Choke for 160 watt bass amp (yet another couple of questions)

[Dave]

Hey there friends,

Here's a question...

I have a Peavey Roadmaster 160 watt guitar amp that I have twiddled the frequency responses and a few other things to be more bass guitar friendly. I have it sounding pretty good as a bass amp on the clean channel.
I thought it could benefit from a choke, so I put in an old choke from a 70's Twin Reverb. It seems to work pretty well, but I am a little concerned that I might be working the choke too hard.
I think the Twin Reverb chokes were rated at 120mA.
The problem is, I don't know how much screen current my 6 X 6L6GC's are pulling (along with the 6 preamp tubes).
I have tried to figure it out, but come up with more questions than answers. The best I can tell is that it is probably drawing around 100 mA but I'm not sure.
Anyone have a tried and true method for figuring that one out?
B+ at the plates is about 470, screens are getting a couple of volts less. I have the amp biased at 100 mA per side. It runs 4 12AX7's and 2 12AT7's.
Is that Twin choke OK?
What happens if you overload a choke?

Dave

[sluckey]

Calculating screen current is easy if you have screen resistors. Just measure the voltage dropped across a screen resistor and divide by the value of the resistor. Repeat for each tube and add all currents for total current.

[FYL]

I think the Twin Reverb chokes were rated at 120mA.

The 125C1A is rated at 4H @ 90 mA, with app. 100R DC

The problem is, I don't know how much screen current my 6 X 6L6GC's are pulling (along with the 6 preamp tubes).

According to the datasheet, you can expect 3 mA min, 11 mA max per tube in AB1 PP. With 3 6L6GC's per leg, screen current draw will be (3 x 3) = 9 mA when idle and (3 x 11) = 33 mA max, for a total of 42 mA (3 active 6L6GC's, 3 idling). Each 12AX7 section will typically draw 1.5 mA or less, 4 tubes, 8 sections, (8 x 1.5) = 12 mA. Each 12AT7 section will draw 10 mA or less, 2 tubes, 4 sections, (4 x 10) = 40 mA.

So we have 42 + 12 + 40 = 94 mA max.

Is that Twin choke OK?

Fender used the similar 022699 on big bore amps such as the 300 PS and 400 PS, so it should be OK. But I'd prefer to use a 120 mA version, just to be on the very safe side.

What happens if you overload a choke?

Two phases : mild overload => inductance drops, the choke becomes less effective; major overload => the choke overheats.

[Dave]

Thank you sir. That's the information I was looking for.

Dave

[Dave]

Here's a couple of other questions about getting this thing to perform better as a bass amp. I have the frequency responses set up the way I want them.
Do you guys think that it would benefit from 220uf/220uf for the first two filter caps?
What is the point of having V4-A running at 135 volts on the plate?
Here is a link to the schematic.
http://www.schematicheaven.com/newamps/peavey_roadmaster.pdf

Thanks,

Dave

[PRR]

> 220uf/220uf for the first two filter caps?

16uFd per pair 6L6 does seem under-capped. But PV rarely skimps where it shows (in this case, as excess sag).

> more bass guitar friendly

Ah, then you may well want "more solid".

I'd take 40uFd per pair, which in this totem-pole feeding 3 pair comes to two 330uFd.

In a similar amp, IIRC, I ran six 470uFd for effective 700uFd, got measurable test-bench power increase and brutal sonic impact. (Hmmm, that may have been a 2x150W chassis.)

> What is the point of having V4-A running at 135 volts on the plate?

Why not?

It is 12AT7, a high-conductance triode, feeding a fairly high load impedance. Where it is at, 135V, it could easily swing 90V peak either way. The next stage needs less than 3V peak.

It seems well-designed to give maximum open-loop gain for a 12AT7 (picked because they needed a high-conductance triode for the reverb driver). This high gain is then used in negative feedback to give about unity gain and super-low coloration. No extra coloration is wanted because there is plenty of color in the preamp, tone, and power sections. Too many layers of color makes mud. (And the Pump stage can do that when desired.) This is just a "glue" stage, to bring together the many signal paths.

