Soft starting big toroids

I want to start at least 4 toroids of 500va+ and wish to leave the circuit breakers in place on the main fuseboard.

Previous projects have used "only" 330va toroids and they vibrated (quite loudly)when powered up, though they were attached well.

I will have a lot of capacitors to charge and don't want to just dump a bucket load of current directly in to them.

I have seen a circuit that simply short circuits a resistor in series with the transformer primary, but it seems a little crude and i will need at least 4 of them (one for each transformer).

Anybody seen anything slicker?
There aren't too many options. Thermistors are one. Relays with resistors are another. If you happen to be planning regulated rails (unlikely) you can make a soft-start regulator. It's a pain, but there's always the old stand-by of bringing up a really big circuit with a Variac.
It's not the toroids that are the problem, it's the caps behind them. You can hook up a huge transformer and switch it on. Nothing happens. The reflected impedance the AC line sees is essentially infinite. Put caps in there and the equation changes, and the more caps, the worse it gets.

Thermistors are easier to use but harder to pick. Maybe the toroid manufacturer has an idea what to use there.
Another solution that works is putting series resistors between caps. This is has 2 functions
1. limits large incoming currents on turn on
2. works as RC low pass filter in the power supply. The resistors are usually 0.1 Ohms each.
This adds to the series output resistance of the power supply but I don´t think it´s so important here. It´s used widely in a lot of amps.
You can use inrush current supressors and they come in a variety of resistances and current handling capability.
radio amateurs use these soft start circuits for their linear amplifiers and you will find these circuits in the ARRL manuals.
I thought I would use relays with a resistance across them and a resistor/cap combination for a small delay.
I had this problem with soft starting two 300VA toroids and
120,000uf of caps but found a project that Elektor UK produced for soft starting . The authors page is :
and the GIF

Of course you will have to change the components slightly
for 115v or contact the author but the project is a dead
simple resistor/relay combo that powers up through the
resistors to stop the surge and then switches over to bypass. I have built several of these and they work well. Also a friend has been using one of these in his amp for nearly two years now.

Regards Kooze
Would it be possible to build a current limiter out of 2 transistors, and put it between the bridge and the cap bank?
I'd set the current at 120% transformer secondary rating.

ASCII art doesn't work too well here, so I'll describe the positive current limiter. A negative limiter should be easy. If Silicon transistors are used, the voltage drop will be 0.6V - 1.2V (current dependant). Germanium transistors should drop this to only 0.2V - 0.4V, but the circuit will respond slower. All transistors, and the pass resistor, should be on a heatsink.

The collector of the pass transistor (NPN high current) is connected to the input and one terminal of the base resistor (1K). The emitter of the pass transistor is connected to the base of the regulator transistor (NPN medium current), and one terminal of the pass resistor. The base of the pass transistor is connected to the second terminal of the base resistor, and the collector of the regulator transistor. The second terminal of the pass resistor is connected to the emitter of the regulator, and the output.

The current limit is set by the pass resistor. With Silicon transistors, the pass resistor should be 0.6/limit. With Germanium transistors, the pass resistor should be 0.2/limit.

Good luck.
You could put transistors between the bridge and the caps
but probably not advisable as you then introduce another component between the two .My preference is KISS ( keep it
simple *) although I have seen a similar Resister/Relay combination in this position on some amp designs.
With regard to the circuit I found the info to run it on 110v

C1=680nF , C2=470nF , R1=120R
(Capacitor working voltage 400VDC/130VAC)

oh and the URL was supposed to be :

(I love the way IBM's drop the extra l at the end)
I have noticed that when i powered A 1000 Watt Plitron Troid with no load that upon startup the transformer humed for a sec and dimed the Lights momentarly. is this not normal?

While were at it is it possible to stack Troids on top of one another without any problems? Will the Bolt used to mount the Two cause any problems? TIA PPL

I looked at the circuit you pointed to, and I have a few (correctable) issues with it.

1. I feel that the on/off switch should directly control the power into an amp. The diagram you pointed to has the power switch turning on a relay. I feel that this is less reliable than a high-power switch.

2. You should be switching both power lines when the switch is thrown. Otherwise, it's possible to have power in the box. This is especially true in old houses without polarized plugs.

3. You'll have to pick your relay carefully. The problem is that you'll need a relay rated for high current, and with a 100% duty cycle. Note: Duty cycle for a relay is rarely published. Good cooling will help, but you should socket the relay, so that when it fails, the repair will be simple.

Personally, I prefer the current limiter, with Germanium transistors. This is primarily because I don't like moving parts in an amp.

Good luck.
When you say you started the transformers with no load, did you mean that there was no rectification & no power supply caps, or did you mean that the rectifier & caps were hooked up, but the amplifier circuit was unhooked?
If you start up a transformer with *nothing* on the other side, then I'd be concerned if the lights dimmed and it hummed. If the caps, etc. were in place, then all is normal; a transformer with rectifier and caps hooked up is not, strictly speaking, "no load." The caps have to charge, and even if the audio circuitry isn't there, you can count on a minute leakage through the caps. Since the caps are charging, you'll get a fairly strong inrush of current while that's going on, then the current will drop off to nearly nothing.

Thats totally normal.
Big toroids have big inrush current even without load on them. They charge too just like capacitors and this current (excuse my spelling) is Demarage current. With toroids it is normal, because of their design they are a lot faster then normal trafos. So at start up when they are totally empty of magnetic field power they make a large fast current opposite to the one we give them (reacting as an inductance they are) making a Chaos in the electricity of the house till it stabilises. Thats why the trembeling of the lights.
Capacitors charge electric power forming an electric field and at start up they are like a short circuit.
Inductors (toroids too) charge power in form of magnetic field and at start up with no magnetic field they try to make an opposite current to the one that tries to get in forming a "problem" that tries to stabalise for a few seconds.
That´s why it is worse with a whole power supply.
Stacking toroids


There is no problem in stacking toroids. Several amplifier manufacturers do this. You have to be sure that you bolt them down with a single bolt down the center. Do not attach the top of the bolt to any part of the chassis, as this results in a shorted turn and will cause problems.


[Edited by jam on 06-06-2001 at 06:38 AM]

Yes, this includes aluminium. The inputs (primaries) should be hooked in parallel with each other. Try to keep them in phase with each other, this will be important if you try and series or parallel the outputs (secondaries), most transformer manufacturer's catalogs will show examples of different hookups.

I also believe there are sonic benefits to using a non-steel chassis such as aluminium or copper.


[Edited by jam on 06-07-2001 at 07:15 AM]