AC current limiting circuit

[kellys_eye]

I'd like to add a variable AC current trip to my variable AC (isolated) power supply.

A variable set point range from 'zero' to 1A would be ideal and the ability to trip in 30ms would also ensure safety.

Can anyone recommend a reliable circuit or device that could achieve this?

[Herald1360]

Hmmm.... maybe an RCD could be modified to do something? Minimum trip current would be about 30mA, though.


Doing it electronically with a current transformer with a variable burden resistor could provide a suitable ac voltage to be rectified and fed to a comparator/relay arrangement at the expense of some cheap(ish) complexity.

[Diabolical Artificer]

Glasslinger over on Utoob built such a circuit, Xraytony further developed the circuit,see here -https://www.youtube.com/watch?v=SkCMgGuQhYU From what I heard the circuit is less than perfect, see Xraytonys videos for more info

Andy.

[kellys_eye]

Interesting link, thanks for that DA. Not sure if it will scale down to the <1A range I need but worth 'playing with'.

I've managed to find a schematic of a professional isolated variable AV PSU - the Sencore PR570 - and extracted the relevant part of the circuitry for experimentation.

I'll post the relevant bits here shortly....

[kellys_eye]

As promised...

The transistor drives the relay whose contacts are in series with the supply. The current sense transformer is an unknown quantity but could be determined by experimentation but the overall circuit seems pretty straightforward.

BTW this diagram is a 'hacked together' version of the original manual and misses out some bits that aren't necessary to deliver the required functionality.

Comments?

Just spotted - there appears to be a short on the input of the second comparator - this isn't true! I didn't edit that connection out properly - it goes to the metering circuitry.

[Terry_VK5TM]

Specs say 0-4A current trip range and as far as I can see the only component that will affect that would be the current transformer ratio.

Other thing to check would be trip time as I notice it specced as up to 300mS in the details of the unit.

[Herald1360]

To get a trip time <30ms the circuit has to be able to respond (including relay dropout time) in less than 1.5 cycles of 50Hz. Not a trivial performance....

[kellys_eye]

Maybe using a triac instead of a relay?

[kellys_eye]

Originally Posted by Terry_VK5TM

Specs say 0-4A current trip range and as far as I can see the only component that will affect that would be the current transformer ratio.

Other thing to check would be trip time as I notice it specced as up to 300mS in the details of the unit.

If 0-4A is represented by 0-4V in the comparator then surely a simple modification of the set point will limit the range to that required?

Are we talking about a current limt or a current trip? Both could be done easily with a cycle by cycle measurement using a microprocessor. Limiting would need a control device with a high power dissipation, a trip, jut a switch (MOSFETs).

[kellys_eye]

Current trip.

Primary use is to repair SMPS circuits and allow for safe design/testing of same.

[G6Tanuki]

If a current-trip, I'd suggest something like a low-value resistor in series with the device-to-be-powered, then something simplistic involving an opto-isolator which - when lit shorts the hold-on current to a power-relay.

My intrinsic worry with any sort of 'trip' circuitry is uncontrolled re-application of power when whatever-caused-the-trip is removed.

You need to ensure it trips-out and stays-solidly-tripped-out until there's some direct deliberate and controlled re-enabling.

[kellys_eye]

I did intend the trip to be of the bistable type with manual reset. I've also sketched out wiring (using a relay and microswitch) that prevents the output from being enabled unless the variac is set to zero.

I have visions of switching on without noticing what the setting is!

Use an RCD as the basis? Remove current windings (sorry about the pun) and rewind to suite.

[kellys_eye]

You have a point about the RCD MM, I have a spare 2A RCD (ABB version) that might be open to adjustment....

[Diabolical Artificer]

Don't RCD's take a long time to trip? And they're not very sensitive. Here's some more info I found -https://sound-au.com/project108.htm The Glasslinger & Xraytony circuits as mentioned had issues with triggering triacs,wouldn't a very simple approach be better using a lamp in series with the mains and an LDR to switch off a relay? Not having played with LDR's not sure how sensitive they are.

Andy.

[McMurdo]

I'd have thought a faulty SMPSU could blow up faster than a trip could cut power. I'd opt for current limiting. I regard a trip as an energy limiter rather than current limiter.

[kellys_eye]

The low variable start-up voltage input along with a preset 'what you consider trouble' current setting level should protect the majority of any SMPS circuitry. I hope!

But you make a point about current limiting. The simple in-line bulb is the KISS solution but maybe a more exotic one would be to control the variac using a stepper and feedback circuit? This would be an interesting idea to investigate after resolving the basics of my variable supply.

[Rubberfingers]

Television Magazine published a circuit in 1997, for a "Lamp limiter with LDR triggered trip". This implementation has quite a long trip time though, to allow for switch on surges. https://worldradiohistory.com/UK/Pra...on-1997-07.pdf

[dave cox]

If you want to repair SMPS you don't necessarily need an AC current trip.

The first part of the SMPS is just converting to DC anyway, so an isolated, current limited ~350V DC supply should be fine. The series element will need to be a N-MOSFET, at that voltage, plus a 'high side' current sense with a latch arrangement to ensure the MOSFET is fully on or fully off!

A potential issues will be a PSU needing the phase locked 50/60Hz input ...

dc