Twin Reverb loss of volume and B+ voltage after faceplate replacement ?!?

[kladmin]

Hello all - I'm hoping you can help me with an issue I'm having with a non-master volume silverface Twin Reverb (not reissue).

I replaced the faceplate on it -- removed the chassis from the cab, pulled 1 power tube and 1 preamp tube to get to the cabinet bolts.  I removed all the nuts screws etc to get the new faceplate on.  The pilot light assembly of course is a pain to get off and maybe I bumped something.  I put the new faceplate on, and replaced all the nuts and screws, and put it back in the cabinet, fired it up.

**The sound was terrible!  Low volume, no sustain, kinda distort-y.

So, I pulled the chassis back out and began troubleshooting, here is what I have checked so far:

B+ extremely low, well under 400v with standby switch ON (B+ looks fine in standby, 475+)  I forget the actual numbers, if you want specifics I'll get them this evening
same problem in both channels
different speakers
different guitar cables
different tubes (replaced them all with extras)
checked 6l6 power grid and screen resistors with ohmmeter, all test good, less than 5% dift, no open or short
replaced tubes one at a time, interesting results, a single 12ax7 in any position (no other tubes in) dropped B+ by 40 volts when flipping the standby switch, and adding a second dropped voltage 80v
measured heater current at 7v ac without tubes from each winding (not to center tap/ground)
measured bias supply voltage at -54v
measured PI resistors the 100k and 82k test good
removed NFB loop, changed tone but problem still there, replaced it
removed faceplate (no change)
wobbled tubes in sockets, I didn't notice any change
tested all B+ diodes and the bias diode with ohmmeter diode tester, all tested good

I'm starting to wonder if somehow the OT died?  I suppose its easy enough to buy another one and replace it to check, but I can't imagine what I might have done to cause that?

Ideas?


Thanks!!

Kurt

[kladmin]

Ok after reading through the "Servicing Fender Amps" section, I've decided tonight's task will be to retension all the sockets.  They all seem loose, so just because rocking them didn't seem to affect any change it won't hurt to tighten them up.

[sluckey]

Did the amp work properly before you replaced the faceplate? Why did you replace the faceplate?

[eleventeen]

"replaced tubes one at a time, interesting results, a single 12ax7 in any position (no other tubes in) dropped B+ by 40 volts when flipping the standby switch, and adding a second dropped voltage 80v"


This is not really a standard test one does on Fenders, and I would ask you a few more questions as to the conditions--- but if you are saying that you can go from having >>ZERO<< tubes in the amp to ONE LOUSY 12AX7 in the amp and this drops your overall B+ 40 or 70 volts, that cannot be right. Four of those 12AX7s draw 1-2 ma, which is "nothing", the reverb driver pulls a little more but still very little, and the last 12AX7 before the set of 6L6 tubes maybe pulls a little more. But you are talking about drawing (with but a single 12A_7 installed) 1-2-3-4-5 ma in a power supply that is probably good for 175-200 ma, which means the power supply should not even know a single 12AX7 is in there.


My suspicion would be that you have a leaky or bad electrolytic cap in the power supply. Or, a combo of a bad e-cap and a bad "intra-node" dropping resistor----the 1 watt resistors under the doghouse, which are usually 2700, 4700, or 10K, depending upon the exact model of Fender you have. You should inspect the caps under the doghouse and look *very* carefully at the resistors. Those resistors can do two pretty irritating things, and you are talking about a non-MV Fender Twin Rev which is (and I know it is hard to believe) a 35+ year old amp, right? One way they fail is bake up in value. Usually not enough to really trash the amplifier, but when a 10K resistor gets to be 16K, uhhh, that just isn't right. The other thing they can do is to crack in half, invisibly. Then you get extremely goofy situations of partial conduction, resistor changes to 100's of K ohms (and that will not work) or opens up completely. If those are discolored, replace with Doug's metal film 3-watters and you are done. But with a screwdriver, apply some sideways pressure to the body of those resistors...sometimes they are cracked in half are the two halves just separate.


