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

Just as a warning and only as a possible bug fix to you all using the Hypex Regulators.

Probably depending on the make of the electrolytic capacitors, some oscillation on the supply voltage may occur, in my case close at 40Khz.
By short circuiting the output sense line to the output line of the Hypex module on the PCB directly at where the module is placed, this problem was fully cured.
It is also possible to make this short circuit directly on the Hypex module itself.

Hans

P.S. Andrew, table to be found on the Hypex site

Hans, is this oscilation related with capacitor impedance minimum at this frequency?
 
Last edited:
Hmm but if you short the sense to the output, then don't you pretty much remove the advantages that the Hypex regs have?

I don't know how much their precision is brought by the remote sense and how much is the design in the regulator circuit itself if the sense feature is short circuited / bypassed...

Out of interest, what was the voltage oscillation you noticed? Are we in very small microvolts or is it at a larger magnitude?
 
Hmm but if you short the sense to the output, then don't you pretty much remove the advantages that the Hypex regs have?

I don't know how much their precision is brought by the remote sense and how much is the design in the regulator circuit itself if the sense feature is short circuited / bypassed...

Out of interest, what was the voltage oscillation you noticed? Are we in very small microvolts or is it at a larger magnitude?

With the physical change of the remote sense point you do no harm at all to the Hypex regulator, the low noise, load and line regulation will stay absolutely the same.
I have also tried standard 7812 and 7912 voltage regulators, and their (noise) performance was very much lower and of course without oscillations.

The remote sense is not at all bypassed, it it simply taken from a point a bit less further away, thereby increasing stability of the feedback loop.

The oscillation in my case was something like 100mV tt or 35mV rms out the top of my head. You will probably never notice it, but I simply did not like it, that's why I made the change.
And again, different electrolytic caps may have different responses, so it was just a warning.

Hans
 
I have added a new column in GB list http://www.diyaudio.com/forums/grou...advanced-volume-controller-2.html#post4696202


I'll take over and produce Hans board for this project. I'll announce price per PCB after I get a quotation.
Only bare boards.

Please subscribe indicating your forum username and number of boards you are interested in.
I'll keep list open till end of May. Afther that you'll be contacted by private message and invoiced via paypal.
Keep in mind that EU people must pay VAT taxes, unless you have a business and provide me EU VAT number.

Regards,
Tibi
 
Last edited by a moderator:
To remove all misunderstandings, hereby a graphical representation of Maya + BPBP

Maya_BPBP.jpg

Hans
 
To remove all misunderstandings, hereby a graphical representation of Maya + BPBP

View attachment 546240

Hans

What components need to be mounted on your board Hans? I see 1 SMD resistor, the through-hole connectors and relays.. do I see a couple of SOIC chips to put on too?

EDIT - oops, those green squares are all SMD resistors, not connector blocks (not sure why I thought they would be.. in fact I neglected to think about that at all).

If so, how's the pitch on those - easy (relatively speaking) to solder each pin or should it involve drag soldering or hot air gun?

Tempted to get a few of your PCBs even without the Maya (but could go for one Maya to see what it's like). Should help the GB to buy more of the PCBs, one for each of the BPPBP boards I have.

Let's make it clear to people that they can use this Maya GB to simply obtain your board too without the Maya (assuming that's ok..?). Others like Billshruv can get six of your boards and use the Maya to control all six for multichannel.
 
Last edited:
But, what you could do is to add a SECOND pot, exactly the same as the volume pot, thus having the same non linearity.
In that way, you can use the first pot for volume setting and the second pot as a variable resistor Z.

The algorithm would then become:

[Y*(A*I + B*I^2)] / [(100-Y)*(A*I + B*I^2)]+Z*(A*I + B*I^2) =([Y] / [100-Y+Z]) * (A*I + B*I^2) / (A *I +B*I^2) = [Y] / [100-Y+Z]

And again, the non-linearity has now been completely removed, and also [100-Y+Z] can never become zero, preventing a gain of infinity.

Hopes this explanation helps to get this problem out of the way forever.

Hans

Not quite out of the way forever, answers bring more questions!

For the second pot idea, judging by the maths/theory, both need to be as identical in their non-linearity as possible? So essentially the same pot?

