1972 Twin Reverb (Guts)

[22uf]

Hi fellas. As a confirmed sucker I picked this up completely untested for a very reasonable fee.
Its a "Castilo" stamp.
I was just wondering if anything stands out as "uh oh"? Before I go collect it! Looks mostly complete and original to me.
There's nothing stamped on the power transformer which is weird?
The codes suggest 1972? Could be a 1973 model though?
The baffle was replaced and the pilot light and footswitch etc.

https://photos.app.goo.gl/cB1ZrpbiCoxetPZd7

[sluckey]

Those "Made in Italy" speakers are not original.

[22uf]

Those "Made in Italy" speakers are not original.

They are probably better than the originals heheh. 
I'm not a fan of Jensens but the alnicos are pretty sweet. (imho)

[sluckey]

I agree. I have one in my 5E3.

[DummyLoad]

export model - very cool!

replace all the bias & power electrolytic caps.

agree italian made jensen alnico sound fabulous and can take the power a twin dishes out.

congrats, nice score.

--pete

[Latole]

Look original except one resistor in the bias circuit ; to adjust at right bias ?

Check if fuse is the right one. Some people may put a too power fuse and amp/ power tansformer is not protected .

Post picture of the filter caps on the dog house. Did you see bubble at one end ?
Best it to replace them all. I use F & T caps

[22uf]

Yup she's a good one. I've added some more pics to the original google drive link up top ^
As far as I can tell the circuit is more or less stock and so are all the preamp valves (minus the Mullard PI).
I had a good old bash on it for a few hours and then the following morning decided to check out what caps the previous seller said his tech had put in the doghouse..

Uh oh.
Lucky me twice I guess!
Will these F&T do the job? I see a ton of guys putting 100uf 350v in instead of the factory 70uf? Just convenience?

I have these .. so no reason I shouldn't use them right?

Also am I right in thinking the doghouse is: 2x 220k, 2k2 and a 10k? I have some carbon film 1W on a shelf somewhere..

[sluckey]

I see a ton of guys putting 100uf 350v in instead of the factory 70uf? Just convenience?

It's because 70µF/350V caps are extinct.

Also am I right in thinking the doghouse is: 2x 220k, 2k2 and a 10k?

yes. I prefer to use 3W metal oxide for all those.

[22uf]

Would 5w be too silly? I have a stash I'd like to use up instead of ordering in 3w ones.

[Latole]

Would 5w be too silly? I have a stash I'd like to use up instead of ordering in 3w ones.

I often use those if I need more than 2 watts

[2deaf]

In addition to that ugly bias thing, I see:

The reverb driver cathode bypass capacitor has been removed and the resistor has been changed.

The mixer cathode resistor has been changed from 820 to 680 ohms.

The .003uF reverb recovery coupling cap may have been changed.

The reverb transformer may have been replaced.

[22uf]

In addition to that ugly bias thing, I see:

The reverb driver cathode bypass capacitor has been removed and the resistor has been changed.

The mixer cathode resistor has been changed from 820 to 680 ohms.

The .003uF reverb recovery coupling cap may have been changed.
=The reverb transformer may have been replaced.

  Thanks for the heads up!

Funnily enough the reverb doesn't work at all. It did briefly, was super weak and barely audible, then stopped about 20m after powering up. I swapped out the tubes and then the tank and leads from a 1977 and still nothing. I did meter the reverb transformer and it does appear to be OK. The previous owner failed to mention that .. but did sheepishly say he used a reverb pedal and squirmed when I asked why he'd use a pedal when he has the greatest spring reverb ever made built in. 



I knew there was something wrong for the price I paid (600). I'm just relieved its not the big transformers (touch wood).

Apparently the original owner (from 72 to about 2010) had used it as a bass amp with 10" drivers and tweeters for 40 years..

Anyway I'm going to sort out the power, bias and rectification first then look at the rest of it. It sounded good (very) scooped.
The master volume has no bright cap on it - I never knew that. It looks to be the simplest MV ever. Just spliced a pot right in hehe.

2deaf once said the 1972 TR was the best amp Fender ever made.

