SF Fender Twin help

[Rich M]

O.K., I've been going through this amp for about 2 weeks now and it is making me nuts. I replaced all the electrolytics and did a blackface conversion and I think that helped the tone . I also removed the MV and put a PPIMV in there. The remaining issue with the tone is this kind of buzz overtone over loud signals,clean or dirty. I changed the tubes, cleaned up some ground wires (what a rats nest!) Nothing. I can't even seem to make it worse. I can tell the amp would sound great if I could get rid of this and I really want to figure this out, but at this point, I'm starting to look longingly at my Superbass on a chair waiting patiently for me. I read G. Weber's book to try and understand Fenders. The only posslble thing that I can find to try is increasing the swamp resistors on the OTs (they are 1.5K). Will this help. Man, I just want this to end ! Please help!

[Shea]

Maybe it's biased too cold and you're getting crossover distortion?

Also, does this happen when the is mv turned up fairly high? If the pots aren't matched very well, then that could cause some buzziness at very low settings.

Shea

[Rich M]

Tubes are at ~37 mA. Buzziness even with MV all the way up.

[Rich M]

I would thing crossover distortion would be worse on quiet passages when the signal itself is smaller. If you bang a chord it goes away once the signal decays a bit.

[mr coffee]

you didn't go all weber/torres and change values of caps around did you?

what kind of preamp tubes are in there..

[slider313]

Check all your grounds on the input and pots. If you removed the shielded cable from the preamp tubes to the input and volume pots, put them back. The grounding is not going to be complete and parasitic osolation is common. Also if you converted the bias supply from fixed/cathode to fixed,go over your connections and make sure the resistors and the cathode bias cap have been removed from the circut. 37 ma. should be about 440 plate volts

[Rich M]

Input cap was changed to 0.001 from 0.01, did this buzz before/after change. I have standard complement of tubes, tried an 12AX7 in PI, didn't change buzz. Haven't changed other caps, checked bias points - all seem o.k..

Slider,
No monkeying with shielded inputs. No fixed/cathode removal necessary (I think it's a '72).

[Swarty]

Are you sure it is not a speaker or something buzzing on the cab? Try an extentsion speaker and put it far away from the amp.

[Rich M]

Tried it both in combo box and into a Marshall 4x12 w/ Weber MASS.

[AL30]

Did you swap out all of the 100K plate resistors? These are common sources of noise. Also, check for drifted resistors. I have a '72 and a few of my resistors had drifted out of spec.

And you are correct - once it's up and running these are fantastic sounding amps - don't forget the Tube Screamer.

AL

[Rich M]

I swapped a couple (V1) but gave up. I'll try that as I have some nice carbon comp 1/2W. I will check for drifted resistors as well. I thought it sounded great with my TS, but the cleans in particular had this annoying overtone. Thanks for the tips, guys. I'll report back.

[slider313]

Does this happen in the normal channel as well? It sound like it could be a cold solder joint on a cathode bypass cap. I check by hitting the eyelet with the iron and if the connection lifts, it should be redone.I use a solder vacuum and clear the eyelet and make sure there's enough lead for a lift free joint.

[Rich M]

Does this happen in the normal channel as well?

Yes. Will check the joints.

[Wakarusa]

Don't forget to check the screen resistors while you're in there. Another common problem with these are weak/dry solder joints on heater wires, particularly if someone has been inside moving things around.

[mr coffee]

without the MV? what year is the twin.. if it's a pull boost one there is some other stuff in the verb circuit that needs to be straightened around also. did you check the mix resitor at the reverb ?

[Rich M]

Hey Todd, been trying to get a hold of you re: that back panel. I will get off my lazy ass and change the screen resistors. One was smoked REALLL good when I got it. Come to mention it, the insulation on one of the heater wires was damaged, probably from THE FLAMES JETTING OUT of that resistor. Maybe replacing that section of wire could be it. Year? I think someone said '72 (Todd?) It had the pull MV. What the hell does that thing do anyway. THe schematic says pull for "pure sound". What reverb stuff needs checking? I should mention, I have the reverb and tremolo turned all the way down and the tank disconnected with the chassis up on the bench. Well I go do a little soldering now and get back to youse. Thanks!