Asking as in terms of control design, you'd want one to be small and out of the way (like a preset) and the other to be the main pot controlled from the front. I guess one can just go on the rear panel.

I guess two Hans' boards controlled by one Maya could also provide a fixed pre-set range by configuring it as a phantom input (that has no input switch element) and the other as the main volume control. That way you can maximise the useful number of volume control steps.

Or is that over the top, complicated? Is the idea with your board that you configure the range via choice of resistor values in the first place?
 
Last edited:
What components need to be mounted on your board Hans? I see 1 SMD resistor, the through-hole connectors and relays.. do I see a couple of SOIC chips to put on too?

Tempted to get a few of your PCBs even without the Maya (but could go for one Maya to see what it's like). Should help the GB to buy more of the PCBs, one for each of the BPPBP boards I have.

Let's make it clear to people that they can use this Maya GB to simply obtain your board too without the Maya (assuming that's ok..?). Others like Billshruv can get six of your boards and use the Maya to control all six for multichannel.
See the circuit diagram below for the used components. Resistors are 0805 SMD's, Diodes are RS1D, Relays are IM23TS.

Relay Board.jpg

Hans

If so, how's the pitch on those - easy (relatively speaking) to solder each pin or should it involve drag soldering or hot air gun?
I solder everything, also the SOIC chips, with a normal small soldering iron.
When something goes wrong, remove the solder with desoldering braid and start again.

Hans
 
For the second pot idea, judging by the maths/theory, both need to be as identical in their non-linearity as possible? So essentially the same pot?

Asking as in terms of control design, you'd want one to be small and out of the way (like a preset) and the other to be the main pot controlled from the front. I guess one can just go on the rear panel.
Yes

I guess two Hans' boards controlled by one Maya could also provide a fixed pre-set range by configuring it as a phantom input (that has no input switch element) and the other as the main volume control. That way you can maximise the useful number of volume control steps.

Or is that over the top, complicated? Is the idea with your board that you configure the range via choice of resistor values in the first place?
I don't get your question.
Control range is always 63dB in 1dB steps, that should be more than adequate for every situation.
The volume control board, with the resistor values R14 and R34 as shown in the circuit diagram, has a control range going from -57dB to +6dB.
By changing R14 and R34 you can shift the 63dB range upwards or downwards, f.i. from -63dB to 0dB (R14/R34 = 2K) or from -51dB to +12dB (R14/R34 = 8K)

Hans
 
Yes

I don't get your question.
Control range is always 63dB in 1dB steps, that should be more than adequate for every situation.
The volume control board, with the resistor values R14 and R34 as shown in the circuit diagram, has a control range going from -57dB to +6dB.
By changing R14 and R34 you can shift the 63dB range upwards or downwards, f.i. from -63dB to 0dB (R14/R34 = 2K) or from -51dB to +12dB (R14/R34 = 8K)

Hans

You answered it anyway! Thanks.

I was just thinking in terms of your board replacing the pots directly, Therefore the twin pot idea mentioned beforehand would be replaced by two of your boards...

However you just clarified that the whole point of your board is that you don't need two pots to set the useable range because the board was of course designed for this very purpose alone... silly me, I forget! :D
 
You answered it anyway! Thanks.

I was just thinking in terms of your board replacing the pots directly, Therefore the twin pot idea mentioned beforehand would be replaced by two of your boards...

However you just clarified that the whole point of your board is that you don't need two pots to set the useable range because the board was of course designed for this very purpose alone... silly me, I forget! :D

Natberg,

I can see you asking lot´s of questions relating to Maya.
Can you be more specific what exactly it is what you want, because I can only see that you want to use more than one BPBP at the same time.
Why is that, for different sources at the same time in a multichannel environment or what?
And what about selection of those input channels, how many input channels, how many at the same time, (different) volume control for how many channels, etc, the whole lot.

So long as the "Hans" board hasn't gone into production, I could still make minor changes to the layout. So be as specific as possible.

Hans
 
Hmmm It would be nice to have a possibility control 3 BPBP boards at once using one Maya board. It would be great preamp for 3-way active DSP stereo system where someone would like to have additional, transparent analogue regulation for example with 16-bit or 24-bit DAC.

I get the feeling you may be talking about steering a 3 way loudspeaker system , and if so why would one need 3 preamps for that, or have you something else in mind?
Paralleling 3 BPBP's with 3 Hans boards is possible without any modification, all controlled by one Maya, but is that what you want, all three doing exactly the same?
And where does the DAC come into the equation?
Sorry, but I need more information to understand what you suggest.

Hans
 
For the above, I would think he means using a digital crossover -> digital output -> 3 x DACs -> 3 x BPPBP -> 3 x powr amps. Usual way of implementing digital crossover via DIY . Commercial, implementations like DEXQ will use digital level control, but when you've such low distortion with BPPBP, analogue volume could possibly have the sonic edge anyway..

All three doing the same would be fine for that application given three identical power amps are used. The digital crossover should ensure correct relative levels are outputted.
 
Last edited:
For myself, I was thinking of a just a couple of things with the Maya:

1) main volume and a tape output both with independent level from the remote/Maya front panel.

2) Stored levels for each input (for A/B comparison of sources, for use with older, non-line level sources etc)

I get the impression that 1) is possible if one could set different I2C serial addresses on your board but 2) isn't really possible unless the ladder/relay scheme was exactly the same as Maya's own attenuation boards.

(forget about my other question equating your dual pot scheme with your boards - I was adding 2 and 0 together and getting 4.. my mistake).
 
For the above, I would think he means using a digital crossover -> digital output -> 3 x DACs -> 3 x BPPBP -> 3 x powr amps. Usual way of implementing digital crossover via DIY . Commercial, implementations like DEXQ will use digital level control, but when you've such low distortion with BPPBP, analogue volume could possibly have the sonic edge anyway..

This is what I meant. One BPBP per one way in active system.

For complete DSP 3-way system there will be following:

digital source -> digital crossover -> digital output -> 6 x DACs -> 3 x BPPBP -> 6 x powr amps.

Max unrestrained digital attenuation for 24-bit output lies near -48dB. If my system has 110dB max output it means that it is still 62 dB loud.
 
For myself, I was thinking of a just a couple of things with the Maya:

1) main volume and a tape output both with independent level from the remote/Maya front panel.
Maya in its current version when I'm right, can only select one channel at the time and control the volume for that channel.
In that case, controlling two volume settings on two Volume boards, will need two independent Maya controllers.

Only with a substantial software change, enabling the selecting of more channels at the same time with different volumes on different volume control boards would this be possible, and that would also need the setting of different I2C addresses on the Hans board.
The latter is no problem at all, but I don't think Tibi wants to take this route just for a few users.
2) Stored levels for each input (for A/B comparison of sources, for use with older, non-line level sources etc
That is a feature that comes standard with Maya, but be aware that the BPBP has only two inputs, so A/B is channel1 vs channel2.

I get the impression that 1) is possible if one could set different I2C serial addresses on your board but 2) isn't really possible unless the ladder/relay scheme was exactly the same as Maya's own attenuation boards.
Both impressions are wrong.

1) is only possible with 2 current and unmodified Maya controllers, and

2) has nothing to do with the relay boards.
Maya CPU uses the I2C bus to select a channel, and to set the volume for that channel.
The controller only sees the PCF chips, and cannot see what type of ladder is used.
It can see though how many relay boards are used.
Remembering the volume per channel is a software feature within the CPU and is not a hardware function.
Switching between channels thus means for the CPU also restoring to the latest volume setting for that channel.

Hans
 
This is what I meant. One BPBP per one way in active system.

For complete DSP 3-way system there will be following:

digital source -> digital crossover -> digital output -> 6 x DACs -> 3 x BPPBP -> 6 x powr amps.

Max unrestrained digital attenuation for 24-bit output lies near -48dB. If my system has 110dB max output it means that it is still 62 dB loud.
In that case, you can simply parallel three volume boards for their input from one single Maya controller.
In case you need differences in gain between the 3 channels, you can change R14/R34 for that reason, giving a permanent difference in gain between the 3 channels.
The volume control will keep the channels in line because each volume step will remain 1dB, independent of the value of R14/R34.

So for this application, the PCB can stay as it is now.

Hans