Mk3 Master Volume fix?

[andykrow]

Hey All,

First off - Hello! This is my first post and it's about my first Mesa, a Mk3 Blue Stripe 1x12 Simul-Class. I absolutely love the tone and can't believe it took me this long to get a Mesa.

My issue is the Master Volume pot. As many people report, it goes from too quiet to WAY too loud between 1 and 2 on the dial. I basically have to just barely nudge the pot to make acceptable changes in the volume without making my ears bleed.

At gigs this has proven to be fairly annoying!

Has anyone successfully modded their amp to fix this situation? I've heard about maybe swapping out the volume pot for something else, or swapping the PI tube out. But I haven't found any reports of anyone actually doing this. It would be great to have this control be way more linear.

Thanks in advance for any help you can give!

-Andy

 

[MrMarkIII]

Well, if it's still an issue if you switch to "Class A" (15 watts), there just may be too much amp.

Try reducing the Volume 1, and/or the Treble, as these controls also affect gain and volume.

If it's too distorted in "Class A", try 6L6's instead of EL-34's, if that's what you have now. That should get a bit more clean headroom in the 15 watt setting (but only a bit).

Don't be afraid of running knobs at "zero" if that gets the tone. 'Specially the Bass knob. If you really need more Bass (you don't), pull the Bass Shift out.

If it's a combo, try plugging the speaker into the 4 ohm output instead of the 8 ohm.

If you have the Graphic EQ, try reducing all the sliders a bit. This will reduce all frequencies across the board, and should take some loudness away.

But what do I know, I still have my Blue stripe I bought in 1990.

 

[loylo69]

I experienced the same thing with my mkIII red stripe.
As suggested by MrMarkIII, you can try to skirt the problem with different settings tricks or by swapping tubes.
But in my opinion, it is mostly a problem of design of those amps by Mesa.

Since the production of mkII to the mkIV, the master pot is configured as variable resistor.
It needs to be configured as a traditionnal volume pot (as a voltage divider), like it is done in almost every amps, and in the mkV.

I have rewired it in the traditionnal "voltage divider" way. The change wasn't drastic first. It still went from too quiet to way too loud between 1 and 2 with a flick of a turn. So I also got rid of the 22k resistor before the pot. Since that, I'm very happy: the change of volume is more progressive. But it still occur between 1 and 2 on the dial.

So, in order to extend the range of sensibility, the pot may also need to be changed, from 250k log (the original value) to 1M log. I haven't tried it yet. I need to order a CTS push pull replacement pot.

 

[andykrow]

@MrMarkIII

It's certainly not too much amp, but there needs to be a happy medium between quiet and full-bore. Otherwise what's the point of a master at all? I feel like mine is basically at full volume at 3 on the dial and increasing from there does little to nothing. It would be great to truly hit max volume at 10 on the dial. The amp still sounds amazing a bedroom levels, it would just be nice to not be accidentally waking the neighbors if I turn that master just a hair too far.

Unfortunately (or fortunately as it may be, haha) I really REALLY prefer the sound of Simul-class w EL34 and 6L6's in there. Class A sounds good but Simul has some magic to it that I'm not willing to live without. I have had quite a few blackface fenders and I need that class A/B 6L6 action..

I do keep the bass dial really low and tend to have the deep switch pulled.

As for Volume one, I have tinkered a bit with lowering that, but I'm a Strat guy who loves amps that are on the edge of breakup, and I've found with this amp I really need Vol 1 turned up pretty hot, usually 8-10.

I haven't messed with bringing down ALL the sliders on the graphic as sort of a second master volume. I do use the graphic so I will try that tonight!

@loylo69

I am not super familiar with amp circuitry, but when you say voltage divider you mean signal comes in on the wiper, and then one side of the resistor track is the output and the other side goes to ground, right? If I'm totally off on that can you describe what you did in a little more detail? I have done tons of rewiring on my fenders so I'm confident I can follow directions

Changing the value to 1M sounds like money to me, please let me know how that goes!

I have submitted this question to Mesa support so I will see what they come back with as well.

Thanks to both you guys for the tips!

 

[andykrow]

Okay, studying up on a voltage divider circuit. Seems I was a little off. It's more like signal in AND out connect to the wiper, and then one side of the wiper goes to ground. Then as resistance builds between wiper and ground, less signal goes to ground and more goes to power section, increasing the volume. (I'd love to hear from some amp circuit gurus here...)

Anyway, it seems to me that a 1M might actually make this worse? Changes in resistance would happen much more quickly.

Is the volume pot actually logarithmic?

