1965 Deluxe Reverb--Filter Cap Issue--Blowing Fuse

[wsscott]

I biased the amp today.  It has 2 6V6GT's and has a plate voltage of 476VDC.  So I used plate dissipation of 14 watts, and so at 70% that's 20.58 mA bias.  One of the tubes is reading 23.5mA and the other is 20.9mA.  I think its running cooler than before with this setting.  My buddy may want it higher.  What bias is generally used/recommended for an original 1965 Deluxe Reverb Blackface?

I also took readings on the filter caps. They are as follows:
A pair in parallel---457VDC
B--455
C--355
D--288

They're pretty close to the schematic, but A & B are both at the upper limits of the 450V caps that were previously installed.  Should they be changed to 500's or fine where they are?

Thanks.

[sluckey]

How can you have plate voltage of 476V when node A cap is only 457V???

[wsscott]

Maybe my method was faulty (most likely). I put red lead on pin 3 and black to chassis ground.

[wsscott]

Ok, with the red probe on Pin 3 and the Black on Pin 8 I read 450VDC.

So do you think a 70% bias setting at 21.8 for this amp is good?  I think it previously was between 13-14 mA.

[jordan86]

Is your 70% based on 12w or 14w of plate dissipation? Most NOS stuff says 12w max on their data sheets. Newer production ones like EH and JJ spec 14w.

That said 21ma is about perfect for 450v on a 14w tube. For 12w valve 70% will be closer to 19ma.

Either way, you’re in good territory.

[wsscott]

Its based on the 6V6GT's Soltek Made in Russia tubes that are being used. I don't know the specs on the tube, or how the owner came about getting them, but 14W is what came up on the Weber Bias Calculator for a 6V6GTA tube, and on the Ron Robinette site that does have the 6V6GT listed at 14W.  I don't know why is had previously been set so low.  Thoughts?

[jordan86]

No clue. Would need to ask owner.

[wsscott]

Interesting that the Fender AB763 Schematic shows “Fixed Bias -35V”, and it’s showing 6V6GT tubes.

Maybe Sluckey can tell us what’s going on.

[Willabe]

Back then, the tubes Fender was using were pretty consistent. It was the golden years of tubes.

Because the tubes were consistent, and he was using the same tube from the same manufacturer, RCA, they could set the -bias a little on the low side and not have to check every amp for -bias, saved time. If they put in a little hotter tube, it would most likely not red plate because they had set them on the cool side. If he was changing tube brands, that would probably been different. That's what we have to do now.

Fender also wanted clean sounding amps, so he was not interested in biasing the tubes hotter.   

[wsscott]

So is one biased at 20 hotter than one biased at 35. I realize they’re negative numbers but nothing I thought about until now.

[sluckey]

Back then... repair shops were not concerned with bias balance, 70%, etc. If the schematic said -35V, the bias pot was set to -35V, and the amp went out the door.

So is one biased at 20 hotter than one biased at 35. I realize they’re negative numbers but nothing I thought about until now.

What are you talking about? I think you may be confusing a couple things. There are two things at play here... the negative voltage on the grid and the milliamps of current flowing through the tube. The negative voltage on the grid controls the current flowing through the tube. Higher negative voltage on the grid results in lower tube plate/cathode current (mA).

-35V on the grid might result in 20mA plate/cathode current.

[wsscott]

Yes I was. Thanks for clarifying it for me. I've learned so much on this forum. Really great!

[PRR]

Truly a 1965 amplifier? Line/Neutral distinction was not generally observed on plugs or inside chassis. Is the chassis outlet even un-equal slots? My 1961 house didn't have that.

[wsscott]

Yes its a real 1965.  The outlet does have different sized openings.

[sluckey]

Hot wire should connect to the narrow opening.

[dude]

Ok, with the red probe on Pin 3 and the Black on Pin 8 I read 450VDC.

So do you think a 70% bias setting at 21.8 for this amp is good?  I think it previously was between 13-14 mA.

The plate voltage should be the about same whether, the black probe is on pin 8 or chassis ground, unless you have a larger than 1 ohm resistor between pin 8 and chassis. On an original 65 Deluxe a stranded wire was soldered to pin 8 and ground no biasing one ohm resistor was used.
I assume you have cut that ground and added a 1 ohm R...? Then the voltage to pin 8 or ground using black probe should be the same, check your resistor value, make sure it's 1 ohm and not 100 or something, very important. 