When if ever was a "cap job" ever done to this Twin? 90% chance it needs it.

[kladmin]

Thanks for the replies guys!!!   I'll try to answer the questions so far:

1:  I removed the faceplate a long long time ago (10 yrs?) because it was thrashed.  I used the amp with no faceplate for years, and  ordered a replica but never had a chance to replace it until now :)  Amp sounded/worked great before the replacement.

2:  Agree that the 40v drop with one 12ax7 indicates something seriously wrong.  The resistors under the doghouse read 10.49 and 2.31 on the 20k scale on my multimeter.  I probed them with chopsticks to see if they would separate but they seem ok.  Caps were replaced at the time I removed the faceplate, which was roughly 10 years ago.  The caps in it were not original so this would be at least the second time the caps had been replaced.  I also replaced the bias supply cap, grid/screen resistors and the cathode bypass caps at that time....

I'm going to try to add some pics

This amp has clearly been through the ringer and has had lots of work done to it over the years, but that is part of its charm ;)

[kladmin]

Under the doghouse

[kladmin]

Overview of chassis, notable changes (not mine):

1:  New (small to my eyes?) wires to the plates and screens, with terminal strips.  (to allow easier servicing?)
2:  Extra speaker jack, banana type
3:  Extra wires ?!  going from added 1/4 jacks (one with "VIB" in sharpie) on the back panel to the spots marked X and Y on the ab763 schematic - any idea what this mod is/does?
4:  Hum balance pot simply removed from the circuit, no pots at all in the bias supply.

[sluckey]

Amp sounded/worked great before the replacement.

In that case I suggest you look closely at what you did during the faceplate replacement. You didn't have to unsolder anything from the front panel, right? Did you lose a screw, washer, or nut during the job? Try to remember if you touched anything else inside the amp. In cases like this, the last things you touched are usually related to the failure.

[kladmin]

Sluckey--

I agree with you 100%, I seriously have been racking my brain as to what I could have touched, here is what I did:

1:  I removed a single tube at each end of the chassis to get at the chassis bolts (1 6l6 and 1 12ax7).  Perhaps while reinserting these later something happened?
2:  Held the chassis by the transformers while removing it from the cabinet
3:  Unscrewed all the input jack nuts and pot nuts, and the bright switch screws.  While re-tightening the nuts back on the input jacks and pots it is entirely possible that a jack or pot could have moved enough to pull a wire off etc., but visually I'm not finding anything.  I'll go back through with the meter and see if I can find anything unusual.
4:  The pilot lamp is difficult (know any tricks?) to remove and reinstall, it is possible that I slipped with a wrench etc and banged around on the heater wires, diodes, B+, center taps, or bias supply, which is why I was checking those items so closely.  So far I can't find anything out of whack over there.
5:  Worth mentioning is that when I screwed the bright switch back on for the normal channel it seemed very tight and doesn't slide freely.  I did check its function though and it works, so I don't think it is causing a problem?  On my "needs replacement" list.

I did not unsolder anything.  I agree the failure has to be related, too much coincedence.  Maybe a washer slid under the board?  I'll see if I can shake anything loose....

[kladmin]

Power tube terminal strips

[sluckey]

3:  Extra wires ?!  going from added 1/4 jacks (one with "VIB" in sharpie) on the back panel to the spots marked X and Y on the ab763 schematic - any idea what this mod is/does?

Those two jacks are connected to the output of the NOR and VIB preamps just before they are mixed together and sent to the power amp. Probably simply used for preamp outputs to send to an external power amp or maybe for recording or line out purposes. Or, they could also be used to connect an external preamp into the twins power amp. None of this should have anything to do with your problem.

[kladmin]

Ah, that makes sense sluckey, agree that it has no bearing on current (no pun intended) issue.  Would it be worth soldering in 1 ohm resistors and checking bias on the 6L6s?

