Fender Bassman 135 (UL)

[Willabe]

Next on the bench for my friend is an ultralinear SF Fender Bassman 135 that he wants me to look at. (He has 2 of them.)

Here's a link to Doug's library for the schemo;  https://el34world.com/charts/Schematics/files/Fender/Fender_bassman_135.pdf

It's soooooo loud he runs the MV on 2. And it's soooooo bright he sets the treb. on 2 or 3 and bass on ~8 and mid on 10. And he faces the speaker cab, 15", towards the wall.   

Kevin O'Connor's TUT 1 has a mod for this amp. He just runs 1 pair of the 6L6's as triodes. There's no choke in these amps so 1 pair in triode is an easier mod. It knocks down the volume some and warms up the amps tone a lot.

The schemo shows 500dcv at the PT CT. There is a wall ACV selector switch with a 127ACV setting. The screens on this amp are fed from the UL taps but they do have 470ohm R's also. I was thinking that for a little extra protection for the 6L6 screen's, and maybe a little less volume and a little compression, I'd raise the screen R values to 1K5 on the 2 tubes still in UL?

He sets his mids on 10, so I think I'll just add a mid pot tail R to give him a little more mids? Why change the pot out when he never uses less mids anyway? There's 2 channels, so maybe a 4K7 R on 1 channel and a ~25K on the other?

I'm going to change the LTPI values to BF values. That should warm it up some too.

And if he likes the lead channel mod I did on his Deluxe Reverb, I'll do something like that on 1 of the channels.

I know someone else has been in the amp and worked on it, have to see what they did or tried to do.

It might need a full cap job. 

I might need to do something with the MV?

And maybe the -FB loop?

Is there an easy way to add a -bias adjust pot to the -bias balance circuit that Fender used in these amps? 
 
Any thoughts? 

[PRR]

> He sets his mids on 10

That makes B and T kinda useless.

The "slope resistor" against the Mid pot reduces everything by the mid-pot value. The B and T knobs add-back B and T from there.

If M is maxed, there's little overall cut, little to add back.

It's also more gain than was probably intended.

Kill some treble somewhere else. Without looking, I'd find a 100K plate resistor and bypass it with 0.002uFd in series with 50K. That will fall above 900Hz, -10dB near 3KHz, but flat above to preserve fuzz-tones. Much experimentation needed. (Might want to do this on the far side of the coupling capacitor so you can clip-lead or switch-box without being shocked.)

[Willabe]

If the triode mod doesn't warm it up enough I'll try that PRR. I think I'll leave the mid controls alone and see how it sounds 1st after the triode mod and cap job.

He said he likes the MV, said it works fine for him, so I'll leave that alone.

It doesn't look like anything has been done inside. It does need a full cap job. 

I'll try and make out a parts list tomorrow so we can order the parts from Doug. One stop shop, fast delivery, I like that!    

[Willabe]

This amp has -bias balance adjust.

Can I put a voltage divider, pot and R, pot in series with -bias and tail R to ground, where the 2K2 series R that comes off the -bias diode to make the -bias adjustable?

Doug sells a 10K bias pot?


 

[silverfox]

Appears that, this is the same as the Super Reverb 135- minus Reverb and Tremolo.
Why are there screen resistors and a UL tap? I suspect to get a further reduction in the screen voltage and that adds to the clean headroom?
silverfox.

[sluckey]

I'd experiment with replacing the 33K that's on the wiper of the balance pot with a 25K pot and 22K resistor connected in series between the wiper and ground. I'd pull the output tubes to do this experiment.

[Willabe]

Thank you. 

[shooter]

I'd sell the six-pac, install a couple KT120's, self biased, done! 

[Willabe]

I'd sell the six-pac, install a couple KT120's, self biased, done! 

It's only a 4 x 6L6. And it is loud! TOOOOOOO LOUD!!!!!!

[DummyLoad]

I'd sell the six-pac, install a couple KT120's, self biased, done! 

It's only a 4 x 6L6. And it is loud! TOOOOOOO LOUD!!!!!!

yer too old then! 


try this: clip out the 120pF cap on the MV tap. disconnect .01uF cap in NFB loop.


