help with 3 channel preamp design

[HTH]

I've recently got round to finishing up my backup rig which happens to be a rack. I'm using a Marshall 9005 50w+50w power amp with a Marshall 9001 preamp. However, the 9001 preamp really isn't up to much and the tone is very processed for want of a better word - not organic in any way. The schematic shows that the signal path is mangled in a big way, so its not surprising really.

That leads me to this thread. I want to build a 2U preamp with 3 channels:

(1) blackface cleans
(2) JCM800 crunch (modded slightly with my usual recipe)
(3) filth gain in the vein of a Bogner Uberschall

To reduce the amount of switch and keep the signal path fairly 'pure', I'm using separate preamp tubes for each channel (i.e. they're not shared between the channels) meaning there will be eight ECC83/12AX7s.

The preamp stages separately are not really an issue. Its the connecting of each preamp signal so that the volumes are fairly evenly matched. In the past when I've tried mating a Fender clean type channel with a high gain channel, the high-gain channel is WAY louder and putting a volume pot in there to lower the dirty channel tends to kill the tone. Whats the best way of matching the volumes while keeping the tone intact?

I've attached a basic schematic of this preamp, but have left out channel volumes. I'd like suggestions on where would be best for each channel to have their respective volume controls for balancing the channel volumes and also general advice on getting the best from this type of setup.

The switch would ideally be via three momentart footswitches to select each channel, but in reality it's probably going to be easier to have two channels with one of the channels switching between two modes. In other words, channel 1 is clean, channel 2 switches between the JCM800 and Uber channels.

Any/all help appreciated.

Also, has anyone got suggestions for getting a suitable PT?.. 8x ECC83/12AX7, 600mA x8 = 4.8A heater current. Thats gonna be quite big, but the HT winding will only need to be around 100mA as each preamp tube is only gonna pull around 10mA.

[HTH]

here's the schematic... (btw, ignore the B+ line, I know this will require filtering between stages, schematic is for illustration only at this point)

[jimboyogi]

I checked the spec sheet and the 12AX7 only draws 300 mA heater current at 6.3V, 150 mA at 12.6V.

So your PT only needs 0.3 * 8 = 2.4A heater current, much easier to find!

[Steve Conner]

You'll need a volume pot at the end of two of the three channels to equalize them. Ideally the two higher-gain ones, they'll be louder. If the pot kills the tone, you just have to find a way round that.

Hint: A high-resistance volume pot feeding a length of cable to a power amp is just about the worst tone suck you can imagine. (google thevenin equivalent, and cable capacitance, scratch your head for a bit to see why)

Use the lowest pot resistance practical, buffer the wiper before driving a cable, and if you still don't like the result, sling a bright cap on there.

Transformer wise, this would probably be a good application for the old back-to-back transformer trick. 240V to 6, and then a smaller 6 to 240.

[HTH]

I checked the spec sheet and the 12AX7 only draws 300 mA heater current at 6.3V, 150 mA at 12.6V.

So your PT only needs 0.3 * 8 = 2.4A heater current, much easier to find!

yep, quite right - it was late when I typed that.

I have a Hammond 369JX (250-0-250 @ 60mA, 6.3vac @ 2.5A), that should be perfect. Maybe a little high on the HT (350v rectified) for ECC83s, but I can pad that down.

[HTH]

You'll need a volume pot at the end of two of the three channels to equalize them. Ideally the two higher-gain ones, they'll be louder. If the pot kills the tone, you just have to find a way round that.

Hint: A high-resistance volume pot feeding a length of cable to a power amp is just about the worst tone suck you can imagine. (google thevenin equivalent, and cable capacitance, scratch your head for a bit to see why)

Use the lowest pot resistance practical, buffer the wiper before driving a cable, and if you still don't like the result, sling a bright cap on there.

Transformer wise, this would probably be a good application for the old back-to-back transformer trick. 240V to 6, and then a smaller 6 to 240.

thanks for that Steve,

in addition to whats shown on the schematic, the summed outputs will feed into 'channel volume' pots (500k) then into a an AC-coupled CF as an effects send and a regular gain stage as an effects return. that will most likely feed into a master volume pot and out into the power amp. would something low like a 100k master volume pot be best? or how about having another AC-coupled CF after the master volume pot to drive the power amp?

[defaced]

The OD3 takes a similar approach to what you've done, and it uses 1M master volumes. You can find the schematic on the SLO Clone forums which may be of some interest for reference. You will need to register to view the forums. Sloclone Forums • Index page

[HTH]

The OD3 takes a similar approach to what you've done, and it uses 1M master volumes. You can find the schematic on the SLO Clone forums which may be of some interest for reference. You will need to register to view the forums. Sloclone Forums • Index page

Thanks for that link, just what I was after. I had really come up with the concept of this preamp in my head to serve the tones I want, but the similarity of the OD3's design to what I've drawn out just goes to show there is nothing new with valves.

In all honesty, I'd probably get away with 95% of what I do with just the JCM800 channel, but if I'm gonna do this, might as well be the last preamp I ever want to build for any purpose.

[rathmann]

quite a number of similarities with this one too - http://www.blueguitar.org/new/schem/...iple_giant.jpg

Switching is accomplished by muting via Vactrols, however I'm not sure what handles the logic.

[HTH]

quite a number of similarities with this one too - http://www.blueguitar.org/new/schem/...iple_giant.jpg

Switching is accomplished by muting via Vactrols, however I'm not sure what handles the logic.

I'm happier with relays and have had good pop-free results with them so far.

[Steve Conner]

A cathode follower driving the cable to the power amp should do the job fine.

You may also end up using the 100k master volume, though, with a larger fixed resistor in series with the clockwise end to limit the range. A direct output from an overdriven tube stage can be over 100V peak-peak, way higher than a rackmount power amp will expect.

I once used a mic input matching transformer backwards in this application, instead of a voltage divider and CF, hoping to overdrive it and add some transformer mojo to the preamp.

I'd be really interested to hear how this sounds when it's done.

[HTH]

A cathode follower driving the cable to the power amp should do the job fine.

You may also end up using the 100k master volume, though, with a larger fixed resistor in series with the clockwise end to limit the range. A direct output from an overdriven tube stage can be over 100V peak-peak, way higher than a rackmount power amp will expect.

I once used a mic input matching transformer backwards in this application, instead of a voltage divider and CF, hoping to overdrive it and add some transformer mojo to the preamp.

I'd be really interested to hear how this sounds when it's done.

I'll be sure to post clips once its all sorted. Thanks for the help Steve (and everyone else who's chipped in).

[chipaudette]

What voltage are you going to use for B+?

If you run on the low side (~240-260V) you may get a very different sound than if you run on the high side (400V).

Just curious...

Chip

[HTH]

What voltage are you going to use for B+?

If you run on the low side (~240-260V) you may get a very different sound than if you run on the high side (400V).

Just curious...

Chip

the transformer I'm using will give me around 350v right after the recto diodes.

I'll be aiming for typical voltages seen in Marshalls and such in high gain designs - around 140v on the plates is always nice and chewy.