TMB + Negative Feedback Loop = Soiled Underwear.

[zaphod_phil]

If you are useing the 220k make sure they are 220k not 220 ohm...these just help to keep the channels from bleeding together.

Well, if you leave the Normal channel connected to the "B" input of the PI, surely you don't need either of the 220k resistors. The NFB coming into the PI's tail resistor should affect both channels equally. One 220k on it's own for the TMB channel going into the "A" input of the PI will help just roll of some of the highs a little. A 470k will do so slightly more.

If you try the newer way of putting the normal channel into the TMB tube,then you won't need the 220ks.

You mean because you're now using the two 470ks instead?

[jersey_aaron]

How exactly do I add a presence control to this circuit? I'm not sure how the pot should be wired..

[zaphod_phil]

The pot is wired between the PI tail resistor and ground. The 0.1uF cap is then connected between the wiper (center lug) of the pot and ground. There you go...

[KHStudio]

Well, if you leave the Normal channel connected to the "B" input of the PI, surely you don't need either of the 220k resistors.

Could you make a rough scetch to help me see it better?
I've allready tried the NFB once & want to collect all my beans before I dive back in and screw something up.

There seems to be several options. Which one do you guys think is the best. (for the 18w / TMB)?
Even a rough scetch of the diffent ones & an idea of what it should do would be killer.


Is this what you guys are saying to do? (hookup wise not the values so much)

http://www.18watt.com/files/nfb_hookup_q1.gif[/img]

[zaphod_phil]

Yes, that's the one I'm suggesting you try first. Different people are trying out slightly different variations, according to personal taste. For example Richie is recommending something like a 47k or 27k instead instead of the 820 ohm feedback resistor and a 1.5k instead of the 100 ohm at the PI tail. The 0.1uF cap across the resistor has the effect of your Presence control turned all the way up. So you could try smaller cap values, like 0.047 to decrease the amount of Presence (and hiss).

Then if this doesn't do it for you, you can try some other variations, like Richies' latest one with both channels feeding into the 2nd pramp tube via 470k resistors. The experimentation is half the fun....

[jersey_aaron]

I'm not getting you, Phil. Are you saying to run a wire from the PI tail resistor to the left side of the 5k pot, and jumper the center and ground with a .1uf?

Do I run the 100ohm resistor seperately to ground from the PI, or does this resistor go to the pot as well?

[zaphod_phil]

It's so hard to explain these things in text And I'm not too good with the drawing packages either. OK, let's say in your case you like 100 ohms as your "presence resistor" at the end of the PI tail. So just replace it with a 100 ohm pot, with the two end legs of the pot connected in where your 100 ohm resistor was - ie between the tail resistor and ground. Then you connect the upper end of your cap to the center lug (wiper) of the pot, and leave the lower end connected to ground as before. That's it - you now have your Presence control.

The only prociso I'd make here is that as you're using much lower resistor values for your two resistors than Fender or Marshall do, I'm not sure how strong the effect of the Presence control will be. But it's certainly worth trying. In Richie's case, as 1.5k pots aren't easy to fiind, if he wanted to try this, he would probably have to use a 2k pot.

[Plexi]

If your going to use a pot, just use a 5k, that is the most common value for a presence pot..and you can adjust that to any setting.. All the resistors values are to set it to a certian point,since you aren't useing a pot..

Richie

[zaphod_phil]

If your going to use a pot, just use a 5k, that is the most common value for a presence pot..and you can adjust that to any setting.. All the resistors values are to set it to a certian point,since you aren't useing a pot.

Actually, you can experiment with different pot values too. And if you're using an 820 ohm feedback resistor, a 5k pot would be out of proportion. For a feedback resistor anywhere in the 27k to 100k range, a 5k pot makes good sense. All that the pot is doing is to allow you to vary the amount of high frequency you filter off, proportional to the position of the wiper, so it doesn't get attenuated by the NFB.

And one other thing, if you use a pot which has a switch, then your Presence control can also be used as your NFB On/Off switch- to reduce the number of extra holes...

[Attilio]

Hi all!
I've been far from 18watt.com due to school problems...
Now I want to convertt my 18w into a tmb...
I have a couple of questions:
a) does somebody have a schematic which shows the switchable negative feedback loop?
b) I really like tremolo channel (i mean with the tremolo off, I usually play with this channell): is it possible to modify the normal channell this way? Will changing the tone circuit and change the 100K plate resistor with a 220K one work?

Thanks fot the help!

[Budso]

I'm going to jump in here and stir up the stew a bit. Phil is probably well aware that I've learned volumes from him over at AX84. But in this case I would wire the presence cap differently.

JA, keep the NFB setup you now have with the two resistors. Remember that the ratio of those resistors to each other control the amount of feedback voltage going to the PI. This feedback voltage is completely out of phase with the incoming signal, so naturally it cancels part of it. You may want to experiment further down the road with different resistor values, so keep these in place.

The cap across the smaller resistor is taking part (middle high frequencies) of this feedback voltage to ground. These frequencies are taken to ground before they have a chance to cancel out their incoming counterpart. So this part of the input signal (middle highs) comes on through strong and unchanged and is known as "presence".

