On Feb 22, 6:20 am, usene... from out-of-phase.de (Christian Wilms) wrote:
> > Why would you use a constant current, vs a constant voltage stimulator
> > (or vice versa) for exciting nervous tissue?
>> My (admittedly simple) understanding is that the current output is less
> dependent on changes in the resistance of the stimulating electrode. As
> we use microelectrodes (glass) for stimulating, changes in pipette
> resistance during longer experiments does present a possible problem.
>> Anybody have a more throrough explaination?
>> All the best, Christian
I agree with this interpretation. Here's how I think about it:
1) To fire an axon you need to create a voltage gradient across its
membrane, but you need to do that from the outside, by applying an
2) The only way to apply an electric field is to create a charge
source/sink pair, i.e., you need to pump charge between the poles of
3) If you want the *field* to be constant, then you need the current
to be constant.
4) HOWEVER, applying a constant voltage to your stim electrodes will
deplete charge carriers (e.g., Cl- ions in the case of a Ag/AgCl wire)
because they need to hop off into the soln in order to carry the
5) As the charge carriers are depleted, the resistance increases,
therefore the current drops, therefore the applied field that the axon
sees drops. Ergo, a constant voltage source will *not* produce a
constant electric field at the axon.
6) In contrast, a constant current source uses feedback to adjust the
voltage in just such a way that the current is maintained regardless
of what happens to the resistance (until you hit the compliance limit
of the device). So it will keep the applied field constant.
Now, having said all that, I don't really think it matters very much
which type you use. Your usually using really brief pulses, and
depletion of charge carriers probably won't be a huge deal. I've used
both types, and they both do the job just fine. You can probably
reduce the problem further by conditioning your electrodes regularly
(e.g., re-chloriding them if they're Ag/AgCl), and by using a bi-
phasic stimulus pulse (so that depletion and replenishment cycle
naturally with each pulse). On a logistical note, constant voltage
circuits are a lot easier to design, and therefore probably a lot