Tube Amps / Music Electronics
For current discussions, please visit Music Electronics Forum. New: view Recent Searches.
New: visit Schematic Hell!
The sunn still shines online!

Listen to great tunes streaming live right now!

ampage archive

Vintage threads from the first ten years

Search for:  Mode:  

Sagging JCM800 pre-amp or blocking distortion problems

8/5/2005 9:24 AM
Sagging JCM800 pre-amp or blocking distortion problems
I've modified a 1983 jcm800 2205 boost channel to 2203/2204 values, minus the cathode follower, I used this tube as the last gain stage (820 ohm cathode) then I tapped the eq in on the plate side. The amp sounds great but doesn't recover quickly. My supply voltage above the plates is only 305v, there are 5 tubes pulling off this supply. The voltage at the plate on V2b is 192v, is this enough? Should I lower the filter cap resistors? if so, how low, what's a good supply voltage? Do both resistors need to match or can I only lower the second resistor to raise the pre-amp tube volts? Would a resistor between the plate and the eq help the tube hold voltage and recover quicker (does that even make sense)?  
Here's an original schematic link which I have changed considerably
8/12/2005 7:59 AM
Ray Ivers

Are you sure your 2205 is this early circuit, and not the later (and IME much more common) one with the 3046 switching chip?  
I'm not sure what you mean by 'not recovering quickly' - during the non-recovery period, what is the amp doing? Is the volume swelling, the overdrive gain varying, etc.? Keep in mind that there's four preamp switching transistors (plus one for the reverb) along with some large-value electrolytic coupling caps, and if they're still in-circuit after your mods they may well be affecting things a lot.  
I don't think your B+ voltages are the problem, but by all means jumper in some resistors and determine that first if you'd like.  
8/13/2005 10:17 PM

Thanks for the interest, it is the early circuit w/o a switching chip. I think I've got the sag or blocking figured out. Voltages are 230, 250 and 210 on the boost channel gain stages. I've left the channel switching alone but I've re-routed the boost channel past the "mix-amp" tube v3a. V3a was a shared gain stage and lent to the channel leaking in the amp. There's still some leak but it's "improved". I tapped the boost channel's output, back "in" after V3a, first through its own .022 cap, then split through a 1meg behind the 680k reverb bypass resistor (that was previously shared) and the other leg of the split through a 1M into the reverb.  
Sounds really good but there is some instability at max gain/channel vol and treble past 5. I've patched it up with a voltage divider between treble and volume. But this seems kinda hokey. Any ideas?  
I've got a schematic that I made from the original marshall, but I can't attach to this form.  
9/30/2005 4:18 PM
Jon Blackstone Re: Sagging JCM800 pre-amp or blocking distortion proble
Doug -  
I've got a 4210 which I have extensively modified. It's that early one with NPN transistors doing the channel switching.  
I rearranged the 'boost' preamp to be like a plexi (tone stack after cathode follower), and replaced the baxandall tone stack in the clean channel with a "leo's ladder". (One with a little less loss than a fender.) The clean channel sounds great.  
At one point I had the boost channel going straight to the phase inverter for a pure Marshall thing, but I guess I decided the extra stages were benign, 'cause I put it back to mixing in before the reverb.  
Interesting that you mention the dry bypass resistor being 680K. I was a little mystified by the schematic saying 6M8 for this. My amp had a 680K in it, but it looked like a previous owner had put that in. I settled on 1M, which gave me ample reverb and still seemed to hit the PI pretty hard. Was there a mistake on my schematic?  
I don't have a problem with leakage betweeen channels. I took out the parallel caps on the mix resistors - maybe those cause the leakage people seem to get.  
My reverb return hums. I might install an FET follower inside the reverb pan to fix that.  
I put larger coupling caps around the post-PI MV. It seemed to me that those .02uF's in series, with a resistance to ground in between, were looking like a textbook HPF.  
Jon Blackstone  
Blackstone Appliances
10/5/2005 9:13 AM