--pete

[Willabe]

yer too old then! 

This things a beast! 

He plays it with the MV on 2 and the speaker cab turned towards the wall.

try this: clip out the 120pF cap on the MV tap.

There is no 120pF cap on the MV and that pot has no tap. 

disconnect .01uF cap in NFB loop.

Ok, I'll try it. 

[shooter]

They are stupid LOUD, too clean, and heavy and 

[Willabe]

I have a question on the diode rating for the FWB rectifier this amp uses. (I should know this already.)

What is the dcv each of the 4 diodes see?

Schemo 400acv per secondary leg, 500dcv at the 1st filter cap and the diodes are listed as 800dcv/1A.

That's kind of low on the diodes voltage rating? Why didn't they use 2 diodes in series for each diode there now to get 1.6Kdcv rating for each diode?

   

[sluckey]

The diodes work in pairs, ie, two diodes will always be forward biased and two diodes will always be reverse biased. When the AC voltage crosses over to the negative half cycle the diode pairs will flip. Since the diodes work in pairs they are effectively in series with each other and the voltage rating of the entire bridge will be twice that of any single diode. So, those diodes are very adequate for that circuit.

[Willabe]

Thank you Sluckey. 

[Willabe]

How much can you increase the screen grid R without excess sag?

The schemo shows B+ of 500dcv with 120acv wall acv input. He does use a variac, but that just keeps the B+ like the schemo shows, and that's still a lot to ask from 6L6's. And it's the screen grid that is probably the weakest link in the tube at these high dcv's?

The amp does have 470R screen grids from the factory, but I'd like to increase the screen grid R's value for 3 reasons;

1. Sag. Give the amp a little sag, amp has SS rectifier B+. (I am also installing some 10w R's in the B+.)

2. Screen grid high dcv protection. Even though it's UL, the amp pushes the 6L6's to/beyond their limits.   

3. Volume decrease. Any volume decrease would help.

Didn't Tubenit play with screen grid R values and end up going as high as 2K2?


 

[sluckey]

I recommend you leave 'em be. It's a UL circuit. You can't make a dirty sounding bedroom amp out of this beast. 

[Willabe]

I recommend you leave 'em be. It's a UL circuit. You can't make a dirty sounding bedroom amp out of this beast. 

Yes, but  - anything - would help.

[Ritchie200]

Is it having problems eating output tubes?  If so, maybe throw some 1K in there, the tubes will thank you.  Otherwise, if it aint broke...  6L6's deserve to die anyway....   Maybe pull two tubes and see how he likes it, but watch your voltages and re-label your speaker jacks.  Might make a little dB difference and allow him to open it up a bit.  Could sound better.  Hey, UL is cool!  Don't be talking trash about UL!

Jim

edit: If he has the master on 2 I doubt those power tubes are even breaking a sweat.  However, if he does an outdoor gig and has to crank it, they may start complaining.  I'm not familiar with this amp.  Does it have a history of eating power tubes - especially new manufacture tubes?

[Willabe]

Is it having problems eating output tubes?  If so, maybe throw some 1K in there, the tubes will thank you.

If he has the master on 2 I doubt those power tubes are even breaking a sweat.  However, if he does an outdoor gig and has to crank it, they may start complaining.  I'm not familiar with this amp.  Does it have a history of eating power tubes - especially new manufacture tubes?

It's not eating power tubes. He has NOS RCA 6L6GC's in it. But like you say, he's playing it with the MV on 2. 

Here's what I've done so far;

1. Full cap job, including the 2 -bias caps, changed those to 100uF/100v, used all F&T caps. (Nice size and good price.)

2. Since I changed the -bias caps, I put in a new ultra fast -bias diode, and took out the -bias series CC dropping R, changed it to a MO 3w R. And Fender had the -bias caps ground to the preamp B+ filter cap ground. And to do that they had a long loop of wire running across the length of the chassis. So I fixed that, those caps are now grounded with the power tube caps, like they should be.     

