Burst Firing for an AC Voltage Controller - What is the supply current rating?

[123MB]

We are working with a heating system which controls heat delivered to resistive elements by chopping out whole cycles of the AC voltage waveform. I.e. the SCRs conduct for one cycle, then stay off for 3 cycles, and then conduct for one cycle, would represent 25% heating power.

It has got me thinking about what the RMS current rating of the supply transformer would be for this machine. Effectively it draws, say, 1000A RMS for 1 cycle, then zero amps for three cycles, then 1000A RMS in the next cycle.

We can ignore other potential issues like flicker and protection coordination associated with this, and assume the load has unity power factor. In theory, could this load be connected to, say, a 500kVA transformer (700A approx)?

I am aware of standards for rating of transformers associated with overloads but most of these are in a much larger time scale, i.e. minutes - for example, the method for 'converting an actual load cycle to an equivalent constant load' in IEEE C57.96 would say that in a 1000A for one cycle 0A for 3 cycle scenario the equivalent load is 500A which means a 750kVA transformer is ok. Do you think this is applicable to this application?

Cheers.

 

[waross]

You must size the transformer for the full, continuous load.
There are operating situations where the SCRs may be conducting for every cycle.
Some-one leaves a door open on a really cold day.

--------------------
Ohm's law
Not just a good idea;
It's the LAW!

 

[jraef]

Zero Cross Variable Time Base SCR firing results in a change (lowering) of the average VOLTAGE that gets to the RESISITIVE load. But no matter what, the maximum CURRENT that can be drawn is dictated by the resistance value. So regardless of the burst pattern, the maximum current seen by the supply transformer cannot be greater than I=E/R, where E is the applied line voltage ahead of the power controller.


" We are all here on earth to help others; what on earth the others are here for I don't know." -- W. H. Auden

 

[Comcokid]

You really need to size the transformer for the full 100%. Yes, the average current may be less, but the issue is the transformer core. To drive the 1-cycle sine the transformer core must not be pushed near saturation.

 

[123MB]

Thanks all -

Comcokid - Really - how can excess current cause saturation? Transformers are typically allowed to drive overload currents without mention of saturation. Excess voltage or DC offset causes saturation right?

Jraef - Understood the maximum current can't be greater than this number, but the equivalent constant transformer load is less than this number.

Waross - Understood - in these systems, cycling for every cycle won't be seen for years, until the heating elements age sufficiently to require more voltage to produce the same heat. The client's need is to run the system with the smaller supply rating until this occurs.

 

[waross]

Remove half of the heaters and fire twice as often.
If you are only triggering one in four cycles, the heater is grossly oversized.
Select a reasonable sized heater and size the transformer at 100%
Don't ever run that system on the standby generator.
You should switch over to a simple thermostat control when running on standby power.
That current pulse hitting the generator at regular intervals can and has destroyed the engine bearings in a short time.

--------------------
Ohm's law
Not just a good idea;
It's the LAW!

 

[waross]

cycling for every cycle won't be seen for years, until the heating elements age sufficiently to require more voltage to produce the same heat.

Paying 400% for something that may never happen would be seen by some as wasteful.
Long before the heaters degrade to 25%, if they ever do, they should be replaced.
And long before they degrade uniformly to 25% they will probably degrade non-uniformly and burn out.

--------------------
Ohm's law
Not just a good idea;
It's the LAW!

 

[123MB]

Appreciate your advice Waross

As I understand, spare capacity in these systems is the norm - referred to as voltage reserve. One could disconnect some elements, but I can see issues like the layout of the furnace and heat uniformity being an issue.

 

[waross]

As I understand, spare capacity in these systems is the norm

4:1 capacity on a mild day is to be expected.
4:1 capacity in anticipation of a possible future failure is not the norm.
I have worked with several time based modulation schemes, but it has been a long time.
I was aware of some of the first cycle controlled heating systems.
The heating was the same heating design as for a legacy on-off controlled system.
Heating is quite simple; The BTUs input must equal the BTUs lost, plus a safety margin.
A 400% safety margin is excessive.
A possible excuse is if the smallest economical heater available is much larger than needed.
If that is the case, then 100% transformer capacity for a rather small furnace is not a serious issue.
If the furnace has the ability to overload the transformer, it is not a case of IF the transformer will be overloaded.
It is a case of WHEN the transformer will be overloaded.
Just ask Mr. Murphy.

A possible exception may be generator capacity for a standby generator.
I have mentioned the possible hazard of cycle based modulation and diesel engines.
I have designed load shed schemes and load selection schemes for standby generator installations when the expected load exceeded the generator capacity.
I would consider a scheme where 2/3 of the heating elements were disconnected and the remaining elements put on on/off thermostat control, automatically when on standby power.

Another issue with an undersized transformer is usability.
If you have other loads on the transformer, or if the transformer upstream is loaded to near capacity, the switching on a small transformer may cause unacceptable flickering of the lights.
Just bite the bullet and specify a 100% rated transformer.

--------------------
Ohm's law
Not just a good idea;
It's the LAW!