Mesa Rocket 44 bias circuit mod?

[unparalleled]

I am having nasty problems with this repair to Mesa Rocket 44......
This amp has just had PT rewound ( the OPT has also been rewound earlier ) and the EL84s are drawing excess current.
The PI cathodes are connected via a resistor to the bias supply and this arrangement draws current from the bias supply.
When the 12AX7 is inserted the available bias to the output tubes drops from -20v to - 3.3v
causing the EL84s to cook.
A design which draws current from the bias supply seems to me to be asking for trouble!
Has anyone successfully remedied this problem?

[techineer]

Check or replace the capacitor 1000 uF 25v and/or check the values of the resistors - the 120K, 15K and 33K. If the capacitor has lost its capacity, it would make the bias voltage load down easily.

[Amp Kat]

You can always scrap that bias supply and derive one off of the plate like so many amps did back in the day. Got a schemo ?

[unparalleled]

You can always scrap that bias supply and derive one off of the plate like so many amps did back in the day. Got a schemo ?

Modding the amp is my last resort, yes I have the full service manual. (That was hard to get!)

[Amp Kat]

Above poster may be right. That bias supply is derived from the plate already and rectified with a 1n4007 diode and filtered and dropped thru the 120k. That filtered negative DC grid voltage at the 1000uf 25 volt cap should be -15 volts on figure F of the schemo. Figure E is 3 volts that goes to the Cathode of the diff splitter for the PI. You should get 3volts and -15 volts at those letter points. The voltage divider drops the 4 volts and feeds the 2m bias resistors. As stated if that cap is bad it may not load down until dragged on. And there is also posibility of the coupling cap after the splitter could be bad also but I'd check that -160 on the anode of that bias diode and again after the 120k dropper to verify you have -15 volts there.
One more thing is if all that checks good and the bias drops after you put the tubes in and it warms up I'd highly suspect the coupling caps. Put your meter on the low DC side and monitor it while running. If it starts fluctuating after the amp warms up then those coupling caps are the problem.

[Enzo]

Normally there is 60v dropped across 120k through that tube, that is half a milliamp of current. There are two sections of the tube, so that adds up to a whole milliamp, one milliamp. That should not tax the bias supply. The tube circuit has a problem, and modifying the bias circuit won;t cure that problem.

I too agree with the bad cap possibility, check it. Even if it is only a milliamp, a bad cap- can collapse. How many different 12AX7s have we tried?

[unparalleled]

Normally there is 60v dropped across 120k through that tube, that is half a milliamp of current. There are two sections of the tube, so that adds up to a whole milliamp, one milliamp. That should not tax the bias supply. The tube circuit has a problem, and modifying the bias circuit won;t cure that problem.

I too agree with the bad cap possibility, check it. Even if it is only a milliamp, a bad cap- can collapse. How many different 12AX7s have we tried?

Thanks for that!
I made the mistake of using preamp valves from the same chassis to substitute instead of a known good tube, voltages are now within spec. using a new 12AX7.
I had already replaced the coupling caps.

[steveward]

Mesa Boogie rocket 44 bias

I have just repaired a mesa Boogie rocket 44 with a very low bias supply . This lead to overheating valves and power transformer. The fault had developed over time and the customer was using up valves at an alarming rate .
All the components in the bias chain were perfect . The problem which was invisible to the naked eye was the PCB which had broken down and was conducting between tracks. None of this was apparent or measurable with test gear until the components were removed . The problem area being around the 1n4007 and the 120k resistor. The track from the AC supply was cut and the the diode and 120k resistor mounted externally. The resistor wa also unsuitable to handle 300+ volts ac and the 120k is now made up of 2 x 60 K resistors . The Bias chain now functions correctly -16.5 volts dc with roughly -12 volts on pin2 of the el84s and the 12-3 volts on the cathode of V1
Hope this helps someone. It took several hours to find this poor design issue. # track spacing wrong for such high voltage Ac and DC in close proximity ..
Regards Steve

[DrGonz78]

Just saw a video on youtube where it was a Mesa DC-3 with runaway bias. In the case of the DC-3 amp the B+ voltage was also dropping down too as the bias voltage started dropping. Was your bias voltage running away as part of the symptom of the problem with your amp too?

[Leo_Gnardo]

Just saw a video on youtube where it was a Mesa DC-3 with runaway bias. In the case of the DC-3 amp the B+ voltage was also dropping down too as the bias voltage started dropping. Was your bias voltage running away as part of the symptom of the problem with your amp too?

I saw that one too Gonz, the Guitologist right? Came to a bad end - he was just a step away from achieving a solution after excruciating hours of leading us - and himself - on a merry chase. Then he threw in the towel, bought the amp, and disassembled it. To me that's like tipping the chess board over, not exactly a professional thing to do. In his case, the bias voltage was failing to one pair of a quad of EL84's. Yes more than likely a conductive PC board, and he could have "air boarded" the bias feed & fixed it for a successful wrap up.

Meanwhile it's good to know steveward's observation. A problem there would affect overall bias voltage, not just one side. Good report, good solution, and welcome aboard steveward!

[DrGonz78]

Yup it was Brad the guitologist. The one pair was red plating. But all the bias was drifting as well as the B+ voltage. Probably the low voltage taps too since there derive from the same secondary source.

[Jazz P Bass]

Hah!
I just had a Mesa MKIII come through.

The one tube had 208 Vdc on the grid!

Yup, bad coupling cap.

[alexradium]

Hah!
I just had a Mesa MKIII come through.

The one tube had 208 Vdc on the grid!

Yup, bad coupling cap.

from boutique to garbage...
bad design,bad boards and components....

[Leo_Gnardo]

Yup it was Brad the guitologist. The one pair was red plating. But all the bias was drifting as well as the B+ voltage. Probably the low voltage taps too since there derive from the same secondary source.

In that case, I'm sure the high voltages were drifting down while the failing bias on one tube pair was sending them into red plate territory, pulling way more current than expected. Nothing unusual about that. I was disappointed this so-called "expert" flailed around aimlessly while broadcasting to the world his faulty troubleshooting technique on the internet. An exercise in "what NOT to do." I s'pose, there's a lesson to be learned there too. But his ego would not allow that: he defended himself with hostile attitude in the comments column. Learn from your mistakes? "Hell no, not ME!!! And @#$! your mama besides." How annoying.

Jazz P Bass, and other repair folks: I've mentioned at least once on these pages finding bad "orange drop" caps in Mesa amps, now Jazz found another. It's getting to be a trend in 20+ year old Mesas. We can no longer assume OD's are "bulletproof" forever. No harm in suspecting you have a leaker, then dig right in and find it. The tough part is replacing the bad cap(s) without removing the whole dam' circus board. Sometimes that's no big deal, other times not so easy.

[drewl]

Had a bad OD a couple months ago on a Dual Rec.