Rescue project: 1972 Twin Reverb silverface

[2deaf]

Thanks! -  For sure the left most one is supposed to be opposite of the rest.   
Was trying to remember that arrow thing ("it points how electrons flow from positive to negative").   gimme a break whats so hard about printing  a couple of "+++"?   

There is a negative symbol in the white bubble and arrows pointing to the lead that is negative.  Better check that electron flow thing.  Ben's conventional current flows from positive to negative.

[jeff_free69]

ok finished the dog house and also the 2 Ecaps on the recitifier before  ( man those new ones look so puny!)

Now, about those diodes...

How to test for diodes?

Use an R-meter with one meter probe at each diode lead (with the amp switched off, discharged and unplugged). Good diodes will measure a resistance one way, by when you switch the probes over to the opposite leads, they will read open

Thats kinda what I thought , but can I do this in-circuit?   I had tried that but got bizarre readings doing so.  Went back again and gave it another shot - still seeing things i don't understand.

For the  lone reverse guy  it measured about 20 k in both directions (very suspicious).

For the string, each measured 380K as shown, and Open when reversed  ( weird thing is the reading changes as you change the scale - it seems to read about 3/4 of whatever the scale is  ????)

[Papa Jim]

I would eliminate the possibility that the rest of the circuit is affecting your readings by temporarily lifting those three red wires on the left side while testing. Does your meter have a diode test setting on it? The link shows how to test with or without a diode test setting.
https://www.fluke.com/en-us/learn/best-practices/test-tools-basics/digital-multimeters/how-to-test-diodes-using-a-digital-multimeter

[sluckey]

You cannot accurately check those diodes in circuit. Do what Papa says. I really doubt anything is wrong with that board. The bias caps have been replaced and are probably good too. In fact, I bet all those caps in the doghouse were good too. Even the one with the broken lead could have been repaired.

At some point you need to just turn the amp on with no tubes and check the B+ and negative bias voltage.

[st]

One of those ugly blue caps must have insulted his mother at some point because he has a completely irrational grudge against them.  He replaced those 0.1uF caps for no reason.  There is a demand for those blue caps, but I doubt he is removing them for resale.

I am not a fan of the man's videos, and his dealings with the bias balance pot are shockingly stupid. However, as far as i can tell he only replaced the two coupling caps in the power amp. In my experience, those blue caps in that position fail frequently, and when they do, the results are arguably more serious than in other positions, so that move isn't completely irrational...

[2deaf]

One of those ugly blue caps must have insulted his mother at some point because he has a completely irrational grudge against them.  He replaced those 0.1uF caps for no reason.  There is a demand for those blue caps, but I doubt he is removing them for resale.

I am not a fan of the man's videos, and his dealings with the bias balance pot are shockingly stupid. However, as far as i can tell he only replaced the two coupling caps in the power amp. In my experience, those blue caps in that position fail frequently, and when they do, the results are arguably more serious than in other positions, so that move isn't completely irrational...

I would agree with you if I had ever experienced a failed blue cap, but I haven't.  I've never seen a terrible blue cap cause 60 cycle hum, either.

[2deaf]

How do you test caps for Leaking DC ?  I thought you need a special meter for ESR.  Mine is V-O-C-L .

Once you get the power supply going, there should be a high voltage on one side of the 0.1uF caps and zero volts on the other side relative to ground.

Where's the peer review?  There would be HT on one side and negative bias voltage on the other for the 0.1uF PI coupling caps.

[SILVERGUN]

How do you test caps for Leaking DC ?  I thought you need a special meter for ESR.  Mine is V-O-C-L .

Once you get the power supply going, there should be a high voltage on one side of the 0.1uF caps and zero volts on the other side relative to ground.

Where's the peer review? 

Now that you've reached the prestigious Level 3 status 
 and obviously always know what you're talking about you will be subject to way less review. We will, of course, expect you to continue to review yourself.
Don't just go around spewing alternative facts.
Anyone with over 1000 posts must be filling the world with facts, so please, be more careful.
Don't make us catch you. We all have better things to do.

[2deaf]

Level 3?!  I had no idea a guy on the no-fly list could get a security clearance that high.  I better clean up my act.

