1967 Pro reverb bias problem

[Michael1]

Hi guys,
 I had a JJ power tube short out last week on a gig-bought a new set of Winged C 6L6's. I have always set the bias with the transformer shunt method and never had a problem till now. I can't get the current draw below 49-50ma. I have changed the resister on the bias pot up and down in value and it makes little difference. I have also cleaned and retensioned the tube sockets. All resisters on tube sockets check fine (470-1500). Voltage on pins 3,4 and 6 on both power tubes is 480v. Voltage on pins 1 and 5 is -54v. Voltage going into bias pot is -63v. Any tips would be appreciated=thanx in advance!


                                                                                                                                                      Michael

[sluckey]

Voltage on pins 1 and 5 is -54v. Voltage going into bias pot is -63v.

With the amp in standby mode crank the bias pot from one extreme to the other. What voltage do you measure on pin 5 of each tube at the extremes of the bias pot? One extreme should be -63vdc which should be enough to drop the tube current down.

I would put a 1Ω resistor between pin 8 and chassis and check the voltage across the 1Ω resistor. This will give you an accurate reading of tube current.

[Willabe]

Pro Reverb 1967, are you sure its a bias adjust pot and not a bias balance pot?

What model exactly do you have? Look on the tube chart glued on the inside of the speaker cab.  

This should be silver face models and both have bias balance;

http://www.el34world.com/charts/Schematics/files/fender/PRO_REVERB_AA1069.pdf

Edit; As Sluckey pointed out this model has -bias adjust.

http://www.el34world.com/charts/Schematics/files/fender/PRO_REVERB_AA165.pdf

Edit; there's 5 pro reverb schemos in Doug's library, all have bias balance. No AB763 Pro reverb which should a bias adjust pot circuit.
some AB763 amps (BF) were built after 65 using old parts in the factory. So you could have a BF 67. My SR was a BF AB763 built in 67.  

Edit; 4 have bias balance, AA165 has -bias adjust pot.

             Brad      

[Willabe]

As Sluckey said verify the available negative bias voltage 1st without any power tubes in. (Well he said in stand by mode so -bias should be there, but no B+dcv applied to the power tubes plate, so you don't damage/burn up the power tubes.)

Then once you do that, I'd still use a cheap set of tubes, if you have them, not the Winged C's until I was sure it was working right.


            Brad    

  

[Michael1]

Thanx for the quik reply's! i just checked pin 5- voltage goes from 30 to 62-standby mode-no power tubes in. The amp is definitely a 1967 model (blackface)-straight bias adj-no bias balance. i plan on putting the 1ohm resister on pin 8-gotta order them and some other stuff from Doug. This puzzles me because i have always set it this way and never had a problem ( with the exception of the JJ's-had to change resister on bias pot to make them bias up) it's back to stock now.

[sluckey]

there's 5 pro reverb schemos in Doug's library, all have bias balance.

That AA165 model has bias adjust (not balance) just like an AB763.

I suggested the 1Ω resistor method so you could get a comparison with the OT shunt method. I have more confidence in the numbers obtained with the resistor method.

[Willabe]

That AA165 model has bias adjust (not balance) just like an AB763.

Yes your right, thanks for pointing that out so I edited it.


            Brad     

[Willabe]

Ok, from the book, The soul of tone, Celebrating 60 years of Fender amps, p. 278, CBS BF Pro Reverb was made from '65 to '67.

Same book, p. 255, almost all models went from brown to black (face plates) in '63.

Same book, p. 309, Starting in mid '68, CBS altered the bias in the Pro Reverb (and other amps) to a bias balance pot. It was abandoned by late '69.


          Brad      

[Willabe]

Voltage on pins 1 and 5 is -54v. Voltage going into bias pot is -63v.

i just checked pin 5- voltage goes from 30 to 62-standby mode-no power tubes in. The amp is definitely a 1967 model (blackface)-straight bias adj-no bias balance.

Something's wrong.  

With power tubes in and in play mode you can only get to -54dcv on pins 5? 8 -dcv difference. Negative bias to the grids should draw NO current to drop any -dcv.

Makes me think the PI coupling caps to the power tubes grids are leaking dcv? Why, when your in stand by no dcv is applied to any of the tubes plates. So the leakage doesn't show up until your in play mode which applies dcv to the PI plates.

I'd try leaving the power tubes out, have the bias pot set for max -dcv, clip the meter on 1 of the power tubes pin 5, meter set for dcv and then flip the stand by to play and take a quick reading for a change in -dcv. Then do the other sockets pin 5. If it shows a change in -dcv then 1 or both PI coupling caps are leaking dcv.

I say quick reading because with no power tubes in drawing current the B+ power supply rail will be way higher than normal and will exceed the filter caps dcv rating. Also the coupling caps take a little time to charge up before they leak any dcv.

So it might be best to do this instead;

Just lift the leg of the PI coupling cap that goes to pin 5 and test for dcv.

Any dcv leakage voltage will work against the -dcv bias.


            Brad    

[PRR]

> Negative bias to the grids should draw NO current to drop any -dcv.

The meter draws current.

Without thinking, I'd suspect his meter is 1 Meg impedance. While many meters are 10Meg, some are 1meg (and some lower).

[Michael1]

Ok=i pulled the leg off of both pi coupling caps and checked for dc--pretty erratic readings but did show dc volts-moving up and down on meter-spiking at 1/2 volt. Constantly moving up and down though. No power tubes in -these are orange drop caps-they have been in there for awhile

[HotBluePlates]

Put a fresh battery in your meter.

