Critique My Schematic

[Daver]

Daver you're right on w/ the Dumble thing in the OD. There is however a coupling cap before the PI - C19 on clean channel and C22 on OD. So leaving the input connected to both grids and switching the output of the tone stacks would be better?

R25 and R26, in clean mode, are in parallel with R5 and R9 of the PI. In OD mode, the Volume control would be in parallel with the PI resistors. You need a blocking cap before the PI to allow it to bias correctly.

I've worked on a "boutique" amp that switches the channels at the grids. It always made a popping sound. After moving the switching to the "end" of each preamp, the popping was gone. You've been warned!

Dave

[Don Moose]

Don do you not think that the 2 parallel tubes in the OD will make it very quiet? Or is this unnecessarily quiet in your opinion.

The noise reduction of a parallel triode stage is something like 6dB, right? I don't know you'd notice that except in the spaces between the notes - and even then, it seems to me a waste of a socket and 300mA of heater current.

The only way to find out, of course is to try it.

[guitician]

there is C19 going into the switch and for the OD there is C22. Do you think those shall be changed?

C19 is used for clean too. C22 is after the distortion. I think to lessen the bass response going into OD you could put a cap between R33&R35.

[lowell]

In regards to the 33k across the volume pot - it's there as part of the voltage divider for the local neg feedback of that stage.

[lowell]

Ok here's a revision guys... thanks thus far and please let me know if there's anything else. Also, I changed the local feedback the only way I could see to do it... with the 10meg after the 2 tone stacks... is this gonna work or is the 10meg/1meg divider values too large current-wise?

[Daver]

You can just connect the 10M feedback resistor to the grid of V2A. Alternatively you can forget about using C19 as part of the feedback and use a separate .047uf and resistor of choice for the feedback like Dumble does. You still need to add a blocking cap before your phase inverter. You don't need to have the "tops" of the 1M volume pots connected together. If it were me, and you're not strapped for real estate, I'd still separate the cathodes on V1. You may need to tweak that once you get the amp running. Carry on!
Dave

[lowell]

so do you think the 10m feedback might just as well be an open connection? I've changed it to 500k and ditched the voltage divider configuration altogether. I will just experiment w/ that feedback once this amp is working I guess.

I'm still at a loss as to why you and others are saying I need another coupling cap before the PI... there is C19 from the driver stage and C22 from the OD stage...?? Am I missing something here I don't see any DC path to the PI grid?

[Daver]

You don't have to have the local negative feedback loop. You can add it later if you want. Resistors from 4.7M up to 44M will give you a variety of attenuation and compression.
"If" the negative feedback loop, as you have it drawn, is removed, then C19 will block any extra DC path to ground when in clean mode. When switching to OD, the 100KA pot from the OD would be in parallel with the phase inverter 1M and 68K to ground. For that mode you would need a cap to block the DC path to ground. We aren't worried about the DC being passed from previous stages. The phase inverter self biases as it is set up, and an extra pathway to ground will screw up that bias. Does that make more sense?
Dave

[lowell]

aha! yes that makes sense. I was not aware anyone was talking about blockin DC to GROUND, thought it was stage to stage. thanks.

[Suusi M]

Hello all,
This is the first design/schematic I've done and would like some advice. I have yet to design power supply, but you can catch the drift. My main concerns are R1 and if leaving both tone stages connected to grid of second tube is a bad idea. I'm also curious of the OD stage should be where it is in the chain. anyway check it out if interested.

Apart from the previous criticisms add the following:

1. the control grids of the output valves are not connected they should be connected to the phase splitter
2. the screen grids of the output valves are connected to phase splitter where they should be connected via the 470 ohm resistors to the HT
3. the suppressor grids are connected to the HT where they should be connected to the cathodes.
4. if you were to build this you would destroy the output valves, almost instantly, and very probably the power supply and output transformer.
5. V2b should have a signal of approx 72volts peak to peak to provide the correct drive for 100 watts out.