[shooter]

no-fly list

You piss of a TSA agent tooo?? they really didn't like a long-hair telling them, just a little left with that hand, ahhhhhh, that's the spot

[jeff_free69]

I would eliminate the possibility that the rest of the circuit is affecting your readings by temporarily lifting those three red wires on the left side while testing. Does your meter have a diode test setting on it? The link shows how to test with or without a diode test setting.
https://www.fluke.com/en-us/learn/best-practices/test-tools-basics/digital-multimeters/how-to-test-diodes-using-a-digital-multimeter

My little Radioshack  meter does not have.  But my other cheap meter does!  (fancier , but I have shied away from using it since it seems to consistently  measure a little on the low side)  Imagine that, after all these years i never measured a diode - most of my work has been passive speaker crossovers. 
So I think there's good news - in open position it reads about 3V.  shorted its 0V.   
On one side of the diode its reads the full 3V (open).  Reversed, it reads about .5V    This is consistent for diodes in circuit and a few other new ones I have out of circuit. 

The bias caps have been replaced and are probably good too. In fact, I bet all those caps in the doghouse were good too. Even the one with the broken lead could have been repaired.

possibly an overabundance of caution - but the big blue atoms start with "96.."  , I think making them 20+ years old.  I'm under the impression that the rule thumb for Electrolytics makes them over the hill.    The bias caps look original to me, making the almost 50. What is the life expectancy?   
Nothing has been tossed - and can be un-replaced.
 

At some point you need to just turn the amp on with no tubes and check the B+ and negative bias voltage.

New power cord is in the house now.
Any comments on that wiring diagram from a couple of days ago?  basically bypassing the polarity switch and going  to fuse then power switch

Thanks for your patience.  I realize this is all an afternoon's work for most of you. 
Same thing happened when I tiled my bathroom.  Learned a lot and the final result was excellent . Even if it did take a ridiculous amount of time.  But now I'm all set to do the the upstairs bathroom  and the kitchen, and won't need to ask so many pesty questions :)

[mresistor]

regarding power cord wiring.   see this link and scroll down to page 16  http://sluckeyamps.com/misc/Amp_Scrapbook.pdf   you will see the best way to wire it.   I have been using x1/y2 safety caps in place of the .047 cap in the stock amp.. and keeping the ground switch function.  There are many opinions on this but I see nothing wrong about it.

[jeff_free69]

Got the nice new Husky 14 ga cord wired in  (for now polarity/deathcap bypassed).

One last check before the smoke test - measuring the resistance on the PT windings.  I have no idea what to expect, other than the Primary (Black-White) to Secondary (Reds) should be open.

I unsoldered the red secondary leads and here's what I measured 

Black - White         : 4.5 ohms
         - Red1,         : Open
         - Red2,         : Open
         - Red /BLue   : Open

Red1 - Red2 :       21
        - Red/Blue:  14
Red2 - Red/Blue:  15

Ground - Red1:     10  (shown in photo)
            - Red2      11
           -Red/Blue   4

Seem reasonable?
Anything else to check at this point?

(just for the record , that huge blob of solder on the ground was there already - i dont think my soldering iron can melt it all at once)

[Papa Jim]

Do you have a current limiter for the first power up to use. It always settles my nerves before switching a new amp one. 

[shooter]

dont think my soldering iron can melt it all at once

the Fender folk got paid by the miles of solder they used!
I took out 3 pots, 2 caps, clogged my solder sucker twice and used 3' of wick 

[DummyLoad]

Got the nice new Husky 14 ga cord wired in  (for now polarity/deathcap bypassed).

saywhat? 14AWG for a 100W amp? (yes, i know it pulls 3x that), but jeeez. 18AWG is overkill @ 10A or 1200W @ 120V.

respectfully,

--pete

[jeff_free69]

Do you have a current limiter for the first power up to use. It always settles my nerves before switching a new amp one. 

yup - thats was the first thing I did.  And found a couple of NOS 100W incandescent bulbs in Grannies apartment.

1 bulb or 2?  I guess this Y adapter puts them in parallel, so not sure if it helps...

say what? 14AWG for a 100W amp? ... but jeeez. 18AWG is overkill

respectfully,

--pete

Ha - I felt bad because I wanted to use the original and just cut off the bad 10" segment.  But punky  47 year old rubber and rope didn't inspire confidence.         
I have a hi-fi background and always use 12 GA on my speakers, so why not beef up the power cord? (there are idiots who pay hundreds and thousands to upgrade their hi-fi power cords - this was only $2 more).  Although the hard plastic grommet did have to be whitled down on its internal "clamp".  Next time 16ga.

