'62 Tremolux nasty oscillation on power up

[seven]

Hi,

I have a friend that bought an old '62 Fender Tremolux (6G9-B) that he
wants to revert back to stock. Some hack put in a Super Reverb type
power tranny and choke, and a Twin output transformer. Also,
aforementioned hack attemtped what looks like a negative feedback
control or effects loop of some kind

Here are a couple of before shots:

http://img.photobucket.com/albums/v3...B/trem1_b4.jpg
http://img.photobucket.com/albums/v3...B/trem2_b4.jpg
http://img.photobucket.com/albums/v3...G9-B/pi_b4.jpg

Looking at the amp and referencing a schematic, I figured I could help
him get his amp back to stock (having worked on a few Fenders of my
own successfully).

Here are some shots of the amp stripped of the unoriginal pieces:

http://smg.photobucket.com/albums/v3...G9-B/trem3.jpg
http://smg.photobucket.com/albums/v3...G9-B/trem4.jpg

He purchased a Mercury Magnetics power tranny FBLT-P (based on the
original #68409) and output tranny FBLTR-O (reference
http://www.mercurymagnetics.com/page....htm#Tremolux_)
in an attempt to get the amp back to its former self.

I installed them in the amp and also replaced the filter capacitors.

I also noticed that some of the resistors in the phase inverter look
to have been replaced by I guess the aforementioned (having incorrect
values and obvious signs of resolder).

Referencing the schem, it looks like the 4700 ohm resistor was at one
point a 100 ohm resistor, the 6800 ohm resistor was at one point an
82k ohm resistor and the 820 ohm resistor was at one point a 470 ohm
resistor. I replaced all of these according to the schematic. I did
notice something weird between the layout and the schem, the feedback
resistor on the 6G9-B layout denotes 56k, whereas the 6G9-B schematic
says 100k. My friend's amp is an 82k, and *looks* to be original, but
it's hard for me to say. Just thought that was weird... not sure if
82k would make a difference here?

Also, I noticed that the 0.0001uF cap across the two PI plates as
shown on the layout and schematic is missing from the amp and doesn't
appear it was ever installed there as the solder joints look untouched
(!).

Anyway, so I powered the amp on and I notice that as soon as it warms
up, there is a horrible squealing coming out of the speakers.

I have a clip of it (m4a/AAC audio file recorded on my iPhone,
playable in WinAmp and I am sure other media players). You can hear my
power on, the tubes warm and the squeal comes in at around 0:14
whereby I immediately shut off the amp:

http://rapidshare.com/files/297631831/Squeal.m4a.html

OR

http://www.megaupload.com/?d=GJOZWI62

So, I pulled the phase inverter (even the power tubes, for that
matter) and the sound goes away, so it seems to be isolated to that
stage.

Yes, I wired the OT primary correctly. Even flipped the wires and got
a motorboating sound so I reverted it back.

The voltages are consistently a little high throughout the amp
(10-20%, I've got 120VAC coming out of the wall), so there's nothing
indicative to me regarding the DC voltages for the various tubes.

Here's the after shots (maybe someone will spot something obviously
wrong?):

http://smg.photobucket.com/albums/v3...-B/trem5_a.jpg
http://smg.photobucket.com/albums/v3...-B/trem7_a.jpg
http://smg.photobucket.com/albums/v3...-B/trem8_a.jpg
http://smg.photobucket.com/albums/v3...-B/trem9_a.jpg
http://smg.photobucket.com/albums/v3...B/trem10_a.jpg
http://smg.photobucket.com/albums/v3...B/trem11_a.jpg

Is there a cap in the PI that could be bad?

How does my layout look?

Any suggestions welcomed and appreciated.

Thanks for your time,
Chris

[Prairie Dawg]

Originally Posted by seven

Hi,

I have a friend that bought an old '62 Fender Tremolux (6G9-B) that he
wants to revert back to stock. Some hack put in a Super Reverb type
power tranny and choke, and a Twin output transformer. Also,
aforementioned hack attemtped what looks like a negative feedback
control or effects loop of some kind



Anyway, so I powered the amp on and I notice that as soon as it warms
up, there is a horrible squealing coming out of the speakers.

