changed plate and cath R and loss of gain

[daz]

On my amp, which all of you by now not only know of but are probably sick of, i made a change today that affected the tone in a way i've been after forever. Someone elsewhere suggested this and i tried it. The cathode follower and the stage before it are typical marshall values and design. I changed the plate resistor from 100k to 220k and th cathode from 820R to 2.7 k. This caused the tone to get rounder and fuller than anything i have ever tried and in a way that is different from anything thats ever come close. But in the process this tweak has thrown out a lot of gain. The amp now has a considerble loss of volume and while it's still loud enough to gig with, the master would have to be way up to the point where the amp is noisy. What i want to know is how i can get that gain back. My first thought is to change the PI because i think it needs to happen after the tweak i did in order not to mess with the tone it created. If i wanted to add gain at the PI, what would be the best way to do that w/o changing anything but the volume of the amp?

[txstrat]

I wouldn't mess with the PI, since it's partly responsible for "your" tone.
What kind of OT are you using? I know it's a bit pricey but I'd think about a bigger OT if you really need to increase the headroom.
I know the OT is also part of the tone, but if you change for a similar one you might get away with it.
Or you could try to fumble around with the values of your menioned resistors (e.g. 150K plate / 1.8K cathode or 180K plate / 2.2K cathode) to find a compromise between gain and tone.

just my 2c

[Chuck H]

There are two considerations to the change you made.

1) you changed the bias of the stage. Probably cooler. This will put that stage into cutoff before saturation. This has the effect of removing the saturation element of clipping and tends to reduce gain unless compensated for in a later stage. It does add a tonal quality that some players like. Many builders (Soldono, Trainwreck, etc.) use a cold biased stage to employ cutoff only clipping.

2) you have increased the impedance of that stage. In this type of gain stage increasing the impedance reduces high end because impedance increases with frequency. The amount of top end roll off has been increased.

You asked about increasing gain in the PI. The PI circuit in that amp is already about as efficient as it can be. Also, since the loss of gain is behind the PI, even if you increase the PI output you will not get back the effect of over driving the PI input that you had before. So you have effectively decreased distortion in two stages. The one you changed by reducing saturation and the PI by reducing voltage input (actual gain entering the PI).

You can get some of the gain back by reducing the value of the 2.7k cathode resistor on the stage you changed. You may still get some of the tonal benefits you've experienced, but this will bias the stage closer to where it was. So it's a trade off to some degree. Since you will be keeping the 220k plate R you will not need to heat the bias of that stage all the way to where it was. The 220k plate R does increase gain when done independantly from the cathode R of that stage. So... Maybe try a 2.2k and a 1.8k cathode R to see (hear) if you get your gain back and the beneficial tone change is still to your liking.

I didn't take the time to explain every detail about what's going on in the interest of being concise. But I hope you get the gist of it. I always say that it's better to understand something than to simply know it rote.

Chuck

[daz]

Thanks. I will try some different values, tho at first i changed the plate R and forgot to do the cathode and if i recall the gain was as low as it was even after i added the 2.7k in place of the 820R. At least i THINK thats the case. But i'll try the cathode first and if that doesn't do much i will gradually drop the plate value.

I'm surprised so few use these values because i have found many marshalls a bit thin and harsh sounding and would have benefited a lot from it. In fact lots of amps that have CF stages would. I have used these values in the first stages and didn't get this kind of result, (similar but not nearly as pronounced) so why here? Because it's immediately B4 the CF?

EDIT: by the way Chuck, i have a pre PI master so the PI really wasn't distorting at the volumes i play at. So it's just that one stage that is cleaner. i never realized how hot that stage was, and now with this major difference in tone i think that has been a problem all along and why it's been so hard to get the changes i wanted even with all the great advice i've gotten.

[daz]

As the court jester of this forum you won't be surprised to hear this. the reason for that volume loss wasn't the 220k resistor....it was the 2.2k resistor. In other words, i thought i had grabbed a 220, but when i looked at it today when i got home it was a 2.2 ! No wonder the volume went away !

So anyways, before i saw that i turned the amp on and played it. It didn't sound nearly as good as i thought because when i tried it last nite it was at low volume. (roommates, late, etc) After work i cranked it and that full sound was there, but it wasn't there in a GOOD way. So much for that tweak. I'll be down in the dungeon if you need me.........

[Chuck H]

Well... Don't abandon the notion just yet. One of my favorite designs is a cross between the Vox and Marshall type circuits that uses a first gain stage with a 220k plate/2.7k cathode resistor. There is a different vibe to the higher impedance circuit that's worth experimenting with IMHE. The circuit I designed uses a dual ganged pot that changes the preamp voicing as the volume is advanced from an AC30 type voicing for jangly cleans to a more Marshally voicing for OD. Very cool! Not sure how it'd work in a push pull amp. But it sounds great in the little VibroChamp I put it in.

Chuck

[daz]

I did use a 220k on the first stage at one point, and you were telling me to go to a 100K. And you know what......you were right. I have tried going back to a 220k since, and in fact just tried it again last nite. But while i DO like it, it is very different from the 100k and the 100 gives me a more alive sort of tone with better lows. So in this amp at least it doesn't seem to be good on any stage. But there are things about the tone with a 220k plate on V1A that i wish i could add to the tone with the 100k. Each has it's own good things, but the 100k overall wins out.

I'm afraid after months of contentment i am entering another experimentation period on my poor abused amp ! LOL! But seriously, i do enjoy the hunt and this time i have new ideas up my sleeve and i think i will make better progress then ever. In fact it already has improved a bit. You should see the turret bgoard tho.....it's horrible ! Like a junkyard of solder blobs and rosin. If i ever get it to where i don't think i will be messing with it anymore (yeah, right) i will get a new board and components and redo it completely, making sure to take lots of pics of the lead dress !

[jrfrond]

That mod will INCREASE the gain of the first stage. Increasing the plate resistor raises the gain and the cathode resistor sets the bias point. You've got something else going on by accident. Was the cathode resistor bypassed and you forgot to put the cap back in? If that is all good, double-check your resistor values and measure them both if necessary. Also try another tube. What you describe should NOT have happened under normal circumstances.

BTW- the reason you find Marshall's thin and harsh-sounding is not the 220K/2.7K combination. There's a lot of low-end roll-off going on, starting with the .68uF bypass cap, and also a series high-pass filter consisting of a 500pF cap paralleled across a 470K resistor after the volume control. Why does Marshall do this? Simply because if you drive a half or full-stack cab at high volumes, you want to keep that bottom end tight (nothing wrong with that! ).

[daz]

John, you probably didn't read all my posts, but last nite i posted to say the reason why that 220k lowered the amp volume a lot. And you're right....it WAS an accident. It was a 2.2k, not a 220k. I thought i had grabbed a 220k but accidentally grabbed a 2.2k. My eyes aren't particularly good anymore, and when combined with impatience things do happen.

[Chuck H]

I did use a 220k on the first stage at one point, and you were telling me to go to a 100K. And you know what......you were right.

Well be sure that I don't use a 220k plate/2.7k cathode combination in all my builds. If I said to try a 100k plate I suppose I had a different reason at a different time.

Chuck

[jrfrond]

John, you probably didn't read all my posts, but last nite i posted to say the reason why that 220k lowered the amp volume a lot. And you're right....it WAS an accident. It was a 2.2k, not a 220k. I thought i had grabbed a 220k but accidentally grabbed a 2.2k. My eyes aren't particularly good anymore, and when combined with impatience things do happen.

Oh yeah..... oops!

Hey, like I said, it DOES happen! I have a sign hanging in our parts department that says double-check you part number or value when pulling from a drawer.