Caps/repair for an old Mitchell Mesa Mk 1 clone(ish) amp

[eurekaiv]

One of my neighbors brought me an old Mitchell Mesa clone that wasn't working. I'm going to do my best to get it fixed for him since I have time away from my job while I'm stuck in the house. I already found two of the original Sylvania 6L6-STRs were totally dead and one is pretty noisy so a new set is incoming. I don't have enough spare 6L6s I'm willing to subject to the near 500v this thing puts on the plates so further testing will wait until they arrive. In the meantime, ordering up some filter caps. The schematic for a Mk1 Boogie shows 3 60uf and 3 30uf filter caps. This amp has 6 60uf caps. So 3 60s in place of the 3 30s shown in the schem. I'm inclined to leave it stock at 6 60s but a 60uf/500V cap is not an easy find. My closest (in-stock and available) options appear to be...

JJ 47uF/500V ~$7 (20% less than 60)
JJ 80uF/500V ~$7 (25% more than 60)

Any reasons for or against going to the Mesa spec of 3/30s and 3 other? My second thought was go 3 80s and 3 30s. I've haven't drawn it up yet (the 60uf caps are covering almost everything on the cap board) but the power supply does look pretty much identical to the MK1 schematic available online.

Also, the pre-amp uses 47uf caps on the cathodes. That seems high to me. I guess it will add more gain? All the mesa schems show 25uf caps there like an old Fender. Would you leave them or go to 25?

[shooter]

looks to be a SS rectifier, so whatever you want to add should be fine.
the modern small's could be "stacked" so you have the vdc covered, but if you go that route you'll need a pair of "balancing resistors" like in BIG Marshall PS's

[Latole]

Your eyes are good to see the value of these filter caps.

I'll go to 30 mfd if power supply is like MKI. 80mfd are too big IMO

Lots of choice here; it is my main store:
https://www.tubesandmore.com/products/capacitors

MK1 reissue schematic; 
https://el34world.com/charts/Schematics/files/Mesa_boogie/Boogie_mki_reissue.pdf

Older (?) MKI schematic
https://el34world.com/charts/Schematics/files/Mesa_boogie/Boogie_marki.pdf

[eurekaiv]

Your eyes are good to see the value of these filter caps.

I'll go to 30 mfd if power supply is like MKI. 80mfd are too big IMO

I measured them (and checked ESR) with an LCR meter and they were all about the same—somewhere between 60 and 70uf.

Do you mean 30s for all 6?

What about all 47s? Plentiful and closest available to 60uf.

I was going to order from Mojo or AES. Mojo is a little cheaper but I'll get the AES order sooner.

[Latole]

Did you remove one side to measure ?

I agree with Shooter, many value will work.
With same values as originals caps amp will sound same as before at loud volume. With other values will you hear the difference ?

[2deaf]

Mine had 60uF/350V for the two in parallel.  100uF/350V is a really standard value in that position.

The three in parallel were 33uF/500V.  33uF/500V or 47uF/500V is fine for those.

Mine didn't have that loose one against the chassis and there is a fair change that yours didn't either originally.

Lots of stuff has been changed in your amp.  Here is what was in mine stock.

[2deaf]

looks to be a SS rectifier, so whatever you want to add should be fine.
the modern small's could be "stacked" so you have the vdc covered, but if you go that route you'll need a pair of "balancing resistors" like in BIG Marshall PS's

Mine already had 150K balancing resistors you can't see between the two 60uF/350V capacitors.

[eurekaiv]

Did you remove one side to measure ?

I didn't lift a leg on the filters since I'd have to cut them out to do so. The leads are un-clipped and wrapped onto the traces on the underside of the board so I visually verified values on all 6 being 60/500 after removing the LIBERAL application of hot melt on them. In fact, this amp is covered in hot melt.

Mine didn't have that loose one against the chassis and there is a fair change that yours didn't either originally.

Thanks for sharing your schem! That will be helpful I hope.

99% sure this amp has never been touched. The soldering looks legit/original to my eyes other then the couple parts I lifted to test. I do a lot of rework for my day job so I can usually tell. Also, the current owner bought it new and it's been broken for ages and sitting in storage. There's more to the story but he said it was never worked on before. I won't get into the rest without his blessing. The loose cap against the chassis is present in the MK1 schematic I found. Attached a pic... 

[shooter]

The "W" white wire looks like somebody played there

[eurekaiv]

Yeah, that bit at the bottom on that first 60uf looks touched but my feeling was that since the part matched, the wiring checked out and the rosin buildup was identical in color, it was probably original to the build.

[2deaf]

There is silicone of different ages.  The last filter cap has newer silicone on it, probably because somebody needed to move it in order to get at the screw that's under it.

Can you post some date codes?

I got my hands on an early Mesa in the 70's and it didn't have that extra capacitor, either.  It also didn't have that capacitor at the edge of the preamp board that would have been on that Mark 1 schematic you linked had the guy done a better job of making a copy.

