AC voltage control in transformer secondary using Triac in priomary

[keshab.patel]

 I am using a step up transformer of input 240V and output 2400V 2A. I am using triac in primary to control the output dc voltage in secondary. The problem is the secondary waveform is not as the same replica of primary. Even I am using two triacs, one for power switch and other to turn on the first triac. The input and output waveforms of transformer are attached.

[oldway]

The only explanation I can find is that there would be a low primary voltage assymetry between positive and negative half cycles, and this leads to a primary direct current that saturates the core.
But nothing visible on the pattern.
Try to measure the primary current with a DC+AC clamp.
First test your triac control with a resistive load. (a 100W incandescent lamp for example.)

Triac primary phase control of a transformer is not an easy task.
Control must be very acurate and transformer core better has to have a little gap for not saturate to easily.

[keshab.patel]

  Its working fine with resistive load. The primary current is around 3A.

[jebcom]

I'm having trouble making sense of the 2 traces. I'd like to see them both on the screen at the same time. It looks like you have a 4-channel scope. How about one channel primary voltage, one channel secondary voltage, and as I think oldway is suggesting, one channel with a clamp-on current probe, showing primary current. At that secondary voltage, I'm assuming it's not an autotransformer so the primary and secondary are isolated, right?

Also, what's going on with that porch -- after the positive going half cycle gets to zero, it keeps going and levels off negative, before the negative half cycle starts. This is part of the assymetry that oldway is talking about, in addition to the different peak values. Have you tried a different triac to see if you can get something more symmetric?

Something else I'd be interested in seeing, just to get a feel for what's happening: Gate your triac for nearly full 360 degree conduction, and verify that you're getting something normal looking on the secondary. Then slowly reduce the conduction time and watch to see where the funny stuff starts. If you can see that plus the primary current as this is happening, I think it would  help  you understand what's happening.

You're probably aware that power transformers are generally not designed to take an instantaneous voltage step like a triac turning on. They prefer to see the time-varying sine wave. You can't really change the primary current instantaneously, so that's why traces including primary current would be interesting to see. I'm not sure what your end goal is, but you can tell us about it if you like. But maybe there's another way to get to your end goal.

[keshab.patel]

 Thanks Oldway and Jebcom. Thing is, I  have only one  differential probe and no high voltage probe. So cant afford to risk the scope. Attached is the primary waveform voltage(ch2) and current(ch3). I couldn,t get how the current is positive even if the voltage goes negative.  I am using transformer (240 pr/2400V sec).  I want to control the dc voltage in secondary. As secondary voltage is so high, no easy idea is getting in my mind to control in the secondary. 

[Circlotron]

The current waveform is only in a positive direction so something is very wrong there. Also maybe you could try putting a light globe across the transformer primary and see if that puts things into line. The magnetising current of your transformer may be causing the triac to unlatch asymmetrically and give you a strange waveform.

[oldway]

Yes, there is something very wrong.

This current waveform is almost impossible.

Should it be true, than your transformer would be fully saturated, primary current would be only limitated by primary resistance and you would blow all the fuses of your main supply.

What is wrong is your current measurements.

When you answered that primary current is about 3A, I understood that you know very little about electricity and power electronics.
Somebody who knows about it should have specified 3 Arms or 3A peak or 3A peak to peak.

Please make a diagram of your measurement circuit.

[jebcom]

I  have only one  differential probe

Ah, differential probe. That can explain the symmetry issue, as you probably had no ground reference. I assume you're doing this for isolation.
But today's voltage trace looks much more reasonable.
Doing this kind of work, you may want to invest in an isolation transformer. I found mine on ebay for a good price.

I'm also perplexed about that current trace. What are you using for a current probe? How is it connected and where are you monitoring?

[oldway]

We are looking for problems who does not exist.
This triac phase control works fine....there is no problem at all.
Only the measurements are wrong.

[keshab.patel]

 The current trace in the attached scope capture is taken by clamp type current probe(100mV/Amp). The current I measured was using clamp meter and of course its 3A and yes its RMS. I think if nothing is mentioned then its consider as RMS. as for 220 Vac system or 110 V system. But thanks oldway for reminding me as an engineer's job is to be specific.  I tested the same circuit with a 150VA transformer and its working fine. And the current trace(Attached) which I have taken by current probe(100mV/Amp) in the primary is not showing any lag. May be the problem is occuring when using it in the 4800VA transformer. Also Attached the circuit I am using.

[keshab.patel]

 The current assymetry(as in the previous scope capture) occurs when I use the differential voltage probe in the secondary of the transformer and clamp type current probe in the primary.Dont know why yet.  Otherwise when use both in the primary side, a little assymetry occurs(the current goes more to positive side and less to negative side. As I increase the firing angle the asymmetry increases. It also causes disturbances in the power supply line affecting computer monitor displays.