I'm probably totally off in my thinking here, but it seems to me that a smaller value of pot would actually be useful here. If there's a 250k in there now, and at 25% rotation (or less) full volume is reached AND it's logarithmic, I'm thinking I can stick a 50k in there and still be able to get full volume. Plus there are different log curves out there to tinker with.

I'm going to dig into this and report back with what I come up with.

 

[loylo69]

Okay, studying up on a voltage divider circuit. Seems I was a little off. It's more like signal in AND out connect to the wiper, and then one side of the wiper goes to ground. Then as resistance builds between wiper and ground, less signal goes to ground and more goes to power section, increasing the volume. (I'd love to hear from some amp circuit gurus here...)

To be more precise, the signal goes into the master pot (at the output of V2B tube, through a 47n cap and 22k resistor) at the 3rd lug of the pot (the right lug when looking at the back of the pot). The signal is then taken out of the master pot at the wiper lug, then going to the GEQ...
And the last lug (the left one when looking at the back of the pot) is connected to ground.

Anyway, it seems to me that a 1M might actually make this worse? Changes in resistance would happen much more quickly.

Not at all. It is counter intuitive, but a bigger pot would divide scale down the signal (voltage division) more than the original pot. So, at the begining of the rotation, you need to raise the 1M pot higher than the 250k pot to have the same volume. The volume with a 1M pot at 2.5/10 is approx the same with a 250k pot at 1/10. (but the volume is the same when they are both at 10).
In fact, a 1M log pot would give you a range of settings from -50dB (pot at 0) to +65dB (full volume).
A 250 log pot (the original pot) would give a range of settings from -30dB to the same +65dB.
So the range is extended with a 1M pot.

Is the volume pot actually logarithmic?

Yes it is. That's because the human ear is a logarithmic sensor, not a linear one.

 

[andykrow]

I love learning this stuff. Thanks!! What you say makes sense, and I appreciate the input.

To be more precise, the signal goes into the master pot (at the output of V2B tube, through a 47n cap and 22k resistor) at the 3rd lug of the pot (the right lug when looking at the back of the pot). The signal is then taken out of the master pot at the wiper lug, then going to the GEQ...
And the last lug (the left one when looking at the back of the pot) is connected to ground.

So you are describing how you rewired for a voltage divider, NOT the stock wiring, correct? And you removed the 22k resistor, correct?

FWIW, the response on this from Mesa was to find a linear 250k pot with a switch. They did not mention converting the circuit into a voltage divider. My intuition can clearly be wrong (haha) but wouldn't this exacerbate the problem??

The great thing about the 1M log pot w switch is that it is easily available in the same form factor as the 250k. A 250k linear switched pot that Mesa recommends is very difficult to find, and only available with a DPDT switch. Not a big deal but the pot is a lot deeper.

Do you think the 1M would have the same sensitivity issues as the 250k, just shifting it further along the knob? For example instead of the current useable range being between 1 and 2.5 it would be between 3.5 and 5?

Anyway, I just ordered the 1M and I'll report back with what I find.

 

[loylo69]

I love learning this stuff. Thanks!! What you say makes sense, and I appreciate the input.

You're welcome, glad to help. :wink:

So you are describing how you rewired for a voltage divider, NOT the stock wiring, correct? And you removed the 22k resistor, correct?

Exactly. The 22k resistor is not needed for the purpose of volume control. But it was added by Mesa to counter the values discrepancy of the pots. By construction, they don't read 0 ohm when dialed all the way down.
In fact, with my original 250k log pot and without the 22k resitor, I can hear a little bit of sound when the pot is dial all the way down. But it doesn't bother me, as getting rid of the 22k resistor allowed me to have a more usefull sweep of the pot.

However, I suppose you wouldn't have any issue with the 1Meg pot. You wouldn't hear anything when the pot at 0 because 1Meg cut the signal even more than 250k. And the sweep should be nice when wired in voltage divider way.

FWIW, the response on this from Mesa was to find a linear 250k pot with a switch. They did not mention converting the circuit into a voltage divider. My intuition can clearly be wrong (haha) but wouldn't this exacerbate the problem??

That surprises me from Mesa. A linear pot would be much worse.

Do you think the 1M would have the same sensitivity issues as the 250k, just shifting it further along the knob? For example instead of the current useable range being between 1 and 2.5 it would be between 3.5 and 5?

I feel it would have a better sensitivity, what you are searching for.
But I can't be sure, it depends of the construction of the specific pot. Some approximate the logarithmic function better than others.

Anyway, I just ordered the 1M and I'll report back with what I find.

I'll be waiting for your report. Thanks in advance. :wink: