BPPBP - Bruno Putzey's Purist Balanced Preamp (well a balanced volume control really)

[Bertel]

I'm currently doing some research on preamp concepts which basically resemble a balanced volume controller, if I may say so.

While doing so, I came across a design by Bruno Putzey which he lays out here: http://www.hypex.nl/docs/papers/The G Word.pdf (originally written for Linear Audio Linear Audio | your tech audio resource , Volume 5 came out in April 2013 with a free PCB for it)

This design includes the following features:
- Buffer as input stage with input biasing network
- Difference amplifier between input stage and variable gain stage, with the output of the difference amplifier being referenced to the cold point of the volume controller
- Linear potentiometer used as feedback element in an inverting amplifier, not as attenuator
- No output stage (almost)
- DC removal circuit (2nd order low-pass filter whose output is subsequently subtracted from the signal)

So basically bal to unbal, followed by unbal active volume control as part of a difference stage, then back to bal with passive balanced impedance output.

Has anyone actually built this preamp? If so, what are your experiences with it? And what do you think of the concept and design as such?



(EDIT: Just saw from AndrewT's post here that the EDN version of the article published here has some very valuable additional comments.)

 

[billshurv]

You need to re-read and get clear the difference between balanced and differential. I will build one one day as it is possibly the lowest distortion DIY design out there.

 

[AndrewT]

You need to re-read and get clear the difference between balanced and differential. I will build one one day as it is possibly the lowest distortion DIY design out there.

Funny you should state that.
I was trying to get that message across recently and some came back arguing how similar the two philosophies are, rather than seeing where they are different and why they perform differently.
I gave up arguing.

 

[billshurv]

It's very easy to get very confused!

 

[AndrewT]

The Bruno Putzeys device has balanced impedance input and balanced impedance output, so can be used in a balanced impedance system.

Note it is the CONNECTION system that is balanced impedance and this is what gives the very high interference attenuation, if you get the cabling and impedances just as they NEED to be.

The difference amplifier just about describes every opamp and every power amp.
They all measure the DIFFERENCE in input voltage and apply their processing to arrive at an output voltage. This part has absolutely nothing to do with "Balanced Impedance Connections".

Unfortunately many corrupt our technical language and shorten to "balanced" and that gets misunderstood by thousands of our Members.

 

[AndrewT]

what do you think of the concept and design as such?

I think it is brilliant.

Jan must have thought so too. He included the project in his "Linear" series of periodicals.

 

[billshurv]

But the amplification stages ARE balanced impedance internally. That is the interesting bit of the design. I also like his choice of words for ground as 'the electron dump'.

 

[Bertel]

That's also my understanding gents, I was obviously too sloppy in my language and probably should have called this "a volume control for use in fully balanced impedance connection systems throughout" or so, thanks for pointing that out (Bruno himself calls it "a balanced volume controller" in his article, by the way).

I think it is brilliant.

Thanks for your assessment AndrewT, I saw you had recommended this approach in the Wire Thread.

Anyone who actually built this?

 

[Joachim Gerhard]

Does anybody know how and when to use J4 and J5 ?
I asume that there is also a posibility for unbal ( RCA ) In and unbal ( RCA ) Out.
I could need bal In to unbal Out.

 

[Pan]

I wonder if J4/5 is simply there for letting your PCB layout software accepting a GND connection. Zero ohm resistor could be used as well.

It's mentioned in the text.

 

[Pan]

Also, unbalanced in = ground one of the input buffer amps.
Unbalanced out, bypass R15/42 and use as GND.

 

[Bertel]

BPPBP - Bruno Putzey's Purist Balanced Preamp (well a balanced volume control...

I asume that there is also a posibility for unbal ( RCA ) In and unbal ( RCA ) Out.

Joachim, instead of using RCA connectors at the amp, would using cabling as sketched in Figure 6 in the article be possible in your application, with RCA at the source and XLR at the input, and vice versa? (Also described in 12.3.1 here: http://www.hypex.nl/component/weblinks/weblink/24-datasheets/28-nc400-datasheet.html )

 

[jan.didden]

I wonder if J4/5 is simply there for letting your PCB layout software accepting a GND connection. Zero ohm resistor could be used as well.

It's mentioned in the text.

Correct. I think he calls it 'net-tie' points. CAD software often insists on connection anything with the same name together wherever IT thinks convenient. If you have two ground 'things' that you want to force to connect at a specific point, then depending on your CAD idiosyncrasies, you might be forced to do a workaround like Bruno did. Clearly explained in the (free) article - no excuse NOT to read it!

BTW I had 1000 free boards which are all given out but the Gerbers are available free on the website (under Downloads) so you can have made some yourself, or set up a group buy. It's a small board so not expensive, BUT it does challenge your SMD solder proficiency though. I managed to get two boards together with no mistakes.

Jan

 

[billshurv]

I'm short of round tuits at the moment, but if anyone is getting boards made I'd be in for a few to help get the costs down.

 

[Turbon]

+1

 

[Bertel]

BTW I had 1000 free boards which are all given out but the Gerbers are available free on the website (under Downloads) so you can have made some yourself, or set up a group buy. It's a small board so not expensive, BUT it does challenge your SMD solder proficiency though. I managed to get two boards together with no mistakes.

Many thanks Jan! I have a visual impairment and can't build myself, so I will have the PCB printed and built professionally.

I will try to figure out how to create an Excel Online sheet, so if anyone wants to join me with either a bare PCB or a completely built unit, please let me know. Will get quotes now and publish cost shortly.

Thanks,
Robert

 

[Turbon]

Hi there Robert.

I want to make my own misstakes so a bare board or two with an recommended BOM would be fantastic 😀!

Regards

 

[jcx]

Putzeys' comment on the THAT Corp patent is dated - actually Whitlock's patent on diff amp input common mode bootstrapping is public domain now

http://www.google.com/patents/US5568561 coverage expired last winter with the most generous interpretation of term == filing date + 20yr

any new design could consider it without IP concerns

 

[Bertel]

Interesting point jcx, thanks!

Yet Bruno also writes (page 10):

I would’ve used the Whitlock’s input chips and implemented the capacitive bootstrap technique as well, except that the distortion performance is not good enough in my view.

 

[Bertel]

I want to make my own misstakes so a bare board or two with an recommended BOM would be fantastic 😀!

Fair enough Turbon 🙂

Bruno has a parts list in his article already, plus there's a more specific one with manufacturing codes in the Downloads section of Linear Audio website here: Downloads | Linear Audio (this is the direct URL to the BoM).