Transformer with unbalanced DC load

Before you dismiss the question as too vague, lacking proper data, let me say that I know it's vague. That's because it's not about a specific product with known parameters. I'm just seeking guesstimates based on your experience.

Take a cheap small mains power transformer, 5VA or thereabouts. Only open-circuit voltage, dc resistances and stated current rating (probably optimistic) are known. Core characteristics, number of turns, flux density, etc. are all unknown.

If this transformer were to be operated with two rectifiers with capacitor filters, one full-wave with a load 10% of the max rating and the other half-wave also at 10% of max. The main concern here is about the unbalanced DC due to the second rectifier load.

I could just try it out with dummy loads but I'd like to be able to estimate the long-term behavior and reliability based on theory and/or your practical experience.

Reply to
Pimpom
Loading thread data ...

My guess would be that there wouldn't be a problem. If you don't have probl em with a half-wave rectifier in the first place, adding the full wave rect ifier in parallel isn't going to make anything worse.

Clearly there's the usual problem of a lot of current running through the w indings in the brief period when the rectifiers are charging the capacitati ve loads, but that's just the wires getting warmer than the average current would lead you to expect. DC imbalance doesn't seem to come into it (or no where where it matters).

--
Bill Sloman, Sydney
Reply to
bill.sloman

The half-wave output is not a problem. It's easy enough to place adequate filtering for the light load and it's to be followed by a 3-T regulator anyway. My concern is about the effect of DC on the transformer. Here's one possible circuit:

formatting link

Reply to
Pimpom

What "unbalanced DC"? There's no DC current component in the power transformer core if I'm understanding the circuit you're proposing correctly.

Most likely result is it will work fine with the transformer hardly breaking a sweat, the DC output voltages will run high as compared to what might be calculated from the loaded RMS output spec voltage because it's so lightly loaded; it hardly even knows the load is there.

I've had two transformers of similar spec, rated for 1 amp at 120 -> 6.3 volts, the first secondary feeding ~500mA worth of tube filaments off a full wave rectifier, then stepped back up using the second transformer's secondary as a primary to 120 and half-wave rectified to around 160V, drawing 10-20mA off that. It's worked fine for years, the transformers barely even get warm.

Reply to
bitrex

Have a look at the schematic in the link I gave in my second post. The DC component is the result of the unidirectional flow in the upper rectifier circuit.

I rather expected that but wanted to be sure.

the DC output voltages will run high as compared to

All of that's a given. The three outputs are to be followed by linear regulators.

Your transformers too have a DC component due to the half-wave rectification.

Reply to
Pimpom

I'll rephrase, I used the term "component" in an incorrect/ambiguous way. There's no way for steady-state DC to "back-feed" the secondary of the transformer in the circuit you posted. There is an unbalanced DC component of the Fourier series of the current waveform thru the secondary due to the current pulses that charge up the output cap on the positive-going peaks of the AC cycle.

But that DC component is finally limited at any given instant of time by the DC impedance of whatever the transformer is driving, which in the case of say half-wave rectifier + filter cap + 20 mA resistive load is clearly much much larger than that of the load by itself for almost the entire half AC cycle. If the DC current thru the load resistor is 20mA the time-averaged DC component of the secondary current waveform can't magically be an amp or something, for ~99% of the time the DC current component thru the primary is just about zero.

Not sure exactly what the latter would be but for such a light load I doubt it would have any effect on the transformer. Driving a big heater or lamp thru directly thru a half-wave rectifier fed from a transformer and I'd imagine that could cause some problems.

Reply to
bitrex

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

** All the electros in your schem are charged with half wave rectified current pulses.
** A DC component in the secondary causes the core to magnetise slightly, so the magnetising current wave becomes asymmetrical. On the primary side, consecutive current pulses taken from the supply are not the same amplitude.

In the example you give the effect will be very minor.

However, with larger transformers and large DC offsets it can be severe, particularly if the steel core is toroidal. In the latter case, the tranny will draw a very large primary current, make a growling noise and overheat.

.... Phil

Reply to
Phil Allison

That's what concerned me. With most of the transformer's characteristics unknown, I was wondering if the effects of the magnetic bias would be severe enough to be a problem for the transformer itself.

On the primary side, consecutive current pulses taken from the supply are not the same amplitude.

At this power level, the effect on the mains supply would be insignificant.

I hoped that this would be the case. Thanks for the confirmation.

I wouldn't dream of doing this at higher power levels, particularly when the DC offset is a substantial fraction of the maximum rating. This is why I gave the loads as percentages of the max rating in my opening post.

Reply to
Pimpom

Do you have a bunch of them? If so, I'd certainly load some down and see. (one of, put a diode temp sensor on it.) George H.

George H.

Reply to
George Herold

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

** Or just measure the primary current and compute the VA.

Folk generally know the rated VA of mains trannies they buy.

.... Phil

Reply to
Phil Allison

SFAIK secondary current is non-magnetising, the magnetising current flows in the winding being used as the primary, any induced secondary current is only going to magnetisise the parts of the core that contrubute to the leakage inductance.

further because the primary has finite resistance ans is being driven from a voltage source, the work current flowing in the primary actually reduces the magnetising current slightly.

That said, Phil Allison is the iron-core transformer expert here. There may be mechanisms that I have not considered.

--
This email has not been checked by half-arsed antivirus software
Reply to
Jasen Betts

The DC current drawn from the secondary during charging half-cycles is almo st perfect compensated by the matching primary current. Magnetising current develops the voltage across the windings, and while the bit of DC current charging the filter capacitors generates an extra voltage drop in the secon dary and primary windings while the capacitor is charging, it is a second o rder effect, unless the windings get hot enough to get the magnetic materia l in the core above it Curie point.

--
Bill Sloman, Sydney
Reply to
bill.sloman

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

** Sometimes the domestic AC supply acquires a DC component, this happens w hen the peak positive and negative voltages on the active conductor are not the exactly the same amplitude. Local half wave loads are the usual cause.

This can have a dramatic effect on the magnetising current of a transformer , particularly toroidal types. The primary current wave becomes quite asymm etrical and shows signs of core saturation every other half cycle. The effe ct is not instantaneous, it builds up over a few cycles and settles down wh en the *average* value of the two half cycle's current waves becomes the sa me.

Exact the SAME thing happens whenever you use a half a half wave load with a similar tranny. During current carrying half cycles, there is a voltage d rop in the primary winding NOT matched by and equal and opposite one during the following half cycle - creating a DC component as above - so the core magnetises ( offsets itself) in one direction.

I tried this with a 30VA toroidal just now and saw a 5:1 difference in posi tive and negative peaks of the primary current. Toroidals are particularly vulnerable due to the lack of air gaps in the continuously wound core.

FYI:

To measure the DC offset of the AC supply, connect a 1M resistor and a 1uF film cap in series across the active and neutral (cap to neutral) and then monitor the voltage across the cap with a DMM set to the 2VDC range.

Take the obvious precautions.

.... Phil

Reply to
Phil Allison

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.