OPA836: Updated complementary to single ended design flow?

Hi Michael,

I haven't been able to find anything, but I have asked around and will let you know if they turn up.

Best,

Sam

There was a small error in my original design spreadsheet that I was pretty sure the Bangolore group had taken on a fix. I was off working on the THS3217 at the time so I had lost interest, but an improved flow would be useful sometimes. 

Ok I'll check with them as well.

Best,

Sam

I would be interested in the updated spreadsheet as well.

Hi All,

Unfortunately I haven't been able to locate a copy. 

Best,

Sam

I did find this...  http://www.ti.com/lit/zip/slyt368

I would guess Jim Karki, the link came from the SLYT368 appnote.

Thanks Nate, I appreciate the help tracking this down.

Best,
Sam

Well I am working with Jim's updated and my original spreadsheet right now, as near as I can tell they give the same R solutions s- so far, I think he just used my solutions - sadly, we both locked the spreadsheets and have no way of opening now.. And the problem that I had noticed later was the swings are offset at the DAC output pins - they have the same amplitude but the side going into the inverting input is shifted positively - I don't think I had accounted for that - but, at least my original spreadsheet does give matching input impedances and gains. 

SLYT368 does at least note the shift you mention - look at Figure 18.  The spreadsheet seems to calculate the same - VDAC+max and VDAC-max are not equal in my design.

Or do I misunderstand?

I'm working to set a DAC and buffer up for single-supply operation, by applying a positive voltage on the grounded side of R3 as a "midrail offset".  In my application I do not need negative voltage.  I can see where being able to start with equal swing and having an unlocked spreadsheet would be useful!

Yea, I noticed that - he is just reducing the target average output swing on the DAC pins to keep the V- dac pin in range, but not really trying to solve for how to get that or how to balance it on the DAC output pins. It would be better to insert the offset voltage into R3 and try to get the swings at the DAC output pins matched in amplitude - that will certainly pull the min value up off of ground for single supply and it might be hard with that one change to get matched gains still, but a gain will result and perhaps it can be set to the overall desired value. It is this kind of stuff that led me to abandon this circuit and go on to what is inside the THS3217 design. 

Well Antony, that password did not work for me to unprotect this spreadsheet - why don't you should send an unprotected version if you think you have it unlocked. 

Hi Michael, here is the unprotected version. the password was "AABBBBABABB>", it appears ">" cannot be used in a filename.

DACdifferential_singledesign_unprotected.xls

Thanks Antony, turns out this lower right equation is one of the more important and I wanted to add this constraint up in the data entry area, Lot's of times the flow finds only negative R solutions. 

Oh and incidentally, way back at the top I had said I thought this was updated - so 

2005 I did the original one that you have attached

2010 Karki updated mine using the same quadratic but shifting the target DAC output swings down

Later, I thought the Bangalore team had done a new one as I had mentioned I thought my original had a mistake in it and was not actually aware of Karki's 2010 update. So there may still be a circa 2016 kind of update that may never have gotten to the public.