I'd love to know why! 

PS: Can you tell me what circuit this one is so I can rustle up a layout? No plans to blackface it until its been restored and appraised.

[Latole]

Yup she's a good one. I've added some more pics to the original google drive link up top ^
 
Will these F&T do the job? I see a ton of guys putting 100uf 350v in instead of the factory 70uf? Just convenience?
 

F & T made 80 mfd at 450 v

[22uf]

The caps in the pic are F&T and are 70uf @ 350v. 
Must be a new line..

The power transformer stamp was hidden under about 1/2" of carbon.



The whole amp looks like its been real warm for a real long time.



Also the chassis stamp is for 1973.

[Latole]

See in Antique electronics 2018 catalog
And today

https://www.tubesandmore.com/products/capacitors?filters=2341a2370

[2deaf]

PS: Can you tell me what circuit this one is so I can rustle up a layout? No plans to blackface it until its been restored and appraised.

This is the schematic.  The AA270 layout is pretty similar, but there are a few things missing.

[22uf]

PS: Can you tell me what circuit this one is so I can rustle up a layout? No plans to blackface it until its been restored and appraised.

This is the schematic.  The AA270 layout is pretty similar, but there are a few things missing.

Thank you very much, sir!

[Latole]

I have those one from Fender ;

[22uf]

I traced out the circuit today using the AA270 schematic from 2deaf and it appears to be mostly right except the tremolo circuit which appears to be AB763 and for some reason one of the 25/25s was moved from the reverb circuit and put across pin 3 of v5 to ground on the vibrato pedal socket. 

There's also a missing bypass cap on the optocoupler which may never have been there to begin with anyway..

Pretty much everything is within 10% except for one of the dropping 220ks in the doghouse which was an impressive 643k. 

[2deaf]

I traced out the circuit today using the AA270 schematic from 2deaf and it appears to be mostly right except the tremolo circuit which appears to be AB763 and for some reason one of the 25/25s was moved from the reverb circuit and put across pin 3 of v5 to ground on the vibrato pedal socket.

Your tremolo circuit appears to be exactly as I drew it.  Does it work when you have a pedal connected?

The reverb driver originally had a cathode bypass capacitor on the board, but somebody removed it.  The bypass capacitor on pin 3 of V5 was installed that way at the factory and it never had anything to do with the reverb circuit.

There's also a missing bypass cap on the optocoupler which may never have been there to begin with anyway..

Huh?  The optocoupler doesn't have a coupling cap.     

[22uf]

I traced out the circuit today using the AA270 schematic from 2deaf and it appears to be mostly right except the tremolo circuit which appears to be AB763 and for some reason one of the 25/25s was moved from the reverb circuit and put across pin 3 of v5 to ground on the vibrato pedal socket.

Your tremolo circuit appears to be exactly as I drew it.  Does it work when you have a pedal connected?

The reverb driver originally had a cathode bypass capacitor on the board, but somebody removed it.  The bypass capacitor on pin 3 of V5 was installed that way at the factory and it never had anything to do with the reverb circuit.

There's also a missing bypass cap on the optocoupler which may never have been there to begin with anyway..

Huh?  The optocoupler doesn't have a coupling cap.     

Sorry! I was looking at the AA270 layout not your schematic. Yes it works just fine. Sounded great.

Again my bad, I was looking at the AA270 layout re: the cap on p3 v5.

Alas again the AA270 layout shows a .022 across the 10m on the optocoupler.

I'll just stop looking at the layout now .. hehe. 

I'll go back and do it again tomorrow with your schematic as a guide instead..

PS: The reverb doesn't work at all.

PPS: I'm new to this, so, does this mean the circuit in this 1973 isn't a AA270 but some other revision?

[2deaf]

The AA270 layout is only similar to the '73.  It is useful for getting your bearings inside the amp.  The main board was very similar for years, even the UL models.

The Fender schematic for a '73 was drawn by an idiot.  He just drew lines where it was convenient for him with no regard for how the clarity of the circuit was affected.  I re-drew the schematic in hopes that it would be easier to understand how the circuit is configured.  Most of the schematics I draw like this are done with the assistance of an actual amplifier.  I have a '73 Twin Reverb sitting upstairs that I used to draw my schematic.

