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.

LMH6702: Interfacing LMH6702 with DAC5681Z

Part Number: LMH6702
Other Parts Discussed in Thread: DAC5681Z, , THS3217, THS3215, THS3491

hi,

I am interfacing Dac5681Z with LMH6702 .Output required is 0v to 3v Unipolar. PL confirm whether circuit attached will work.

Thanks

SeemaBEL_LMH6702.tscDAC.pdf

  • Hi Seema,

    why don't you follow the recommendation given in section 7.3.13 of datasheet? The DAC5681Z has differential outputs and it's wise to use the recommended transformer, for instance for optimally suppressing common mode noise.

    Kai
  • Hello Seema,

    Yes, if you have an AC coupled requirement a balun is best with your DC offset brought into the centertap on the secondary. If you need DC coupled output, LMH6702 is a good solution. I would note that for best HD suppression, you need to use both sides of the DAC output. The THS3217 was developed for this application. What you showed looked reasonable for a single sided DAC output solution.

    A range of output solutions was featured in this TI Design, 

    http://www.ti.com/tool/TIDA-00684

  • Hi Michael,
    Can you suggest modifications required in TINA circuit if both outputs of DAC are to be used?Output range is 0v-3v .
    Thanks
  • Hi Kai,
    Instead of using the transformer , If I use both outputs of DAC and feed to the LMH6702,will it help?In that case will there be an offset in output?
    Thanks
    Seema
  • Well the single op amp complementary to single ended is surprisingly messy algebra wise. Part of the difficutly is that the inverting input side will show an active impedance due the dependent signal from the V+ side. To introduce an offset, that would come in from a wideband buffered DC level on the V+ side that would normally go to ground for a bipolar output signal. There was an update to an earlier app. note I did - let me see if I can find it,  that original app note appears to be gone SBAA135

    But - 

    we fixed all these issues with the D2S stage of the THS3215 and THS3217 - it would be much easier (for you) to just use that? 

  • Hi Seema,

    I think, if replacing the transformer by a differential amplifer would be a good idea, TI would have recommended this in the datasheet. But I don't find any hint that a differential amplifier can be used instead of the transformer. Even the EVM has the transformer.

    Seema, Michael is much more skilled with this sort of amplifiers than me. Maybe he can help you further.

    Kai
  • Ok all,

    these comms DACs are always characterized with just a balun output to get the best SFDR for datasheet reporting. If, however, you need a DC coupled output you have to go an amplifier solution. And yes, I have suggested for decades that the DAC folks support that in the datasheet - no luck so far, so we amplifier folks have to do it. The most ubiquitous requirement is of course AWG outputs that usually need to support single ended DC coupled outputs - those are also complementary DAC's with some kind of amplifier solution - in defining the THS3217 and THS3491 I visited probably 15 of those kind of guys to extract those definitions

    We used to do single op amp diff to single, but that is pretty hard to get good SFDR with - hence the THS3217. In our testing, the LMH6702 was the best at that industrywide and I do have a slightly flawed design flow for that - waiting to hear back on where the updated one is. I vividly recall discussing this with the guy working on that, and he did do something better.
  • It appears that this app note is now removed from the TI web site.There was an excel design tool that allegedly went with it, but back in that time I password protected those things and forgot the password to update - so I suspect that is removed also. But you can get a preview of how unobvious this task really is due to the active impedance on the inverting input side. So what the new update was supposed to to, match the loads on each side of the DAC output (for SFDR suppression) give the same gain magnitude for each side, and DC level shift if need be. I kind of doubt there was a tool built to go with it. Adding the buffers at the input of the D2S stage in the THS3217 essentially isolated the DAC load problem from the D2S problem.

    I found it here,

    docplayer.net/63068968-Wideband-complementary-current-output-dac-to-single-ended-interface-improved-matching.html
  • hello all,



    We have not found the updated version of this analysis yet, but here is a better link to the original one. This shows the algebra involved for this apparently (but not) simple interface,



    www.ti.com/.../sbaa135a.pdf