Mine had the 680k resistor as well, it was a different color and looked as replaced, I think it's original though, a work in progress at the plant before it went out the door. This circuit didn't last long before Marshall reworked it. I've made the boost channel a jcm 800 2204 minus the cathode follower. I've tapped the eq off the plate side of 2b, then around 3a (previously the channel's third gain stage, shared with the normal channel) through a .022 cap that pairs up with 3b's .022 cap. This was my first amp mod, but it's turned out surprisingly well. The boost channel screams, and I put in a switch on the normal channel so I can lower the cathode resistors to 820ohms (plexi-ish)(1a & 3a) or up to 2.7k and 4.7 k very clean. My problem now is some instability in the boost channel, strangely not at full gain but between say 4 and 8, I get a squeal. from 9-10 it sounds great, and I can back the guitar off but I'd like to have a full usable range on the gain control. I put in a 100pf cap after boost volume to ground and it helped narrow the squeal from say 5-7 on the control, but I'm just stabbing in the dark.  
By the way what I thought was "blocking" distortion was a problem with rewiring the cathode follower to a gain stage, the switching circuit connects the plate to the grid between 2a and 2b. I moved the switching blue or purple wire to the plate on 2b and that fixed the problem. This channel still hits the post reverb tube really hard, I upped the 680 ohm to a 820 ohm to settle the gain a little.  
What's an HPF?
10/5/2005 12:17 PM
Jon Blackstone JCM800 4210 - post-PI MV, switchg Xtrs
> Mine had the 680k resistor as well, it was a different color  
> What's an HPF?  
Sorry. High-pass filter. A 'T' of 2 caps in series with a resistor to ground between them is a formula for a steep bass cut. In the case of the post-PI MV, I think the 1M R to ground (the pot) is too large to hurt. But the bias resistors are 220K, and cutting the capacitance before them (by sticking in the 2nd .022 cap) could make an audible difference. 'Don't really want to be cutting bass at this point in an amp with a small, open-back cabinet. I just replaced 2 of the .022s with .1uF's.  
Hey, BTW... has anyone here talked about how a post-PI MV defeats the negative feedback?  
Another 4210 thing to watch out for: Beware the switching transistors. They can act as diode clamps on negative signal swings. I had a nasty distortion that took me a long time to figure out, and that's what it was. I think the transistor had failed. But when you think about it, isn't there a diode from collector to base in any simple NPN transistor? I may replace mine with darlingtons.  
> I've made the boost channel a jcm 800 2204 minus the  
> cathode follower.  
So you don't see the cathode follower as part of the Marshall 'magic'?  
> I've tapped the eq off the plate side of 2b, then around 3a  
> (previously the channel's third gain stage, shared with the  
> normal channel) through a .022 cap that pairs up with 3b's  
> .022 cap.  
You mean the boost channel bypasses the reverb?  
Of course none of this addresses your current problem. Sorry no insight there. Do you have a resistor between the gain wiper and the grid of v2a? I found this to be necessary. I do not recall the symptom that necessitated it. But the objective of my mod is to make the boost channel like a plexi, and a plexi would have such a resistor. In fact, a plexi would also have a resistance to ground at the grid of v2a in the form of the mix resistor from the 'normal' input. This cuts the level here quite a bit. Maybe your configuration is just pushing to much signal into v2a, and your 'bright switch' cap (my Fender roots showing) may be aggravating it.  
Jon Blackstone
10/5/2005 1:48 PM

I don't have a resistor after the gain pot before the grid, might try it out, I yanked out the bright cap.  
As far as the cathode follower, I would have used one but I wanted that tube to be the third gain stage before the eq. The old circuit sent the boost channel and the clean channel through valve 3a (a shared gain stage) I routed the boost channel around 3a right into the .022 coupling cap off the 3a plate.  
I also put a 3n3 cap in series with the presence, I found that mod somewhere on the web, now it acts like a deep bass resonance. I don't know the science, I just tinker til it works.  
Did you pull out the clipping diodes? I put in a toggle switch so I can add them back between stage 2 and 3 of the boost channel.  
Also stage 1 bypass cap is now a 22uf, that sweetened the whole thing up and I put a .68 bypass on stage 3.  
   Page 1 of 2 Next> Last Page>>