I want to measure the resistance of the resistor that oscillates with time:R(t)=R0+RAcos(ωt).R(t)=R0+RAcos(ωt).The idea of measuring resistance is to use constant current supply and observe voltage change, as it is shown in the figure. The voltage drop across the resistor is proportional to the resistance as long as the current is constant:R(t)=u(t)i.R(t)=u(t)i.The described method works as long as the current remains constant. However, I don’t know how to achieve a constant current supply that doesn’t oscillate with resistance. The original idea of the current source was transistor circuit with resistors. LTspice simulation of such circuit showed, that current oscillates with resistance oscillation. Is there any relatively simple way to achieve constant current supply that doesn’t oscillate significantly? The oscillation frequency range is 0-2 kHz.
- 1search precision constant current sink with opamp and you’ll definitely find plenty. – Marcus Müller 39 mins ago
- Ah, let me see. (1) A Constant Current Source (CCS) by definition supplies a constant current, say 10mA, no matter what. So of course it’s current i(t) = constant, won’t fluctuate with, in your case, it naughty resistor neighbour R(t). (2) A good reference to CCS is the “Current source -Wikipedia” en.wikipedia.org/wiki/Current_source, describing the following implementations: (a) Passive, (b) Active without negative feedback, current source with negative feedback, … / to continue, … – tlfong01 2 mins ago Edit
- (c) Constant current source with thermal compensation, using two NPN BJT (as a negative feedback current mirror, I think, 80% sure :), … (3) Versions (a) and (b) with with no negative feedback current mirror has a poor performance, for sure would damage the reputation of EE SE Forum. (4) Me poor lazy hobbyist would recommend newbies to try (c), and if you are also interested in (c), you can give me the user requirements, like Vcc = 5V, Constant current = 100mA etc, I can try to design a simple circuit for you. Cheers. – tlfong01 57 secs ago Edit