[Wakarusa]

Awww crap. My humble assistant said the panel was taken care of, yet I see it tucked away in the corner over there. I'll get it out. Next year we don't suspend the periodic assistant beatings for the Holidays ;)

[Shea]

I really doubt the reverb circuit has anything to do with it. Yes, the reverb circuit is a little different from blackfaces, but that doesn't cause buzziness, unless you somehow rigged it so that the pull boost is on all the time. But with the pull boost off, there should be no buzzing and the reverb should sound fine. It does in my mv twin.

Shea

[Rich M]

Todd,
Before you launch the panel, are you sure that it is "wing-shaped" in that era amp? The chassis looks similar to this :
http://boston.craigslist.org/msg/121632022.html
I'm concerned the raised end will interfere with the power chord.

I changed the screen resistors - no change. One thing that has me concerned is that the grid of V4B (reverb mixer?) has about 100 mV on its grid. On the other side of the 3.3M (between the 3.3M and the 0.022 uF coupling cap) I get ~660mV. I initially thought this was the problem, but since these nodes are leaked to ground through 3.3M + 470K, this is a very small current and I thought more reflective of poor design. Is this incorrect? Is the 0.022 too leaky?

[Rich M]

This works out to an equivalent series resistance of the 0.022 of 1.3G ohm also, it wouldn't explain that same buzziness on the Normal channel (I think).

[Rich M]

While I'm at it, seems like G.Weber's advise is questioned here. What value you'all recommend for the PI input cap? Also, any advantage to splitting all the shared cathodes?

[Wakarusa]

This works out to an equivalent series resistance of the 0.022 of 1.3G ohm also, it wouldn't explain that same buzziness on the Normal channel (I think).

Assuming your readings are DC and not AC, then I'd replace the .022uF coupling cap. There shouldn't be any DC on V4B's grid.

The "winged" panel might fit, might not. The AC cord is in the same place on older amps, it just makes a sharp 90 depgree turn and tucks behind the panel. The later style, I believe, was just a cost saving measure. Up to you whether you want to try it.

[Rich M]

Todd,
I 'll give it a try. I think it looks cooler.

I'll replace the 0.022. In the meantime, I decided to test the output coupling caps and did this by disconnecting the bias supply (and removing output tubes!) and ground bias feed through 1.3M resistor. Measured about 1V. Changed one coupling cap and it seems much better - I am reassembling it now. :AOK

Any thoughts on the PI input coupler? 0.001? 0.01? 0.02?

[Rich M]

Well, I found a leaky coupling cap on the PI, replaced it and the voltage went from 1V to 0.25V. Moving the Grid leads away from the HT leads seems to have lowered it into the noise. So I restored the bias ckt and tubes and fired her up.

Still does it!, But I've got to say the tone otherwise is substantially improved. I went after the 0.022 next. and lifted the lead on the side of the 3.3M - NO leakage. I'm still getting ~0.5 on the eyelet node. It isn't grid leakage as the grid of V4b is at ~0.07V, which is probably just resistive dividing. I guess next I need to pick the board apart to see where the problem is. I've got to say, I'd rather be working on my PCB Marshall than this thing! What a whacked construction technique (sorry). I can't wait to start on my Fender 75 (LOL!).

[Wakarusa]

Any thoughts on the PI input coupler? 0.001? 0.01? 0.02?

Really a matter of taste. The twin has enough reserve power to take the extra low freq response without going muddy. If you're happy with the tone as it stands (and with the range you can get from the tone stack) then leave the stock value. Otherwise, I'd try incremental changes.

[Rich M]

Thanks.

Sooooo, anyway, I got ~1.5V on an eyelet with nothing connected to it! Even more curious is there really isn't anything with any appreciable voltage in the vicinity. I lifted the grid ckt to V4B (leaving it grounded through 470K) buzziness is still there.

