Tube Pre-Amp Pedal (JCM800) Voltages Help

[ddog8]

Hey All,

I'm building a JCM 800 Pre-Amp clone, a GrindCustomsFx STM800. Here is the schematic:

https://i.ibb.co/wBLnNX7/STM800-schem.jpg

Before I put the 12AX7 tubes, I have confirmed that the plate voltages are 190V (A1 and A6; B1 and B6 on the schematic).
After I put the tubes in, the plate voltages are:
A1: 145.7V
A6: 124.5V
B1: 105.1V
B6: 190.0V

As far as I know, there should not be a large discrepancy between B6 and A6, (should be around 20V). I am not certain what the discrepancy should be between A1 and B1.

What I have done:
-Confirmed that the heater voltage is around 6V (A9 and B9). This is what it should be as per the build doc
-Confirmed that all the tube socket connections match the right location on the PCB
-Tried a new set of 12AX7 tubes


I am about to reflow the entire board. Based on the schematic are there any specific areas (or components) that I should pay attention to, that may be causing this huge voltage drop? Thank you for the help!

[ferdinandstrat]

Hey dude, I built several tube preamps like this, I dont remember specific voltages but I do remember that the tubes and plate resistors cause a voltage drop because they put a load on the power supply

Is the preamp not working? Is it not producing sound?

[ddog8]

I have the pedal plugged into the return of my fx loop and it is working, but I am just not getting the 'growl' / gain (for example like in this sound test - https://www.youtube.com/watch?v=gJaJspvdYMQ). For me it sounds more like a boost unless I max the gain and it comes more alive.

I read over in the madbean forums that B6 and A6 plate voltages should be within 20V of each other and was thinking that this could be the cause of my issue (poor gain).

[ferdinandstrat]

It's a silly question but did you use the correct components? Specifically the pots and plate resistors?

[PRR]

Welcome.

> I read ...that B6 and A6 plate voltages should be within 20V

You already stated that B6 is 190.0V. (apparently with or without tubes, which says your supply is stiff.)

B6 is a plate directly connected to the supply.

A6 is under a plate resistor, which ALWAYS makes a drop. We tend to want a "fair fight" between tube and resistor. Which would be halfway or 90V-100V, but in fact we like to bias high. I'd expect this plate to sit at 75%-80% of supply. Around 140V-150V. Which is what you see, so close we have no cause to fret. All your DC voltages look good. Which IS the first step in making (or distorting) audio, but you also need the AC/Audio parts right.

There's at least a gain of 1,000 to V2B output, so unless you strum with a limp noodle it ought to distort significantly. Signal tracer should find a BIG rise of gain from stage to stage (and a loss through the tone network).

[ferdinandstrat]

Correct me if I am wrong but is the voltage at the plate of the final stage, the cathode follower, supposed to be the same as the power supply output? Kinda makes no sense to me, doesnt a tube with no resistor still supposed to load the power supply?

[ddog8]

Sharp eyes with the voltages 

I have originally run my power supply higher than 190. I don't have the number, but it was somewhere 190 - 210. The tube voltages in my post is when I have run it with the higher voltage (and actually wrote them down). However, a post in the madbean forums said that the PSU is unstable at over 190, so I have dialed the voltages down to around 185 - 190 V. I can re-measure and post the (newer) voltages. However, the B6 and A6 discrepancy is still over 20 V and the other numbers are similar to what I have posted (albeit lower)


I am going to box up the pre-amp and try it gain. I am running the pre-amp into the return of an Egnater Tweaker 15W head, apparently the fx loop is line-level rather than instrument level - could that be causing my gain issues? It is also tube buffered and I haven't tested the tube there (I don't normally use an fx loop).

[GibsonGM]

The CF isn't doing a lot of loading in a quiescent state, no.  Look at its cathode...(100k...)   I would not run this in the FX loop, try it direct at the amp input! 

That thing should make your Egnater scream.  By voltages, it doesn't look like there's anything wrong.   You could audio probe it, but you'd have to know how to do so safely AND the cap in the probe must be rated higher than B+ voltage, as well as being totally insulated.   There is a process for doing this, and one best know it pretty well or risk toasting themselves permanently...don't need to test the tube in the amp or anything else in there, let's keep it simple (and not get electrocuted)     

[ferdinandstrat]

I cant help but feel as though there is a wrong component somewhere

I noticed that the schamatic you posted has a presence control while the demo doesnt, as far as I know presence does remove a bit of gain in theory but I dont think that it should remove as much as you described

[PRR]

........doesnt a tube with no resistor still supposed to load the power supply?