I have a clip of it (m4a/AAC audio file recorded on my iPhone,
playable in WinAmp and I am sure other media players). You can hear my
power on, the tubes warm and the squeal comes in at around 0:14
whereby I immediately shut off the amp:


So, I pulled the phase inverter (even the power tubes, for that
matter) and the sound goes away, so it seems to be isolated to that
stage.

Yes, I wired the OT primary correctly. Even flipped the wires and got
a motorboating sound so I reverted it back.

The voltages are consistently a little high throughout the amp
(10-20%, I've got 120VAC coming out of the wall), so there's nothing
indicative to me regarding the DC voltages for the various tubes.


Is there a cap in the PI that could be bad?

How does my layout look?

Any suggestions welcomed and appreciated.

Thanks for your time,
Chris

The only thing I can suggest is to take and print out a schematic and tape it to the wall next to your workbench, and then proceed to validate every component and circuit path from one end to the other, marking off each path and component with a highlighter as you go. This'll do two things-one, it validates the build and second, it gets you into a real up close and personal visual inspection that can reveal all sorts of surprising things.

From what you've described the power section works. Does the volume control have any effect on the volume? That could narrow it down even further.

Did the amp work at all when your friend got it? Did your friend buy this amp as a project that somebody gave up on? I have had a couple amps that were like that-nobody could figure our how to fix em so they get passed around. There's no real answer except start from scratch, because 9 times out of 10 the problem we encounter can be traced back to the last person who was in there. That implies no disrespect, it's just a fact of troubleshooting, which is really an application of probabilities.

I think if I was you I would locate the guy who did the "mod" and break his knuckles-that hole in the chassis is hideous.

[loudthud]

Disconnect the 100K (or 56K) feedback resistor and see if that kills the squeeling. Looking at the pdf file for the 6G9-B on Schematic Heaven the schematic says 100K and the layout says 56K. This tells me there was problems in that circuit and changes were made to try to fix it. Normally the higher value would tend to be the one that didn't oscillate. YMMV. Does the amp have 0.1uF caps between the PI and power tubes?

[seven]

Originally Posted by Prairie Dawg

The only thing I can suggest is to take and print out a schematic and tape it to the wall next to your workbench, and then proceed to validate every component and circuit path from one end to the other, marking off each path and component with a highlighter as you go. This'll do two things-one, it validates the build and second, it gets you into a real up close and personal visual inspection that can reveal all sorts of surprising things.

OK, I've done this twice before, but the only thing I haven't done was verify the under-board to make sure all jumpers are in place. I *ass-u-me-d* they were all there in the places where the solder looked untouched.

Originally Posted by Prairie Dawg

From what you've described the power section works. Does the volume control have any effect on the volume? That could narrow it down even further.

You know, I didn't because the noise was so irritating and deafening. Something tells me I don't think it would make a diff: in between attempts at powering on, I had the volume full counterclockwise and then midway, and no difference in the oscillation nor the loudness.

Originally Posted by Prairie Dawg

Did the amp work at all when your friend got it? Did your friend buy this amp as a project that somebody gave up on? I have had a couple amps that were like that-nobody could figure our how to fix em so they get passed around. There's no real answer except start from scratch, because 9 times out of 10 the problem we encounter can be traced back to the last person who was in there. That implies no disrespect, it's just a fact of troubleshooting, which is really an application of probabilities.

The amp didn't work when he got it. My friend got such a good deal on it that it would have been foolish (!) had he not.

No disrespect taken, I wasn't the one who F'ed this amp up!

Originally Posted by Prairie Dawg

I think if I was you I would locate the guy who did the "mod" and break his knuckles-that hole in the chassis is hideous.

Preach on, brutha!