[kladmin]

Just remembered, I also tightened the reverb input jack which was loose.  Reverb is working though, I don't think that would cause a problem?

[sluckey]

The pilot lamp is difficult (know any tricks?) to remove and reinstall

http://missionamps.com/hardware.shtml

[eleventeen]

You mention these added jacks to "pick off" the preamp signal(s)


Jacks can, depending upon whether their built-in switching capability is used and how it is used, DEFEAT the preamp signal going forward into the power amp---or simply serve as a "pick off" or "insertion point". When you insert a 1/4" plug into your "preamp out" jack, does the amp go quiet, in other words, does that jack DEFEAT passing the signal to the power amp?


If so, then passage of your preamp signal to the power amp (eg; the "stock" configuration) is dependent upon a nice clean switch connection being made when a plug is NOT inserted into your "preamp out" jack. It's very common for that switch to get a little funky over time and require some VERY GENTLE abrasion to start working well. I would test for this condition by running a steady-state signal (could be an oscillator or a radio signal---ideally NOT you strumming a guitar) through the amp and repeatedly sticking a 1/4" plug into those jacks.


You don't really know what nominal, stock, unmodified voltage levels are on those "X" and "Y" points that are brought out to those external "preamp out" jacks. And I don't, either. As a matter of elimination of possible causes, I would strongly consider disconnecting those external preamp jacks. Shouldn't be any big deal, you have the pic you took above, make a sketch, take some notes. Should be two solder joints >>>UNLESS <<< inserting a plug into those jacks DEFEATS the signal moving forward into the amp's power section. The difference is important. If those jacks DEFEAT signal passage, everything in the signal path is suspect. If they are simply "pickoff" points, then it should not matter.

[sluckey]

You don't really know what nominal, stock, unmodified voltage levels are on those "X" and "Y" points that are brought out to those external "preamp out" jacks.

DC voltage is zero. Signal voltage is zero to probably a max of 1V RMS. The jacks appear to be wired parallel to the signal path to simply pick off the signal, not to interrupt the signal. Breaking the signal path would require more wires.

[eleventeen]

[kladmin]

Thanks guys, yes, that wire is insulated..  I don't think that particular mod is causing a problem, but I will check the jacks.

On a different note, the bias resistor is reading like 2.2k, unfortunately I don't remember what it was before but I'd like to get the bias sorted - any place other than radio shack open on Sunday that might have 1 ohm resistors?  N Dallas (frisco tx)

[DummyLoad]

Thanks guys, yes, that wire is insulated..  I don't think that particular mod is causing a problem, but I will check the jacks.

On a different note, the bias resistor is reading like 2.2k, unfortunately I don't remember what it was before but I'd like to get the bias sorted - any place other than radio shack open on Sunday that might have 1 ohm resistors?  N Dallas (frisco tx)

Fry's


--pete

[sluckey]

On a different note, the bias resistor is reading like 2.2k

What are you calling the bias resistor? Picture will help.

[kladmin]

I am talking about this resistor - keep in mind that there is neither a hum balance pot or bias adjust pot in the circuit.  I am going to add the pot back in as a bias adjust pot now that I have a 1 ohm resistor on one of the 6L6 cathodes.

So the reading across the cathode 1 ohm resistor is .4 mv - can that even be right?  So in addition to little voltage being fed to the 6l6s there is also hardly any current at all.  Something is stealing all that voltage and current, right?  Like a short feeding it to ground or something?

[kladmin]

Some actual voltage measurements WITH TUBES so that we have specific numbers - 343 B+ is waaay to low, am I wrong?

B+ - measured at standby switch on standby 492

B+ - measured at standby switch active/on  343

6L6 Plates  343v

Bias voltage at divider (common lead for 220k resistors)  -54v

PI 100k and 83k junction  327v

[kladmin]

Forgot to add heater voltage measures 6.76 v AC measured at the pilot light connections

[kladmin]

The pilot lamp is difficult (know any tricks?) to remove and reinstall

http://missionamps.com/hardware.shtml

Ha!!  Love!!  That is definitely going to be the next tool I buy that I can't find when I need it!!