Taking a break to chainsaw up some firewood before it cools off again.

[slider313]

Todd,
I 'll give it a try. I think it looks cooler.



Any thoughts on the PI input coupler? 0.001? 0.01? 0.02?

On silver panel amps I use either a .01 or a .002. The silver panels seem to have more bass response than a blackface. In the blackface circut I use a .02 which works well with the vibrato mod ( 50k switch pot ).

[mr coffee]

I really doubt the reverb circuit has anything to do with it. Yes, the reverb circuit is a little different from blackfaces, but that doesn't cause buzziness, unless you somehow rigged it so that the pull boost is on all the time. But with the pull boost off, there should be no buzzing and the reverb should sound fine. It does in my mv twin.

Shea

but he changed the type of MV.. and the pullboost/reverb circuit is different than say a blackface/early silver face. I've had both the 3.3 and 10pf cap go south on these amps and it has reflected in additional noise/harshness which may be construed as a buzz.

If this is also a model with the "hum balance" pot on the rear I've had the go bad.. I go back to the 70's schem and tie the 100 ohm resitors to ground off the pilot lamp to create the artificial centertap

and I go .001 on the pi cap

[Rich M]

I think my biggest problem on the reverb circuit is that eyelet that is floating a 1.5V. I seems there is this striated deep blue stuff on the tag board from a nearby HT tie point for some plate resistors. I removed this (along with the wax) with some "Goof-off" to no avail. Anybody have any suggestions (short of replacing the tagboard)? The fuzz I'm hearing is definitely gated in the sense that it only happens during the loudest transients, it is on both channels and seems to be independent of MV position. It was there before the PPIMV. With the replaced PI coupling cap, the overdriven sound of this amp is amazing, even with the MV at bedroom levels.

[mr coffee]

not using the eylet but making/soldering the connection supported above the bad spot.. I hate that Blue showers stuff..:BITCH

[Shea]

Rich, sounds like your eyelet board is conductive. I've read about different ways of fixing this, but the simplest one was to leave the amp on for 24 hours. I haven't tried it, I'm just passing on something I read.

Shea

[Shea]

but he changed the type of MV.. and the pullboost/reverb circuit is different than say a blackface/early silver face.

Yeah, I know. I have one of those amps. I replaced the master volume with a post-pi kind, and left the reverb circuit stock. No buzziness. The differences that exist between the blackface and master-volume silverface reverb circuits don't cause buzziness.

Also, when you remove the stock master volume, if you simply disconnect the the 4 leads that activate the boost from the eyelet board, the boost will be off permanently.

Shea

[Rich M]

not using the eylet but making/soldering the connection supported above the bad spot.. I hate that Blue showers stuff..:BITCH

Is it common?

[Rich M]

Also, getting back to an earlier question, is it possible that I need to increase the swamp resistors? The tubes are old GE's with faded getters, but test good on my emissions tester (same buzziness with a set of Reflektor wafer base 5881's).

[slider313]

Check all the input jacks and see if they're tight. A slightly sprung jack can have all sorts of overtones and paracitics. Sometimes the hardest things to diagnose are the simple and overlooked.

[Rich M]

Back into this thing. It's going out one way or the other next weekend, so I want to try and beat this thing. Jacks are tight.

I have a question. Is it normal for V3 to get hot? I mean real hot. Not glowing, but too hot to handle? I have the tank disconnected so I didn't know if that was screwing up the loading on it. Looking at the schematic, I figure 440V plate voltage and about 8.5 mA total current (cathode: 4V/470 ohm) I'm dissipating about 3W. Doesn't seem that bad.

Anyway, it not the problem. I pulled V3 and V4 and plugged into the normal channel. Same s***. It appears not affected by the MV position so I think it might be the phase inverter. Makes sense? Since I changed most of the resistors, I might just swap out the caps. I don't understand the tremelo ckt so I don't know if I can pull V5 to diagnose. Any thoughts? I really appreciate the help, guys.

[RussB]

The late SF scheme runs the snot out of the reverb driver tube..."lesser" tubes will eventually fail. What cathode resistor does you amp have on V3?