3. Rebuilt both preamp channels, the bass channel I modded as a lead channel, 220K plate R, 2K7 K R, 2.2 uF bypass cap and a .0047 coupling cap. He said the amp was hissy so, I change all the 100K CC plate R's and the 2 - 100K TS slope R's to new CC's.  (I did this on his '67 BF DR's normal channel and he loved it. It really upped the sustain without getting gritty, very smooth.) And while I was in there I put in a silver mica 500pF for the lead channels TS treb cap, and a silver mica 250pF in the normal channel. On the normal channel I put in a 10uF K bypass cap.

4. Rebuilt the PI to BF specs, for a little more drive to the output tubes and more touch sensitivity. And I used a 1000pF (.001) silver mica cap for the PI input. (This amp can use all the chime it can get.)

5. Fixed some other ground loop wiring problems. Wires running back and forth. 

6. I've taken out the CC screen grid R's, disconnected the power tube K ground and cleaned up the socket pins.

I'm ready to install the new screen R's and put in 1 ohm bias sense R's. And I'll wire 2 of the 4 6L6's as triodes. Found it in a Kevin O'Connor book he says is/was a standard mod he did with Fender UL amps. Supposed to add some even order harmonics and decrease the volume from full to 3/4. I'm going to try it.

I'm going to add a 10ohm/10w R in series in the B+ before the 1st filter caps, hopping for a little sag.

And I'm going to see how disconcerting the -FB loop sounds. That might be very helpful to this amp? We'll see.

Adding all these things together might/should help to warm up the amps overall sound.

[Willabe]

Is biasing the the 6L6's the same with 2x wired as pentodes and 2 wired as triodes?

I have 1ohm K bias sense R's installed. 

[shooter]

datasheet shows 3ode and 5ode both 30W Pdis, tweak 'em to ~70% and watch for red

[Willabe]

Thanks shooter. 

[tubenit]

IF the amp were my own,  …………………..   I'd make it cathode biased with each pair of 6L6's having their own cathode resistor and cap.  That way I could simply remove a pair of 6L6's when I wanted to for reduced power.  And being cathode biased, I think you'd have significantly reduce volume from that?

DaGeezer made an incredibly sweet sounding amp out of a Twin Reverb using a pair of cathode biased 6V6's along with a pair of cathode biased 6L6's.

With 500v on the plates on the amp you're working on, you won't be able to use 6V6's but you could still use two pair of 6L6's cathode biased.

Just a thought …………

With respect, Tubenit

[Willabe]

It's the UL taps on the OT that's killing the even order harmonics and makes it thin. That's why I'm going to try a pair in triode.

Jeff how high have of a screen gird value have you used?

[Willabe]

I measure only ~ 80K from 6L6 grid to ground on each 6L6 instead of the normal 220K.

Isn't that throwing away a  - lot -  of drive signal to the 6L6's?

They wanted these amps clean, so they used an UL output stage, with a -FB loop, but that was not enough for Fender on this amp. 

They also cut the drive on the PI, 47K plate R's and 330K grid returns. And cut the drive further to the 6L6's with them only seeing ~80K total for grid return.   

[2deaf]

If you don't like the UL thing, then don't use it.  You could connect the screen resistors to the 410V end of the 2.7K 10W resistor and lift one leg of the 30K 20W resistor and see what happens.  That should give it some sag/compression. 

[Willabe]

If you don't like the UL thing, then don't use it.  You could connect the screen resistors to the 410V end of the 2.7K 10W resistor and lift one leg of the 30K 20W resistor and see what happens.  That should give it some sag/compression.

I almost did that, but there's not enough B+ filtering there for the screens. Only 20uF after a 2K7 R, instead of a choke.

Fender must have added that B+ node, with nothing on it, as an extra B+ filtering stage before the PI and down stream preamp tube stages because they didn't use a choke. 

I even thought of using the screen taps for the plate dcv to the 6L6's to lower the dcv a little if I put in a screen B+ node. 

I've installed 3 x 10 ohm/10w R's, in series, to the FWB, 1 in each leg from the PT's B+ wind going to the FWB and 1 right after the FWB. And I'm increasing the screen grid R's. Both of those things together will give some sag.

[tubenit]

I've used as high as 2.2k/3w screen resistors in a handful of amps.


