set the threshold for which parts will be distributed and which parts will be discarded. Since harmonic distortion is not tested prior to distribution, the AC specs we display in the datasheet are derived from lab characterization or simulation therefore Min/Max is generally not applied. 

    Best Regards,


  • Many years ago, we made an effort to provide detailed over temp and process AC spec limits. That was using process case limit libraries and their temp effects. Designers and management hated this, so it was dropped, but if you want to see an example - look at the OPA2677 datasheet. 

  • Yea, I was just reminding myself how insanely dense the information was here - a big part of how tight the limits are is the supply current trim and then thorough DC testing going ou the door - if the supply current and Zol and input error tests are good, it is very likely these AC limit specs will be met without testing, the 25C limit column is just process case, then the over temp spans are process and temp variation simulations. Helps to have very good process device models. ANd yes, the THS3491 does have a trimmed supply current which helps everything 

  • Also one very important statistical nuance on this type of spec. Things that are report in dB terms need to be converted to uV/V terms before you do any statistics - incredibly common mistake in development groups. If you use dB, the spread is erroneously wide leading to false min values. just in case you ever needed to deal with raw data. 

  • Hi Ignacio-san, Michael-san,

    Thank you for your support.

    I understand the AC and DC characteristics.

    I will tell the customer what you said.

    If I ask you an additional question, please let me discuss it again.

    Best Regards,