Mark V Design Defect - Channel Switching Noise

[L5man]

I have a Mark V 40th Anniversary wide-body Combo. I also have a Mark IV, and a Mark IIC+, all bought new with EVM-12L speakers.
They are all great amps. I have primarily used the Mark IIC+, and the Mark IV, but more recently the Mark V. It is a wonderful
amplifier, and very flexible. I can get many great sounds, and when using it in a Wet / Dry setup it is wonderful with a Black Face
twin as the other amp.

However...this amp has a serious design flaw. I have researched the channel switching noise issue in great detail. I also have significant education
in electronics and >30 years chip design experience. There is something wrong in this design with the switching, and I believe that
Mesa Engineering needs to come clean with the error and a plausible explanation. It's not acceptable for anyone to just say -
Just live with it - That's the way it is. This is a clear design error...

I took my amp into a great repair center. We suspected the JFETs in the Mute circuitry. The outcome was that the mute control circuit is functioning properly as are all the mute FETs. An external DC supply was hooked up to the mute buss so we could positively measure the JFETs without removing any and voiding the warranty - it was hard to tell for sure if they were turning on all the way with the brief mute pulse.

After trying to slow down that pulse a bit, one can actually ground the input to the power amp (where the last mute FET is), so the amp is effectively in permanent mute mode. Even this does not kill the loud pops experienced, meaning the noise is leaking in some other way, likely due to a PCB layout/power supply grounding flaw.

The switching is quite loud at first yet it quiets down considerably once you switch the channels 5-10 times, to the point at which it’s probably not noticeable when playing. Mesa Engineering Folks - We shouldn't have to do this for this amplifier. This is a serious design flaw - but it appears that the amp is 100% functional. If you had this in your PC do to a Microprocessor or motherboard chip design issue, and you had to reboot 5 times to get to a start-up screen, you would not deem that acceptable.

Mesa Boogie - Your explanation??? Lots of Mark V owners need to know the truth...I'm sure that your Engineers do...

 

[mace]

Hello. Could the channel switching noise be a manufacturing issue? My late 2011 Mark V does not have excessive pops, or pops that I notice.

And therein lies part of the issue…what is ‘excessive’ or ‘loud’? It is subjective. Do you have an audio recording of the pops and how they get softer after repeated channel switching? I could do the same and that would be a start for direct comparison. Maybe what is too loud for some is OK for others?

 

[Newysurfer]

I have a Mark V 40th Anniversary wide-body Combo. I also have a Mark IV, and a Mark IIC+, all bought new with EVM-12L speakers.
They are all great amps. I have primarily used the Mark IIC+, and the Mark IV, but more recently the Mark V. It is a wonderful
amplifier, and very flexible. I can get many great sounds, and when using it in a Wet / Dry setup it is wonderful with a Black Face
twin as the other amp.

However...this amp has a serious design flaw. I have researched the channel switching noise issue in great detail. I also have significant education
in electronics and >30 years chip design experience. There is something wrong in this design with the switching, and I believe that
Mesa Engineering needs to come clean with the error and a plausible explanation. It's not acceptable for anyone to just say -
Just live with it - That's the way it is. This is a clear design error...

I took my amp into a great repair center. We suspected the JFETs in the Mute circuitry. The outcome was that the mute control circuit is functioning properly as are all the mute FETs. An external DC supply was hooked up to the mute buss so we could positively measure the JFETs without removing any and voiding the warranty - it was hard to tell for sure if they were turning on all the way with the brief mute pulse.

After trying to slow down that pulse a bit, one can actually ground the input to the power amp (where the last mute FET is), so the amp is effectively in permanent mute mode. Even this does not kill the loud pops experienced, meaning the noise is leaking in some other way, likely due to a PCB layout/power supply grounding flaw.

The switching is quite loud at first yet it quiets down considerably once you switch the channels 5-10 times, to the point at which it’s probably not noticeable when playing. Mesa Engineering Folks - We shouldn't have to do this for this amplifier. This is a serious design flaw - but it appears that the amp is 100% functional. If you had this in your PC do to a Microprocessor or motherboard chip design issue, and you had to reboot 5 times to get to a start-up screen, you would not deem that acceptable.