So nothing alarming about those measurements, I take it?
I just measured everything in sight.   I was a little suprised at the Ground - secondary measuring so low (not that I know what I'm yapping  about)

[sluckey]

You really need to take this to someone that has a clue. Or take it to the curb!

[Papa Jim]

Quit doing this  and just power the sucker up before all the help just ceases dude.

[PRR]

> I have a hi-fi background and always use 12 GA on my speakers, so why not beef up the power cord?

Speaker is 8 Ohms and the wire IS the signal path.

The amplifier is like 120V/2.4A or 50 Ohms, far higher than 8; and isn't directly in the signal path.

Simple proportion suggests that you could use 8/50 as much copper. Scaling from #12@8r, you could use #20.

Actually the Code suggests using a wire which will pretty-sure pop the fuse in the cellar before it sets the carpet on fire. This will be lamp-cord, formerly #20 but IIRC now #18.

However for a lousy 6 feet on a 40 pound appliance, around a bunch of drunks, I have never objected to OVER-kill. Hell, I use #14 for instrumentation carrying 0.050V 0.010A, just because it is readily available (mowed extension cords) RUGGED.

[Ed_Chambley]

oh, forgot then I smoke my wine 

I find that combustable wine to go well with Cream, even their reunion DVD.

[shooter]

combustable wine to go well with Cream

thanks for the suggestion
I picked up topographic oceans to test drive my latest build, works well!

[Saransk]

While most, if not all, the posts have been on track to get this amp back up and working, the suggestion to just gut it and but new board and such in it has a lot of merit.  This is a not a collector amp, and to restore it to that condition with the proper type parts is an extremely expensive proposition.  The great news is that the "Blackface/Silverface" Twin Reverb is one of the great amp circuits ever designed and the various "upgrades" done to the classic AB763 circuit are easily reversed.
Hoffman, as well as other vendors, make brand new eyelet or turretboards, that are a drop in fit.  You could also purchase a "small parts" kit that has all of the parts to rebuild it using only the chassis and transformers.  There are a couple of really good "how to build a tube amp" instructions that can walk you through how to get it built - and get it built to work.
I rebuilt my first Fender almost 50 years ago and unless someone walks in with "Grandad's Tweed Twin that was in a closet" old amps are just that - OLD!  When I do a rebuild I am perfectly happy to spend someone else's money if they just have to have some particular "magic component" but many of the $10 caps out there are rebranded $3.00 ones.  But usually they want an amp that works, works reliably, and can handle gigs, etc.  It's like owning a 1965 Vette and want to drive it - some stuff needs to be rebuilt/replaced.
If you want to use this for any real playing get the old stuff out of there - it will still sound better than 90% of the newest amps.  Metal Oxide Plate resistors, Metal film Cathode resistors, new film coupling caps make for a solid & quiet amp.  I've even started using large film caps for the Power Supply caps.  They are about the same as the better electrolytics.
Nothing electronic that is over 20 years old is worth saving save the transformers.
I have a 1970 Bassman that I rebuilt with a new board and sockets.  It is dead quiet and should last another 20 years with only an occasional tube replacement. 
Good luck

[Papa Jim]

That being said by Saransk then power the thing up because you have nothing to lose really.

[Saransk]

AS Alan Shepard said "Light this candle."
I've never seen a Fender catch fire - some smoke but no fire like older Marshalls are sometimes prone to do when abused.
Use a cheap speaker for the initial test - even massive Electro-Voice speakers don't like pure square waves and such.

[Papa Jim]

Thread is going on like 26 days and still no power testing.

[jeff_free69]

You really need to take this to someone that has a clue. Or take it to the curb!

Done!

For anyone interested, it is by a dumpster behind  the old Sears at the CrossRoads Mall in Omaha.   I figure you have a few days before the truck comes...

[jeff_free69]

Just kiddin (jeez - where 's your sense of humor?)

Clues, I have aplenty.  Experience, which varies directly with equipment ruined, not so much.

