As ENZO says, I would suspect super sonic oscillation - Particularly because the schematic does not show grid stoppers on the triodes 2,3&4. Again, you can read up on grid stoppers at the valve wizard site.
As for your gain staging question;
The vol after the tone stack is fine. I think the 100k cathode resistor on triode 3 is suspect. I would tack a 10k in parallel with the 100k and listen to the result to see if it is closer to the sound you want.
(How about some voltages as previously asked for?)
The vol after the tone stack is fine. I think the 100k cathode resistor on triode 3 is suspect. I would tack a 10k in parallel with the 100k and listen to the result to see if it is closer to the sound you want.
(How about some voltages as previously asked for?)
Last edited:
I don't want to deflect the conversation from the hum/noise issue, but I just noticed the comment about plate resistor values.
My understanding is that to increase (voltage) gain a.k.a. drive, you need to increase the plate resistor value, not lower it.
Adam's Amplifiers: Triode Calculator
is a useful little simulator which has helped me.
The triode calculator (Adam's Amplifiers: Triode Calculator) shows that the 47k plate and 100k unbypassed cathode resistor will have a 'gain' of less than 1.
OK, this has given me a bit to work on thanks 🙂
There are too many comments to quote from in one go, but i will try and find some 100 ohm wirewound resistors for the heater supply and i will measure the anode voltages on the tubes.
I was told by someone else some time back to change the anode resistors on the second tube to 47K from 100K, at the time i thought it made an improvement, but i should change back to see what results i get.
Ive got a scope, i dont remember seeing oscillation at the time, but the 50 hz hum was visible.
It will be interesting to see what i get on my scope with this sound, if i post a photo of the display on the scope, it may be some use.
There are too many comments to quote from in one go, but i will try and find some 100 ohm wirewound resistors for the heater supply and i will measure the anode voltages on the tubes.
I was told by someone else some time back to change the anode resistors on the second tube to 47K from 100K, at the time i thought it made an improvement, but i should change back to see what results i get.
Ive got a scope, i dont remember seeing oscillation at the time, but the 50 hz hum was visible.
It will be interesting to see what i get on my scope with this sound, if i post a photo of the display on the scope, it may be some use.
Also, even if that 100K is what was intended, then with the level of feedback you get with that 100K, a grid stop resistor is absolutely essential to prevent oscillation. You could do with some other grid stop resistors too but start with this stage.
Cheers,
Ian
What would you recommend for a grid stopper between the pot and grid of the third triode?
For the 'artificial center tap' for your heater string, you could use higher value resistors (220R, 470R, etc.) if you have a pair of resistors in your parts bin. And, since they don't draw much current even 1/2 watt resistors will do for experiments (especially if you move up from 100R). I've never used anything bigger than 1 watt resistors for that job.
OK, i thought they might draw a bit of current.
I probably have some carbon film or carbon comp floating around in my draw.
There is no current to ground, unless there is a shorted power tube or something. In normal operation, the two resistors are in series across the 6v, so use Ohm's Law to determine current through them.
So for 100 ohms, we have 200 ohms across 6.3vAC, which I get 32ma. 200milliwatts the pair. 100mw each. That is why even half watt is enough.
The value is not important. if you chose 200 ohms, then cut the above numbers in half.
So for 100 ohms, we have 200 ohms across 6.3vAC, which I get 32ma. 200milliwatts the pair. 100mw each. That is why even half watt is enough.
The value is not important. if you chose 200 ohms, then cut the above numbers in half.
Yeah, its not that much after all when you do the math lol.
I assumed it would draw a fair bit of current, as 100 ohms sounds low resistance, but i guess the heaters on a tube are even lower than 100 ohms.
I will try and work on the amp this evening and see what improvements i can make.
You should move the tonestack from the first volume control to the last one.