Of course. But does the power supply sag? In this case, not clear.

[ferdinandstrat]

........doesnt a tube with no resistor still supposed to load the power supply?

Of course. But does the power supply sag? In this case, not clear.

Well in theory the voltage at the plate of the CF should be lower than the actual supply voltage, no?

[Rob Strand]

Well in theory the voltage at the plate of the CF should be lower than the actual supply voltage, no?

B6 is a plate directly connected to the supply.

[ferdinandstrat]

Yeah but doesnt the CF load the supply? Correct me if I am wrong about this

[PRR]

Yes it loads the supply.

How much drop is that?

And you should measure the working B+ with the load applied, so everything is accounted. You can balance the numbers with multiple measurements and scrap paper, but usually easier to measure the amplifier.

[vigilante397]

Yeah but doesnt the CF load the supply? Correct me if I am wrong about this

It loads the supply, but that means the whole supply voltage is going to sag, and since the plate of the CF is connected directly to the supply, it should be the same voltage as the supply voltage, regardless of what that voltage may be. So you're not wrong, but I think you're overlooking that part.

[ddog8]

Just a small update:

I've boxed the pedal up and tried it again. I'm pretty sure the gain knob isn't working. It seems to be always set on the minimum (ie it has no effect as I turn it). Pedal comes somewhat alive if I turn the Master on max, but it isn't even close to the effect on the video I linked above. I tried the pedal both into the effects loop's return and directly into the input.


I think I either have a cold solder joint or a bridged / grounded  connection somewhere near the gain potentiometer. I've checked the schematic and it seems that these components are related to the gain knob (other than the tubes):
- R1
- R2
- C1
- R4
- C2
- R3
- C3
- C4
- R5

Am I missing any components? (Sorry I am not much of a schematic reader). Are there any of the above components that I can ignore (ie are not likely to cause the gain knob to be set to the minimum). Thanks for the help!

[GibsonGM]

First I would check my component values, be sure they are as shown on schematic (470k etc).  Then test that the gain pot is doing what we want...resistance test the thing (always w/power disconnected).  Make sure the pot works, test resistance to ground, be sure nothing like the following grid is grounded when pot's turned up and so on.   You're sure the cold end of the 1M gain pot IS grounded?  No solder bridges at the pot?  From "A2" to gnd, you should see increasing resistance as you turn the pot. 

You can't 'ignore' any of them, but you can systematically rule them out as causing the problem!   

Re: schematic, there's not much to read: there is a branch sending "very highs" to gnd via C2.  Remaining signal (the vast majority) flows thru C3, R5, gain pot, to next stage grid.  Any problem there will cause...a problem.  If you keep working on them you'll learn to read them too.

Here is where a 'proper' audio probe would come in handy to isolate where the problem occurs...a simple signal generator, too (you can input a known AC voltage and read the output of each stage to determine what's happening inside it).    Until you get addicted you can stick to easter-egging, tho  

[MetalGuy]

One shouldn't be expecting too much gain from a JCM800 "bare" preamp. It just isn't there.
I remember many years ago when I was a newbie in this filed how disappointed I was when I built a JCM800 clone  preamp only. It sounded nothing like a real Marshall JCM800. Actually it sounded like sh*t as all Marshall preamps only from that era. Later I learned why.
My advice is to remove the Presence control completely. It's not suitable for this preamp at all and will only choke the sound.
Correct and according to many best voltages for the first stage are 260-270V (VB, not the plates). There are Marshall voltage chart all over the internet.

[vigilante397]

One shouldn't be expecting too much gain from a JCM800 "bare" preamp. It just isn't there.
I remember many years ago when I was a newbie in this filed how disappointed I was when I built a JCM800 clone  preamp only. It sounded nothing like a real Marshall JCM800. Actually it sounded like sh*t as all Marshall preamps only from that era. Later I learned why.
My advice is to remove the Presence control completely. It's not suitable for this preamp at all and will only choke the sound.
Correct and according to many best voltages for the first stage are 260-270V (VB, not the plates). There are Marshall voltage chart all over the internet.

My experience is a little different. I agree the presence control is useless on a build like this, but my JCM800 preamp build sounds fabulous. But like most preamps the overall sound is going to depend on what you're doing with it. Running it straight into a mixer or recording interface, sure it won't sound great. But running it into a tube power amp or even into the front end of another amp, it sounds like a JCM800.