Push-Pull Amp Balanced Input?

[Cressy Snr]

I was cleaning around my equipment this afternoon and came across a pair of XLR balanced analogue outputs on my Musical Fidelity M1 DAC.
I hadn't given them any thought until now.
I've been single ended through and through and have no experience of balanced equipment.

So...the question is: could I simply take each channel and wire the positive to one half of my differential driver stage and the negative to the other,
so that there is no need for a phase splitter transformer ?

Of course it would mean buying a pair of XLR leads and some XLR sockets for the amp, and learning how to do a balanced shunt volume control.

Could it really be that simple?

[Cressy Snr]

Something like this?

[Cressy Snr]

And what values of series resistor and pot?

Lots to think about but I reckon this has the potential for awesome sound.

[pre65]

Would not a balanced volume pot have 4 sections, two per channel ?

Something like this.

http://www.ebay.co.uk/itm/Balance-XLR-2 ... 43ac452417

[pre65]

Like in the Raven preamp

http://www.google.co.uk/imgres?imgurl=h ... g&dur=1362

[Nick]

And what values of series resistor and pot?

Lots to think about but I reckon this has the potential for awesome sound.

Yep, its what I have planned for the 300b push pull I will get around to one day.

[Cressy Snr]

Would not a balanced volume pot have 4 sections, two per channel ?

You'd think so wouldn't you and most people would do just that. But since I posted a couple of hours ago I have had a read of Morgan Jones.

Here's what he has to say:
It may be that you want to build a balanced preamplifier, in which case the volume control should be balanced too. Curiously, people persist in placing notionally identical controls in each path of each channel. This is wrong. The two mechanically ganged attenuators cannot have perfect channel matching, so a common mode noise signal such as hum will be attenuated unequally, converting a portion of it to differential mode, to which a balanced preamplifier is sensitive. The correct way to construct a balanced volume control, using the minimum number of components, is to use a balanced, type C, attenuator having a fixed series resistor in each leg.

The reason that the balanced type C configuration is superior is that the degradation of common mode rejection is solely dependent on the matching of the fixed series resistors. Therefore they should be of 0.1 percent tolarance or better.

Unfortunately the type C attenuator has the disadvantage of a high output resistance and in combination with the input capacitance of the following stage, this causes high frequency loss if ignored.

So the pot value and the series resistor values have to be carefully worked out so as to not roll off the HF, but the balanced shunt pot is both reasonably inexpensive and better at rejecting hum.

[Cressy Snr]

Here's one I made up for a 5WPC feedbackless triode-connected EL84 amplifier.



Simple isn't it? (for a pp amp that is)

Being feedbackless, and only two stages, it is surely screaming to be direct-coupled,
provided we make sure the driver stage is made to come up to voltage before the output stage

Push-pull balanced input, no coupling caps, no phase splitter.
What's there not to like?

This looks like one for Owston methinks.

[pre65]

The pot before the 5687 grids, is that a balance pot ?

If it'a volume attenuator I don't see how it works.

[Cressy Snr]

If it's a volume attenuator I don't see how it works.

It is a volume pot. I don't get it either. My primitive brain can't see it.

OTOH Morgan Jones does know what he is doing and is emphatic that this is the way to control volume with balanced gear.
And as an ex BBC engineer, he must have been using and repairing almost exclusively professional class balanced line equipment.

It is one of those cases where I am just copying MJ. I'd like to understand
Maybe someone on the panel could explain it, as there is very little on the web about it.

If I could understand what a type C attenuator is, then maybe I might have a chance

I mean because the inputs are floating either side of ground, then where exactly is the signal being shunted to when the control is turned?
Can the volume ever be completely zeroed?

[Cressy Snr]

AHA!

The penny has just dropped.

Here's how I think it works.

The fact that the two incoming signals are of equal and opposite phase is the key to understanding this arrangement. It means that with the wiper at one end of the shunt pot the incoming signals will be completely shorted together and being of opposing polarity, will thus cancel out, leaving zero sound provided incoming balance is perfect.

As we go towards the other end the ratio of of cancelled to allowed signal decreases and decreases until the full signal is allowed through to the grids.

Seems logical enough now.

[pre65]

That make sense, even to me.

[Mike H]

It works because in the centre of the pot value is a 'virtual earth', about which the two opposite polarity inputs swing.

The signal level is the Voltage dropped across each half of the pot value.

So if the pot value was zero, no signal input to either input stage.

It's quite a neat arrangement must be said.

[Cressy Snr]

It works because in the centre of the pot value is a 'virtual earth', about which the two opposite polarity inputs swing.

The signal level is the Voltage dropped across each half of the pot value.

So if the pot value was zero, no signal input to either input stage.

It's quite a neat arrangement must be said.

Aye....that 'n' all

[Cressy Snr]

Well I have got myself some twin screened microphone cable in a lovely shade of midnight blue and the appropriate XLR plugs and sockets.

Time to have at the EL84 amplifier I think.