[2deaf]

The reverb doesn't work at all.

Probably start by posting voltages for V3 and V4.

[22uf]

The reverb doesn't work at all.

Probably start by posting voltages for V3 and V4.

I will do. I'm going to replace the caps and sort out the bias and rectifier circuits first.
I'll be back! 
Thank you for your help. I'd be lost without it tbh.

[22uf]

Just ordering replacement diodes. Will these suffice?

https://www.hificollective.co.uk/semiconductors/mur4100e_ultrafast_diode.html

4a but that won't hurt?

[sluckey]

1N4007s are much cheaper and just as good for that amp.

[2deaf]

Just ordering replacement diodes. Will these suffice?

https://www.hificollective.co.uk/semiconductors/mur4100e_ultrafast_diode.html

4a but that won't hurt?

Prices like that'll make a grown man holler.  Mouser has them for $0.79 and $0.49 for MUR2100EG's here in the States.

Whatever diodes you get, make sure that you heat sink them really good when you solder them in.   

[22uf]

Another silly question. I spotted a cheap 2A variac for sale locally and was going to pick it up.
Am I right in thinking that this would be perfectly OK to use for initial power up testing the twin (which is 3A at full power).
Just to use with no power tubes at 40-60v or so to make sure nothing is shorting/badly wrong/I'm getting OK voltages across the nodes etc?

[sluckey]

A lightbulb current limiter would be better suited for that purpose. A 2A variac would be useful for light duty things but a 10A variac would be much more useful.

[Latole]

A lightbulb current limiter would be better suited for that purpose. A 2A variac would be useful for light duty things but a 10A variac would be much more useful.

I agree.

[22uf]

Thanks. It turns out the variac is a 5a one so I picked it up anyway.
Now to find someone that sells 150w light bulbs in 2021 hehe..

I'm going to be playing this amp pretty hard and want to get some power tube distortion going on, likely with a UA Ox.

With that in mind would it be wise to replace the screen resistors with some higher values: 470 1 watt screens > 1k 5w wirewounds (I have some Mills ones).
And the 1.5k 1/4 watt with 1.5k 1/2 watt metal films?

I presume going 470r to 1k 5w will cut a few watts/db off the amp and give the power tubes an easier life?

That said others say the screen resistor on 6L6's doesn't need to be any greater than 2W. "6L6's are a beam pentode. Because of this design they are much more efficient than a standard pentode because they don't draw nearly the screen current of a standard pentode."

Unrelated different question: is increasing the bias caps to 100uf/100v going to change the tone any? I can get 50uf/100v electrolytics and wondered if using the original capacitance would be a good move? Some "internet talking heads" say increasing bias filtration does change the tone.

I knew I had to replace these components and I want to try to get my BOM in one order to reduce shipping etc.

[Latole]

 
Now to find someone that sells 150w light bulbs in 2021 hehe..

 

75 to 100 watts will do the job, no need 150 or use 2 X 60 watts bulb.
Surplus stores may have these.

Amazon have some 100 watts 230 volts

https://www.amazon.co.uk/Leuci-Standard-Bulb-B22-100-W-230-V/dp/B00NBSR1NW

[PRR]

> Amazon have some

Be careful. The ones called "vintage" with long straight yellow "filaments" are really LEDs.

"Halogen" is a good word. While LEDs are eating halogen's lunch, you can still buy tubular halogen lamps for a buck because replacing a large outdoor fixture with an LED thing is real labor.

[22uf]

Hehe thanks fellas. Waiting for parts, I thought I'd crack on with troubleshooting the non-functional reverb



Much relieved the transformer appear OK.

On to V3 and V4..



So these values are all wrong. I'll start by putting them right based on 2deafs schematic.
I hate that feeling when you're not so arrogant to presume the previous guy was a moron but you'd really like to but then worry you are missing something critical...