[2deaf]

It turns out I was mistaken--my Mitchell did have that capacitor glued to the chassis and it was the Mesa that didn't have it.  I guess I should have checked my notes before I posted.

[Latole]

Mine had 60uF/350V for the two in parallel.  100uF/350V is a really standard value in that position.

The three in parallel were 33uF/500V.  33uF/500V or 47uF/500V is fine for those.

Mine didn't have that loose one against the chassis and there is a fair change that yours didn't either originally.

Lots of stuff has been changed in your amp.  Here is what was in mine stock.

I see preamp not power supply

[Latole]

Did you remove one side to measure ?

I didn't lift a leg on the filters since I'd have to cut them out to do so. The leads are un-clipped and wrapped onto the traces on the underside of the board so I visually verified values on all 6 being 60/500 after removing the LIBERAL application of hot melt on them. In fact, this amp is covered in hot melt.

 

Measuring old and leaky filter cap could be false, you have to look what is write on.

If you want to put new filter caps, you have to remove olds ones anyway .
You can cut both side of the wire half way to make easy solder for new caps.

[Latole]

From book; "The how and Why of Guitar Tube amps as"Gar" sees it " page 18
Gar Gillies is the builder of Canadian made Garnet amps

A great book everybody who like to fix amps may have.
_________

"Vintage amps generally had smaller filter capacitor sizes than the amps produce today.
More recent products have higher wattage output and place larger current demand on filter caps.
Newer amps often use 100 mfd or larger.
Usually these amp use silicon diodes rectifier and combined with larger power supply filter capacitors, make for a very stable, well regulated P.S.

This desing contributes to tight bass response and a sterile audio-like sound

If it is what you like, great but many players prefer more vintage sound created to some extent by filter capacitor with much lower capacity . "

[eurekaiv]

Knowing the owner I think both he and I would prefer what’s closest to schematic. Parts are ordered, will update with some progress shots when they come in.

[Latole]

Knowing the owner I think both he and I would prefer what’s closest to schematic. Parts are ordered, will update with some progress shots when they come in.

If the owner like the sound, original values are the good choice.   

[2deaf]

I see preamp not power supply

[Latole]

Thank you.

[eurekaiv]

Got my parts today and started in getting the stuff I knew needed replaced swapped out. I had a couple drifted carbon comps on the grids so I replaced them all. Also a couple 2.2M resistors were high as well.

In doing all this, and checking over the map vs. the schematic I noticed some parts are different versus the schematics 2deaf kindly shared. They seemed original to me.

The one I have the most interest in is the 220k resistor to ground in between the 3rd and 4th (switched) gain stage. In my amp, it's a 2.2M—quite a difference. I'm curious if this is a mistake in the schem, a mistake in the build or possibly an on purpose change.

In the power section, the bias is fed via a (1N400X) diode and 470Ω resistor in series. These parts are swapped in the amp.

I have all 12AX7s instead of 3 12AT7s. I assume they're original to the amp as all the tubes are Mitchell branded Sylvanias. Obviously this will work but should I at least replace the 12AX7 in the reverb stage?

I have 100k resistors to the plates of V1A and V1B instead of 68k.

Attached my working schem with notes for reference.

[2deaf]

Got my parts today and started in getting the stuff I knew needed replaced swapped out. I had a couple drifted carbon comps on the grids so I replaced them all. Also a couple 2.2M resistors were high as well.

In doing all this, and checking over the map vs. the schematic I noticed some parts are different versus the schematics 2deaf kindly shared. They seemed original to me.

The one I have the most interest in is the 220k resistor to ground in between the 3rd and 4th (switched) gain stage. In my amp, it's a 2.2M—quite a difference. I'm curious if this is a mistake in the schem, a mistake in the build or possibly an on purpose change.

In the power section, the bias is fed via a (1N400X) diode and 470Ω resistor in series. These parts are swapped in the amp.

I have all 12AX7s instead of 3 12AT7s. I assume they're original to the amp as all the tubes are Mitchell branded Sylvanias. Obviously this will work but should I at least replace the 12AX7 in the reverb stage?

I have 100k resistors to the plates of V1A and V1B instead of 68k.

Attached my working schem with notes for reference.

The 220K to ground between the 3rd and 4th stages is not a mistake on the schematic.  I reverse engineered this amp in the early '90s when I was at the top of my game.  They were copying an early Mesa that had the 2.2M/220K combo.  I will attach a schematic from a Mesa that I reverse engineered in the late '70s.  Randy just took the same old Fender AB763 type thing and put a gain stage in front of it.  That put a 3.3M/150K or so combo between the 3rd and 4th stages.  He upped it a little for more gain and Mitchell copied that.  Oh yeah, I also have the circuit board in my hand right now and it has the original 2.2M/220K combo on it.

[2deaf]

It looks like your OT only has a single secondary and the impedance selector hole has a plug in it.  It also looks as though you have no choke.