> how much screen current my 6 X 6L6GC's are pulling (along with the 6 preamp tubes)

A ruff answer is right there on the plan. Plate-feed at 490V. Screen-feed at 475V. In between is a 400 ohm resistor. (490V-475V)/400= 15V/400= 0.037,5A = 38mA.

How much of that is preampery? From screen feed to driver feed is 475V-445V across 1K, or 30mA. 10mA runs in the reverb driver. However the next tap down seems to be 30V/2.7K or 11mA.

Note: "Voltages may vary". I do not think these numbers are supposed to be exact. In debugging, it usually isn't "little" errors like 415V versus 410V, it is way-off like zero or 50V where should be 400-some V.

Anyway..... That suggests 8mA for six screens. Is that possible? On 6L6 the idle screen current is ideally zero (the grid wires are supposed to "hide" the screen wires), and may be plus or minus. The screen current will rise when clipping. 6L6 does not have hi-volt G2 data but 7027 is a very similar devices with some high-volt data. We may take 5mA-22mA/pair as reasonable. 15mA-66mA per 6-pack.

This suggests the R96 point flows 30mA-90mA more or less.

However 90mA in 400r gives only 36V drop, just 21V drop from idle, which is "nothing" compared to the nominal 475V G2 voltage. I would not expect a choke here to do any good, just more arm-pain. Supersize the first cap-stack first and beat on it a while; that may be what you think you want a choke for.

[Dave]

Thank you sir. Your generosity is much appreciated.

Dave

[Dave]

Two more questions,

Two 10pf caps tie the plate to the grid of the V1a. I assume this is some sort of local feedback loop and I know that it doesn't work very well if you remove them. They seem to steal some of the high end content of the tone and I would like to get rid of them, but its not that simple. The overdrive channel gets all irritated if you just take them out, although the clean channel sounds better.

Also, everytime the signal path comes out of a new gain stage, it runs through a "coupling cap" and then through a cap paralleled to a resistor .0015/100k in series with the signal path. It looks like to me that it is an attempt to squeeze off low frequency content and allow easier flow of upper mid-range and higher. Yet, again, that's just me looking at it and trying to decipher the function. Am I right about this? If so,,,, I want to jumper those. Any objections?

Dave

[sluckey]

That .0015/100k is part of a voltage divider used to lower the signal between stages. The cap bypasses the higher frequencies around the 100K with little attenuation. Bass frequencies will be attenuated more. You may have too much gain if you just put a parallel jumper across those. Just try it and see if it's still stable.

You could also just remove the .0015 so that all frequencies are attenuated equally. This would accomplish the same as the jumper, but would leave the voltage divider intact, so the gain should not increase.

[Dave]

I tried using a jumper and found that it is not only perfectly stable with the jumper, but sounds significantly better. I can't really see why they did this in the first place unless they were trying to accentuate upper mid frequencies and retain that sort of signature peavey sound.

As far as the little local feedback loop, I moved the whole thing to the V1B. It seems to have been included to keep the crunch channel stable, but when used on the V1A it just sucks all the chime out of the clean channel. It seems to work just as well on the V1B and there has no effect on the clean channel. So, now, the clean channel is much nicer.

Dave

[Dave]

OK, I have been trying to work this out, and I am not having much luck.
The V4-A is the mixer stage and I don't like it at all.
There is a 1meg grid stopper there that is a real tone sucker.
Here's the trouble. If I reduce that grid stopper to 100k, for example, the clean channel really comes to life and sounds fantastic (with extensive modifications). However, at anything significantly less than the 1meg, the dirt channel gets really unstable and suffers from tape worms and other horrible parasytic problems.
So, right now, I have settled on 1meg bypassed with a 120pf cap, and it works pretty well like that, but I am far from satisfied.

If I take the signal off the grid of the V4-A and bypass it altogether, It is not unstable at all, so the wierd oscillations are coming from the V4-A but I don't know why. Any thoughts?

Dave