[wsscott]

Since I’ve not owned Fender amps, I’m confused on the Standby switch. The amp I’m working with has you flip the switch to Standby when you want to power up the output tubes and play. That’s counter-intuitive to me. Is that the way it’s supposed to function?

[shooter]

the music world should have NEVR used the word standby WITHOUT the word operate!
Standby typically removes HV from the PA section, when you go to operate the HV comes on, ready to play/transmit etc.


in guitar world standby ON should be HV OFF
standby OFF should be HV ON
which IS counter intuitive!

[wsscott]

Whew, I thought maybe there was a wiring problem. Glad it’s not just me!

[shooter]

 
that's what happens when you give an artist, technical terms 

[Willabe]

You can also think of it as in standby mode and in play mode.

[wsscott]

Since the A Node caps are reading 457 VDC in Play mode, and 500 VDC in Rest mode, and the 22uf caps are rated at 450 VDC, do you think they should be replaced with 500 VDC caps, or lower the voltage someway?

[HotBluePlates]

... has a plate voltage of 476VDC.  ...  One of the tubes is reading 23.5mA and the other is 20.9mA.  ...  What bias is generally used/recommended for an original 1965 Deluxe Reverb Blackface?

I also took readings on the filter caps. ... They're pretty close to the schematic ...

The plate voltage with the new PT is crazy-high.  But you probably won't find better.

Until Nov. of last year, I had a 1964 Deluxe Reverb.  Its 6V6 plate voltage was just under 400vdc (about 394vdc) with 120vac applied to the PT primary.

[jordan86]

Did you use a Hammond 209BX? Wondering why the secondary is so high? Is your wall voltage high? Hammond specs 330v secondary. Are voltages pretty high on the preamp plates as well?

You could try a a different rectifier? 5V4 or 5U4 would drop you to more “period correct” voltages. If you have the Hammond it would support a 3A rectifier, so 5U4 is an option. But the 5V4 is a always great to me.

I’ve had good luck finding NOS 5v4 tubes on eBay for less than $10 each.

[wsscott]

My buddy wanted to use the Mercury Magnetics FDP-20 which the company said was the replacement they make.  It is rated for 350V secondary.  So that's certainly part of it.  The current rectifier tube is a GZ34.

[jordan86]

I’d probably try a different rectifier. That way your buddy can feel the satisfaction of buying the swanky expensive components and still have usable voltages 

[wsscott]

I suggested the Hammond you mentioned, in part because I’ve used it in 2 of my amps, but he was impressed with its reviews, has a really good ear, and uses it in his studio for guest musicians who really liked the sound before the PT died. So he’s hoping this will preserve the old sound. Other than this and a complete recap of all the electrolytics before he got it, it’s circuit and components are original.

[dude]

457v is way high, higher voltage will make the amp sound sterile. Either the new PT is  faulty, or your wall voltage is way high, like over 130vac…? The voltages on the schematic were marked when wall current was 115vac, now most wall current is around 120, so expect maybe up to 10vdc B+ more.
You shouldn’t have to screw around changing rectifier tubes to lower voltage. Contact MM and talk to them, might be a bad PT.
I also suggested you check your biasing resistors, make sure it’s 1 ohm. If they are, then any mA reading more than 5 points off means your tubes are likely unmatched, the more the worst.

[wsscott]

I just checked again.  The bias resistors that were previously added each read 1.3 ohms.  And the A Node caps read 457 VDC.  I connected the black lead to the chassis and the red lead to the positive terminal of the cap.  Both of the A Node caps read the same.  Wall voltage is 121.7 VAC.

[mresistor]

As Sluckey reminds us..  the notes state +- 20% .   That would mean +/- 83vdc.  It appears to me and in my experience 457vdc plate V is not really high.  Also with standby engaged there is literally no current flowing through the power supply, so the higher voltage on the paralleled caps shouldn't be a problem.  If you wanted to you could put the two caps (reservoir caps) in series with balancing resistors like the Super Reverb, then you'd have 900vdc voltage rating and that wouldn't be hard to do.  Or get some 600vdc rated caps from JustRadios.  Also wall V in this country can vary as high as 123vac, but 120 is spec.  The .3 on your reading for the 1 ohm resistors is probably due to your meter.  Touch both leads together first and see what the meter reads and then account for it in your readings.

[wsscott]

Oh, and I checked the voltages again at Pins 4 & 6 of the rectifier where the red secondary connect, and they each read 353 with red lead to the pin, and black to chassis.

