So I've been trying to figure out a way to regulate the current of an H-bridge and can't seem to wrap my head around it. Basically, I need the current across the load to be at a current (say, 5ma), given a changing load resistance. This could be easily achieved using and an H-bridge built using SPDT relays (or a DPDT) to switch polarity at set time intervals and a digital pot to regulate the voltage... but ideally the circuit will not be using relays due to their noise.
Am I right in saying that this is unachievable using a digital pot and H-bridge IC since the voltage has to be taken from the wiper - which always provides positive voltage? Or would controlling the voltage source of the H-bridge work? (sorry, I've never used them before!)
I've thought of using a current regulator like the LM334z but that won't work in series with the load either since they can't handle reversing polarity......
Any help would be mucho appreciated . Hopefully there's a simple solution I'm missing.
Normally an H-bridge is either full on one direction or the other, or off altogether. For lower drive capability, the H-bridge is pulsed on/off very quickly for different durations, with a PWM signal for example controlling the width of the on-pulse. The pot can be used as an analog level, read it and change the PWM value accordingly.
The supply voltage is not changed; nor are the transistors that make up the bridge only turned partially on, as that makes them operate in their linear region where their on-resistance is higher, vs their saturation region where their on-resistance is lower and they can run cooler.
Diculino:
Ah, of course! Is the math as simple as multiplying the current by the duty cycle?
So for example if the load fully on is measured to be 10ma, then a PWM signal with 50% duty cycle would result in 5ma average across the load?
Approximately yes, and the practical value will not be much different.
Though if you were willing, you could wire a current sense device in series with the supply somewhere either at ground or at VCC, and then bias and/or rectify the output and actually measure your current (Keep in mind that it can go negative (Device dependant) when you are going in reverse. Really depends on your application. If you were just trying to roughly control your output (say you just wanted to roughly control motor speed) then you wouldn't need to do what I am suggesting, and would just have settings at various percentages.
For a current control bridge you compare the actual current with a preset value using a comparator with hysteresis,
and use this signal to control PWM duty cycle... The winding inductance is doing the work for you. Microstepping
stepper controllers are basically two current-control H-bridges for instance.