How to isolate one circuit area from the next

[12AX7]

After asking about the master on my JCM800 style amp and why it changes tone so much, i investigated further on google and i realized that as someone at another forum said, the pre PI master forms a filter with the treble control that feeds it from the tone stack. That makes perfect sense. But a resistor off the master's center lug doesn't rectify the issue, so i thought about ways to isolate the treble control output from the stack from the master that immediately follows it. I cannot simple do a post PI master because i would have to use 1M resistor as a ground reference where the pre PI master was and that would form the same filter as when it's turned up giving me the same bright harsh tone. using a voltage divider or smaller resistor would limit the volume much too low. So is there another way to isolate the treble control from the master without putting another gain stage in between them?

[John]

I hope I've done this right; this is the tone stack and PPIMV from the HoSo that a few of us built. I don't notice any difference in tone with the PPIMV, no matter where I've got it set, and with the Trim Pot there is a very slight perceived treble increase. Emphasis added because I've not really paid attention, and so am going on memory right now.

Not sure if that is helpful to you at all, but sometimes looking at different schematics helps me understand things better. I'm very new at all this, so I figure if I was able to do the TS and the volume and it works good, it must be pretty idiot proof. 

[12AX7]

Thanks, but that won't do what i need. I've been trying things and looking at this all morning and i think i've come to believe the only way would be to set the gain where it works best, low, then put another gain stage after it. But i'm not going to for many reasons including logistical ones.

[chocopower]

I cannot simple do a post PI master because i would have to use 1M resistor as a ground reference where the pre PI master was and that would form the same filter as when it's turned up giving me the same bright harsh tone.

You can avoid the 1M resistor with no perceptible effect.
Look at a Plexi schematic:
http://www.drtube.com/schematics/marshall/1987pljp.gif

Change the 1M pot for a double 250k at mount a post P.I. MV.

[12AX7]

I tried that today along with most every imaginable thing i could think of to eliminate the master there or change it's effect. Nothing works without radically changing the tone in one way or another. I believe this is an insurmountable problem unless i added a tube which can't happen due to space. The big issue here is i have this tone that i'm 100% happy with but is only available at low volume and it seems at this point theres no way to get it even close at higher volumes.

[kagliostro]

May be you can insert a mosfet like Tubenit do in some of his projects

Kagliostro

[12AX7]

I've actually thought of that. I considered throwing a couple jacks between the treble pot and master and putting a pedal or something in there so i could see how that works. Then if it did do just what you are saying. I may consider it, but i really feel like theres gotta be a simpler way. I keep looking at schematics trying to get ideas. One thing i keep thinking about is putting the tone stack further upstream, tho i know that will change the basic tone a lot. But then again it could be for the good.

[tubenit]

You might try a 22p or 47p silver mica cap from the (MV) pot's wiper to ground.  I have done that on master volumes and liked the results. That will drop some of the very high end to ground.

With respect, Tubenit

[12AX7]

I'll try it but I'm pretty sure it's not going to do what i need for a couple reasons. But i'll give it a go anyways. Just gotta find a cap that small. But i won't have a mica, just ceramic disc.

EDIT: no go on the cap. Did little to help when turned up, and took away a bit of what was good when lower. Pretty much what i expected. Upon further  tweaking i think it's true that theres too much gain going into the PI. problem is the master doesn't seem to be the best way to dump some of it or i wouldn't have started this thread in the first place. But i can't figure out any other way because the only other place i can see is going to drop the gain in the pre and kill the overdrive drive i like. Already tried that, but where the gain is now on the overdrive stages isn't a heck of a lot, and if i drop it there it gets too clean.

[12AX7]

This seems to help, tho hard to tell at the moment since due to neighbors i won't be able to turn it up enough to be sure for a while. But i put a 250k pot in place of the 1M and a 100k to ground before the pot and a series resistor off the out lug as has already been mentioned. These things together seem to work much better. It now feels like a more appropriate amount of gain is hitting the PI without changing the tone i liked at low vol much.  I can tell because there was some blocking when i turn it up quite a bit and that seems to have gone. I will test it more by tomorrow after work at the latest to see if the higher volume tone is good. I tested it somewhat using a post PI master too so i could raise the master w/o getting too loud for the neighbors and it seems pretty good. But the real test without that second crossline master is the only sure test. 

[HotBluePlates]

"Isolation" isn't the problem, hi-pass filter theory is the problem. Further, the stock type of master volume will always have this issue.

The problem is that the bass roll-off frequency is determined by the coupling cap's interaction with the resistances around it. As you turn down the stock master volume, that resistance gets smaller and the bass roll-off frequency gets higher.