So moving forward ( 1 agonizingly slow step at a time.  sorry my day job sux and I can only get an hour here or there for this)...

Since there were no unexpected shorts or opens seen on the PT, I then tested its voltages, while still disconnected.    As far as I can tell things are good :
The receptacle tester is happy with the rewired outlet.

The VOM seems happy too: The red secondary reads at just over 340VAC.     
The RevC schematic  (no push-pull, like mine) is pretty fuzzy, but it looks like it could be saying 340VAC.   RevF/G are clearer and also shows 340VAC.

The bias voltage is a little amibiguous.   Mine reads 62V on the red-blue wire.   
RevC shows  what looks like -69V, but on the other side of the resistor and diode.   However the resistor value there is 470, whereas mine is actually 1K.   
The RevF drawing (has  push-pull) has the 1K and shows -63V. 

The truth probably lies somewhere in the mythical Rev D-E (any sightings?).  But I'm hoping  it means its all in the ballpark and the PT is OK.   .

Also took another look at the diodes out of circuit and they measured OK

The  next step is to reconnect the PT and get the B+, Bias Voltage  and readings at other points of interest. 

[mresistor]

jf69  hope you read the notes on the schematic.. 

[jeff_free69]

jf69  hope you read the notes on the schematic..

If you're referring to the vibe pedal ground, then yes; and thanks for the reminder.  I've been meaning to make a shorted RCA  jack, which I think is whats needed, correct?  ( I do not trust the actual pedal).   So thats in place now.
If something else - then please elucidate.

With the PT reconnected I powered up and took some measurements (No tubes). 
Some good news and a clueless question, of course.

1) The B+ came in pretty much spot on at 478VDC.
2) The bias right in there at -64VDC
3) no smoke or sparks

I took a couple of other measurements, but my notes are based on having the tubes in.  What I'm seeing does not match those:
In the doghouse (see image)  the red wire, which is connected to the diodes reads correctly as expected.  But the others are way off.  Is this simply because of the lack of tubes, or indicative of a problem?

On the 6l6 sockets my notes show that pin 6 should read 446V it reads 270 .  So, as above.

What else is relevant / meaningful without the tubes?

Thanks

[2deaf]

I took a couple of other measurements, but my notes are based on having the tubes in.  What I'm seeing does not match those:
In the doghouse (see image)  the red wire, which is connected to the diodes reads correctly as expected.  But the others are way off.  Is this simply because of the lack of tubes, or indicative of a problem?

You said you took measurements with the tubes in. Then you asked if the voltages are wrong because the tubes are not in. 

You got a problem, alright.  Take the power tubes out and leave the little tubes in.  Post the voltages without the light bulb limiter.  If the choke/screen node is still 175V, shut it down immediately.

[jeff_free69]

> You said you took measurements with the tubes in. Then you asked if the voltages are wrong because the tubes are not in. 

Sorry for the confusion, but what I was trying to say was that i took notes from studying the schematics.  Then traced out where those points were in the layout and measured there. 
But my measurements are way different, and I'm hoping thats because I still have not put the tubes in, and that the schematics wouldn't account for that.

[2deaf]

What about the standby switch?  With no tubes and the standby switch closed, all of the voltages should be the same after all the cap's charge up.

[Papa Jim]

If you really want to isolate the problem then just disconnect the wire after the standby switch that goes to the center tap of the output tranny, and each individual wire that goes from the different caps B+ nodes to the amps different circuits or stages. Measure and they should all be pretty high. Connect one wire up at a time and determine which one is causing a large drop down to 170vdc that you see. Troubleshoot the component and wiring in that stage of the amp. That would just be my own personal method. Maybe not everyone's but it is a start to doing something. Post the voltages before and after connecting back each B+ node and output tranny center tap.

[mresistor]

Sorry it appears that cbs in their wisdom didn't include the note on this schematic that voltage readings are + or - 20% - also today's wall voltages sometimes cause many volt readings to be higher than indicated on older Fender schematics.

[jeff_free69]

What about the standby switch?  With no tubes and the standby switch closed, all of the voltages should be the same after all the cap's charge up.

Excellent question!
I was playing with the Standby while testing directly at the diodes ( there's one place on the schematic that shows that : 478V  "off" , 450V "on".  )
But by the time I flipped her over to get into the doghouse it may have been off (can't be sure) .

