Replacing TL072 in Mesa/Boogie TriAxis: OPA2134 or OPA2604 ?

Cub · Post by **Cub** » 20 Sep 2011, 07:19

Hello lads, hope you can help me with this one.

I've been having a little trouble with the TriAxis and found out a opamp that takes care of the "Dynamic Voice" (sort of a graphic equalizer with a limited number of pre-set unchangeable settings) might be a fix. The thing is, when things get a bit loud, the chip starts to clip in a rather unpleasant way.
Now, I read some people have changed it for a Burr Brown OPA2134 or OPA2604 and was just about to order some, but then I read these words of caution on a M/B forum:

I researched the OPA2134, and in a lot of cases the OPA2604 came up along with higher praise. It's reputed to have better sound quality and even higher dynamics, but it's not a direct drop in from what I've read.
I read that the higher slew rate could cause oscilation, and it has a much higher current draw. So with the higher power, the more heat it will generate. Plus, I don't know how it will affect other parts in the circuit, and if this will cause a chain meltdown. I feel confident in replacing a TL072 with an OPA2134 as the DIY community accepts that this is a safe swap.

The TriAxis is the most important and most expensive piece of gear I own, I wouldn't want to destroy it but if I could get rid of the nasty clipping, it would be perfect !
What do you lads think ? Thank you for your time and help in advance !

All TriAxis schematics are on Tubefreak, click on "schematics" in the left frame, then on "Mesa/Boogie Triaxis Preamp" in the main frame. Sorry, it won't let me link to that section directly.
http://tubefreak.com/tubefreak.html

This is the bit with the Dynamic Voice:

phatt · Post by **phatt** » 20 Sep 2011, 11:12

Hi Cub,,
A darn tricky circuit to work out but,,

My guess is IC3a is being slammed way to hot,, so changing opamps won't do anything useful.
I would just simply add some series resistance in front of IC3a and/or reduce *R233* from 1 meg down to at least 470k.

This is the problem when shoehorning opamps in the middle of Valve circuits.
Opamps are great but headroom is a big ask.

A lot less hassle and cheaper to turn something down,, Even a couple of Zener's across the input might help, Fender use zeners in there deville circuits.

Phil.

deltafred · Post by **deltafred** » 20 Sep 2011, 12:26

Is the input DC coupled from "V3 A" or is there a decoupling cap?

If it is DC coupled (or a leaky decoupling cap) then any offset is going to reduce the usable signal level before asymmetric clipping occurs.

I don't see how changing the opamp is going to improve matters that much. IC3 is running on +/- 15v so there will be a good voltage swing available with the TL072.

Edit
Just a thought, have any of the previous stages been modified for more gain at any time.

Cub · Post by **Cub** » 20 Sep 2011, 16:00

Thanks for the replies so far, lads. My TriAxis is a version 1, no mods whatsoever. All chips are socketed, so swapping an opamp isn't as big of a deal as it may seem. I'm a novice pedal guy at best and it's a bit intimidating to look under the hood of this midi pre-amp. If changing the fizzy opamp will fry this otherwise fine unit, I'll be ampless ! So I'm cautious after reading the post I quoted above about what to put in there.
V3 is the mixer tube, there are decoupling caps at V3A and there's also the presence and master before it goes to the Dynamic Voice section.

Below are all the schematics from Tubefreak:

Mesa/Boogie Triaxis preamp
First stages with additional info (89kb)
First stages (66kb)
Dynamic Voice + loop (33kb)
Outputs (33kb)
Mode switching (86kb)
CPU/Memory (42kb)
Port decoder (25kb)
System clock/reset/display/midi (40kb)
Mod notes1 (21kb)
Mod notes2 (32kb)
Presence mod notes (28kb)
Regulator mod notes (18kb)

Groovenut · Post by **Groovenut** » 20 Sep 2011, 17:26

The most simple fix is turning the master volume down until the clipping on the offending patch stops. If you need further overall volume you can turn your PA up. This will also probably mean changing the master volume setting on your other patches to match the lower volume level of the altered patch. I ran into this exact situation early on with my use of the Triaxis and this was the fix Mesa recommended. The large variance is Master Volume settings is there to accommodate the clean modes, which will have higher MV settings by nature, as well as the lead mods which will have lower (sometimes very low) MV settings. As deltafred said unless you can find a way to increase the supply voltage to the opamp, your not going to gain any headroom.

Intripped · Post by **Intripped** » 20 Sep 2011, 19:27

probably an insane idea... IC stacking?

Cub · Post by **Cub** » 22 Sep 2011, 11:32

Groovenut wrote:The most simple fix is turning the master volume down until the clipping on the offending patch stops. If you need further overall volume you can turn your PA up. This will also probably mean changing the master volume setting on your other patches to match the lower volume level of the altered patch. I ran into this exact situation early on with my use of the Triaxis and this was the fix Mesa recommended. The large variance is Master Volume settings is there to accommodate the clean modes, which will have higher MV settings by nature, as well as the lead mods which will have lower (sometimes very low) MV settings. As deltafred said unless you can find a way to increase the supply voltage to the opamp, your not going to gain any headroom.

Thank you, I think I'll just try that and leave the insides unmolested.

Just for curiosity's sake though, was that bloke on the M/B forum onto something ? Will things get messy when a 2604 replaces the TL072 in the IC3 socket because its higher slew rate and current draw ?

kleuck · Post by **kleuck** » 22 Sep 2011, 12:27

2604 eats a lot of current, almost 10 times the TL072's one at idle.Depending on the PS design, it can be an issue yes.
Slew rate are almost the same.

Hides-His-Eyes · Post by **Hides-His-Eyes** » 22 Sep 2011, 12:48

Intripped wrote:probably an insane idea... IC stacking?

That's not going to increase the headroom is it?

Intripped · Post by **Intripped** » 22 Sep 2011, 13:43

not big on theory, that's why i said probably an insane idea
i juts read something about headroom improving when you make this mod in a TS-like pedal... i thought that maybe it could work here too.

mjones99 · Post by **mjones99** » 02 Oct 2011, 12:17

If you want to replace the opamp I would go with something rail to rail, like a TLC2262, TLC2272 or something similar. Stacking does increase headroom a bit in my experience, but rail to rail i/o will be the best option, of course you could stack a couple of rail to rails.
j

Groovenut · Post by **Groovenut** » 02 Oct 2011, 15:00

Cub wrote:
Groovenut wrote:The most simple fix is turning the master volume down until the clipping on the offending patch stops. If you need further overall volume you can turn your PA up. This will also probably mean changing the master volume setting on your other patches to match the lower volume level of the altered patch. I ran into this exact situation early on with my use of the Triaxis and this was the fix Mesa recommended. The large variance is Master Volume settings is there to accommodate the clean modes, which will have higher MV settings by nature, as well as the lead mods which will have lower (sometimes very low) MV settings. As deltafred said unless you can find a way to increase the supply voltage to the opamp, your not going to gain any headroom.
Thank you, I think I'll just try that and leave the insides unmolested.

Just for curiosity's sake though, was that bloke on the M/B forum onto something ?

This was from a service tech at Mesa Boogie. I cant remember his name now, maybe Mike.

phatt · Post by **phatt** » 03 Oct 2011, 08:50

It matters little who said what.

The laws of fizzy stuff should be obvious,,,the opamps can only handle 30 ish volts signal swing (and that's pushing it)

Meanwhile Valvey stuff can swing even up to TEN times that

Yikes,,,, go figure!!
Phil.