There couldn't be a problem with "not enough isolation" from the treble pot, because the fullest sound from this type of master volume is typically when the master is full up; that is electrically the same as connecting the master volume wiper directly to the treble pot's wiper.

An added series resistance at the master volume wiper seeks to keep the changing bass roll-off low enough even when the master is turned almost all the way down. You can either try a bigger resistor than 100k (maybe 220k), or if that doesn't work, try a different type of master volume.

[12AX7]

What is odd is that when i add a resistor to the wiper the tone gets brighter. Plus, and this also goes against what i commonly hear, is that it sounds fullest down low and the bass roll off happens as i turn it up.

[kagliostro]

The problem is that the bass roll-off frequency is determined by the coupling cap's interaction with the resistances around it. As you turn down the stock master volume, that resistance gets smaller and the bass roll-off frequency gets higher.

I think I'm near to understand but don't

the MV pot is connected between the Treble wiper and ground so if you turn the MV the wiper of the MV will insert a lower or higher resistance as seen from the grid of the following tube, but the resistance between the wiper of the treble pot and ground will not change .........

what I'm missing  

Thanks

Kagliostro

[HotBluePlates]

What is odd is that when i add a resistor to the wiper the tone gets brighter. Plus, and this also goes against what i commonly hear, is that it sounds fullest down low and the bass roll off happens as i turn it up.

I'd need to see a schematic for your amp (especially if you've altered anything from the stock form of the amp). You're right that your experience is opposite what happens with everyone else's JCM800.

Know also that the treble pot does not just pass treble, but also bass and midrange. After all, it has to because that's the only connection from the tone stack to the rest of the amp that follows. As you move the treble pot wiper "up" it is closer to the end of the pot that is connected to a small cap (highs), and further from the end of the pot connected to bass and mids (the big caps).

the MV pot is connected between the Treble wiper and ground so if you turn the MV the wiper of the MV will insert a lower or higher resistance as seen from the grid of the following tube, but the resistance between the wiper of the treble pot and ground will not change .........

The circuit components involved are the coupling cap, and whatever resistance it sees.

The coupling cap after the master volume sees a fixed resistance from the input of the phase inverter, but a changing resistance from the master volume wiper to ground.

The formula used to calculate the -3dB frequency for this cap uses the sum of both resistances as the single "r" used in the formula.

[12AX7]

I'd need to see a schematic for your amp

if you look at a 2204 from the cathode follower to the power tubes you're seeing my amp except for a few component values and a switch to lift NFB. The values are just a 47k slope n the stack and 1k/47k cathode/tail resistors on the PI. And lower filtering.

Heres what I'm considering doing. Maybe you could tell me if you think this would be a good idea and also if anyone knows of a amp thats  done like this please tell me. I have removed the cathode follower and run the amp with the tone stack fed by the plate of the 3rd stage instead, leaving 1/2 of V2 unused. It sounds good like that, tho i do prefer the CF, but not by a lot. So what if i play the amp and see whare the master sounds it's best, which is around 1 or 2. Then leave it there, unslder it and measure the in to wiper and wiper to out so i can put a voltage divider in it's place. Then take the signal from there and go into the unused 12AX7 side now vacated by the CF and put a master after that stage going to the PI. Is there anything wrong with that idea?

[kagliostro]

Thanks HotBluePlates

I was looking only to the first page of the schematic

http://www.el34world.com/charts/Schematics/files/marshall/jcm800_lead_mstvol_50w_2204.pdf

and don't see the capacitor which is on the second page and even stupidly I did not remember that could be one there  

is the same problem there is with PPIMV if they are used in fixed bias amp where a resistor is put in series to minimize the effect of turning the pot

Kagliostro

[kagliostro]

Something like this is totally wrong ? 

may be a cap in parallel to the second pot as to minimize the high lost by the insertion of the "in series" resistance ??

Kagliostro

[HotBluePlates]

is the same problem there is with PPIMV if they are used in fixed bias amp where a resistor is put in series to minimize the effect of turning the pot

That particular schematic has something going on which may be an error, or may be intentional.

There is a series resistor from the wiper to reduce the effect of changing the master volume setting.

If the master is all the way up, the two coupling caps are in series. Caps in series behave like resistors in parallel; the effective value of the capacitance is lower than the smaller of the caps (in this case, it is half the value). Normally, the first cap would be 10 times as big as the second cap. Here, they are the same value. The effect would be to reduce bass the most when the master is full-up.

