Mastering console

[Bear]

Hey, I'll gladly use someone's IC mastering console design - it'd be far from the weakest link in any mastering process I would undertake. With really high quality IC's well implemented, I don't see as much potential difference to disctrete circuitry as with a heavy lifting task like a mic pre or a summing bus. I'm ready to be schooled on this point, though.

Bear

 

[TomWaterman]

I'm totally with you bear,

I think the BB chip version should work well if implemeted properly. The chips I mentioned will keep the cost down no end. A PGA2310 is about $10 and has 255 steps, a rotary switch for a passive will run around $200 and only have 20-30 positions. A relay based attenuator may be around $150 but adds a great deal more steps and can remain passive.

I have been trying to incorporate various elements into mine and the passive system is just so appealing. I'm planning on using Forssells for my small active circuit as they are so clean and will also run on bipolar 24V = higher headroom (only a small bit but its there). With a passive atty in front of them, the whole console should have almost 'infitite' headroom.

The one argument I can think of against IC's is that the majority are in need of large doses of NegFeedback as they have rediculously high open loop gains. Something like a Forssell992 however, has a relatively small open loop gain <100dB, as such requires less feedback.

Although something I have not had time to investigate thoroughly, many many 'hi-fi' people prefer designs with low or no neg feedback and low openloop gains.

Something to think about. Plus of course there a far fewer components within a F992 compared to say an OPA627.....I think I prefer a simpler signal path for this job......

Cheers Tom

 

[PRR]

> its an inverting opamp stage that is used to change volume.

The PGA2310 is non-inverting.

The WM8816 is inverting. (You can fix this elsewhere in the chain, if you really care.) It uses an external opamp.

> there is something in the datasheets (can't remember off hand) but I think you'll need to run the device into an impedance way above 600ohms otherwise THD starts to rise.

No, it is rated to drive 600 out, but if the source impedance is high-ish you get THD in the switches used to set the input gain.

If you are using a chip to unbalance a pro-interface input, "INA134/7 on the front of it ", that will swamp the gain-switch THD.

The killer problem on the PGA2310 is that digital input, and devising a nifty user interface. Plain old POTs have some nice features: they don't lose setting when the power goes off, the control is the readout and the readout is the control. With digital knobs you end up with a cluttered central readout or a lot of little LCDs next to each control. Or develop something around the new touch-sensor chips, plus an LCD, so you rub a strip and it shows where it is set.

 

[smorphet]

[quote author="PRR"]
The killer problem on the PGA2310 is that digital input, and devising a nifty user interface. Plain old POTs have some nice features: they don't lose setting when the power goes off, the control is the readout and the readout is the control.[/quote]

What's involved in getting a digital output from an analogue pot? It must be one of the least demanding ADC applications, and 8 bit ICs start at a couple of dollars. Anyone tried to DIY something like this?

 

[TomWaterman]

[quote author="PRR"]The PGA2310 is non-inverting.

The WM8816 is inverting. (You can fix this elsewhere in the chain, if you really care.) It uses an external opamp.[/quote]

Ahh yes you are correct sir. I knew one of them was for sure.

[quote author="PRR"]No, it is rated to drive 600 out, but if the source impedance is high-ish you get THD in the switches used to set the input gain.[/quote]

I was thinking it may have been the other way around. Doh! So of course the low output impedance of the receiver stage is a bad idea. Any recommendations for a way to keep the source impedance high? 10k-10k transformers?

Thanks for the clarification PRR - (note to self * read the dataheets before posting!)

:thumb:

-Tom

 

[PRR]

PRR wrote:
No, it is rated to drive 600 out, but if the source impedance is high-ish you get THD in the switches used to set the input gain.

So of course the low output impedance of the receiver stage is a bad idea.

Read it again. You WANT a low source impedance. A chip is ideal.

 

[daArry]

Hey PRR

So with say a INA134 receiver on the input - a buffer of some sort would still be needed after, before goin to the PGA ye? then again on the output of it...

Like implemented on this page:

http://www.mhennessy.f9.co.uk/preamp/analogue.htm

Also some good info there on the PGA in the research section...

Ta

 

[Svart]

What's involved in getting a digital output from an analogue pot

a couple fets, some diodes and a few resistors set to turn each fet on as you turn the pot to lower resistance. that would give you a step type effect. other than that most people use rotary encoders for this type of setup. those "pots" in digital mixers are really encoders..


I'm working on some touch sensor devices for work, I've seen the new slide touchsensors(both the circular and the strips..) and they are really nothing but multiple touchsensors in a single IC, the sensor pads are segmented, and as you move you trigger a different section of pad. they only output digital.. no analog.. :sad:


:thumb:

 

[TomWaterman]

[quote author="PRR"]Read it again. You WANT a low source impedance. A chip is ideal.[/quote]

Oh geez , talk about getting me confused.....I'm scared by your posts PRR - I get intimidated like a little girl and then try and justifiy something that doesn't even exist!! hahaha

My brain was saying that somewhere the impedance must be high but it is in fact the other way around......I knew there was a reason why I was planning on putting a receiver berfore the PGA if I used it. Lets just forget I ever spoke eh? :wink:

I have a question, say you were going for a fully differential path, using one PGA for left and another for right, do you think that they would track reasonably accurately, differences between two chips? I meant to get another PGA and test it in the evaluation board but I haven't got around to it.

Cheers Tom

 

[Svart]

how about lets say, 4 simple opto( or whatever else..) compressors each with a simple gyrator based paragraphic EQ on it's sidechain? you could sum all bands up with a simple discrete gain stage or trafo or whatever..

multiband compression on the cheap.. this would be easy besides being a staple of a mastering setup.

 

[PRR]

> So with say a INA134 receiver on the input - a buffer of some sort would still be needed after, before goin to the PGA ye?

Why? What is the difference between an INA134 and an OPA2134? To a first approximation, both are low-Z outputs. And in this specific case: I think a INA134 is a OPA(2)134 with some standard resistors in the package, so they are the SAME at their outputs.

> then again on the output of it... Like implemented on this page:

He has a good, to him, theoretical reason to buffer the PGA2310 output. Personally I think it is a frill: few loads are really 600Ω, and as good as the 5532 is, I'm not sure why we want an added step of distortion, not to mention complication and board space.

SIMPLIFY: If your source is unbalanced and reasonably low-Z (1K is no big problem), and you are not obsessed with triple-oh THD numbers (board layout can degrade THD more than the test-specs suggest), then just use a naked PGA2310.

> say you were going for a fully differential path, using one PGA for left and another for right, do you think that they would track reasonably accurately, differences between two chips?

Tracking specs are on the datasheet.

 

[TomWaterman]

[quote author="PRR"]Tracking specs are on the datasheet.[/quote]

Ahhh thanks I'll check it out.

Cheers Tom

 

[ijr]

Any pics from the inside of that Crookwood?

Not easy( studio is far from me, and if I have nothng to do
else, I will not drive to this studio just to take pics...)
but I will try.

 

[TomWaterman]

Hey thats cool Igor! :grin:

Don't go out of your way - I'm not asking you to drive all the way to a studio...!! I was just curious how they were put together inside.

If you ever have to service it again then pics would be cool but don't make a special trip for me.....

Thanks for the offer though.

Cheers Tom