Adjustable AC current limiter for output of Variac for safe circuit testing

Question

I would like to add a variable AC current-limiting circuit to the output of my Variac to use when testing circuits so that I can limit the maximum AC input current in case of shorts or other issues.

Even low cost, variable DC power supplies typically have a separate knob for adjusting the maximum DC current output -- I would like to create the same thing for the AC Variac output.

I know that in the past people have used a combination of incandescent light bulbs of various wattages to create current limiters with a discrete number of steps. However, such contraptions are big, fragile (bulbs break), and hard to source now that incandescent bulbs are being phase out.

So, is there an all-electronic circuit that could sense and limit the current output to a variable value based on say a potentiometer. Ideally current would be clamped but I could also tolerate current being shut off.

Any suggestions?

I would be more worried about the lack of isolation from a variac. What are the currents you're interested in working with? 10s of mA? 10s of A? How sensitive do you need the cutoff to be? How quickly does it have to react? I would take a second look at incandescent bulbs, they're still available, still cheap, and yes they break but so do a great many other things on your workbench. — vir, Mar 22 '23 at 20:55
Self resetting or manual reset? Based on RMS current or current threshold? What about factoring in the unusual shape that a lot of AC loads impose on the current they draw? — Andy aka, Mar 22 '23 at 21:07
Was just reading about PPTC (Polymeric Positive Temperature Coefficient) fuses which are known colloquially as "resettable" or "self-recovery" fuses -- but they really seem to be very similar to the light bulb behavior except that they (1) are smaller (2) have a steeper resistance vs. temperature curve (3) are less delicate.
Couldn't I even make a nicely selectable one using a rotary switch connected to an array of different values?

Indeed on Aliexpress, it seems like I can get an assortment kit for pretty cheap and have a large range... — JK1, Mar 31 '23 at 21:37
Presumably one could put several identical current devices in parallel to get higher current shutoff values (and indeed once one device starts shutting off power, the current to the other parallel devices will increase, thereby accelerating them shutting off too!) — JK1, Mar 31 '23 at 21:43
And one could add an LED indicator across the PPTC (together with a diode and a resistor) which would then illuminate when the PPTC starts cutting off and has a sufficient voltage across it — JK1, Mar 31 '23 at 21:56

Transistor · Answer 1 · 2023-03-22T21:26:19.397

1

The circuitry required would be non-trivial. An obvious idea would be to use a triac control but at 90° conduction you've already got peak voltage and anything with a rectifier will at that point have full DC voltage. That makes it useless for SMPS supplies, etc. In addition the triac won't turn off until the end of the half-cycle.

It might be possible to use an audio amplifier driving a step-up transformer but again this is complex.

The incandescent lamp solution is rather elegant in its simplicity. At low currents the filament resistance will be 25% to 10% of its hot resistance so it doesn't cause too much voltage droop. At high or fault currents its resistance quickly rises with the filament temperature giving the desired limiting response and much better than a simple resistor. You might be able to get the same type of response with nichrome heater element wire but I haven't done any calculations.

Another possibility is to monitor the AC current and compare it with a scaled copy of the AC voltage.

schematic

^{simulate this circuit – Schematic created using CircuitLab}

Figure 1. By adjusting R2 the comparator can be set to trip if the resistance of the load is too low. (This is a block diagram only. A real limiter would need some detailed design.)

Note that this might work well with a resistive load but be problematic on an inductive or capacitive load.

edited Mar 22 '23 at 21:26

answered Mar 22 '23 at 21:14

Transistor

175,532
13
190
404

You could rectify the CT output and compare that to a DC level. Also, a latching circuit would be needed to avoid reclosing the relay when the current is interrupted. You could also use a pair of MOSFETs back-to-back, driving the gates with an optocoupler. The current detection and trip operation could be nicely handled with a microcontroller. – PStechPaul Mar 22 '23 at 22:01
1

@PStechPaul, I would want to rectify both the CT output and the VT output and compare them so that it would trip early in the half-cycle. About 20 years ago I designed and built a circuit very similar to what you are describing with MOSFETs, opto-coupler and microcontroller. In normal operation it switched on at the zero-cross and by means of analog squaring circuits compared switched off the output when the mean-squared current value reached the square of the RMS current setpoint. It worked well. – Transistor Mar 22 '23 at 22:19
I'm not getting why you need to compare the output current to the voltage. I assume the PT and CT are on the output of the Variac, but they will be out of phase unless it's a pure resistive load. The OP seems to want to be able to preset a given maximum current value and trip when overcurrent is sensed. If very fast response is needed, it should be possible to monitor sequential samples and trip on rate of rise (di/dt), like current limiting fuses that protect within 1/4 cycle. – PStechPaul Mar 23 '23 at 00:35
@PStechPaul, by comparing the ratio of voltage and current we're really doing a resistance (or impedance) measurement. That means that we can detect the overload condition very early in the half-cycle. This would work with non-resistive loads as well but it could struggle with rectifier loads which will have current peaks when the diodes conducts. – Transistor Mar 23 '23 at 22:39

Adjustable AC current limiter for output of Variac for safe circuit testing

1 Answers1

Linked