Master volume problem.

[Alan0354]

I tried putting a 1M pot across the plate of the PI stage of both my Fender Pro Reverb and the modified Bassman100 as master volume. It sounded awful when adjusting the resistance down( lowering the gain of the PI stage). This is a circuit used by Matchless, it cannot be that bad. But it just sounds terrible. Any idea?

[cbarrow7625]

I have never had luck with that type of MV. My experience has been the same as yours. That being said, I think the problem is that type types of amps on which I have tried it are just not the right design to make use of that type of MV. Not every "trick" out there for tube amps is universal. If you start a design with the idea of using the cross-plat MV, then you can design the rest of the circuit to work well with it. Slapping it into a Fender amp like those is almost never going to sound good.

What tends to work well in those is a dual ganged 1M pot on each PI output between the PI & power tubes, with blocking caps before and after it of course. For my own amps I usually just replace the power tube 220k grid leak resistors with a dual ganged 250K pot. One less cap in the signal chain. Although I would never do that on a production amp. I also make sure to put a 1M resistor between the pot wiper and the bottom lug of the pot to help maintain the bias to the power tubes should the wiper on the pot fail.

[pdf64]

Maybe it works better on LTPs using a tube with high plate resistance, eg 12AX7, rather than lower plate resistance, eg 12AT7?
And a low degree of (or no) global negative feedback may also help.
Pete

[Chuck H]

I think it's most likely that the folks at Matchless never intended that master volume to sound good at bedroom practice volumes. Real players understand the compromise. It's probably only intended to sound good taking "a little" volume off the full output power. Matchless amps were designed to be gigged.

cbarrow's suggestion of replacing the bias resistors with a dual 250k pot is a good one.

EDIT: My 2C is actually that post PI MV's don't usually sound very good. A simple resistive attenuator would sound better IMHO.

[J M Fahey]

The 3 distinguished gentlemen above are right.
1) our friend Alan did *not* do the same as in Matchless.
It has no negative feedback, so as an example, the first 10dB of attenuation lower output volume by 10dB, a very useful range.
The Fender, excellent as it is, does have heavy NFB (for a guitar amp) and so the first 10dB (or so) of attenuation are wasted in killing NFB without controlling volume much, if at all.
2) expensive and not easily available Matchless tend to be used by Pros .... who gig , at least in a Club situation, with a few dozen patrons ... so the actual attenuation range is, say, to lower volume somewhat, so as not to annoy the drummer or make the Soundman job easier ... not *much* attenuation by any means.
Agree that at bedroom or appartment levels, sound will be bland, even in an original Matchless.

[big_teee]

I like the Lar/Mar PPI-MV Mod.
I have it in my clone 2204s.
If the PPI-MV is turned above 10 oclock, it sounds pretty good, but you do sacrifice some Presence Control.
I run both the pre and post MVs.
It makes the amp useable at medium low volume.
I wouldn't have a 50w and bigger without one.
T

[Chuck H]

Good catch on the NFB Juan. I knew that too but failed to consider it. Post PI MV's within a NFB loop are most often doomed.

[Alan0354]

I have to try to open the NFB.

[J M Fahey]

Fine, but then remember it won´t be a Princeton any more

[Chuck H]

Indeed! IMHE BFish type amps have a mid scooped tone and lifting the NFB makes them harsh and flabby. I still contend that my simple attenuator is a good solution.

[Alan0354]

I tried opening up the NFB, it did not help, still sounded crappy. But opening the NFB give me a new sound that is pretty nice.

[Chuck H]

Attenuator

[Alan0354]

Attenuator

Yes, I got a THD Hotplate already. I am just monkeying around since Matchless did that.

For reducing to half power, I think replacing one of the two speaker with a 10 ohm seems to sound better than the THD set at -4dB( attenuate a little over half power). The resistor sound more natural, the only noticeable thing is I get a little less bass with the resistor. But the mid and highs are pretty much the same. The reason I call out 10 ohms is because I tried 8 ohm and particular 4 ohms, it was more obvious the bass got attenuated. I wonder why.

[Chuck H]

Maybe because the LF impedance of a guitar speaker has a dramatic peak (usually between 30 and 100 ohms at 70Hz to 85Hz). If you series a resistor of 10 ohm most of the LF power shared by the two elements (the resistor and the speaker) is going to go through the resistor instead of the speaker. So the lower your resistor value the more LF is diverted through the resistor. Especially at that peak frequency. Some resistive attenuators actually run the amp into a higher load than the speakers rated ohmage to average the dynamic impedance curve. I think the Airbrake is 22 ohms (intended as an 8 ohm speaker replacement) and the Ultimate uses a 32 ohm load for all impedances. The only way around it is to build an active load as the resistive element. My personal attenuator uses an active load in place of a resistor. Because of it's design that makes it a combination resistive/active load and seems to be a tad better than just a resistor. I've never played through a Hotplate.

[Alan0354]

Maybe because the LF impedance of a guitar speaker has a dramatic peak (usually between 30 and 100 ohms at 70Hz to 85Hz). If you series a resistor of 10 ohm most of the LF power shared by the two elements (the resistor and the speaker) is going to go through the resistor instead of the speaker. So the lower your resistor value the more LF is diverted through the resistor. Especially at that peak frequency. Some resistive attenuators actually run the amp into a higher load than the speakers rated ohmage to average the dynamic impedance curve. I think the Airbrake is 22 ohms (intended as an 8 ohm speaker replacement) and the Ultimate uses a 32 ohm load for all impedances. The only way around it is to build an active load as the resistive element. My personal attenuator uses an active load in place of a resistor. Because of it's design that makes it a combination resistive/active load and seems to be a tad better than just a resistor. I've never played through a Hotplate.

The resistor is in parallel. The Pro Reverb is 2X12",8 ohm each. So I replace one speaker with a 10 ohm resistor to keep the impedance.

But your logic might still hold, if the impedance of the speaker peak at 100Hz to 30 ohm, then when you have both speakers, the two still share the same power. BUT if I use a 10 ohm resistor as one speaker, at resonance, the speaker is 30ohm but the resistor still 10 ohm. This means most of the power around 100Hz goes to the resistor and the bass gets attenuated.

You should try this when you have time and tell us your opinion. To me, it's the closest sound to reduce to half power for tube amp ( SS doesn't work). I can hear the obvious attenuation. The THD change the sound ( colored) even at -4dB.