opamp dc buffer

[lincoln]

Hello,
I am working on The http://m.eet.com/media/1126947/11805-view_download_the_schematic_featured_in_the_clip_pdf_.pdf]original design (PDF)[/url] has some features that I'm not interested in so I have simplified it some what. So far I have sourced a 10K Kelvin Varley divider and prototyped

.

The issue is that I am seeing an voltage offset out of the buffer. If I tie the non inverting input of the lt1881 to ground  I see a vout of -19uV. Tying the non inverting input to ground through a 10K and 1M resistors, the output goes to 12uV and 87uV respectively. Are these reasonable numbers for the lt1881? The numbers for 0 and 10 K ohms look reasonable but I cant explain the 87uV  for 1meg. This makes me think I am doing something wrong. This is after washing the flux from the board.

Realistically the max input impedance is 10K  and the offset voltage in this range is acceptable but if there is some thing i'm getting wrong i would like to correct it.

Thank you,
Link

[Artraze]

The LT1881 has an input offset voltage of ~30uV typical, so -19uV is well within spec.

Similarly, the input bias current is ~100pA, so that through a 1M resistor would give an additional 100uV offset, which is pretty much what you're seeing.

So all in all what you're seeing is about what one would expect to see for a LT1881.  It's maybe not the greatest device, but the design is only meant for 10V to 5 decades, which means a basic uncertainty of +/-50uV so it'll certainly do the job.

[David Hess]

The LT1881's CMRR and PSRR could add up to another 6 microvolts per volt of offset error as well.

The LT1881 is a good choice for this.  It looks like Linear Technology took their LT1008/LT1012/LT1097 series and added a rail-to-rail output stage.  I wish they had included a schematic in the datasheet.

[lincoln]

Well after months of procrastinating I hit send on the Gerber files. Looks like the internal routing didn't make it through the export,  opps. here is the first board fixed:

[lincoln]

Well, f^&*,  put it together, probalby should have soldered it in stages,  the board trips current limit on the power supply. Anyway, reference is beautiful. Dekapot is fantastic.

link

[Alex Eisenhut]

OK, what's a "Tecktronics"?

[lincoln]

this:


plus this:


EEVblog #480 - 300MHz Owon SDS Oscilloscope

[LukeW]

Wow, for a second there I thought that was a new leader in the cheap Wun Hung Lo DSO market.