I don't know if that is a mistake, or if the designer was trying to offset the increased treble as the master is reduced (the changing resistance) with the increased treble as the master is turned up (the series caps). They might have been trying to gain a consistent sound at all settings.

[kagliostro]

If the master is all the way up, the two coupling caps are in series. Caps in series behave like resistors in parallel; the effective value of the capacitance is lower than the smaller of the caps (in this case, it is half the value)

May be that the "in series" resistor is add also for the purpose to decouple the two resistors EDIT Capacitors ?

Kagliostro

[12AX7]

The problem is that the bass roll-off frequency is determined by the coupling cap's interaction with the resistances around it.

So do you mean theres no problem with the master's interaction with the tone stack, but instead with the cap between the master and PI?

[HotBluePlates]

Yes, exactly.

Unless there's something crazy going on as a result of the changes you've made in your amp. But I can say for certain, in the stock circuit, the reason the sounds gets thin at low master settings (opposite what you seem to have) is the interaction between the master and the coupling cap between the master and phase inverter.

[12AX7]

Weird. Well, i tried a resistor in the INPUT side of the master and that does just the opposite ! More like what you and others ay it's supposed to work. I also assure you it's wired right. I'm sure thats going thru your mind and i understand that not knowing me. But it is correct i assure you. The differences from the CF to the PI are as i say are just value changes and otherwise just like a 2204. However, i DID recently remove the switch that disconnects NFB and the associated components, and i can add that back easy enough. But that wouldn't cause this reverse action would it? Anyways, like i said i tried the input lug and it works fairly well like that. I also added a 150k across the outer lugs after seeing a soldono that has the same setup, IE: CF/tone stack/master/PI. It has a 150k across a 1M master with a 470k at the input lug. So i did it just like that. I still feel it's not the same turned up and still displays some of that harder brighter tone w/o bass missing, tho not nearly to the degree there was. And of course i also can't be sure how much of that is hearing at higher volumes.

Anyways, i will have to test this further when i get a chance. As i've said, neighbors make it hard for me to be able to turn up enough to evaluate it more than several times a week. But if this doesn't cut it i think i'm going to do as i said earlier and remove the CF stage, feed the stack from stage 3's plate, and add a gain stage after the tone stack with the former CF's unused triode side and put the master between the 3rd and new 4th stage. I have a gut feeling that may make for a much better overall tone anyways in my estimation.

[12AX7]

Heres an interesting find that may give HPB or someone a hint as to whats going on. I got a huge change in the amp to the good by putting a resistor in series from the cathode follower's cathode to the tone stack. I tried several values and a 470k seems best. The master now has a much more normal range where it gets progressively louder in a linear fashion as i turn it up. Before it would jump to loud as he|| around 1 or 2 on the control and just start howling if i turned up more than about 9:00. Now it's not only linear in volume, but no more abrupt harshness and loss of low end. It still leans that way as i turn it up enough, but 1-not as much, and 2- it now works like it's supposed to when i put a resistor in series with the wiper and stays even more normal as i turn up.

I could leave it like this i suppose, but i know it's a bandaid on a bigger issue and i was hoping someone might have an idea after reading this. It may also help to know i have a switch that removes the CF stage and sets the tone stack as plate fed from V2a, and that does NOT do what this resistor did. It just drops the gain a tad and tonally sound a little different.

[HotBluePlates]

The master ... would ... just start howling if i turned up more than about 9:00.

That could be the clue...

You disconnected the NFB, so does it still howl if up past 9? If no, swap OT plate or secondary wires (pick one set, the easiest, and swap them). If yes, you probably have a feedback problem (not related to the loop), which may be causing the weird tonal effects.

For this to be the cause, and the master to affect it, there would almost have to be a feedback loop going on between the phase inverter and the stage ahead of it.

A good clear picture of your amp's innards would probably tell if this is likely.

The only other possibility that came to mind (and it didn't seem highly-likely) was a bad pot.

[12AX7]

No, it's not that. Probably my fault for a bad description. When i said howl i meant it just sounds like you stepped on a clean boost pedal set too high with a already distorted amp. Blocking distortion too. But it's just too much gain. Anyways, i tried another approach and removed the resistor i mentioned and instead put a split load on the cathode follower to lower gain, and this works even better. I need to play around with things, but at this poi t the master works infinitely better if not perfect. I need to put a pot in place of the cathode and use that to find the perfect split load values. The series resistor of the master's wiper is still working well. So apparently it had something to do with the excessive gain that caused it to act differently than normal. This opens the door for a lot of  fine tuning and hopefully a final version that works as I imagine it should.