[DummyLoad]

B+ - measured at standby switch active/on  343


something is loading the power rails. badly - that should be around 440-450V. 


6L6 bias looks right.


--pete

[kladmin]

B+ - measured at standby switch active/on  343


something is loading the power rails. badly - that should be around 440-450V. 


6L6 bias looks right.


--pete

Agreed - any idea what I might be looking for?  Would it more likely be a loose washer or something like a pot connection got accidentally removed during reinstallation?  Could a blown OT cause this?

[DummyLoad]

divide and conquer.


discharge caps & take resistance measurements from each PS node to ground. should be several K ohms at each filter cap.


how do you blow the OT just changing the face-plate? seems this is coincidental, but i doubt it's the OT. are the guide pins of one or more of the 6L6 you removed broken? if yes, is it possible you installed a 6L6 incorrectly?


perhaps you bumped or broke something. are/were any of the tubes red-plating? sockets OK? what's the B+ without power tubes installed and standby switch in play mode? it should be four hundred-ninety-something still, if not, it's a filter cap gone bye-bye or a short.


are you getting -54V at all pin 5 of the 6L6 sockets? 


are the 470ohm screen resistors OK?


how about the 1.5K grid-stoppers on the 6L6 sockets?

[kladmin]

divide and conquer.


discharge caps & take resistance measurements from each PS node to ground. should be several K ohms at each filter cap.


how do you blow the OT just changing the face-plate? seems this is coincidental, but i doubt it's the OT. are the guide pins of one or more of the 6L6 you removed broken? if yes, is it possible you installed a 6L6 incorrectly?


perhaps you bumped or broke something. are/were any of the tubes red-plating? sockets OK? what's the B+ without power tubes installed and standby switch in play mode? it should be four hundred-ninety-something still, if not, it's a filter cap gone bye-bye or a short.


are you getting -54V at all pin 5 of the 6L6 sockets? 


are the 470ohm screen resistors OK?


how about the 1.5K grid-stoppers on the 6L6 sockets?

So in backward order:

1:  I replaced the 1.5k grid stoppers and 470 screen resistors with Doug's parts at the time of last cap job about 10 years ago and they have held up great, haven't drifted much (all 470 read 461 and all 1.5k read 1.491k)

2:  I am getting -53 at all pin 5.  **One thing that I noticed - I get 344 at both the plate (pin3) and the screen grid (pin4) - this is not normal I don't think?

3:  I am getting 491v on the plates with no tubes, and also 491 on the screen grid!!  uggh, are they shorted together somewhere?

4:  6L6 guide pins are all good

5:  I checked the doghouse cap spots for resistance to ground earlier on but I will revisit this.


Sooo...   Safe to say screen grid voltage should not always be the same as plate voltage, right?  Must be shorted together somewhere?

[kladmin]

with no tubes resistance between pin 3 and pin 4 of all 4 sockets is 590 ohms, not sure what it 'should' be

[sluckey]

Safe to say screen grid voltage should not always be the same as plate voltage, right?  Must be shorted together somewhere?

with no tubes resistance between pin 3 and pin 4 of all 4 sockets is 590 ohms

This is all perfectly normal for that amp. Voltage and resistance readings are normal. The only thing separating the plate from the screen is one half of the OT primary and the choke. Both are low resistance as far as dc ohms is concerned.

[kladmin]

divide and conquer.


discharge caps & take resistance measurements from each PS node to ground. should be several K ohms at each filter cap.

Ok when I do this on most of the doghouse caps + side they initially read a very high resistance and then start clibming, usually past the scale of my meter.

The first node (pre-choke, 2 100uf caps + 220k resistors starts lower than the others and then climbs, but there's still lots of resistance.  Took one of the resistors out of the circuit and it meaures 195k, so it has maybe drifted a bit but should still be ok?

