Influence of DC-currents on TXs ?

This question originated in another thread.
Best formulated by Juergen himself:

from JH:

What I really don't know (and maybe I should ask this in a different, "newbie" thread!):

What *is* a good corner frequency for an audio transformer to protect against DC, in general ??

(Not just speaking of compensation; also speaking of ordinary DC blocking with a capacitor.)

For an uncompromised low end, I'd set the corner frequency as low as possible. But if it takes 10 seconds for the currents to be settled after turn-on or after plugging something in, will this long transient be as hurtful as "real" DC?

Click to expand...

So it's about the unkown... like what happens with an
output-TX when DC-currents of tenths of mA's flow through a
non-gapped transformer during power-on, especially
when that takes several seconds
(for reasons as in that other thread, but for now
let's just assume they are that long).

Will there be remaining magnetism ? Will inductance remain lowered ?
Will they catch the bad guys ?
Will the primary & tertiary winding find each other in the end and live on happily ?

Answers in the next episodes (we hope...)

Thanks,

Peter

I brought this up in the other thread - I'd love to know it.

It's obvious that DC is harmful to an ungapped transformer, but how long a transient does it take to count as "DC" in the sanse of being harmful to a transformer?

Will the resonant circuit that is made of an AC-coupling capacitor in series to a transformer make sure that any transient will fade out in a way that possible magnetisation will be demagnetised again? Too good to be true?

How to choose the lower frequency corner with transformers? (It's clear how to calculate it - but how to choose it?)

JH.

from JH:

I brought this up in the other thread - I'd love to know it.

Click to expand...

So far no luck... let's bump.

It's a good question - but I don't know the answer.

I can say, though, that I've never had any problems with transformers that were capacitor-coupled - no matter how big the capacitor.

My guess is that you'd need several-seconds - maybe minutes - of core DC saturation (that is, not merely magnetisation) before you have ill effects.

JAkob E.

Thanks Jakob,

My guess is that you'd need several-seconds - maybe minutes - of core DC saturation (that is, not merely magnetisation) before you have ill effects.

Click to expand...

This sounds encouraging, and that all kinds of fancy start-up schemes are not needed.

Hmm, just when I got a good feeling about this, that
'dont-DMM-a-mic-TX' comes to mind again...
(that warning against measuring the DCR of TXs with a
resistance-meter). But that's probably still a rumour
and/or not relevant for the larger output-TXs we're talking about here.

Thanks,

Peter

But that's probably still a rumour

Click to expand...

I don't think it was only a rumour.

The warning was written in application notes for the great old danish "JS" type mu-metal core input transformers. It even was printed on every one of these transformers: "Do not expose to DC!".

But this might have been a specific issue for these mu-metal-core input transformers. We don't see warnings like these with modern-day transformers.

Jakob E.

from Jakob:

I don't think it was only a rumour.

The warning was written in application notes for the great old danish "JS" type mu-metal core input transformers. It even was printed on every one of these transformers: "Do not expose to DC!".

Click to expand...

Thanks, this makes the story complete. If anyone feels otherwise please let it be known, but fearing all possible problems and let circuit-complexity grow because of it might be overdoing things.

No tricky slow starts etc, just going for a compensation of the static situation. Now I need to get me some BA283-PCBs or put it on perfboard and try.

Bye,

Peter

Most outputs are 50/50 or M6 which has less reminance than 80 Ni.