Redoing everything today, I get very different results. 
Not sure I have great explanation, but  yes  i think I see what you suggest now

1) Powering On with the Standby OPEN  (opposite the power switch)
   those  3 test points on the caps towards right start off essentially 0 V ( actually 6)
   The red wire  towards left  measures 476

2) Closing the standby (same direction as power):
all test points in doghouse  measure 473V ( after a quick charge up)

3) Opening the standby again   a discharge begins  - counting down at a noticeable rate, pretty much 1V sec.

With this being the case is that looking better?

Possible explanations of the discrepancy:
1) Last night I had standby open by the time i went to the doghouse.  When testing I did not notice the slow but consistent discharge.   possible but tonight it was quite evident.
2)  Today I reset the the probes and ground point.  Did flipping it over last night  disturb ground point?  possible , but the red wire still measured normally last night, just not the other 3.

I guess #1 is more likely a better explanation .

We'll see what happens tomorrow ( 2 outa 3 aint bad)

 

[2deaf]

I knew what you did when I looked at the numbers you posted.  I say you put the tubes in and hook the power cord up to that wang-dang limiter with one 100W bulb.
 

[jeff_free69]

yep - a common rookie mistake, no doubt


Time to get tubular.  First I had to build one of those stands and tacked on a dummy 8 ohm load.

With the wangdang special 100 W limiter and tubes in 

 I noted  the bulb glowed a noticeably brighter than without the tubes, even after it settled down after the initial flash.
 is that normal?

                           B+       revbias   
standby open       362         -50   
          closed        262        -40

No smoke , sparks or me going 
so taking off the training wheels...

Without limiter

standby    open      470      -64
               closed    450      -64

I'm expecting either 478 (revc)  or 470 (revF)   open
              and either 450 (revc)  or  445(revF)    closed
              and either -69 (revc)  or  -63 (revF)   rev bias

So mine's a little of each
 close enough for Rock n Roll?

What next?
I have the 1 ohm resistors but didn't install yet - are we good to go ?

[2deaf]

Shut it down.  Plug in speakers.  Plug in guitar.  Power up, no limiter.  Hit it.

Attached are the actual voltages from a '72 Twin Reverb.

[jeff_free69]

Finally hit it! 
Plugged in the Strat and the results weren't half bad (not counting my rusty playing).

Both channels worked.  No exceptional buzz or hum.

The Normal channel was normally clean and bright

The Vibrato channel had a quite a bit more bite and drive.
But the vibrato wasnt vibrato-ing.  The opto thingie was flashing and you could hear a slight ticking, but no change in volume whatsoever.

Note:  the reverb tank is not connected that this point

Tubes
Giving things the chopstick test, some preamp tubes are slightly microphonic.
But V4 is highly microphonic (this is for the Reverb i believe).    And the .003 ceramic cap connected to it is even more so.  if you gently  rub it, it sounds like the chopstick is on your eardrum

I have to double check which are which, but some tubes are RCA and others had faint markings  that vanished when I wiped off the dust.  not sure what to make of that - very old or very cheap replacements?.

Power Tubes
3 out 4 make a weird grunting sound when tapped.  Like a distorted, thuddy reverb.  the guts rattling around I presume?

Bias check
I took a crack at calculating the Plate Dissipation  using the Output Transformer measurement .
Please feel free comment if I got any of this wrong:

Measuring  from  (diode string /OT center tap)  to Plate Pin 3   
   Voltage drop                            2.2 V                2.3 V
   R                                            36.4 ohms         38.5 ohms
   calculated current (V/R)    =    .060 mA            .059 mA
   B+ voltage                                         x 456 V
   power dissipatation  (VxI)        27.3W               27.2W

While these numbers look impressively close, isn't that too high for a grid-biased 6L6GC tube?
My understanding is the recomemended max power is usually considered to be 30 W x .7 = 21 W
These tubes are matched Ruby's - are they made to handle this?

Your thoughts ?
Thanks
             

[sluckey]

power dissipatation  (VxI)        27.3W               27.2W

Remember there are two tubes on each side so you need to divide those numbers by two. So, 27W for two tubes means 13.5W for each tube. That's a bit low.