[Latole]

You must always take reading with one leg disconnected, never on the circuit

[22uf]

You must always take reading with one leg disconnected, never on the circuit

Hi the pictures are a bit crap. Sorry! I'm doing my day job and this simultaneously hehe. All legs lifted and reverb transformer desoldered completely. 

[22uf]

Had some time this morning so I started on the doghouse.






Does everything look tickety-boo? Never done this before! 

Also all that insulation melting was factory! I'm pleased to say I didn't melt anything!

[22uf]

Put the v3 and v4 circuit back to AB763 spec this morning.



2.7k/820 & 2.2k/820 with the 4 bypass caps instead of the 70s unbypassed/680r setup.

680Ω:
8.8mA total current / 2 = 4.4mA per triode
4.4mA * 389v = 1.71 watts per triode

2.2kΩ:
3.73mA total current / 2 = 1.865mA per triode
1.865mA * 392v = 0.73 watts per triode​

So the 680r unbypassed drives the tank ~4x harder?

Is that really necessary? Was blackface reverb so much weaker?

[pdf64]

Put the v3 and v4 circuit back to AB763 spec this morning.
...
2.7k/820 & 2.2k/820 with the 4 bypass caps instead of the 70s unbypassed/680r setup.

I’m confused by the above, it doesn’t seem to align with TR AB763?

680Ω:
8.8mA total current / 2 = 4.4mA per triode
4.4mA * 389v = 1.71 watts per triode

2.2kΩ:
3.73mA total current / 2 = 1.865mA per triode
1.865mA * 392v = 0.73 watts per triode​

So the 680r unbypassed drives the tank ~4x harder?

Is that really necessary? Was blackface reverb so much weaker?

The lower value bias resistor increases the valve’s idle dissipation by that amount; anode dissipation at idle does not equate to signal power output.

The bypassed cathode causes the preamp to overdrive in an unpleasant manner. Due to V3 grid clipping at lower signal levels than it occurs at the 3rd stage (V4b). With the master volume of the later SF TRs, that became apparent. Previously it had been masked by output valve grid clipping.
With a 2k2 bias resistor, removing the cathode bypass may have caused excessive degenerative feedback, reducing the driver’s gain / power output too much. Reducing the bias resistor value would restore much of that, whilst still preventing grid clipping at V3 from contaminating the signal path at too low a signal level.

[22uf]

I’m confused by the above, it doesn’t seem to align with TR AB763?

Sorry. I had the hangover from hell this morning.
I have:

V1B & V2B cathode: 820R
V3 cathode: 2.2k
V4 cathode: 820R
V5A cathode: 2.7k
All bypassed with the usual 25/25's.

I think thats correct? I know its not silverface values obviously. I'd read that the BF values were a lot easier on the 12AT7?
~30% dissipation etc.

2deaf's schematic has a 1.2k for V3.

The lower value bias resistor increases the valve’s idle dissipation by that amount; anode dissipation at idle does not equate to signal power output.

The bypassed cathode causes the preamp to overdrive in an unpleasant manner. Due to V3 grid clipping at lower signal levels than it occurs at the 3rd stage (V4b). With the master volume of the later SF TRs, that became apparent. Previously it had been masked by output valve grid clipping.
With a 2k2 bias resistor, removing the cathode bypass may have caused excessive degenerative feedback, reducing the driver’s gain / power output too much. Reducing the bias resistor value would restore much of that, whilst still preventing grid clipping at V3 from contaminating the signal path at too low a signal level.

Ahh. Ok. Would this be most apparent when you crank the channel volume and have the master on 3-4 ish? I just set the master to 10 and forget about it if I am honest.
What do you recommend?

[pdf64]

With those typical settings, don’t worry about it 

[22uf]

With those typical settings, don’t worry about it 

 

Uhm quick question since I have your attention..

This blackface doghouse board has "extra" jumper wire.



Mine didn't have..

Do I need to add the ones circled in blue? NB the missing jumper down the other side IS present on mine.