[oldway]

It does not look like at all with what I can measure on the same circuit with inductive load. (transformer)
Upper trace (Y2): current in the primary of the transformer measure with a Fluke 80I-410 AC+DC probe.
Nominal sensibility is 1mv/A but I used 10 turns so sensibility is 10mV/A.
Lower trace(Y1): primary voltage measured with probe /10.
For security, I used a medical grade isolation transformer.

[Circlotron]

  As I increase the firing angle the asymmetry increases. It also causes disturbances in the power supply line affecting computer monitor displays.

Does your transformer make noises or get hot? Does it work okay with the full 240V connected directly to it?  If not, there is something wrong with the transformer.

[eneuro]

Even I am using two triacs, one for power switch and other to turn on the first triac.

In my spot welder I use AC mosfets switch (very small gate charge needed) to switch on triac (or two thyrystors in parallel) and no problem

Switching time is controlled by small MPU (ATTiny85) which additionally detects mains 230VAC zero crossing, sets time offset for phase offsets, etc..
Nice thing with AC mosfets switch (2xBUZ78 in parallel) is triac/thyristors gate can be activated exactly when needed while N-channel mosfets in such AC switch are ON/OFF on demand 

BTW: I've also microwave a few kV transformer and it will be used to test ESD gap PCB, so I hope spot welder driver can be used to set pulses -in this case of high voltage secondary output 
In spot welder it drives 2V-3V high current output, but primary windings of both trafos are  230VAC, so it should work.
Also nice HV diode can be found in microwave teardown and HV capacitor, but  of course one have to be carefull if trying to do such teardown of already working microwave due to this HV cap.

[keshab.patel]

@Circlotron Yes it makes noise when controlled by triac. the noise increases as i increase the firing angle.

[oldway]

No way to control the primary of a 5Kva transformer with such a simple and imperfect circuit.

[oldway]

In the same Application note of ST, AN308, you have a better circuit figure 11, page 10/15.
C1 is reset every half cycle by DB1, R- and R7 so C1 start always from the same initial voltage.
http://www.fingers-welt.de/gallerie/eigen/maschine/sauger/AN_Triac_InduktiveLast.pdf

But this circuit is still far from ideal, some asymetry exist in DIACS and the gate pulse is very short ( 2 or 3 µs).
With such a short trigger pulse and with inductive load, triac can fail to stay on after triggered because current may not reach the value of holding current at the end of the trigger pulse.
http://www.st.com/st-web-ui/static/active/en/resource/technical/document/application_note/CD00003854.pdf
To connect a 100W icandescent lamp in parallel with primary of the transformer may help.
Also choose a low holding current (Ih) and snubberless 3 quadrants triac.

[eneuro]

To connect a 100W icandescent lamp in parallel with primary of the transformer may help.

In spot welder added such icandescent lamp but in parallel with triac/thyrystors 

[oldway]

In parallel with primary of the transformer, it helps to solve the problem of trigger fault because holding current is not reach with short gate pulses with inductive current. (We add a resistive current in the incandescent lamp)

In parallel with triac, it helps to reduce risks of saturation of the transformer by providing a symetrical ac pre-magnetising current.
But in this case, it's not possible to reduce secundary voltage to 0V with no load.

As I already wrote:

Triac primary phase control of a transformer is not an easy task.
Control must be very acurate and transformer core better has to have a little gap for not saturate to easily.

[necessaryevil]

Triac control on the primary of a transformer is not a good idea. The dutch Elektor published a bench supply with triac control -it caused the transformer to overheat.

It might be a better idea to:
1) do a similar regulation at the secondary side
          The Elektor Magazine published such a circuit in nov. 1991 called "dissipation limiter"
2)Switch between transformer taps

[oldway]

Yes, I agree, that's not a good solution if you don't use the right control circuit.

For example, AEG use it in his industrial battery chargers without any problem...But it's not that kind of cheapy control circuits that keshab.patel is using !

With 2400V secundary voltage, secundary regulation and taps switching seems not to be possible.
To use a variac could also be a solution.

[Circlotron]

Or just use a TDA 2086 phase control IC. They are light years ahead of any do-it-yourself circuit.

[oldway]

Or the oldie TCA785, better version of the TCA780.
I used them in the 80'ties in a project of high power 3 phases industrial rectifier/battery charger. (330Vdc - 600A)

[Circlotron]

Yeah that TCA 785 looks good. One reason why I like the TDA 2086 is because it has an extra input from the top of the triac used for measuring the zero crossing of the current waveform. This is useful when you have an inductive load because of the lagging current.

[oldway]

That's not necessary if you choose the option "broad pulse" of the TCA785. (Pin 12 and 13 to GND)