200nA Precision AC Constant Current Source

Eduardo,

It's not clear how 50nA of DC current would allow single supply operation. Did you mean 500nA? What is the impedance of your load?

A Howland type current source is probably good solution. There are various postings on Howland current sources that may help. If you provide more details we may be able to provide further help.

Regards, Bruce

Hello Bruce,

Above I mentioned using DC current of 50nA for single supply but what I meant to say was "If you can recommend a circuit that allows for DC pedestal current, I will use a single supply, If it cannot supply a DC pedestal current, then I would use dual +/-3.3V supplies." - Just wanted to mention that I have that flexibility available.

If I could use a DC pedestal current, then it would have to be 100nA, with an AC current of 200nA(p-p) in order to achieve the 200nA(p-p) spec using a single supply.  I view as an AC current of 200nA(p-p) riding on a 100nA DC current.

The load impedance is high and will vary from about 180k to about 600k.

To elaborate more on what my final goal is to have a stage where I provide an AC voltage in, and get an AC current out with the specs mentioned above for the current out.  I am providing a similar function to the "AC leads-off detection" feature provided by the ADS1298 where it injects the patient with an AC current waveform to detect the presence of a lead on the patient.  I want to provide this functionality but using discrete parts.

If more convenient, you can call me at the office at 407-455-6189 to discuss, or we can continue through TI E2E.

Thanks for your time,

Ed Rey

Ed,

I'll give a call to explore possible approaches. I'll also suggest to other possible readers that the precision data converter group may have explored various ways to implement this technique with the ADS1298.

Regards, Bruce.

Ed,

I'll post this schematic here and discuss it with you off line. In short, I'm thinking this approach may be better than a Howland at your very low current. The sense of the input is reversed... a 3.3V input from the DAC gives zero output current. To reverse it in analog seems unnecessary from your description. Leakage current may be an issue. We will discuss off line.

Regards, Bruce.

Hello Eduardo,

Bob Pease wrote an Application Note about the Howland Current Pump. You may find it enlightening...

Application Note 1515: A Comprehensive Study of the Howland Current Pump

 http://www.ti.com/lit/an/snoa474/snoa474.pdf

Regards,

Paul Grohe

SVA Precision Applications