Building a quieter JCM 800

Fatfingers

I own an old vertical input Marshall JCM 800 50 watt combo (4010). It sounds great but it's ridiculously loud. Louder, in fact, than an Orange Overdrive 120 I recently sold. I mean absolutely deafening. I keep waiting for the call from Angus Young to do a stadium tour on rhythm guitar for AC/DC, but he's got his cousin in the band now so it looks like I've missed the boat again. So I don't need the level of volume it's capable of.

I don't want to start chopping a vintage Marshall up (I did, but I've been talked out of it) so I've decided I'd like to build another one, with a less deafening power amp. (I don't want an 18 watt job. I have an 18watt Ampmaker P1800 I built. It sounds great, but it's not quite loud enough.) Just a pair of EL34s or similar turned down to something more like 30 or 40 watts flat out. I know this is rated at 50 watts but someone on here, possibly ICBM, says that's the undistorted rating, and that flat out he's seen the 50 watters churning out 80 or 90 watts when fully cranked. Which is how I use mine, albeit with an attenuator. But it's still too loud.

ecc83

Just a thought but you could try part cathode biasing?

You probably want about 330-360 V HT and maybe play around with the screen supply voltage.

Top Men will surely be on the case post wheatybangs...

Dave.

digitalscream

http://www.yellowjacketstc.com/

?

Fatfingers

Yeah, I've seen Yellowjackets. But as I'm building a new amp it seems a bit daft to just copy a JCM 800, with its hoofing great transformers and massive output, simply to then strangle it. I'd rather make one that's not as loud in the first place.

ICBM

The big beefy tone is mostly to to with the big transformers. If you use smaller ones you'll strangle it too, and it will sound more like the 18-watter again.

Cathode-biasing is a good call. Although it will only reduce the power a bit, it will make the amp much more compressed so it will seem less overpowering.

Also, here's an odd thought - try disconnecting pin 8 (the suppressor grid) on each EL34. I recently came across an amp someone had tried to convert from 6L6s to EL34s and had forgotten that you need to connect these to ground. To my surprise since I always assumed it *must* be done, the amp worked and sounded more or less normal but was low on power.

You could also run them as triodes - connect the screen grid resistors to the plates instead of the B+ supply - but that produces quite a muddy sound as well as less power, to me.

ecc83

ICBM said:

The big beefy tone is mostly to to with the big transformers. If you use smaller ones you'll strangle it too, and it will sound more like the 18-watter again.

Cathode-biasing is a good call. Although it will only reduce the power a bit, it will make the amp much more compressed so it will seem less overpowering.

Also, here's an odd thought - try disconnecting pin 8 (the suppressor grid) on each EL34. I recently came across an amp someone had tried to convert from 6L6s to EL34s and had forgotten that you need to connect these to ground. To my surprise since I always assumed it *must* be done, the amp worked and sounded more or less normal but was low on power.

You could also run them as triodes - connect the screen grid resistors to the plates instead of the B+ supply - but that produces quite a muddy sound as well as less power, to me.

Ah! Glad you agree about cathode biasing, just a hunch!

I would guess dissing the suppressor grids would turn the pentodes into "kinky" tetrodes and thus distortion start earlier?

Strapping as triodes is going to drop the anode impedance quite a bit I shall see if I can find out just HOW much.

Fatfingers, PM me.

Dave.

ICBM

ecc83 said:

Ah! Glad you agree about cathode biasing, just a hunch!

I forgot to say that you need to do it with separate resistors and caps per valve to keep the same sound, too - if you use a shared resistor it will turn it into a Vox .

(Not really, but it will take it a bit in that direction and remove some of the distinctive big Marshall crunch, I think.)

ecc83 said:

I would guess dissing the suppressor grids would turn the pentodes into "kinky" tetrodes and thus distortion start earlier?

Yes, I think so. I'd never even heard of it being tried before, I assumed it either wouldn't work or would damage the valves or something. Actually the amp ran fine and produced about 60% of the normal power.

ecc83 said:

Strapping as triodes is going to drop the anode impedance quite a bit I shall see if I can find out just HOW much.

Not enough to matter. Marshall do this on the Jubilee and JCM900 amps - it gives about 50% of the power, but to me it changes the tone in a not-good way as well as reducing the volume, so I'm not really a fan.

ecc83

"Not enough to matter. Marshall do this on the Jubilee and JCM900 amps - it gives about 50% of the power, but to me it changes the tone in a not-good way as well as reducing the volume, so I'm not really a fan."

Plus of course the PI and other stages have to provide substantially more drive voltage.

Dave.

jpfamps

I'm not a huge fan of triode strapped pentodes either

Your best bet, in my opinion, would be to use EL34s at lower HT voltage and possibly experiment with cathode biasing.

We use EL34s at 320 VDC, with a 1k dropping resistor in place of a choke running fixed bias into 5k6 loading. This gives up to around 28W unclipped depending on the valve type; some current production EL34s only give around 22 W in this configuration.

Using a choke would give more power, and this configuration would benefit from lower loading.

Cathode bias will give less power, and could be made user switchable.

Fatfingers

jpfamps said:

I'm not a huge fan of triode strapped pentodes either

Your best bet, in my opinion, would be to use EL34s at lower HT voltage and possibly experiment with cathode biasing.

We use EL34s at 320 VDC, with a 1k dropping resistor in place of a choke running fixed bias into 5k6 loading. This gives up to around 28W unclipped depending on the valve type; some current production EL34s only give around 22 W in this configuration.

Using a choke would give more power, and this configuration would benefit from lower loading.

Cathode bias will give less power, and could be made user switchable.

That sounds like the way to go then. Now I need to work out how to active that.