[slider313]

Check the 2200 ohm resistor on pin 8 of V3.I had the same problem with a silverface vibrolux. In that amp there was a 470 ohm and the tube was very hot to the touch. I changed it to a 2200 and it cooled it down and tamed the reverb down a notch.

[Rich M]

The cathode resistor is 470 ohm. You saying change it to 2.2K?

I noticed that in the PI, there is no FB from the 82k plate resistor (RC FB). Is that o.k.? It not used in Marshall's, so i figure it's not critical. I swapped 0.1/200V in PI, disconnected FB from OT all to not much avail. Can I run this amp on two output tubes to test for errant swamp resistors? I still wonder if 1.5K are big enough. Any thoughts?

I scoped the vibrato. Oscillator works seems must be opto coupler. I convinced myself it was ok to pull V5 - made no change. Seems Fender likes to run their tubes hard - that one was pretty hot. Who the hell puts 400V on the plate of a preamp tube??

[Wakarusa]

Agreed that most Fender reverb ckts run the drive tube very hard. Highly recommend swapping out the 470 Rk for a 2K2. You can also bypass the 2K2 with a 25uF cap if there isn't one there and if you need more dwell. There's a transition period between the AA270 (2K2 bypassed) to later 100W/135W (unbypassed 470) where you can find a bypassed 470 that goes through 12AT7 like most folks go thru Kleenex. (You occasionally find a 470 or 820 Rk in blackface Fenders too... get rid of 'em). Eihter way, the reverb driver tends to get hot to the touch.

It's fine to pull the trem oscillator, as you've discovered.

Dumb questions du jour:
-- coupling cap leaking DC?
-- oscillation (in the preamp or the power section..) checked for this on an o-scope? Have an o-scope?
-- couldn't tell from the thread so far, but have you tried running it through an extension cab (internal speakers disconnected) to see if this is a vibration problem?
-- tried cleaning tube socket pins? A bit of pro-gold can work wonders.
-- does it do it if you back off of the PPIMV just a bit (say 20 degrees knob rotation)?
-- which PPIMV circuit did you use? ground referenced and isolated by extra caps? Referenced to the bias supply? Single pot between the PI output legs?

[Rich M]

Hey Todd,

"coupling cap leaking DC?"

Which one? I checked the bias points on the pre's and they seem close. The problem is on both channels, so it is unlikely to be V1 or V2 (or V3 or V4 or V5 by their removal)



"oscillation (in the preamp or the power section..) checked for this on an o-scope? Have an o-scope?"

didn't see any - my scope is a little f-ed up - it works though I can see the vibrato oscillator going.



"couldn't tell from the thread so far, but have you tried running it through an extension cab (internal speakers disconnected) to see if this is a vibration problem?"

Most of the testing was on the bench hooked up to a Marshall 4x12 thru my Weber MASS. Does it in combo box as well

-- tried cleaning tube socket pins? A bit of pro-gold can work wonders.

Yeah, I used some clean evaporating stuff because I didn't want any sticky residue collecting junk.

-- does it do it if you back off of the PPIMV just a bit (say 20 degrees knob rotation)?

I does it throughout knob rotation at about the same amount (as much as you can quantify things.).

-- which PPIMV circuit did you use? ground referenced and isolated by extra caps? Referenced to the bias supply? Single pot between the PI output
legs?

Rich mod. I don't like the double cap setups as they tend to thin up as you decrease the volume. The only time i've had a prblem with this setuip is when using a combo of yellowjackets and EL34's. Because the yellowjkts seem to draw grid current, they would reek serious havoc on the bias points of the EL34's.

I put this back into the headbox and it is shrill as all hell. I have a PI input coupler of 0.01. Any idea how to get some midrange and high-bass (don't want it to get farty) without swapping speakers (my b-i-l is on a low-budget)

[sixstringslut]

Sounds like frustration. May be worth a trip to the local amp doc. I had an OT do this to me once. Maybe get a second opinion or push it aside for a while.