[jeff_free69]

power dissipatation  (VxI)        27.3W               27.2W

Remember there are two tubes on each side so you need to divide those numbers by two. So, 27W for two tubes means 13.5W for each tube. That's a bit low.

Ah! Of course, I completely forgot about that (paying too much attention to the math - where's that forehead-slap smiley at?)
Most of the vids / tutorials have a single tube per side and aren't geared towards the twin.

I'm traveling this week, so no time to do anything. But  when I come back it looks like I have some biasin' to do.

I've reviewed the Blackface bias conversion links (change from balance to level adjust).  But again, the examples I've seen so far are geared towards schematics with a single tube per side.  Is this also recommended for a TR?

Is it important to balance the tubes within each side somehow (how do you balance 4 tubes)?

BTW  - I did make a change before I started:  For the (2) 68k resistors coming from the bias balance pot, one measured 66k and the other 75k ( so way more than 10%).  I didn't know how much range the balance pot was going to have , so i replaced the 75k  with a new ~67k.  Haven't touched the balance pot yet

Thanks!
 

[2deaf]

I've reviewed the Blackface bias conversion links (change from balance to level adjust).  But again, the examples I've seen so far are geared towards schematics with a single tube per side.  Is this also recommended for a TR?

Is it important to balance the tubes within each side somehow (how do you balance 4 tubes)?

The number of tubes makes no difference to the bias circuit because there is no grid current, so circuits geared toward single tube per side are valid for a TR.

That balance thing is for idle so that hum is minimized.  It is not necessary to balance the tubes within each side (although some folks do), only the balance between sides.

[PRR]

Fancy-folk try to get all four bottles gushing the same at idle.

A TWIN is not made to idle but to ROAR. Anything over a whisper, the idle bias is overwhelmed by signal, the tubes conduct as needed. Anything under a whisper, you should have brought a smaller amplifier.

With four tubes, two each side, "balance" is not real important, differences average-out. Try not to have one tube under 50% Pdiss and another at 90% Pdiss, but anything half to two-third Pdiss is plenty "balanced" for any Twin use.

(And yes, I knew a guy played jazz cello through a Twin, and I would not change my advice for him.)

[jeff_free69]

OK - back in action.  Only had a little time last night, so before I started making any real modifications, I wanted to check with another measurement.   In the interest of Socratic endeavor I added the 1 ohm shunt resistors  and took some measurements there.

All 4 resistors checked out very close to 1 ohm, as far as I could tell.  The shorted leads on my meter measure .2 ohm.  Each R measures 1.2      I even tacked them all together and they measured 4.2 (close enough to use mr ohm's law and pretend the meter is reading milliamps instead of millivolts)

V10         V9              V8           V7
32mv      29mv          31mv       29mv

So in terms of A-B balance , the sides  are within 1 mv
today my B+ was 479V,   VxI 's are
15.3W   13.9W        14.9W      13.9W

Slightly higher than the OT calculation (as expected) but still too low.


Also, wanted to double check all my doghouse voltages:
                     red     yel2        brn     yel4  (per 2deaf's image)
mine              454     453       416     343
Expected        450     446       420     345
2deaf's           450     448       418     348

[2deaf]

There are lots of things that can be done to increase the dissipation.  You could change the 1K resistor to 1.5K in the bias circuit and see what happens.  Robrob has an adjustable bias mod for silverface Fenders.  Or you could change the bias balance control to a bias adjustment control. 

[PRR]

I don't think 14W-15W is "too low", especially on a Twin. It's not a whisperer.

I'd leave it be.

[Papa Jim]

Looks like your done and it was a sucess.

[mresistor]

  Or you could change the bias balance control to a bias adjustment control.

I mentioned this to him in reply #11 and one of the best discussions with graphics for changing over to adjustable bias is right here https://el34world.com/charts/bias_conversions.htm


I think its one of those things CBS did to Fenders circuits that doesn't make much sense to keep in todays environment and I'm not a fan of the bias balance circuit. Can be done easily and doesn't take long to do it.

[PRR]

Also remember the '72 Twin is really based on older plans which assumed the original 19W Pdiss of a 6L6. Yes, in 1972 they were using the "30W" 6L6GC, but they may have had no reason to re-bias since it worked fine. And nobody had invented any "70%" rule yet. (70% of 19W is 13W.)