[seven]

Originally Posted by loudthud

Disconnect the 100K (or 56K) feedback resistor and see if that kills the squeeling. Looking at the pdf file for the 6G9-B on Schematic Heaven the schematic says 100K and the layout says 56K. This tells me there was problems in that circuit and changes were made to try to fix it. Normally the higher value would tend to be the one that didn't oscillate. YMMV. Does the amp have 0.1uF caps between the PI and power tubes?

Hi,

What I did before I read your post was simply unsolder the lead that comes off the output jack that connects to the 82k resistor (100k on the schematic, 56k on the layout) on this amp's board. And guess what? No oscillation. So I need to investigate and see why. BTW, it looks like the 82k resistor replaced whatever the original one was when the butcher modified the amp with a master volume or some kind of NFB control...

Maybe I'll try a 100k there and see what happens?

Yes, the amp has the two blue 0.1uF caps as it leaves the PI for the 6L6s...

Thanks for your help. I'll let you know what I find when I get a chance to check it out again in the next few days.

[Prairie Dawg]

Originally Posted by seven

OK, I've done this twice before, but the only thing I haven't done was verify the under-board to make sure all jumpers are in place. I *ass-u-me-d* they were all there in the places where the solder looked untouched.



You know, I didn't because the noise was so irritating and deafening. Something tells me I don't think it would make a diff: in between attempts at powering on, I had the volume full counterclockwise and then midway, and no difference in the oscillation nor the loudness.



The amp didn't work when he got it. My friend got such a good deal on it that it would have been foolish (!) had he not.

No disrespect taken, I wasn't the one who F'ed this amp up!



Preach on, brutha!

I guess what got to me was that it was so inartfully done-particularly when you figure if you wanted to do such a thing for some unknown reason you could get a surface mount power transformer and mount it by drilling some small holes.

It sort of simplifies things a lot, though, when you figure that your problem is somewhere in a 10 inch x 24 inch piece of real estate. Plus you've got some dandy iron to play with now.

[MWJB]

One thing jumps out at me as being wrong. You have changed the NFB load resistor to 4.7K from the 100ohm that was in there, BUT the NFB dropping resistor is stil 820ohms. Either go back to 100ohms for the NFB load, or up the value of the NFB dropper to 27K/56K.

[MWJB]

"BTW, it looks like the 82k resistor replaced whatever the original one was when the butcher modified the amp with a master volume or some kind of NFB control..." There is no 82K resistor in the NFB loop, grey/red/brown = 820ohms

[seven]

Originally Posted by Prairie Dawg

I guess what got to me was that it was so inartfully done-particularly when you figure if you wanted to do such a thing for some unknown reason you could get a surface mount power transformer and mount it by drilling some small holes.

Something tells me this was done sometime in the early 70s... all the EIA codes on the pieces that were changed were from that era +/- a year or so.

Re: the inartful work, I think whoever did this just used what they had. Not trying to justify the damage done, but no one back then imagined they'd be ruining a "vintage amp"...

[seven]

Re: the "82k", you're right. It was a 820 ohm resistor. I saw the layout diagram had 56k and the schematic had 100k and then I saw the grey and red bands on the resistor and for some reason ignored the brown band and in my mind concluded it was an "82k"... I'm getting old...

To make it a little simpler to understand, I've marked on this schematic in red what the resistor values were before I started working on the amp:

http://img196.imageshack.us/img196/3...emolux6g9b.gif

Yesterday I replaced the NFB dropping resistor with a 100k as per the schematic and it still oscillated, though not so bad (a little more quieter). It seem to be more of a motorboating type of oscillation (slower cycles) than a squealing, though there was still a little squealing.

So it seems like what I can do is to try a 56k for the NFB dropping resistor and see what will happen, although if what loudthud said is true, it probably would still oscillate with the 56k if it did with the 100k (?).

Or I can put the 820 back and change the load resistor to 100 ohm. This would be deviating from the schematic, though...

