I picked up a project recently, a guitar tube amp in need of some repairs. So this is my fist time working with tube amps, so I'm still learning.
I was able to fix most of it up myself but I'm getting very harsh Hard Clipping in the preamp stage. I have all the toys (oscilloscope, DDMs) and was able to track it down.
I believe it is from Grid Current Blocking distortion (is that correct?). This is a snip-it of the schematic and a representation of the wave forms I see.
View attachment Untitled.bmp
The sign wave is perfect on pin "1" but the positive side is very clipped on pin "7".
How can I stop this but keep the good distortion? I have tried changing: C31 to a 2.2nf, R42 to 470k, swapping Valves, R43 up to 4.7k, R41 to 100k. With no luck.
might be just too much gain going into V2B?
I was able to fix most of it up myself but I'm getting very harsh Hard Clipping in the preamp stage. I have all the toys (oscilloscope, DDMs) and was able to track it down.
I believe it is from Grid Current Blocking distortion (is that correct?). This is a snip-it of the schematic and a representation of the wave forms I see.
View attachment Untitled.bmp
An externally hosted image should be here but it was not working when we last tested it.
The sign wave is perfect on pin "1" but the positive side is very clipped on pin "7".
How can I stop this but keep the good distortion? I have tried changing: C31 to a 2.2nf, R42 to 470k, swapping Valves, R43 up to 4.7k, R41 to 100k. With no luck.
might be just too much gain going into V2B?
An externally hosted image should be here but it was not working when we last tested it.
This actually belongs in the Instruments & Amps forum per forum policy where I suspect you will get the help you need.Edit: Moved
Try changing the second stage cathode resistor value. Best to get a 10k pot and adjust it while watching it on the scope. You get the clipping on the second stage but because of the big grid stopper it is not dragging down the voltage at the first tube. The large grid stopper limits the blocking distortion so the cap should be fine.
Check the voltage on pin 6, the anode resistor R44 could be high resistance - a VERY, VERY common fault on valve amps.
I have checked all resistors already, first thing I did. Pin 6 has 230V (228v), R44 96k, so that is at 0.7mA at -1.6v bias.
I think the issue is that I’m getting 150v AC peak to peak on pin 1. Should that not be like 2v ptp?
This thing is like a square wave generator!
I think the issue is that I’m getting 150v AC peak to peak on pin 1. Should that not be like 2v ptp?
This thing is like a square wave generator!
Fox, I guess the question is what are you using to drive this input, the 12AX7A is a high mu triode and that circuit will provide quite a lot of gain, it's probably intended for a low output passive pickup. This input is probably intended for a couple of hundred mVrms at most. I suspect you are overdriving the input by a couple of volts.
There is no grid stopper on the 2nd (or 1st) stage in the scheme. R41 is dropping the signal about 1/3rd since it is in a voltage divider with R42.
To add a grid stopper to the 2nd stage you would "replace" the wire going to pin 7 with a resistor. Values vary dramatically. I would try 20k or so.
To add a grid stopper to the 2nd stage you would "replace" the wire going to pin 7 with a resistor. Values vary dramatically. I would try 20k or so.
That measurement it was taken with 90mV ptp at the input. Far less then my guitar can put out. even if I put in a ECC81 it still is about the same.
I think I need to dump some gain.
The amp is close to this, but 100W with some additional changes.
https://www.dropbox.com/s/cfdxedhffyypdlm/RG50TCSchemat.pdf
I think I need to dump some gain.
The amp is close to this, but 100W with some additional changes.
https://www.dropbox.com/s/cfdxedhffyypdlm/RG50TCSchemat.pdf
Last edited:
If that is the case and you are seeing 150Vpp on the plate you must have oscillation occurring. - the maximum available gain would give you maybe 6Vpp. Try to determine whether that is the case with an oscillator rather than the guitar.
I get the same thing with a signal generator. This is on V2 so there is another 12ax7 in V1. Since I cant find anything wrong, it might be just a P.O.S. and needs some improvement rather then fixing to make it sound good.
What level are you inputting from the generator? - the first stage has a gain of about 20 (100K/4k7) and the second stage about 40 (100K/2K2), giving an overall gain of 800.
So if you're getting 150v p-p out of the first stage, then you must be inputting about 7.5V p-p 😱
Thank you all for your time! OK, I'm a Chemist, no an electronics engineer, I like electronics, but I'm all self-taught. My experience is with transistors, and not Valves. So I think I’m missing something.
So let’s forget out this amp and it’s problems, I would like to learn more so I can fix it myself. I thought tubes distort with soft clipping and produce even order harmonics. With this amp, it's distorting with hard clipping and odd harmonics. So how is nice smooth distortion created with tubes?
To my understanding, when the grid signal is driven past the area of linearity the output signal starts to distort and compress (in a nice way). So what happens when you push the gird really hard? Will the signal will go from cut off to grid current limiting, making the signal at "cut off" clip hard and distort with odd order harmonics?
So how do Marshall/ect create heavy but pleasant distortion?
So let’s forget out this amp and it’s problems, I would like to learn more so I can fix it myself. I thought tubes distort with soft clipping and produce even order harmonics. With this amp, it's distorting with hard clipping and odd harmonics. So how is nice smooth distortion created with tubes?