If we want to vary the frequencies that come through, we could change the cap out, substituting a larger or smaller value. Or, we could simply wire a pot as a variable resistor between the cap and ground and call it a "presence control".

Run a wire from the cap to an outside lug of the pot. Wire the wiper to ground. Voila! If the pot runs backwards, put the wire on the other outside lug. Or use an DMM to figure out which outside lug to use. Sorry, I can't recall. Hope this is of some use.

Bud

[Budso]

oops. Led you astray on the pot wiring. Try this:

cap to -> outside lug of pot

other outside lug to -> ground

Then jumper the wiper lug to one of the outside lugs. This then varies the resistance as you turn the pot, thus affecting the effectiveness of the presence cap. Sorry for the confusion.

[zaphod_phil]

Dang, my cover's been blown!

Or, we could simply wire a pot as a variable resistor between the cap and ground and call it a "presence control".

If you do it that way, you will be varying the *total* amount of NFB in the amp, rather than just the bypass of higher frequencies. So what you will have there is a Feedback control, rather than a Presence control - but some folks may like to try that as well...

[zaphod_phil]

Hi all!I've been far from 18watt.com due to school problems...

Hey welcome back! What have you been eating, or have you been having plastic surgery? You used to look like Jimmy Page, now you look more like Hendrix!

a) does somebody have a schematic which shows the switchable negative feedback loop?

This is still very experimental, with different ideas and resistor values being tried. The "Dookie Mod" is JA's version - http://www.18watt.com/modules.php?name= ... oad&id=506 but you need to follow the discussion in the thread. If you place a switch in line with the feedback resistor, it's switchable. My comment was that you can leave the TMB and Normal channels connected to the PI as they were before and not bother with the 220k mixing resistors. Richie is doing a completely different version with the first stage of both channels being fed into the TMB, much like an original Plexi.

b) I really like tremolo channel (i mean with the tremolo off, I usually play with this channell): is it possible to modify the normal channell this way? Will changing the tone circuit and change the 100K plate resistor with a 220K one work?

No not quite, you would need to copy the whole circuit pretty much - with a 330k load resistor, followed by the high-pass filter consisting of C15, R29 and C12 in the PedalMonkey schematic, going into the tone control circuit.

[Attilio]

Hi Phil!
Do you meant something like this? (switch and component values are not shown)?

[jersey_aaron]

Ok Gents, a few things..

First, the presence control with a 5k pot is a no-go. It just doesn't work with the 820 ohm, 100k, or 47k. It took me a long time of trial and error, swapping components back and forth to discover the "magic" combo of the 820ohm and 100ohm resistors for this feedback loop to work. I guess this is one section of the TMB that does not resemble the "standard" behavior of a Marshall circuit. Which brings me to my second point -

REMOVE the .1uf cap and replace it with a 25uf. This "opens" up the amp like you would not believe. An amazing amount of clarity, definition and LOUDNESS will appear. Like a veil is being lifted from the entire circuit. Everybody MUST, at the very least, try this. You will be amazed.

Third, Phil is right- there is no need to send both channels into one side of the PI. You can leave both channels "as is".

So, here is the updated "Dookie Mod", version 2.0, with all of the above "enhancements" added.

[Budso]

Or, we could simply wire a pot as a variable resistor between the cap and ground and call it a "presence control".

If you do it that way, you will be varying the *total* amount of NFB in the amp, rather than just the bypass of higher frequencies. So what you will have there is a Feedback control, rather than a Presence control - but some folks may like to try that as well...

Phil, you gotta help me here. I don't see how my setup differs from yours other than the placement of the pot. You're using the fixed resistance value of the pot to replace the resistor that normally goes to ground. I'm just leaving that resistor there as is. You then use the wiper's sweep (very clever I might add) across the pot to vary the resistance between the cap and the circuit. I'm also doing that with the variable resistance pot between the cap and the circuit. Both of these circuits vary only the frequencies going thru the cap.

So with both circuits having a fixed resistance to ground, how does my circuit vary the total NFB? Thanks again.

Bud

[zaphod_phil]

Do you meant something like this?

Yes, that's it!

[zaphod_phil]

I don't see how my setup differs from yours other than the placement of the pot.

Sorry, I think I misunderstood your description. I now see what you're saying.

You then use the wiper's sweep (very clever I might add) across the pot to vary the resistance between the cap and the circuit.

Nothing clever there actually - it's just how Fender, Marshall and most of the rest do it....

[zaphod_phil]

First, the presence control with a 5k pot is a no-go...

Well, 5k (or 4.7) is pretty much the universal value used by Fender, Marshall and most of the rest, which is why I think Richie was recommending it. However, I like the fact that you've obviously tweaked your amp very carefully by ear. However, there's one thing about this particular variation that's a little unusual, if you use a 25uF cap to bypass the 100 ohm resistor, that will ensure that only the very lowest frequencies get fed back and all the rest of the lows, mids and highs will go through the amp just like before. But maybe that's what you want - tightened up low end and everything else going through full throttle....