Mesa Boogie - Your explanation??? Lots of Mark V owners need to know the truth...I'm sure that your Engineers do...

Right... great research you've done except.......my Mark V is silent when switching channels.

I've also played another five 25, 35 and 90W models and none of those have any noise either.
On TPG forum, and many Mark V threads there, I've not heard anyone else complaining about noise either

 

[KRichter]

I have a Mark V 40th Anniversary wide-body Combo. I also have a Mark IV, and a Mark IIC+, all bought new with EVM-12L speakers.
They are all great amps. I have primarily used the Mark IIC+, and the Mark IV, but more recently the Mark V. It is a wonderful
amplifier, and very flexible. I can get many great sounds, and when using it in a Wet / Dry setup it is wonderful with a Black Face
twin as the other amp.

However...this amp has a serious design flaw. I have researched the channel switching noise issue in great detail. I also have significant education
in electronics and >30 years chip design experience. There is something wrong in this design with the switching, and I believe that
Mesa Engineering needs to come clean with the error and a plausible explanation. It's not acceptable for anyone to just say -
Just live with it - That's the way it is. This is a clear design error...

I took my amp into a great repair center. We suspected the JFETs in the Mute circuitry. The outcome was that the mute control circuit is functioning properly as are all the mute FETs. An external DC supply was hooked up to the mute buss so we could positively measure the JFETs without removing any and voiding the warranty - it was hard to tell for sure if they were turning on all the way with the brief mute pulse.

After trying to slow down that pulse a bit, one can actually ground the input to the power amp (where the last mute FET is), so the amp is effectively in permanent mute mode. Even this does not kill the loud pops experienced, meaning the noise is leaking in some other way, likely due to a PCB layout/power supply grounding flaw.

The switching is quite loud at first yet it quiets down considerably once you switch the channels 5-10 times, to the point at which it’s probably not noticeable when playing. Mesa Engineering Folks - We shouldn't have to do this for this amplifier. This is a serious design flaw - but it appears that the amp is 100% functional. If you had this in your PC do to a Microprocessor or motherboard chip design issue, and you had to reboot 5 times to get to a start-up screen, you would not deem that acceptable.

Mesa Boogie - Your explanation??? Lots of Mark V owners need to know the truth...I'm sure that your Engineers do...

I'm not expert by any means, I know the JFETs in the Mute circuitry, had a bulletin in 2009 from Mesa, but it seemed that that was basically for general switching, and the symptoms did not occur when the pedal was attached. At the time, that was for Serial #'s prior to 1224. I believe, and don't quote me, but the relay's were being half on. I don't think that's the problem in your case. Initially The pedal will pop (softly), when you first power up and cycling thru 1-3, one time will discharge static, after that it will be quiet. I am actually sending my chassis to Mesa this morning for some deep unrelated issues.

 

[steve Rhenius01]

I have a Mark V 40th Anniversary wide-body Combo. I also have a Mark IV, and a Mark IIC+, all bought new with EVM-12L speakers.
They are all great amps. I have primarily used the Mark IIC+, and the Mark IV, but more recently the Mark V. It is a wonderful
amplifier, and very flexible. I can get many great sounds, and when using it in a Wet / Dry setup it is wonderful with a Black Face
twin as the other amp.

However...this amp has a serious design flaw. I have researched the channel switching noise issue in great detail. I also have significant education
in electronics and >30 years chip design experience. There is something wrong in this design with the switching, and I believe that
Mesa Engineering needs to come clean with the error and a plausible explanation. It's not acceptable for anyone to just say -
Just live with it - That's the way it is. This is a clear design error...

I took my amp into a great repair center. We suspected the JFETs in the Mute circuitry. The outcome was that the mute control circuit is functioning properly as are all the mute FETs. An external DC supply was hooked up to the mute buss so we could positively measure the JFETs without removing any and voiding the warranty - it was hard to tell for sure if they were turning on all the way with the brief mute pulse.

