Hoffman Amplifiers Tube Amplifier Forum
Amp Stuff => Tube Amp Building - Tweaks - Repairs => Topic started by: 12AX7 on February 27, 2012, 11:09:30 pm
-
In a 2204 type amp, is there any reason why you couldn't use split loads for the PI plates? I want to reduce the gain to the PA and everything i have tried so far changes tone too much. I have tried nearly everything you can imagine, and i've come close enough to know thw amp is better with a lot less gain to the PI. But this is about the only thing left i havent tried. Before i do i wanted to see if theres some reason i am missing that this would not be a good idea. (and yes, i've tried tubed from AT7 down to AU7.
-
Anything you do to reduce gain to the power tubes will mess with the feedback ratio, which a) may or may not be that noticeable until a certain point, and b) may not be a bad thing2 depending on the tone you're going for.
You could do the split load thing, but it seems a better and variable approach would be a post phase inverter master volume. Maybe the PI "limit" control found in the Carlsbro T60 and more recently the Mojave amps would be even better.
-
Thats one of the things i tried and doesn't work well. It changed the tone a lot. Like i said i've tried most everything, so if it's remotely common figure i probably tried it. Anyways, PI split load plates, yes? No? Any reason that it wouldn't be a good idea compared to regular gain stages? As to the NFB issue, are you saying it would change that even with both sides of the PI having split load, and if so, couldn't it just be changed so it's more balanced between the normal signal and the nfb signal? i have a NFB pot that goes from 22k to 122k.
-
Are you saying you tried a PPIMV or the PI limit control?
If you tried a PPIMV, which one/type did you try? There's a _ lot _ of guys here and on other sites that love the Lar/Mar PPIMV.
If you tried a PPIMV (or a pre-PIMV), what is wrong with the sound? Is it thin as you turn it down?
Have you tried a boot straped pre-PIMV? There's also boot straped pre-PIMV that uses a dual pot, so both grids of the PI are boot straped. Some like/love those as well.
Brad :icon_biggrin:
-
Have you tried simply reducing the value of the PI plate load resistors? I'm using 15K in my November amp. The November is a Plexi preamp driving EL84s.
-
Have you tried simply reducing the value of the PI plate load resistors? I'm using 15K in my November amp. The November is a Plexi preamp driving EL84s.
Wow, really?! I had no idea you could go that low. But what about split load? I mean, if you can use 15k plate resistors i suspect a split load wouldn't be out of the question.....no? Maybe i'll try paralleling them with the same values to half them and see if that takes me closer to what i'm after. Then if so, try going lower till it feels right. Still, i'd like to know whether split loads are doable here tho.
-
Yeah, I was saying the any attenuation inside the feedback loop will change the tone, which is perhaps what you're hearing that you don't like.
Maybe it'd be helpful if you could tell us what you're after by reducing drive to the power tubes? Less volume, less power tube distortion? More PI distortion?
-
There's no problem with doing split loads. I would expect the results to be very similar to a PPIMV that is permanently set to one position. They are both voltage dividers working on the same signal. The only difference is one is on the left side of the coupling cap and the other is on the right side of the coupling cap.
What PA tubes are you using?
-
Try to reduce those 220k grid load resistors to 100k to 150k and increase the grid stoppers to the power tubes to about 22k. That will not take away much of the loudness but will reduce the signal to the power tubes a bit.
Cheers Stephan
-
Maybe it'd be helpful if you could tell us what you're after by reducing drive to the power tubes? Less volume, less power tube distortion? More PI distortion?
just to reduce the signal strength to the PA without changing the tone much, unless of course it's to the good. (which i what i feel will happen if i can do this in a transparent way) there is too much signal, and if i cut it down in the preamp it ruins the amps tone which i fought long and hard to arrive at. But theres so much signal that when i turn up the master it's loud as hell long before you even get the master close to 10:00. having owned many marshalls i know this just isn't right, and the tone suggests it's causing oscillation when it gets up there just a bit.
