Are you really fine with IC voltage regulators ?

[tombo56]

OK. As there was too many orders of magnitude between specified 0.01 CV leakage value and measured results, I removed all parts from the bredboard and assembled again voltage reference with RC filter. While reassembling, one probe with spring hook, jettisoned its parts around. Hmm?

Now, with 10 KΩ, voltage drop is 0.035 mV. With 100 KΩ, voltage drop is 0.35 mV. So, Panasonic FC 330 uF/35V measures 30 – 40 nA leakage, comparable to other capacitors. Still very low but it seems OK.

 

[Mark Johnson]

Great, play around with variations. 2R+2C+opamp Sallen Key lowpass --> followed by --> passive RC , might be a candidate.

 

[chris719]

Thanks for your inputs.

I think that with filtering I can get sufficiently low noise and a low drift with an LM/TL431 and slightly lower noise with the SPX2431.

I have made some simulations and it seems like a first order filter with e.g., 10k and 22 uF + an OPA1612 buffer will achieve a lower noise level than the LT3042, even with a TI TL431. At low frequencies the SPX2431 should achieve a slightly lower noise. At high frequencies it will be dominated by the op-amp.
I plan to make the filter with two capacitors in series and a resistor to the midpoint to almost eliminate the leakage current of the capacitor as done in Walt Youngs Ultra Low-Noise Voltage Reference.

Do you know what the converter needs as far as performance from the buffer? I'm not sure about Delta Sigma converters, but SAR application notes often have fairly fast amplifiers to attempt to settle as fast as possible.

 

[JensH]

It will need a low noise and low impedance, but the exact requirements are not specified. I plan to use the ES9822PRO.

 

[JensH]

I just measured the noise of a TI LM431ACM3/NOPBCT-ND, SOT23.

I measured a noise of 14.7 uV, equal to 99.1 nV/sqrt(Hz) at a current of 1.15 mA. The noise at 10 Hz was around 3 dB higher than at 1 kHz.
At 10 mA the noise increases a little (as it was also the case with the other devices), to 15.2 uV, equal to 102.8 nV/sqrt(Hz).

Not quite as good as the Sipex/Maxlinear SPX2431, but better than the onsemi devices.
Once again it was only one sample, so perhaps I should try to compare a number of devices of the same type.

 

[MarcelvdG]

The lowest frequency of interest is somewhere below 20 Hz. It is for test equipment, so in principle as low as possible, but of course there will be a diminishing return of investment somewhere. And other noise sources will dominate at some point.

I plan to use the ES9822PRO.

Please keep in mind that reference noise of an ADC causes noise sidebands (skirts) around the signal, so if you are interested in what the spectrum looks like, subsonic noise could be important when you want to measure non-subsonic signals.

 

[HalFoster]

Bonnie baker (with TI) has written numerous articles on using voltage references, especially in relation to various ADC topologies - well worth the time to Google and read her material.

Hal

 

[jan.didden]

IIRC she wrote a column 'Baker's Dozen', as you say, well worth looking up.

Jan

 

[chris719]

IIRC she wrote a column 'Baker's Dozen', as you say, well worth looking up.

Jan

Another extremely good set of articles is from Derek Redmayne formerly of LT. They refer to working with the LT2387 ADC, which is not an audio converter, but contain a lot of interesting bits.

https://www.analog.com/en/technical...rt-iv-how-to-drive-the-ltc2387-reference.html

 

[PRR]

IIRC she wrote a column 'Baker's Dozen'

Bonnie Bakers are singers, educators, real estate agents, and on and on.

This is probably what you mean: Baker's Best

She seems to have moved here ?

 

[jan.didden]

Yes that's her. As it says in the bio in the About:
"She is also the author of A Baker's Dozen: Real Analog Solutions for Digital Designers as well as coauthor of several other books."
She's not very well known in engineering circles but I've learned a lot from her.
She can pick complex stuff apart and make you understand how it works.
Recommended.

I've also learned a lot from MarcelvdG , had the pleasure of spending a few hours with him in his home town, years ago.

Jan

 

[dreamth]

I've also learned a lot from MarcelvdG , had the pleasure of spending a few hours with him in his home town, years ago.

Jan

He's the kind of guy you never own...especially in a matter of a few hours.I'd probably consider a conversation with him a pleasure after digesting his knowledge for a full decade...a few hours...would be torture for me! His surgical precision feels more like Predator's invisible spear cutting through one's skull to me.And I only knew four people I can say that about , but that's because they indulged into revealing themselves to us all.It was not ment for each of us it was done for everybody as most of us can't really understand an entire phrase of what these people say and they become increasingly tired of talking to walls.Just a kind reminder of true knowledge status...