After trying to slow down that pulse a bit, one can actually ground the input to the power amp (where the last mute FET is), so the amp is effectively in permanent mute mode. Even this does not kill the loud pops experienced, meaning the noise is leaking in some other way, likely due to a PCB layout/power supply grounding flaw.

The switching is quite loud at first yet it quiets down considerably once you switch the channels 5-10 times, to the point at which it’s probably not noticeable when playing. Mesa Engineering Folks - We shouldn't have to do this for this amplifier. This is a serious design flaw - but it appears that the amp is 100% functional. If you had this in your PC do to a Microprocessor or motherboard chip design issue, and you had to reboot 5 times to get to a start-up screen, you would not deem that acceptable.

Mesa Boogie - Your explanation??? Lots of Mark V owners need to know the truth...I'm sure that your Engineers do...

I seem to have a similar problem - amp left unused for the duration of the pandemic. When switched on after that all the diodes in the 12V heater supply failed. I replaced these but the amp signal does not get as far as V7 - could this be failed J175's ?

 

[bandit2013]

I seem to have a similar problem - amp left unused for the duration of the pandemic. When switched on after that all the diodes in the 12V heater supply failed. I replaced these but the amp signal does not get as far as V7 - could this be failed J175's ?

It could be J175M1. that one is located on the input of the phase inverter.
If you can get signal out of the send jack with the FX loop turned on, that would be J175M1. The other JFET in question would be J175M4 which is in front of the FX loop just before the GEQ circuit.

Since you stated the bridge rectifiers blew out on the 12V supply, I would also check the voltage regulator for function, LM2940. If not mistaken, I believe the J175FETS are on when the gate has zero volts in respect to the source pin (Vgs), and off at 12V. J175 only needs a 6V Vgs to get it into pinch off. If the gate got damaged due to transient, it will be stuck in conduction mode.

 

[steve Rhenius01]

It could be J175M1. that one is located on the input of the phase inverter.
If you can get signal out of the send jack with the FX loop turned on, that would be J175M1. The other JFET in question would be J175M4 which is in front of the FX loop just before the GEQ circuit.

Since you stated the bridge rectifiers blew out on the 12V supply, I would also check the voltage regulator for function, LM2940. If not mistaken, I believe the J175FETS are on when the gate has zero volts in respect to the source pin (Vgs), and off at 12V. J175 only needs a 6V Vgs to get it into pinch off. If the gate got damaged due to transient, it will be stuck in conduction mode.

I did check and replace the Voltage regulator, which brought the rectified heater and 12V supply back but signal seems to disappear between V3 and V7 (nothing at the grids of V7). I guess it may be failure of J175 but which one - how to check ?

 

[bandit2013]

do you have any signal on V6A? that is the last gain stage of CH3 but does pass through V3A first. There are no J175 FETS that shunt the control grid before V3A.

J175M4 is a mute FET just in front of the GEQ input. And J175M1 is a mute FET on the input circuit of the phase inverter.

Is the cooling fan running? That is also powered by the 12V regulator.

 

[woodbutcher65]

I have a Mark V 90 and its channel switching is essentially noiseless.. Oh, there's some click, but it's low in level and in particular it would not be noticeable if you change channels while playing. However it's fair to point out that I neither gig with the amp nor record with it. What's acceptable to me may not be acceptable in a live or recording situation.,

 

[steve Rhenius01]

do you have any signal on V6A? that is the last gain stage of CH3 but does pass through V3A first. There are no J175 FETS that shunt the control grid before V3A.

J175M4 is a mute FET just in front of the GEQ input. And J175M1 is a mute FET on the input circuit of the phase inverter.

Is the cooling fan running? That is also powered by the 12V regulator.

Apologies for the delayed reply - hospital duties for family making life errratic. The cooling fan is running so voltage regulator must be ok now. If I inject a signal at the input I can find it on pin 2 of V5. Not sure which of the J175's to check or exactly how ! Possibly a failure there is muting ongoing signal ?

