Small valve mixing console

[C.B. - Boudio]

This looks like a very nice circuit.

Do you mean the EAB or my revision too?

That is a NFB tone control as used in many of the Mullard/ Philips in their project books. The low Z drive requirement does not apply.

Oh okay, that makes sense.

That is not feed forward, it is a totally standard NFB circuit.

But the signal at the anode of the second triode is in fase with the grid/cathode of the first triode, no?

You could do a lot worse than copy this schematic.

The reason I don't really want to copy it as-is, is because the output is so low, and if I'd use an SRPP or something like that as an output stage, the input won't be in fase with the output (well, I could wire the IPT accordingly, but I'll still run into the same problem for the instrument input), and I fear that even more gain stages will make it too noisy even for me. Otherwise, I would probably do it that way.

 

[Disco Volante]

The first cathode lacks a capacitor, which is effectively local feedback meaning lower gain. A pentode output stage might also help get you where you need to go?
I'm sure DaveP would know how to calculate all of that ;-)

 

[DaveP]

But the signal at the anode of the second triode is in fase with the grid/cathode of the first triode, no?

Yes, it is in phase and it makes the un by-passed cathode more positive, more positive than the grid which is the same as the grid becoming more negative, i.e. NFB.
best
DaveP

 

[ruffrecords]

View attachment 146104

Since you say a baxandall has to be driven from a low impedance source to a low impedance load, I'm a bit confused then by what they did here, with such high impedances.

It is all relative. The bass and treble pots are most likely 1Meg and the output impedance of the driving stage is about 50K which is low compared to 1Meg

Cheers

Ian

 

[DaveP]

Do you mean the EAB or my revision too?

I mean this VE 12 German amp. They are nearly always very well designed.
best
DaveP

 

[C.B. - Boudio]

Yes, it is in phase and it makes the un by-passed cathode more positive, more positive than the grid which is the same as the grid becoming more negative, i.e. NFB.
best
DaveP

Now I see! I thought NFB was always with one signal in the opposite phase.

It is all relative. The bass and treble pots are most likely 1Meg and the output impedance of the driving stage is about 50K which is low compared to 1Meg

Cheers

Ian

Ah, that makes sense. Bass and treble are actually 510 and 470k respectively (see the resistor ladders at the bottom).

 

[C.B. - Boudio]

This is the state of the circuit now, with a line amp and some of the routing.
I won't be recreating a VE12 because of it's very low output (and associated needed extra stages and thus valves), the IPT's I have access to, and last but not least the cost of ECC83's (and most double triodes) compared to pentodes (I thought for a split second of using EF9x's, maybe triode connected, but I don't have the space and I may or may not run into problems concerning heater current) - except for if there are workarounds for all of the above.



There are inserts before and after the fader, mainly to be able to use each channel as an independent preamp, with or without a master volume, or if I'll ever find a party of fancy EAB/Telefunken/... faders, that I can leave the rotary fader out (then the faders will sit fixed in a desk, and the console itself could come out of it easily to record on location, for exemple).
Why? I don't really know yet, but it seemed handy so I might as well do that. (I fear that that's one of my main philosophies...)

The third bus switch is for the monitor section, which will consist mainly of a transistorised headphone amp.

The ECC81/double EC92's (well, any triode will do, I think) buffer send and return, but I won't implement those immediately.

And below is the line amp, for the two stereo buses (so I will eventually have four of those). The transformer is a 10k:150 1/2W Edcor. PL95 triode connected, fed by an EF80 (or if needed, something higher gain, I can't tell yet).

There are also 2 stereo line inputs as an add-on, switchable between off/bus 1/bus 2, as simple as it's drawn there. If switched out, should the pot get replaced with a 100k resistor?

I hope this is all a viable design, with the limiting factors I have. If not, don't hesitate to correct!

 

[DaveP]

C.B.
There are a few problems I can see:-
The two EC92's and the PL95 are "Bootstrapped" but not in a good way!
Read this article to work out how to calculate the resistor and cap, https://valvewizard.co.uk/accf.html
R19, R24 and R27 should not be as high as 2.2M, they will cause noise and instability, they should be more like 330k/220k and even then the input Z will be around 5M. C11, C13 and C16 should be calculated to give (with the new input Z) a suitable high pass frequency and the caps will probably only need to be around 10nF.

Be very careful how you wire up the heaters as some are 6.3V and one 4.5V, it will be easy to burn out something with a simple wiring error.
best
DaveP

 

[C.B. - Boudio]

C.B.
There are a few problems I can see:-
The two EC92's and the PL95 are "Bootstrapped" but not in a good way!
Read this article to work out how to calculate the resistor and cap, https://valvewizard.co.uk/accf.html
R19, R24 and R27 should not be as high as 2.2M, they will cause noise and instability, they should be more like 330k/220k and even then the input Z will be around 5M. C11, C13 and C16 should be calculated to give (with the new input Z) a suitable high pass frequency and the caps will probably only need to be around 10nF.

Thanks, I didn't know that. I'll correct them.

Be very careful how you wire up the heaters as some are 6.3V and one 4.5V, it will be easy to burn out something with a simple wiring error.

The plan is to wire the heaters in series per module. 2x PF86 and a PL95 will work at 12V (should be 13,5 from the top of my head but it's probably not bad to underrun them slightly).
An ECC81 or 2x EC92 can work at 12V (the send/returns), and then the EF80s and EL95s in pairs, per (stereo) bus.

 

[C.B. - Boudio]

@DaveP For the bootstrapped cathode followers, I took the 1M from one of your designs:

(ah, seems like I misremembered Rk and Cout a bit)

 

[C.B. - Boudio]

This is now the most recent revision.

 

[MidnightArrakis]

This is now the most recent revision.

[This is now the most recent revision] -- J6/Pin-2 isn't connected to anything:



The schematic symbols for J5 and J6 are typically used for showing either BNC or RCA connectors. What are these connectors?

Also.....is J2/Sleeve, that appears to me as being a common "GROUND" connection going to the rest of your entire circuit, supposed to be -- ISOLATED -- from connecting to any actual "GROUND"? I don't see any "GROUND" symbols here.

And, I'm not completely sure, but.....I don't think that INPUT J1/Pin-1 and OUTPUT J10/Pin-1 are supposed to be connected together. Other forum guru's can chime in about this connection, but I'm under the impression that INPUT and OUTPUT "GROUND" connections are a "NO-NO" due to causing ground-loops.

/

 

[C.B. - Boudio]

These are RCA connectors, unbalanced inserts so to speak, and bridged by a bent solid copper wire, just like you see on those older smaller Teac/Tascam mixers for example. Their rings (pin 2) go all to ground.

Sleeve of J2 goes directly to ground indeed. I'll put a grounding symbol somewhere at that bus.

Pin 1 on XLR is always shield/ground and the signal is between 2 and 3, that's why J1 and J10 are connected together at their pin 1's.
Signal ground/0V/chassis is one bus bar, as it were, connected at only one point to mains ground (in the power supply enclosure; from there, the 0V connects to the mixer chassis. The RCA jacks and such are all mounted directly onto the mixer enclosure, ground connections not insulated from it).
There are probably slightly better ways of grounding, but at least I'm confident that this is feasable and virtually impossible to mess up.