To my understanding, when the grid signal is driven past the area of linearity the output signal starts to distort and compress (in a nice way). So what happens when you push the gird really hard? Will the signal will go from cut off to grid current limiting, making the signal at "cut off" clip hard and distort with odd order harmonics?
So how do Marshall/ect create heavy but pleasant distortion?
If it's like in that picture where majority of one half wave is "chopped off" quite harshly then it's pretty much what each and every preamp tube in this kind of gain stage does when it clips asymmetrically.
Soft clipping? Not in these circuits. Don't believe all the hype. You will have to design it in. Cutoff and saturation, for example, can have different clipping characteristics.
If it LOOKS too hard to you try some low-pass filtering, it will "smooth up" the edges of that clipped wave, or at least turn the waveshape so obscure that it doesn't LOOK that harsh.
But this is pretty much how your average preamp tube gain stage clips. Yes, it clips much harder than, say, a solid-state diode. But they don't say that in those books where they hype tube amps or portray that in those DRAWN waveforms where solid-state clipping is always shown jagged and hard and tube clipping is nice and smooth.
Now you just saw the real deal on oscilloscope screen. Consider yourself more enlightened now.
Important point is that we shouldn't be listening with our EYES.
Are you sure it's not actually working as intented, as this kind of clipping and behaviour is very typical for tube preamps. And yes, often considered the "good" distortion. The great tube distortion. Forget the idea that tube clipping is supposed to be soft and start with that new mindset.
How bad does it really sound? Is it due to pre and post distortion EQ or the clipping itself? Because usually an EQ makes a HUGE difference to tone of distortion whereas softness or hardness of the clipping is barely audible and you can't usually even distinct the two from each other.
Actually, that kind of asymmetric clipping would have very prominent amount of even order harmonics. The more symmetric the clipping the more the harmonics are odd order. In truth, when an amplifier distorts there will be plenty of both even and odd order harmonics in the signal. You won't just get one, it's always both.
Oh, forget also most of the stuff about harmonics you read in the context of tube amps too. The theory mostly applies to clean signals and is entirely misapplied in case of circuits that are deliberately designed to introduce huge amounts of clipping distortion.
Odd order distortion doesn't sound bad or good as is. It just sounds like distortion.
Even order distortion doesn't sound bad or good as is. It just sounds like distortion as well.
Preferably using pentode tubes with very "curvy" transfer characteristics, e.g. EL34, in push-pull and in open loop. For preamps, topologies like differential (LTP) or SRPP might work fine. Not typically with anything single-ended.
In something like JTM45 the grid of the second stage is slightly overdriven into grid conduction giving a bit of assymetry to signal. The second gain stage is overdriven so much it practically produces a square wave output even when overdriven moderately. Power amp will smooth things up with asymmetric soft clipping if driven hard enough. Most of the "magic" (no, not really) actually lies in careful EQ:ing of the signal throughout the signal path, from input to speakers the signal is constantly EQ'd. Post and pre distortion EQ has an astounding effect on tone of the distortion. Much more than the clipping itself.
Why not study schematics of some Marshall (or those etc.) circuits for reference...
Last edited:
My mind just melted. Let me just think about that for a while.
Thanks for pointing that out, I was only going by what I read in books, and with my inexperience I was confused.
I really appreciate the help, and I hope that can help more people.
I will try some experimenting, and if I find anything that might help others in my situation I will post. Thanks again!
So's my daughter, she's got a Masters Degree in Chemistry, and is currently doing her Phd.
Interestingly during her Masters Degree they had to build an occasional electronic circuit, which she had no problem with as I taught her to solder while at Primary school 😀
I was at college in the early 70's, valves weren't mentioned on the course at all - apart from one afternoon when we had a temporary teacher, and he taught us about valves.
As I asked before, what level of signal are you inputting from your signal generator?.
😛
Similar to me, my education did involve some electronics and my father taught me how to solder when I was able to hold the iron. Good for her for working towards the Phd.
Sorry, ~90mV pp I match the amplitude from the guitar signal I was getting while it was oscillating. I did not want to overload the input. I checked the amplitude on my guitar signal, while doing that I found if I lean my guitar against my cab and keep the volume low it would feedback nicely at around 200hz at ~90mVpp. Now of course I can strum and get more, but that was a nice number to work with.
90mV p-p is pretty low, and shouldn't overload the first stage - which should only be outputting about 2V p-p at that level.
If you're getting 150V p-p it sounds like (as already suggested by kevinkr) that it might be unstable and oscillating?.
This is in the second stage. So How would I see that the amp is unstable, what am I looking for?
Ditto what teemuk said, plus google the valve wizard. Welcome to the journey (addiction).
(I was gonna be a chemist, then I was gonna be an electronics engineer, and I ended up a psychiatric nurse with a degree in accounting - go figure)
(I was gonna be a chemist, then I was gonna be an electronics engineer, and I ended up a psychiatric nurse with a degree in accounting - go figure)
If it's 150v p-p out of the second stage that sounds pretty reasonable, it's hard to know what sound you're getting, and what sound you're actually hoping for?.
Do you have a complete schematic you could post?.
- Status
- Not open for further replies.