Continue to Site

Welcome to our site!

Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

  • Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

Looking for advice on low voltage AC current limiter

Not open for further replies.

I am trying to design a current limiting circuit for a low voltage AC power supply for use with my toy trains. The output is 18 VAC and I would like to electronically limit it to 10 amps. I would rather not use fuses or a circuit breaker but an electronic limiter. Am I crazy or is this possible?

I am aware of the vast numbers of current limiting regulators for DC but this is AC output. Could I use a triac in series with a resistor on the output and a circuit that would turn off the triac when the voltage drop across the resistor reached a preset level but turn it on when the voltage was below that level? Is there a simpler and more eloquent way to accomplish this?

Breakers just don't seem fast enough and I would rather spend my money on a circuit than a bunch of fuses. (They blow as soon as there is a derailment or other glitch).

What are your esteemed opinions? Thnaks in advance.

Use a small transformer with 1...2 turns as primer, the secondary not critical.For breaking a triac good choice, but You need additional timer for reset. If not enough the idea i can draw.(for me simplyer as writing)
Hi Sebi. Thanks for your reply.

Are you suggesting that I use the transformer primary as a a reactor in series with the current flow? And then control the current flow with the secondary?

Perhaps, if you would be so kind, it would be better for you to draw out your idea when you get the opportunity. I would appreciate it.

Thanks again.

Hi Mark,

Off hand i dont recall any circuits for limiting AC current.
So ive made one up.
I havent built it.
Its quite simple as such circuits go,
you would have to experiment a bit with the values.
Controlling low voltages can be a pain.
You may only need one diode in the base,
you may find that 2 ohms is a bit high,
you may find the dummy load is unnecessary,

Its only a skeleton to show how i would approach this
current limiting question.

I hope its self explanatory, its made around a diode
bridge rectifier so as to use a transistor to pass AC
through the load, the load current is supposed to start
affecting the transistor at around ten amps.

Best of luck with it,


  • 10AmpLimiter.jpg
    36 KB · Views: 1,789
John, Your idea is good,but serial 2ohm and 10A is 20V loss. If an shortcut occured the dissipation 180W! I've modifyed Your circuit, hope works fine. First i've thinking about current transformer and triac (as switch), but my circuit can only switch off by overload,but no dissipation.
In case of current
limiting need great heatsink for transistors.


  • CurlimAC.gif
    104.5 KB · Views: 816
Hello John & Sebi,

I set out to breadboard John's circuit and decided early that the resisitor should be more like .2 ohms instead of 2.

Sebi's revisions appear to provide a lower total voltage drop across the regulator cicuit and does not require a power resistor. If I can find a transformer that puts out 20-24 volts, there would still be close to 18 VAC at the output.

Sebi, you are right about providing plenty of heat sinking for the pass transistors (2N3055's). I will still use circuit breakers as a safety measure but hopefully they will not be required to trip if all goes well with the regulator.

You folks have been very inspirational. Thank you.


PS: Someone told me there was a circuit shown in an older (early 1990's) issue of Electronic Design News. Reportedly it has current limiting and short circuit protection for low voltage/high current circuits such as these. Unfortunately it is too old to be on EDN's online archive. Where would one look for older issues of this magazine? Libraries? Schools? Any ideas? Thanks.
Hi Sebi,

You are quite right, 2 ohms was a mistake.
I was thinking 2 volts (at 10 Amps)
and put it wrong without noticing,
so i should have put 0.2 ohms.

As to the dissapation, yes a short limited
to ten amps would give 180 watts,

unless you made the cicuit cut the supply for
an interval, then retry.

I see you have recommended two 3055s to cope
with that presumably.
I reckoned that one would do as the supply
probably would drop from 18v to about 10v in
the event of a short.

Thats ok, two is probably better than one.

I see you have a 0.2 ohms in each leg, which
should give a volt at 5 amps each, by putting two
further stages of gain to the feedback, you have
made quite a sharp step or cut-off to the point
on the 'knee' where the current limit occurs.

However the emitter-base voltage at the point
where the small transistor starts to conduct,
is probably around half a volt (dunno exactly)
So maybe the resistors for that, the 0.2 ohms
could be a bit lower value.

Just by the way, i would have joined the emitters
also on the 2N3055's and used a single resistor
as a current sensor. Making low value resistors
is a pain, making two the same would be awkward.

I usually cut up an old food tin
using household scissors and make a long thin
strip by cutting in a spiral. I find they accept
copper wire soldered on quite readily.

Thats how i make very low values.

For ultra-low values i use Mig Welding wire,
and put ceramic beads on it, and use crimping
to mount it.

Unfortunately at higher currents the heat affects
the resistance, so if its for an indicator it has
to be large so as it doesnt get too hot.

Thank you for the modification to my circuit,
it looks fine to me.

Maybe a small preset could be included as the
small transistor would take very little current
in the base lead?

Maybe as shown here?
What do you think ?

I think it gives a degree of control to the maker,
so the 10 amp point could be set without changing
the low value resisistor too many times.

Regards, John


  • Preset.jpg
    6.2 KB · Views: 773
Hi Mark,

Ive just seen your last post,
Let us know how you get on.

I was wondering about a circuit to cut the supply,
as Sebi said earlier, then to re-try, and cut off
again if the excess current still existed,
and to continue to try every few seconds untill
normality was restored.

This would give minimum dissapation.

It depends how far you are willing to go
with your electronic construction.

Regards, John
Hi Sebi,

I wonder if you would mind looking over this
It may not be possible to get an SCR with suitable
hold-on currents,
i havent fiddled with circuits like this for years,

I would value yor opinion, maybe guess some values?

I hope it is clear how i mean it to function.

Regards, John


  • SCRtriac.jpg
    66.7 KB · Views: 694
Not open for further replies.

Latest threads