I would like to build a circuit to stimulate metal electrodes with biphasic (square wave) current pulses (in the uA - mA range).
The stimulation would be a negative pulse (+40uA, 200us), then 10 us (0A, inter-phase), then a positive pulse (-40uA, 200us). The frequency would be 1500 PPS (pulse per sec), so one "pulse" in this case would be 666.66 us (negative, inter-phase, positive, then 0A the remaining 256 us). Similar to the uploaded picture. The measurement time should be long enough (hours).
Also I somehow would like to measure the current during the pulses and the potential response of the electrode, then save it to txt files.
How would I even start with it? Is it possible to build such a device? I already have an UNO and an L298N module.
OK, so I am curious (as most around me know only too well ).
What do you propose to stimulate with the metal electrodes? I doubt the electrodes themselves will care much either way for a little current flowing through them.
Yes. Something like a current mirror could work, or a shunt & opamp constant current source. You'd have to work it out for both the positive and negative leg, which could be as simple as basically the same circuit being mirrored and both halves being triggered alternatingly.
Google "constant current source schematic" and take it from there. Getting good accuracy at 40uA may be something you'll have to think about for a minute but should be feasible.
Is it a problem if there's a DC offset on the signal? It would make component choice and circuit design a little easier. Although it can also be done without much problems with a 0 bias.
Heck, if the DC resistance of the coil is constant and known, you could get away with a voltage regulator, two resistors as current limiters and two mosfets as switches.
The characteristics of your electrodes needs to be known (even if it has large variability).
One will need a voltage supply adequate to generate such currents. The variability will also determine if the current source can be passive (resistor) or does it need to be active.
It will be for platinum microelectrodes for brain stimulus. I cannot say more about it. I need a current pulse with the aforementioned parameters. I can stimulate the Pt with potentiostat (AUTOLAB), but our system is slow: 30 min net stimulation time (1500 pulse per second, current pulse 60 uA, biphasic) takes 9 hour which is insane, that is why I would like an alternative method, how to do it. I thought about a signal generator, but that cannot offset the two channels (one channel for positive pulse, then one for negative pulse).
).