[pdf64]

The 0V common layout on Fender amps with the remote doghouse is so ‘non-ideal’ that it’s kinda like there’s no point trying to knock it into shape.
If it works fine (which it most likely will) then don’t worry about it.
If it doesn’t, then it’s something of a turd polishing exercise, ie pointless, trying to make it fit how it ‘should’ be. eg there’s most probably a bigger problem somewhere.

[22uf]

Swapped out the old bias caps and added a bias adjust. I swapped the 15k for a 10k and added a 50k linear pot in series.




Is this all looking OK? The pot itself is not grounded, just the wiper. Is that OK or should I run a ground to the pot (its held on with the tension of the two solder joints and a velcro pad).

Edit: Meters out at either end at between 10kohm and 60kohm so I guess that's WAD. Am I right in thinking that if the wiper failed, the bias would just fall back to 10kohm and therefore I wouldn't loose bias completely in that event?

[22uf]

Had the amp up on the variac today.
445v on the plate to cathode.
I put in some 1ohm resistors on pin 8 instead of the braids to ground.
The bias pot tops out at 30.4ma accross those.

I calculate that using Rob's tool to 52% dissipation.

It sounds great!

I'm not sure I can be bothered to change the 10k resistor to get more bias..

I have a Hantek 2D42 on order ($100 why not). I guess I'll change that and try 60-70% dissipation then. When I can see it on a scope.

I finally also fixed the reverb. After testing *everything* it turned out to be a combination of V3 being more or less dead and some heavy corrosion on the tank jacks. Lots of contact cleaner later.. reverb! Not as lush as the modern tank I pulled from the 77 twin to test with. But very "vintage" sounding.

One oddity was the 220k resistor on the reverb output socket to ground had dropped to 25k. Never seen that before.. 

The power transformer does have a bit of a buzz. It really doesn't seem to like 242v out the wall socket. It falls silent at 230 on the variac.

Final task will be to figure out some kind of additional fusing. I'm not liking the idea of relying on 1w screen grids to act as fuses.

Then I'll blackface it I guess.

The amp has a lovely sweet sound that the 77 doesn't have. The 77 however has a HUGE amount of punch and presence this one lacks.
Swings and roundabouts. For rock n roll I think I'd prefer the 135w amp.. ! I have to say, the hate for the UL amps is crazy. The clean tone on BOTH amps is so similar as to be indistinguishable in a mix. Both are superb amps and I think the hate driving down the prices makes them a bargain if you're in the market!

Thanks again for all the help/advice. 

[22uf]

Just doing the blackfacing of the PI and I'm not sure what this extra 0.1 cap does?
Its not on the schematic for ab763 or the aa270?



It seems to be attached to the power tube grid leak resistors and the NFB resistor?

[sluckey]

Talking about that disc cap? Sure it's not a .01?

[22uf]

Derp! Sorry. I'm still only two weeks old when it comes to schematics. 
What function does that cap serve?

[sluckey]

no idea

[22uf]

no idea

The amp sounds better with it in.
Just for fun I doubled it up with second .01uf cap and it was a low pass filter at that point.

Amps all done now. I only have one remaining issue..

Please excuse the terrible phone footage.

Power T is really audibly buzzing like an aquarium and getting VERY hot (I have a thermocouple on order to get some precise measurements but too hot to touch) after 30 mins or so.
This is with basically no signal going through it (volume on 2-3 ish).

The amp voltages are all OK and stable. The (supposedly) 100r resistors on the back of the pilot light read 50ohms each - I though they may be open.
I have new power tubes installed and have checked all the draw on the preamp tubes and its all tickety boo.
Nothing is "stressing" the PT as far as I can tell.

The amp idles at ~133w and never draws more than 0.8a-1.0a at the wall even cranked.

I can't find any electrical reason for the PT to be getting so hot .. other than perhaps .. because its an export model the switch is faulty and its not set to 240 even though it says it is. I guess I'll hardwire that for 240 and re-check..

Any thoughts? How much "buzz" is too much? I really don't want to have to buy a PT.. they are mega expensive here ($300). But likewise I don't want to risk a blow out or meltdown on stage.. 

The 77 twin buzzes a bit (not as loud) but that PT never gets hot to the touch.