[dude]

Meters don’t read 1 ohm’s alone very well.  Use a 100 ohm, check reading, then add the 1 ohm R in series read the two, when you get only 1 ohm increase your good to go for an accurate 1 ohm resistor.
Check the B+ with the amp S/B in operation mode. That is your B+. You can always use a bucking external transformer to get vintage wall current.

[72Blazer]

No wonder I alwas seem to get 1.4 Ohm readings when measuring a 1 Ohm resistor.  I'll have try this.

[PRR]

Just read a "no resistance". Rub the probes together. Many meters will read "0.3Ω" for a dead short.

That's actually "valid", there's a part-Ohm in the leads, switching, and fuse. Some meters let you calibrate it out, many do not.

ALWAYS cross-check your meter for a near-zero reading before any small measurement. Low battery, worn fuse, many problems can throw-off the reading. If 0.3Ω is the "normal" reading on a short, that's fine, just remember it.

[HotBluePlates]

So you might not be too "crazy high" but definitely higher than the upper end of vintage Fender amps.  The 121.7vac at the outlet will raise B+ some, but I still suspect low-ish winding resistance.

... Mercury Magnetics FDP-20 which the company said was the replacement they make.  It is rated for 350V secondary.  ...

Mercury Magnetics makes a laundry-list of power transformers as "Deluxe Reverb replacements."

My 1964 Deluxe Reverb had a 125P23B power transformer.  Mercury Magnetics says their model FBFDR-P corresponds to that transformer, but their specs show an output of 350-0-350v.  My amp's B+ winding had 334-0-334v on the B+ winding (and resulted in 394vdc at the 6V6s with an idle current of 24.9mA per tube).  I'll have to look through my notes to see if I have the PT winding resistance recorded somewhere.

Someone with a late-64 125P23C power transformer in their 1965 Deluxe Reverb measured 348-0-348v on their B+ winding and got 442vdc at the 1st filter cap with a wall voltage of 121vac.  The resistance of the B+ winding (Red-to-Red) was 122.3Ω.

Someone with a late-66 025130 PT in their silverface late-67 Deluxe Reverb measured ~350v a.c. output from the B+ winding.  With a British Mullard GZ34, their B+ at the 1st filter cap was 437vdc.  Their B+ winding measured 134Ω from Red-to-Red.

[wsscott]

Given the higher voltage this amp now has, is there a substitute for the GZ34 that's currently being used that would put out lower DC Voltage to the A Node.  My buddy is having me change the caps to 500V, but I thought I would also see if a change in the rectifier tube would be a good idea.  Thanks.

[jordan86]

Any of these would work with your PT, as the spec sheet says it provides 4A on the 5V secondary for the rectifier.

5V4: Will drop roughly 10-15 more volts.
5U4: Will drop roughly 30 more volts
5Y3: Will drop closer to 40-50 volts.

5V4 are NOS. Others can be found in current production. A good NOS 5Y3 may be a cool place to start if you’re looking for more vintage correct voltages.

Most of the big AB763’s (twin, showman) operate at 460V on the power tubes, so it’s not terribly off. The Deluxe was usually down around 410-425v though.

[66Strat]

To counter the high B+ voltage, you could stack two electrolytic capacitors with balancing resistor across them. This would double the effective voltage rating at half of the capacitance rating. As it is now the two 16uF capacitors are paralleled to achieve an effective capacitance of 32uF. To maintain the same 32uF capacitance, one would need to stack two 64uF capacitors. Since 64uF is not a standard value, it is common practice to stack 70uF, 80uF, or 100uF capacitors to achieve effective capacitance values of 35uF, 40uF, and 50uF respectively.

Our host carries F&T 100uF/350V capacitors in his online store. These capacitors stacked would provide an effective voltage rating of 700V DC. They are located near the bottom of the page in the attached link.
https://hoffmanamps.com/MyStore/perlshop.cgi?ACTION=enter&template&thispage=Capacitors&ORDER_ID=!ORDERID!

[wsscott]

Thanks guys for all the leads you’ve contributed. I’m going to upgrade the caps, and then just see how it goes for now. But now I know what options are available.

The other issue with the amp before the PT blew, was that the tremolo sounded like a helicopter landing when it was turned on.  Well it doesn’t do that anymore and sounds great. You can hear some ticking in the background, but that seems to be part of the design. I don’t know what “fixed” it, unless it might have had something to do with the second A Node cap not being grounded. Anyway, it’s working now.