 

[steve Rhenius01]

do you have any signal on V6A? that is the last gain stage of CH3 but does pass through V3A first. There are no J175 FETS that shunt the control grid before V3A.

J175M4 is a mute FET just in front of the GEQ input. And J175M1 is a mute FET on the input circuit of the phase inverter.

Is the cooling fan running? That is also powered by the 12V regulator.

Can't see anything on grid of V6A - fan is clearly running now.

 

[bandit2013]

You can check them without any power applied to the amp. Since I am not well versed on JFETS, I mostly use MOSFETS in product designs I develop. Been long time since I considered testing for failure. Here is a basic way to test the JFET P-Channel with a multi-meter. The ohm measurement may provide enough voltage to bias the PN junction between the GS or GD. You should have what appears like a low resistance between D to S. If your DVMM has a diode test, that may also work. Note the DVMM does have polarity on the probes. Try using it both ways in resistance measurements testing. Diode test should be red as + and black as -. I thought this may be helpful. No need to remove the JFET off the board.



For testing CH3. Signal path is as follows

V1A -> tone stack -> V1B -> V5A -> V4B -> V3A -> V6A -> CH3 master note that signal is passed through relay RYC3E to connect CH3, CH1/CH2 (location where J175 M4 mutes) This is in front of the GEQ circuit.

You can find an older schematic for the V, it is a 2009 date code but is very much on par with current design.

Another way to find out if J175 M1 (the muting JFET) that follows the FX loop: Turn on the amp, channel does not matter here, make sure the FX loop is not in bypass mode. Enable the FX loop. Plug your guitar into the return. If you cannot hear anything from the guitar signal, M1 is stuck in conduction mode or the amp has the mute feature turned on. You can adjust the volume by using the global output or solo boost. It will not be very loud but if the M1 is not damaged, you will hear signal at the output. You can do this without the footswitch connected, just set the FX channel on the back panel.

 

[steve Rhenius01]

I did check and replace the Voltage regulator, which brought the rectified heater and 12V supply back but signal seems to disappear between V3 and V7 (nothing at the grids of V7). I guess it may be failure of J175 but which one - how to check ?

Will do as suggested and report back. Thanks, Steve

 

[L5man]

I did send my Mark V 40th back to Petaluma. They said that they did a thorough check out of the amp, but nothing was wrong with it. The channel switching noise is inherent in the design due to the excessive voltages being switched. Do the lower power Mark V versions or the Mark VII exhibit this phenomenon?

 

[steve Rhenius01]

Thanks, plenty to explore here will proceed tomorrow and report results !

 

[bandit2013]

I did send my Mark V 40th back to Petaluma. They said that they did a thorough check out of the amp, but nothing was wrong with it. The channel switching noise is inherent in the design due to the excessive voltages being switched. Do the lower power Mark V versions or the Mark VII exhibit this phenomenon?

I do not have the Mark VII yet. Planning on getting one soon. JP2C and Triple Crown do not have any pops associated with channel switching that I have noticed. This is only apparent when using the footswitch. Running those amps without the footswitch does create pops as the mini toggle switch for channel selection does not directly trigger the strobe mute as the Midi controller will when using the footswitch as that too has a midi chip set inside it. I would assume the Mark VII would have similar characteristics as it too is a midi controlled switcher.

 

[steve Rhenius01]

It could be J175M1. that one is located on the input of the phase inverter.
If you can get signal out of the send jack with the FX loop turned on, that would be J175M1. The other JFET in question would be J175M4 which is in front of the FX loop just before the GEQ circuit.

Since you stated the bridge rectifiers blew out on the 12V supply, I would also check the voltage regulator for function, LM2940. If not mistaken, I believe the J175FETS are on when the gate has zero volts in respect to the source pin (Vgs), and off at 12V. J175 only needs a 6V Vgs to get it into pinch off. If the gate got damaged due to transient, it will be stuck in conduction mode.

Thanks - it doesn't seem to be M1 or M4 FETs - both test ok. No signal appears at V7 and no sound at all from speaker. Not sure where else to look ?