If that doesn't work, try a new bias cap (although you didn't say you had hum, which a failed bias cap would cause).

With no tubes and the coupling caps lifted, there should not be any a.c. or any variance of the d.c. bias voltage at pin 5 of the output tubes.

[Willabe]

Sluckey and PRR know what they're talking about so....

Put in the 1R's and see what you get.

1/2 a dcv leakage is nothing even with a great meter. All caps have a very small amount of dc leakage as far as I know. I don't think those caps are bad and are not causing a -bias problem.

All power tubes will draw a different amount of current. Might be a hot set of tubes and a somewhat false reading using the shunt method and your meter.


            Brad    


Edit; While typing HBP posted, I would defer to him. I always forget the fresh battery.         

[Michael1]

Thanx again guys for the help! i just checked pin 5 on both power tube sockets (no tubes-Pi caps disconnected at one end) Im getting -30 Vdc.

[HotBluePlates]

What model meter are you using?

[sluckey]

Im getting -30 Vdc.

So crank the bias pot to the other extreme where you get -62vdc.

You never said if you have any red plates or any other problems with the high current. Do you? Or are you just concerned about the numbers?

[Michael1]

Thanx again everybody! Ok-no tubes in-Pi coupling caps disconnected-I get -30 vdc at one end and -62vdc at the end of sweep/ Meter is an older micronta digital. I have a Fluke coming though. Mainly concerned about the numbers as i have had this amp and meter for several years and i have changed power tubes 5 or 6 times and it always would let me adjust to 32ma. With the exception of the JJ's-had to change the resister on pot so they would bias up. I will be changing the bias cap and putting in the 1ohm resister this week like you guys said to do.  Thanx!

[sluckey]

Plate current at 50ma and plate voltage at 480V is only 24w static dissipation. 6L6s can do 30W so you're at about 80%. I like to set fixed bias amps at 70%. But I do like to be able to adjust up or down from 70%, rather than be at one extreme of the bias pot.

[Michael1]

Reconnected the Pi caps and put new tubes back in- got 60vdc on pin 5. Pins 3 and 4 are at 475 vdc

[sluckey]

How much plate current?

[Michael1]

I'm getting 46 ma at lowest setting with pot - transformer shunt method-dont have the 1 ohm resisters yet for checking voltage drop

[sluckey]

I'm looking for a PAIR of numbers----, ie, plate current and plate voltage. So, when you have 46ma current what is the plate voltage? And when you have 475v plate voltage, what is the current?

[HotBluePlates]

... Meter is an older micronta digital. ...

Then it is probably a 10MΩ input impedance, based on some manuals I was able to find online.

... I get -30 vdc at one end and -62vdc at the end of sweep ...

I'm getting 46 ma at lowest setting with pot ...

I just don't see the new tubes being so high Gm that you'd get 46mA with -62v of bias. It would be interesting to see you go back and forth with each meter to see how they compare.

I'm wondering if your milliamp function of the Micronta is out of calibration, and just reading higher than the true current.

Still, 42mA and 475vdc is still only 19.95w and ~66% plate dissipation, which is fine for 6L6GC's. But as Sluckey is implying, when current goes up plate voltage comes down, so you have to take both measurements consecutively.

[Michael1]

It's 475v plate @ 46ma. it's entirely possible that meter is out of calibration-Ill check it again when i get my fluke-and i'll also check with the 1ohm resister installed. You guys are great-thanx a million for sharing your knowledge. I left amp on for  a few minutes and watched for red plating-didnt see any.I was also confused why changing the 27k resister on bias pot didn't make much difference

[sluckey]

It's 475v plate @ 46ma

OK. That's 73% of max dissipation. I'd call it good. It will still be interesting to compare to readings with the 1Ω resistors. Let us know when you do that.

I was also confused why changing the 27k resister on bias pot didn't make much difference

You never said how much you changed the value of that resistor. Making it smaller will only increase your current reading and that's the opposite way from where you wanted to be. You have -63v coming into the bias pot. That's the max voltage you can get which will result in the lowest current reading you can expect. So, making that resistor larger is still not gonna get the bias voltage up much more than -63v.

But, I think that -63v on pin 5 should bias those 6L6s really cold (much less than 50ma). I'm still not confident in your shunt method readings. Stay in touch.

[Michael1]

Thanx so much everybody--ill give an update when i get the fluke and the resister in. I went down to 10k on the bias pot and up to 30k-made very little difference.

[Michael1]

Thanx again for all the help everybody! Got a new meter (Fluke 27/FM) and installed 1 ohm resisters on pin 8 of power tubes. Readings I get are:

Pin3-484v
pin4-484v
pin5-66v
Pin6-483
Pin8-43.5mv-after resister .2 mv

Second power tube:
Pin3-484v
pin4-484v
pin5-66v
Pin6-483
Pin8-46.9mv-after resister .1 mv

This is with the bias pot all the way ccw

Thanx again!

[Willabe]

Glad you got it fixed.

Did you try swapping the 2 power tubes with each other to see if they match closer? Their fine as is but why not try and see?


            Brad      

[sluckey]

Your current readings with the 1Ω resistors are right in line with your readings for the shunt method.

[Michael1]

Swapped the tubes and the readings followed the tubes-they were supposed to be a matched pair-is that close enough? Can't thank you guys enough for the help-it is much appreciated!


                                                                                                                     Michael

[Willabe]

I would say yes, their fine.


           Brad     

[Michael1]

Thanx again everybody-a fixed amp is a wonderful thing!