Do you think that same grounding issue could have caused the PT to fail, or maybe was it just due to old age?

[pdf64]

the notes state +- 20%  .   That would mean +/- 83vdc.

Given the normal loadings, surely common sense tells us that note can’t reasonably be intended to be applied to every voltage on the schematic?
My interpretation is that it applies to nodes whose voltage will be affected by normal valve variance, eg preamp anodes and cathodes, output valve Vg1-k bias.
Whereas voltages at / closely tied to PT secondary windings should be, by necessity and basic good engineering / manufacturing practice, within a couple of % of nominal.
As it would be ridiculous for a transformer whose voltages were 20% off nominal to be accepted by QA. And it would result in the other voltages noted above tended to vary by 40%.
Consider that valve heater voltage has at most a 10% tolerance.
And that mains supply voltage has at best a 5% tolerance, most regions are worse than that, getting on for near 10%.
Hence the spec on PTs has to have a fairly tight tolerance on loaded secondary voltages.
And it follows from that the tolerance on the rectified HT VDC at the reservoir cap can’t be much worse, a few % at most, when the PT is being supplied with its nominal intended mains voltage.

[PRR]

....surely common sense tells us that note can’t reasonably be intended to be applied to every voltage on the schematic?...

Agree.

Some things can be 20% out no problem. A grid-leak biased triode plate can set almost anywhere and work as good as ever.

Those voltages are about fault-finding. The most likely values in a specific amp are "about right" or "way out!" Like zero or full B+. Don't get hung-up on "small" discrepancies, is how I read it. Being 19% out may be odd but is probably not "sick".

Anyway we do see "wrong" values on schematics, because humans have errors.

[pdf64]

… we do see "wrong" values on schematics, because humans have errors.

Also the schematics available in the public domain are just a snapshot in time, designs get revised, but the revised document might not survive, or perhaps it was only ever an improvised amendment to production line copies.
Some Fender models only lasted a few months, hardly worth spending a draughtsmans time documenting revisions.
Whereas other models lasted years and we know there were changes (eg 5F6A Bassman), but only the initial documentation is available.

[sluckey]

Those voltages are about fault-finding.

And don't forget... probably more than 90% of the "electronic voltmeters" of the time have never been calibrated since they left the factory. That statement may have had more to do with CYA than actual voltages.

[mresistor]

To counter the high B+ voltage, you could stack two electrolytic capacitors with balancing resistor across them. This would double the effective voltage rating at half of the capacitance rating. As it is now the two 16uF capacitors are paralleled to achieve an effective capacitance of 32uF. To maintain the same 32uF capacitance, one would need to stack two 64uF capacitors. Since 64uF is not a standard value, it is common practice to stack 70uF, 80uF, or 100uF capacitors to achieve effective capacitance values of 35uF, 40uF, and 50uF respectively.

Our host carries F&T 100uF/350V capacitors in his online store. These capacitors stacked would provide an effective voltage rating of 700V DC. They are located near the bottom of the page in the attached link.
https://hoffmanamps.com/MyStore/perlshop.cgi?ACTION=enter&template&thispage=Capacitors&ORDER_ID=!ORDERID!

And I actually said that in reply #79

[1blueheron]

Have been following this thread with interest.  Seems like a great tutorial on troubleshooting and replacing a faulty PT, along with all the potential pitfalls one might run into along the way.  I like it!

As for this amp ever sounding the same as it did to a person who thinks they have golden ears... that might be wishful thinking but it never hurts to wish upon a star, particularly a star that would otherwise be dead forever.

I find it interesting how many of the amps that have a "sound to die for" often do die shortly after that sound to die for is obtained.    I believe the famous "purple-plexi" met with this demise?  Went from hero to zero in one puff of smoke and the mystique could never be reproduced.   Psychological or real remains to be proven.

Regardless, changing to a higher voltage PT, and changing bias I would think, will lead to a very different sounding amp.  Which leads me to ask the question... Why change more than is necessary if you want it to sound like it used to?  Sometimes it is the deviation from "factory" or "right" that makes an amp special in the ears of its owner.  Or perhaps it is how many beers were spilled in it, or the special resistive qualities of nicotine rivers that have coursed through its pots.

[wsscott]

Yes, you’re suggesting what I suggested to my buddy. Leave everything else alone, but change the filter caps to F&T 500V since the originals were replaced anyway with a brand I never heard of.