The tubes that were in mine weren't necessarily the originals, although the 6L6's were Mitchell branded.  The reverb transformer in mine was more suited to a 12AT7 driving the 150 Ohm tank that was in it.  Yours may not be the same.

[eurekaiv]

Yep, no choke. Forgot to mark that down. The speaker in this thing is a HEAVY 8ohm EV and there is one speaker jack wired in what I think is parallel.

Thanks for the clarification on the 2.2m. Do you think it was a mistake at the factory? Or might there have been some reason to increase this value so much.

[2deaf]

Yep, no choke. Forgot to mark that down. The speaker in this thing is a HEAVY 8ohm EV and there is one speaker jack wired in what I think is parallel.

Thanks for the clarification on the 2.2m. Do you think it was a mistake at the factory? Or might there have been some reason to increase this value so much.

Mine had an Altec 917-8A in it that had been re-coned.  It wasn't jumping up and blowing away in the breeze, either.

One speaker jack is a shorting type in parallel with an open type.  The shorting jack is like the "Speaker" jack on a Fender and the open jack is like the "Ext Speaker" on a Fender.  You might want to note this on the back panel somehow so that folks will know which jack to use once the chassis is back in the cabinet.

I don't know about the 2.2M/2.2M divider, but it seems odd that they would copy Mesa so closely and then make such a radical change.

Would you be up for determining the OT turns ratio?  Does the OT have any numbers on it?

[eurekaiv]

Would you be up for determining the OT turns ratio?  Does the OT have any numbers on it?

I'll check on the numbers when I get back on it tomorrow.

I've never tested this on a transformer but I think I would need to apply AC to the primary and measure the AC at the secondary to come up with this number. If so, will say 6V from a filament winding be sufficient? Unfortunately don't have a variac. It was on my swap meet list of things to look for until this lockdown took hold.

[shooter]

say 6V

PRR gave an example in another thread, believe it was 6vac 30:1 (ballpark) = 200mV

[2deaf]

I've never tested this on a transformer but I think I would need to apply AC to the primary and measure the AC at the secondary to come up with this number. If so, will say 6V from a filament winding be sufficient?

1.  Remove the OT red wire from the 120 Ohm resistor and tape the red wire off.

2.  Insert a plug into the shorting speaker jack.  This plug needs to be open circuit, so a guitar cable with nothing hooked to the other end is fine.

3.  Remove the power tubes.

4.  There is a black wire and a blue wire connected to the middle of the power tube board.  These wires have been spliced to the OT wires.  Connect one side of the heater circuit to the black wire and the other side of the heater circuit to the blue wire.  The heater circuit is easily accessed right there by the black and blue wires.

5.  Switch the Standby switch to non-op and turn the power on. 

6.  Measure the heater voltage.

7.  Measure the voltage at the speaker.  We're expecting something like 280mV or 400mV.

8.  Post the heater voltage and the speaker voltage.

[eurekaiv]

I must not be doing something right... I have 0V (AC measured across the speaker lead. Heater voltage is 6.28

[shooter]

not positive but try shorting the leads together, then hit the triangle button n see if they go to 0.

is your meter range set to lowest?

a hot pickup should get you ~>100mV.

[eurekaiv]

With just the red lead connected I get about .245V, or 245mV. Amp on or off measured the same.

The amp is working perfectly fine now FWIW. It sounds ok I guess, I can’t really crank it up today but its not really the kind of amp I would use much and I don’t think this EV speaker is too hot.

I also need to check the bias still, make sure its at least in an acceptable range. Do people mod these to adjustable bias? The fixed bias seems like it could be a problem with the wrong set of tubes.

[2deaf]

With just the red lead connected I get about .245V, or 245mV. Amp on or off measured the same.

I don't understand this.  Just the red lead of your meter without the black lead connected to anything?  The red center tap wire of the OT?  How can you get any AC voltage if the amp is off?

[shooter]

+1

I would leave the amp til you can prove your meter is capable of measuring <500mV

pickups, multi_use_phone, AUX out WITH volume control........and music:) should be poor mans signal sources. 

[eurekaiv]

Confirmed everything was properly connected by buzzing it out and... one of my alligator clips was not getting to a pin. Had to push away some more of the hot mess—I mean hot glue—to get it on there good. Anyway, 400mv is what I’m measuring. Confirmed with the BK and a Fluke 117 meter.

[2deaf]

I also need to check the bias still, make sure its at least in an acceptable range. Do people mod these to adjustable bias? The fixed bias seems like it could be a problem with the wrong set of tubes.

This one shows a bias circuit I put in when I re-did my Pro-100. 

[eurekaiv]

Thanks for the bias schematic and again for all your help. It's sitting just under 41ma after it settles in a bit. That seems barely to one side of hot so I think I'm gonna call this one a day. I should probably stop fiddling with it since it's not mine and it's sounding pretty good right now. Pic attached after the re-cap and a few replaced resistors.