DC Offset effects on Transformers

[T2SE]

Can anyone explain to me the mechanics of mis-firing SCR's (poor/missing gating signals)causing a DC offset in a transformer and its effect on the core saturation and heat in the transformer

 

[Skogsgurra]

Sure. There are three things that make triac operation with transformer load tricky.

1 The transformer primary has a certain inductance that makes the impedance equal to wL, where w is frequency in radians/second and L is the inductance in Henries.

2 It has a very low resistance, which makes even very low DC voltages push a high current through the winding.

3 The core iron saturates at about 1.8 Tesla. Sometimes a little bit higher, but not much.

A misfiring triac produces a DC component (positive and negative half-waves not equal) and this DC component drives a substantial DC current through the primary winding. This DC saturates the core and a saturated core is no longer magnetically active. I.e. it has a very low inductance, which results in a high current drawn from the AC source.

The usual reason for this is that the triac is unsymmetric and reaches holding current at different levels in positive and negative operation. There are also other reasons like poor timing and such.

A remedy (in the former case) is to connect a resistive load across the transformer primary. If the extra power loss cannot be tolerated, a capacitor (AC!) can be connected in series with the resistor to form a classic snubber. It will help the triac reach holding current before current in primary has built up.

If the problem is poor timing accuracy, the remedy is to make the trigger circuit more accurate.


Gunnar Englund

http://www.gke.org

 

[T2SE]

Maybe I was not clear in the first question, this is the feed transformer for an SCR drive that drive a motor and I'm wondering about the mechanics of inducing a DC offset into the secondary of the transformer.

 

[alehman]

Do you mean common mode DC voltage (entire winding at DC potential with respect to ground) or DC applied between the terminals?

 

[T2SE]

DC as you would see if you had a missing gating pulse(s) on one of the six SCR's in the load side of the circuit.

 

[alehman]

That's not really DC from the transformer's point of view, but 1/2 rectified current. I don't think that would bother the transformer.

 

[Skogsgurra]

Yes, it bothers the transformer. Just as primary DC does.

As soon as you have a DC component (pulsating or constant doesn't matter much) in the core, you will get core saturation. So a bad trigger circuit will cause exactly the problem you describe. And the reason is that the core get DC saturated so that the inductance is reduced and primary current increases.

On a side note, this is very similar to how a magnetic amplifier works. A DC control current control current in the main winding.

Gunnar Englund

http://www.gke.org

 

[Warpspeed]

SCR misfiring can cause flux doubling and some fairly massive momentary inrush currents, but it is not going to be any worse than that which occurs at initial transformer turn on.

A very great deal depends on the saturation characteristics of the particular magnetic core, and the original designed operating flux density of the transformer.

E/I transformer cores with the inevitable small (unavoidable) built in airgaps between E&I laminations, and designed to run at relatively modest flux densities are going to be SCR friendly.

Tape wound cores run fairly hard right on up to high flux density are going to produce some fairly impressive current spikes in the primary at initial power up, or if occasionally an SCR fails to trigger.

Unless you designed the transformer yourself, it is difficult to know how much of a beating the SCRs and transformer are going to see. All you can probably do is hook it up to a storage CRO, monitoring primary current, set the trigger threshold to something beyond full load operating current, and marvel at the occasional very high current spike.

As long as it is well below the I^2T rating of the SCR, it should not harm the transformer, provided it is not happening continuously.

 

[T2SE]

Thanks for the answers and bsaically what you describe is what I'm seeing. Do you know of any web sites or technical information that I can look at to fully understand the mechanics of the introduction of the DC offset into the transformer?

 

[Warpspeed]

http://relays.tycoelectronics.com/app_pdfs/13c3206.pdf

http://www.allaboutcircuits.com/vol_2/chpt_9/12.html

 

[ScottyUK]

Ferroxcube have a good applications guide for soft ferrites on their website as a download. Obviously it is quite biased toward their products, but the general information is worthwhile. And it is free!

If you want a good reference book, 'Soft Ferrites' by Edgar Snelling is worth finding. It is an old publication but still in print because there is little published since that betters it. Some of the newer materials aren't listed, but the maths is perfectly valid if you like to do things the hard way or want to understand the fundamentals behind the things you observe.


----------------------------------

One day my ship will come in.
But with my luck, I'll be at the airport!