What PA tubes are you using?
El34X2, and the entire amp from the CF on back is very much 2204 except the adjustable NFB, dual bias and maybe a few other very minor things.
Try to reduce those 220k grid load resistors to 100k to 150k
I have but the amp never sounds quite right like that compared to the 220k's. Tried that several times, the last being just recently.
By the way, last nite i tried a voltage divider before the master. I don't know if theres anything unorthodox about 2 VD's in a row, but coming out of thr treble i put a 1M (ground ref) and a 330k together and came out of that to the master. I had started with 1M/1M but it seemed too muddy. The 330k and 1M worked better and still lowered gain enough. But the tone changed and it was getting late so i never tried a bypass cap. i was going to pass the 330k with maybe a 250pf. I will try that tonite. But while i'm waiting to head home from work and try that, can anyone tell me what they think i should use as far as a cap value across the 330k of a 1M/330k divider, and also what value if i use a 1M/1M VD? there must be a calculation that would give you the best results for getting it as transparent as possible, which is my goal, but even if i can find a way to do the calculation, my math skills are literally non existent. So any suggestions would be very welcome. Thanks.
-
Have you tried simply reducing the value of the PI plate load resistors? I'm using 15K in my November amp. The November is a Plexi preamp driving EL84s.
Wow, really?! I had no idea you could go that low. But what about split load? I mean, if you can use 15k plate resistors i suspect a split load wouldn't be out of the question.....no? Maybe i'll try paralleling them with the same values to half them and see if that takes me closer to what i'm after. Then if so, try going lower till it feels right. Still, i'd like to know whether split loads are doable here tho.
A 12AT7 or 12AU7 will work well with a plate load of 20k (which is 2 x 10k when you are looking at a cathodyne PI).
-
and the tone suggests it's causing oscillation when it gets up there just a bit.
Here are things that you can try. Note that the snubbing cap is AFTER the coupling caps.
With respect, Tubenit
-
and the tone suggests it's causing oscillation when it gets up there just a bit.
Here are things that you can try. Note that the snubbing cap is AFTER the coupling caps.
With respect, Tubenit
The only thing i haven't tried is the big grid stoppers. But don't those really only attenuate highs? I don't find them to drop gain to any noticable degree unless they are used as 1/2 of a voltage divider. the caps i have endlessly tried and found that using a 50pf across the plates or after the caps is a huge part of the marshall sound and it's one change i've made that i would never remove. It makes the amp ! I don't know how it works because as far as i know it should just remove highs via phase cancellation. But it does something else that i can't really describe, and in fact almost seems to increase highs rather than lower them. Or actually not so much add, but "change" them in a way that there seems to be no less than before, but they are world sweeter. It really changes the character of the amp big time. It's funny how one little 50pf cap can do so much ! I did try values up to 250 too, but 50pf seemed to capture that sweetness with no unwanted loss of highs. Anyways, thats besides the point of the thread, but it's ok because i decided just to leave it as is. It sounds so good that anything i do to it now changes the tone for the worse. So I can live with a master thats usable to 10:00 and after that becomes nasty. But that fact is, at 10 it's so loud theres not a venue i ever played where i could have used it that loud. Maybe thats part pf the reason it sounds so good....at a given volume the signal at this amp's master is set closer to ground than others would be. I can tell you that in my preamp i found that a 100k to ground after the 2nd stage rather than a voltage divider has proven to be the best sounding solution at that point of the circuit. I have tried voltage dividers there and even a pot to be able to find the perfect VD values. But nothing matches the amp's tone with that 100k ground ref right there after the coupler. I think i tried a similar thing at the master too, ( resistor across it to change pot value to around 100k) but again tone was changed for the worse. Been a long time tho and with the latest changes i think i may try that again before i zip her up and just play instead of tweek.....at least for a while. :icon_biggrin: Thanks all.