Some Quick Mic Pre Input Transformer Questions

Hi,

I remember hearing someone mention how input transformers with a centre-tapped primary were good for phantom power reasons - can anyone explain why?

If given the choice, is it best to chose an input transformer with a centre-tapped primary if it is at no extra cost?

My second question is, can anyone explain what a mic pre input transformer with "two split bifilary wound primaries" is?

Thanks,

Roddy

[quote author="rodabod"]
I remember hearing someone mention how input transformers with a centre-tapped primary were good for phantom power reasons - can anyone explain why?[/quote]
Better phantom power simmetry, smaller chance for the core magnetization...

If given the choice, is it best to chose an input transformer with a centre-tapped primary if it is at no extra cost?

Click to expand...

Yes, IMHO.

My second question is, can anyone explain what a mic pre input transformer with "two split bifilary wound primaries" is?

Click to expand...

A transformer with bifilary wounded two windings for the primary side, then usually connected in serial ("splitted").

Regards,
Milan

[quote author="moamps"][quote author="rodabod"]
I remember hearing someone mention how input transformers with a centre-tapped primary were good for phantom power reasons - can anyone explain why?[/quote]
Better phantom power simmetry, smaller chance for the core magnetization... [/quote]

That makes sense. One other thing I was wondering, if the centre of the primary is "tapped" then is it just shorted to earth? Would this not short out the phantom power?

My second question is, can anyone explain what a mic pre input transformer with "two split bifilary wound primaries" is?

Click to expand...

A transformer with bifilary wounded two windings for the primary side, then usually connected in serial ("splitted").

Click to expand...

So, is this a good thing? Would this be the better thing to chose if it didn't cost any more?

[quote author="rodabod"]
One other thing I was wondering, if the centre of the primary is "tapped" then is it just shorted to earth? Would this not short out the phantom power?[/quote]
The phantom power is connected to this tap. Like in PP tube amps.

Regards,
Milan

For really canceling the DC, the transformer should be wired in PP fashion. There are many single coil transformers with center tap, and those do not have balanced DC resistances. Only dual coil transformers offer balanced DC resistances, and those are the ones you should use, if you really want to feed 48v throught the transformer CT. I don´t see a really good reason for doing it, anyway.

Better phantom power simmetry, smaller chance for the core magnetization...

Click to expand...

There is in fact a qood chance that right the opposite is true. Two 1% resistors are very likely to be a much better match than (any) two transformer windings. I strongly agree with rafafredd that you better don't do it.

One other thing I was wondering, if the centre of the primary is "tapped" then is it just shorted to earth?

Click to expand...

Can you elaborate? Do you mean that the center tab is internally connected to the faraday shield/can?

Just to make sure: you feed the 48 V through a 3k4 resistor to the center tab, not directly.

Samuel

[quote author="Samuel Groner"]
There is in fact a qood chance that right the opposite is true. Two 1% resistors are very likely to be a much better match than (any) two transformer windings. [/quote]

Did you ever measured this?
Anyway, my old 414s work much better using this approach, especially in a higher distances.

Then why is it that NO Jensen or Reichenbach trans. do not have a center tap on any of their Mic trans.
And they are some of the best available...
What would be the purpose?

[quote author="rafafredd"] There are many single coil transformers with center tap, and those do not have balanced DC resistances. Only dual coil transformers offer balanced DC resistances, and those are the ones you should use, if you really want to feed 48v throught the transformer CT. I don´t see a really good reason for doing it, anyway.[/quote]

Right, so is it better to have one of the "bifilary" ones which have two coils on the primary?

I'm a bit lost as to what the advantage of the centre-tap was for phantom power in the first place. Surely if +48V is connected to each side of the primary, then no current will flow through it anyway since there is no potential difference?

[quote author="Samuel"]

One other thing I was wondering, if the centre of the primary is "tapped" then is it just shorted to earth?

Click to expand...

Can you elaborate? Do you mean that the center tab is internally connected to the faraday shield/can?

Just to make sure: you feed the 48 V through a 3k4 resistor to the center tab, not directly.
[/quote]

I made a stupid assumption (yes, I thought you earthed the tab to the shield). So, you apply the 48V to the centre tap and will then there will be 48V at each end of the coil. But... Why don't you just have the 48V at each end of the coil without using the centre-tap?

With respect to coil matching, etc, the reason I am asking is because I have a choice of some Beyer transformers and can choose to have a centre-tap or a bifilary (twin) primary coil. I just don't know what is best.

I can check the % tolerance for coil matching etc. on the datasheet if anyone is interested.

[quote author="Mike Keith"]Then why is it that NO Jensen or Reichenbach trans. do not have a center tap on any of their Mic trans.[/quote]
Did you check Cinemag tranies (modern Reichenbach)?

And they are some of the best available..

Click to expand...

Then, the SOWTER TYPE 9045 is the worst?

[quote author="rodabod"]Surely if +48V is connected to each side of the primary...[/quote]

It's not. It's connected to the center-tap through a 3.4K resistor.

Peace,
Al.

The only advantages I can think of for applying phantom power to the center tap through a 3k4 resistor rather than to the ends through a pair of 6k8 resistors are...

1. Saves you a resistor. (dumb)
2. Won't reduce the input impedance like the 6k8's will (generally a non-issue)
3. If the two halves of the primary ARE perfectly matched, you'll get better CMRR with any 10% 3k4 resistor than two random 1% 6k8 resistors. However, two 1% resistors can easily be hand-matched to 0.1% or better.

Can anybody think of any other advantages?

A 600 ohm primary is going to be a few ohms, so if going thru a 3.4k into the CT, then out the ends to the mic, draw the circuit and analyze 3.4 k + 8 ohms vs 3.4 k + 10 ohms and see what the differential is.

I would be more worried about what is inside the mic. If, for some reason, pin 2 draws more phantom ma than pin 3, you could have the potentioal for some micro amps on the windings. (with the CT connected)

Yes, the disadvantages are abundant. Plenty of opportunity to magnetize your transformer core, thereby semi-permanently increasing your 2nd order harmonic distortion. I know I've heard of popular modern microphones that are known to pull signifigantly more phantom DC from one leg than from the other. I want to say the TLM103, but I'm going by a vague memory of hearsay so don't quote me on that.

Of course the easier way to cause a problem will be to use a defective cable or otherwise short one of the input pins to ground.

It won't make anything stop working, but it'll degrade audio quality.

If one 6.8k resistor is pulling more than the other I would say something is wrong...

Thanks for the replies.

So, if given the choice of either single primary, center-tapped primary, or twin-primary, what would you choose?

Twin-primary (bifilary)?

The advantage to a dual primary winding is you can connect them in series for a 600 ohm input Z or parallel them for 150 Ohm input Z, or 150/50 depending on the application. As a bonus, when they are series coonnected there is a center tap available. Some applications have the center tap grounded - this is what is called a true balanced input, it balances with respect to ground. Nowadays usually the input is floating with respect to ground reducing ground loop and isolation problems.

I would think it would effect the CMRR in a negative way.

A realization, if one leg is pulling more than the other, it does not matter if it's thru the CT or not, you will still have a voltage difference across the windings.

[quote author="CJ"]A realization, if one leg is pulling more than the other, it does not matter if it's thru the CT or not, you will still have a voltage difference across the windings.[/quote]
I was referring to the typical (2) 6.8K resistor config.