 

[bandit2013]

I will take a trip to work today while on vacation to get some gear I need.
I will leave the rest in the private message.

Wow, still no sound by December. Sorry if I left you hanging. How did you test the JFETS? They will be different than what you may think unless you are familiar with the P-CH FET. I will see what I can dig up today. Will take pictures too. Sorry for the blurry images.

 

[bandit2013]

I do not have the Mark VII yet. Planning on getting one soon. JP2C and Triple Crown do not have any pops associated with channel switching that I have noticed. This is only apparent when using the footswitch. Running those amps without the footswitch does create pops as the mini toggle switch for channel selection does not directly trigger the strobe mute as the Midi controller will when using the footswitch as that too has a midi chip set inside it. I would assume the Mark VII would have similar characteristics as it too is a midi controlled switcher.

As for the Mark VII, I got one. No pops with that one either. Actually it is a decent amp and I really like it. Very on par with the Badlander and the JP2C.

I was reading back through the posts, you lost the voltage regulator and such. That may have crapped out a few parts along with it. I plan on finding the JFETS and trace it back to the strobe mute circuit. Not sure which diode in the circuit connects to what set of JFETS. I know where the one for the reverb is but never tried to find the one's for the phase inverter or front of the GEQ. Not having the full schematic is a pain, not even sure where the +5V runs too. I assume it is associated with the chip used to manage the channel change. At the moment I think I broke my back, I can barely walk at the moment. I will dig into the amp a bit but for now I need some pain meds and a bit of rest.

 

[bandit2013]

May as well share this with others. If I am wrong on this, please respond with some corrections or comments. I do not want to misguide anyone having the similar issue.

I was able to get the Mark V90 chassis out. I did not realize I already mapped out the strobe mute circuit as the ribbon cable was marked with a black stripe. Here is a different view. At the top of the image below is the strobe mute marked with black and highlighted in yellow. It is the 5th wire from the right side of the ribbon cable. (4th wire from the left). This wire connects to the collector of the 2N6426 NPN Darlington Transistor.

Here is a closer look at the strobe mute circuit. The diodes in this image may be part of the trigger circuit connected to the base. None of those components are connected to the collector. the resistors above the transistor are part of the trigger circuit tied to the base. The tantalum cap or yellow bullet cap may be one of the timing caps. I did not probe it to see if it was connected to the collector.

I traced the strobe signal line from the 5th post on the pin header to a resistor close to the V7 mute circuit. That was a bit difficult as the J175 was pushed down so close to the PCB it was hard to get the test leads in there. The gate has a 330k resistor as marked. The strobe connects to the other side. The black blob next to the plastic board support between V6 and V7 is the J175M1 transistor. You probably found this part already.

The other mute for the signal path that is in front of the GEQ is actually tucked right next to the row of inductors. It is easy to find with the array of Zener diodes. J175M4 is sort of hidden by some wires. I could not find the gate feed circuit. I only focused on the V7 mute (phase inverter mute). Note that the front of the amp will be at the top of the image.

I also found a diode where the cathode was connected to the strobe mute bus. I could not find where the anode was connected too.

The strobe mute should at least have +12V dc on it most of the time. As for the actual voltage on the gates of each of the J175 FETs I did not measure. The only time the two JFETS will be shorting out the signal is when a change occurs on the channel. the signal will get pulled low and remain low for a timing period based on an RC time constant. I could not locate the pull up resistor on the collector of the strobe transistor. Just for S&G's if you want to find the chip used to decipher the voltage from the footswitch, it is located under the ribbon cable that connects to the two rotary switches on the back panel. it is on the same board as the Strobe mute circuit. Good luck if you needed to remove it. This layout is a mess, Almost like the RA100 as things are not where you would expect them to be located.

So if you are looking at voltages while the amp is powered up, measure the G side of the 330k resistor above in reference to ground. You should have a voltage above the pinch off voltage of 3V to 6Vdc. It may be higher than that as the resistor is probably just a current limit. On the strobe mute side it may be closer to the +12V. Sorry if I did not take this any farther. Not doing so well at the moment.