I put the pre amp into an electronics simulator (ltspice) and:

1. the drive to V2b is hopelessly inadequate at 2 Volts peak to peak, should be approx 72 volts peak to peak.
2. it is prolly better to use a 12DW7 for the phase splitter (1/2 12ax7, 1/2 12au7), the 12ax7 half to provide the gain to the 12au7 phase splitter half.
3. It looks like C19 / R26 junction should be connected to R24 and R20 so that the tone controls become 'active' as opposed to passive.
4. As a general rule noise is proportional to gain and temperature. 12AX7's use hum canceling heater windings so by doubling the overdrive valves is probably adding something called shott noise

Don't take this criticism too hard as a first attempt it ain't bad.

[Suusi M]

Ok here's a revision guys... thanks thus far and please let me know if there's anything else. Also, I changed the local feedback the only way I could see to do it... with the 10meg after the 2 tone stacks... is this gonna work or is the 10meg/1meg divider values too large current-wise?

Apart from the previous criticisms add the following:

1. the control grids of the output valves are not connected they should be connected to the phase splitter
2. the screen grids of the output valves are connected to phase splitter where they should be connected via the 470 ohm resistors to the HT
3. the suppressor grids are connected to the HT where they should be connected to the cathodes.
4. if you were to build this you would destroy the output valves, almost instantly, and very probably the power supply and output transformer.
5. V2b should have a signal of approx 72volts peak to peak to provide the correct drive for 100 watts out.

running the preamp circuit on the simulator (ltspice)

1. the drive to V2b is hopelessly inadequate at 2 Volts peak to peak, should be approx 72 volts peak to peak.
2. it is prolly better to use a 12DW7 for the phase splitter (1/2 12ax7, 1/2 12au7), the 12ax7 half to provide the gain to the 12au7 phase splitter half.
3. The Overdrive circuit is best before the tone controls, so they can color the overdriven signal.
4. lose the 10 meg resistor
5. change c19 to 100n
6. connect c 19 to r20 and r 24 so that the tone controls become 'active' as opposed to passive.
7. add 33k resistors in the missing arms of the tone controls
8. change r19 and r27 to 33k
9. lose c1 and c21
10. lose r18 and R21 and c14 and C9
11. change c7 and c8 with 100n 250V caps

see the attached circuit Draft16.pdf

Added a more conventional pre amp design that would work with almost any marshal / laney / ampeg power amplifier stage. Its an idea that I have been playing with for a while. your post made me finish the design

[lowell]

Are you saying that v2A is not driving the PI enough? Not exactly clear on that. And how would you suggest increasing it? I'm also going to look into this.

[Suusi M]

Are you saying that v2A is not driving the PI enough? Not exactly clear on that. And how would you suggest increasing it? I'm also going to look into this.

yes that is exactly what I am saying. It needs a gain stage with a gain of at least 36

Tung Sol 6L6GC data

check 55 watt amplifier class ab1 which is the analoge of the 4*6L6GC 100 watt - the p to p input is 70 volts.

I added a bit for luck and coz of losses due to the extra loading loading of the phase splitter

[lowell]

the V2B needs to receive a 72v p-p signal on its grid? I have just calculated the V2A stage as having a gain of 40 though. I'm using these values:

Vp=150
Vk=1.5v

Av=(mu*Rp)/(Rp*ra)
Av=(100*100k)/(100k*150k)
Av=40

In assuming V2A is receiving 2v on its grid that gives us an 80v drive signal to grid of PI. Am I missing something?

[Suusi M]

the V2B needs to receive a 72v p-p signal on its grid? I have just calculated the V2A stage as having a gain of 40 though. I'm using these values:

Vp=150
Vk=1.5v

Av=(mu*Rp)/(Rp*ra)
Av=(100*100k)/(100k*150k)
Av=40

In assuming V2A is receiving 2v on its grid that gives us an 80v drive signal to grid of PI. Am I missing something?

Look at the schematic Draft 16a.pdf which is your revised circuit without the Overdrive switch. As per my electronics simulator (ltspice), which allows me to test almost any circuit.

I have put the AC signal levels at the key points so we can agree on the key points