I'll report back in a few days unless someone else has some words of wisdom to schlep my way...

Thanks for the help!

[olddawg]

I have a bone stock 63 Tremolux in mint condition. As far as I know it has never been touched except to replace a few tubes. If a few gut shots would help you let me know. I'm not sure if I could do it instantaneously but at certainly in the near future if you still need a reference point.

[tboy]

Originally Posted by seven

To make it a little simpler to understand, I've marked on this schematic in red what the resistor values were before I started working on the amp:

The values in red would be correct for the AB763 blackface version of the amp, and should have worked just fine, so I'd be looking elsewhere for the problem.

Originally Posted by seven

Yesterday I replaced the NFB dropping resistor with a 100k as per the schematic and it still oscillated, though not so bad (a little more quieter). It seem to be more of a motorboating type of oscillation (slower cycles) than a squealing, though there was still a little squealing.

So it seems like what I can do is to try a 56k for the NFB dropping resistor and see what will happen, although if what loudthud said is true, it probably would still oscillate with the 56k if it did with the 100k (?).

Looking at other brownface era schematics for comparison, I think 56k is probably the correct value, but I still don't think that's the problem.

[seven]

Originally Posted by tboy

The values in red would be correct for the AB763 blackface version of the amp, and should have worked just fine, so I'd be looking elsewhere for the problem.

OK. You know, I didn't even bother to look at the AB763 schematic for this amp. I was under the impression that there were some fairly significant changes between the G and the AB series. Apparently not so much in the long-tail PIs both of these amps used... besides the fact that a 12AT7 was used in the AB763 instead of the 12AX7 used in the 6G9-B.

Note that while I reverted some of these resistors back to what the 6G9-B schematic indicates, I didn't revert the 820 back to a 100k (or 56k).

Originally Posted by tboy

Looking at other brownface era schematics for comparison, I think 56k is probably the correct value, but I still don't think that's the problem.

OK. What do you think the problem is? Or are you saying you're not sure what it is, but that it's probably not with the PI?

[tboy]

Originally Posted by seven

Note that while I reverted some of these resistors back to what the 6G9-B schematic indicates, I didn't revert the 820 back to a 100k (or 56k).

But you have done that now and it still oscillates, is that right?

Originally Posted by seven

OK. What do you think the problem is? Or are you saying you're not sure what it is, but that it's probably not with the PI?

If I knew what what the problem was I wouldn't hold out on you. I'm just saying that the blackface part values were OK, and the brownface values you have now are OK, so the problem isn't being caused by incorrect PI/NFB part values. It oscillated with more feedback (when you had the 820 in), and it also oscillated with less feedback (when you put in the 100k), and even though 56k would seem to be the correct value, I expect it will still oscillate if/when you make that change.

Since the oscillation stopped when you disconnected the NFB, I'd be inclined to try swapping the OT primary leads again, now that you have the correct values in the feedback loop.

[MWJB]

+1 on Tboys suggestion.

[seven]

OK, I've been meaning to reply to this last week, but forgot.

After restoring the 820 NFB dropping resistor to 100k, I still got oscillation. At this point, it just didn't make sense why the amp still oscillated. So I tried something that in my mind was crazy: I swapped the primaries on the OT.

Oscillation is gone!





Yes, I know many of you are rolling your eyes now. But I followed the MM wiring diagram and the official Fender layout diagram wrt the wire colors on the respective plates. I guess somebody at Mercury Magnetics was winding that output transformer at 4:50pm on a Friday... They were either tired OR thinking about

Needless to say, the amp is sounding great. Probably the best it's sounded in 35 years, given the EIA codes on all the previous components the "butcher" put in.

Thanks again for your help. I learned a lot.

[MWJB]

Following the wiring diagram means nothing when it comes to OT primaries...even when fitting say a later, Vibrolux OT to a Tremolux you may have to reverse primaries, lost count of the number of RI Bassmans I have seen with switched OT secondaries....if you have to reverse them, then you just have to reverse them