A tonestack gives some signal attenuation. When you put this between the first and second gain stage, you kind of neutralize the gain from the first stage and you won't be able to overdrive that stage. Move the tonestack and you'll have enough signal to overdrive the second stage.
Some already questioned the 100k cathode resistor of the third stage.
Besides the low gain (<1) and high internal feedback, the biggest problem is that a cathode resistor of this magnitude will basically cut off the triode (no current at all). I tried to sim it but I can't get the current low enough to match the 100k: < 0.2 mA and a gain of <0.3. In other words: this stage will be very hard to overdrive as well, but it won't be clean either since it operates in a very non-linear region.
Get the current up to >1 mA and you'll have some gain (try 10k instead of 100k). However, the gain will still be very low (~2) until you put a bypass cap on the cathode (~13).. More current will improve the clean sound further.
So in short: input stage -> vol control -> gain stage -> gain stage -> tone + vol -> driver stage -> output EL34.
Hmmm, perhaps too much gain...
OK, I will take that onboard, there are so many things for me to try here thanks to everyones input!
I do question some of the bypass caps on the cathodes.
I think i should lower the bypass cap on the first cathode to perhaps 25uf.
50uf does seem rather high.
Not too sure why most marshalls only run at 0.68 on the first stage.
Ive been simulating the curves and there is a significant difference between such values, i wonder if this is preference to the musician, as alot of fender amps have 25uf/1.5K on the first stage.
If i move the tone circuit between the second stage and the second volume, will I need a grid stopper on the second stage? Since that volume pot kind of acts as one. I expect i would just simply put a 22uf coupling cap in place of the tone stack?
I got too busy yesterday to do much, but ill give you the voltages between the anode and ground on each tube for your reference.
What size bypass cap would you recommend on the third stage?
would 25uf be OK on all 4 stages?
A cathode bypass cap increases the gain of a cathode-biased triode stage. Values such as 25uF, 50uF, or even 250uF will 'fully bypass' the stage (depending on the value of the cathode resistor, etc.). This means that the gain is maximized for all frequencies of interest. For guitar, I don't think you will notice any difference between 25uF and 50uF. But for low values such as 0.68uF the maximum gain will be retained for mid and high frequencies, but bass frequencies will be cut.
You can see some typical curves here (page 25-27):
http://www.valvewizard.co.uk/Common_Gain_Stage.pdf
High gain pre-amps, which expect to produce a lot of distortion in the pre-amp, often include some bass-cut prior to the distortion, to avoid a 'muddy' overdrive sound.
You can see some typical curves here (page 25-27):
http://www.valvewizard.co.uk/Common_Gain_Stage.pdf
High gain pre-amps, which expect to produce a lot of distortion in the pre-amp, often include some bass-cut prior to the distortion, to avoid a 'muddy' overdrive sound.
Last edited:
Yes. A grid stopper should be soldered directly to the tag on the valve socket. A volume pot is physically too far away from the tube to benefit as a grid stopper, and equals zero at full volume when you need it most.
Cheers
JimG
PS. Did you get my PM?
OK, thanks for that, im really only familiar with marshall circuits, but they seem to only have a grid stopper on the first stage that ive seen (well as far as my 2203 goes anyway)
Im working on it at present, i will get back with the anode voltages soon.
I cant see any PM showing here (yet)
Last edited:
OK, here are the voltages when measuring across each anode to ground:
From left to right in the schematic.
Triode 1 - 287V
Triode 2 - 293V
Triode 3 - 353V This is the suspect one with the 100K cathode resistor
Triode 4 - 97V This seems rather low?
From left to right in the schematic.
Triode 1 - 287V
Triode 2 - 293V
Triode 3 - 353V This is the suspect one with the 100K cathode resistor
Triode 4 - 97V This seems rather low?
Last edited:
Yes, that's way too low. Check the connections around that triode. If the connections look OK, measure the voltages: cathode-to-ground and grid-to-ground for that triode.
- Status
- Not open for further replies.