[eleventeen]

Measuring e-caps with an ohmmeter requires some interpretation, because they acquire and retain charge from the voltage supplied by the ohmmeter. The gross fault occurs when the cap reads like a short. This does not appear to be the case with your situation.


Sure would like to know what's eating your B+.

[kladmin]

Sure would like to know what's eating your B+.

Lol indeed, I'm honestly stumped.  Maybe I'll do the 1x 12ax7 test again, seems like there must be a component shorted or slightly shorted to ground?  Probably something that carries pretty good voltage?  Around the PI?

[2deaf]

If you are tired of taking shots in the dark, you can always take an empirical approach.  F'instance, start at the PT/rectifier and move on down the line if they are OK.  The PT/rectifier is real suspect anyways, with such a large voltage draw-down without blowing the fuse or melting the PT.  Disconnect a wire from the standby switch and connect a 1 ohm resistor between the switch and the wire.  Now do whatever you do to get 343V on the plates.  These transformers can put out 450 ma while barely breaking a sweat, so it would take something way in excess of 450 ma to draw the voltage down over 100V and it would also indicate that the PT/rectifier is OK.

If, instead, the current is way lower than expected, you may have a defective PT or, more likely, a rectifier that is half-wave instead of full-wave.  You're going to replace the diodes if you are at this point, so take them out and see what AC voltage is coming out of the PT relative to ground -- should be 330V to 350V on each side.  If it is significantly different, you need to devise further tests to ascertain why.  If they are normal, put new IN4007's in (don't get them hot when soldering) and see what current it puts out now.

[eleventeen]

"Lol indeed, I'm honestly stumped.  Maybe I'll do the 1x 12ax7 test again, seems like there must be a component shorted or slightly shorted to ground?  Probably something that carries pretty good voltage?  Around the PI?"


I think I would re-try this, myself, more in an effort to reset my brain. Maybe you dropped a locknut or a blob of solder somewhere where it's making a partial short. I think that is farfetched, but stranger things have happened. It's even possible that a wire cut-off stub left over from when the amp was manufactured is stuck in there, somewhere. 


Maybe you need to unscrew the screws that hold the parts board to the chassis, see if you can pry the parts board a teeny bit up off the chassis --- including separating the parts board from the insulating board underneath----and see if there's a random piece of metal or the stub of a wire under there.


Does installing the 12AX7 in the very first preamp tube socket (normal channel first preamp tube which I will call "V1")  have the same effect on the whole B+, including the B+ fed to the power tubes as:


installing the tube in V3, reverb driver?


installing the tube in V6 PI?

[2deaf]

6L6 Plates  343v

Bias voltage at divider (common lead for 220k resistors)  -54v

PI 100k and 83k junction  327v

It dropped 16V across the choke and a 1K resistor, which is totally normal at this voltage and clearly indicates that nothing in the pre-amp, reverb, or PI is drawing the voltage down so radically and anyways, if they were, the doghouse resistors would have burned-up a long time ago.  It also shows that the screens aren't doing it because the choke would drop the voltage a lot more than that.

Something on the proximal side of the choke is malfunctioning.  That doesn't mean that there are not other problems beyond the choke, but they are not causing the 343V at the plates.
   

[kladmin]

Hey guys, I know I'm reviving my dead thread but I got the Twin Reverb back up and running and just wanted to update you on what the solution was!!

I took it to David Smith at Dabeck amps (who does amazing work), and it turned out the problem was.....

The power transformer.  That's it.  Thank you ALL for your help troubleshooting!!  New power transformer and everything is right where it should be.  Must have just been a coincidence that the transformer shorted itself to the 240v wiring while I was working on the cabinet, or maybe I looked at it funny ;)

Replaced with the Hammond Twin PT and it works and sounds like a twin should now!!


Kurt

[Willabe]

Thanks for letting us know.