With respect,  Tubenit

[Willabe]

That's what I thought.

Did the larger valued screen grid R give too much compression?

And did they drop the output volume any?

[tubenit]

I could not discern any volume drop.  I liked the compression "feel" that it gave. It wasn't too much at all. Perhaps added some smoothness to edge of breakup tone.

[Willabe]

Ok, that's good info for this project, thanks Tubenit. 

[Willabe]

It was suggested to me to 'change the ratio' of the -FB loop by changing the R values in the loop.

Is saying 'changing the ratio' just another way of saying more/less -FB? Or does it just mean changing the ratio of the -FB's voltage divider? Which would be more/less -FB, same thing. Or is there more to it involved with the OT, power tubes and PI all being in the loop that I'm missing?

I'm going to try it and see how it sounds, but I'd like to understand a little about what is actually happening circuit wise  -IF-  'changing the ratio' is more than just more/less -FB. (I'm also going try it with out any -FB too.) I know it also effects speaker dampening along with cleaning up a power amps distortion.   

I seem to remember PRR and HBP explaining -FB loops as extra gain that can be sacrificed in the loop and they did the math to show how much gain was lost? And I get the gain loss part. 

Stock values are; 820ohm for the series R and 100ohm for tail R to ground.

Suggested values are; 8K2 for the series R and 1K for the tail R to ground.

Is that 10x less -FB injected into the PI? Or is this just changing the -FB ratio? Or are they the same thing?

[sluckey]

The ratio is the same.

820/100 = 8200/1000

[Willabe]

So changing to those values wont do anything to the sound of the amp's output? 

The PI see's the same voltage.

[sluckey]

The NFB voltage will still be the same because the divider ratio is the same. But the tail resistor is also part of the cathode bias circuit so the dc bias voltage will change  and that may affect the sound. 

To change the amount of NFB you would only change one of the resistors, usually the  820Ω.

[Willabe]

To change the amount of NFB you would only change one of the resistors, usually the  820Ω.

Just to make sure, increasing the value of the 820ohm R will decrease the -FB, right?

[sluckey]

I'm sorry Sluckey, I deleted your post.      

Sluckey said correct to my question and suggested using a 10K or 100K pot in place of the 820ohm -FB series R. 

Yes, to dial it in, thank you.   

[sluckey]

Sluckey said correct to my question and suggested using a 10K or 100K pot in place of the 820ohm -FB series R.

That's not quite what I said. I said put a 10K or 100K pot ***IN SERIES*** with the 820Ω resistor.

That way you can easily dial the amp back to the original NFB value or turn the pot the other way to decrease the NFB.

[Willabe]

Sluckey said correct to my question and suggested using a 10K or 100K pot in place of the 820ohm -FB series R.

That's not quite what I said. I said put a 10K or 100K pot ***IN SERIES*** with the 820Ω resistor.

That way you can easily dial the amp back to the original NFB value or turn the pot the other way to decrease the NFB.

Again, sorry I deleted your post, thank you for correcting what I quoted you as posting.

 

[Willabe]

It was suggested to me to 'change the ratio' of the -FB loop by changing the R values in the loop.

I found out what I missed/misunderstood on this.  ^^^^^^   

It was suggested to me to NOT change the ratio of the -FB R's, but to make them 10x of what they are now.

It has to do with the 10x values interacting with the OT's primary parasitic's in the output stage.

It's supposed to keep the bass end tight, (keep the speaker damping?) but soften up the mid/higher frequencies so the amps not so stiff in playing response.

If I had time I would wire up the 2 different value R pairs on a switch and listen for any difference. But I'm busy getting the house ready to list for sale and packing to move to Rochester N.Y.

I worked on this amp and the BF DR just for the $$ to help with the move costs. It was a real blessing to have these 2 amps fall onto my bench right now.  

I did install the 10x values, and the owner should be able to hear/feel any difference as this is his main gigging amp.

[Willabe]

I finished all the wiring and mods and I'm going to fire it up with the lamp limiter, then I'll check the dcv's and -bias dcv.

If all goes well, I'll plug in a speaker and see how it sounds. 

I took before pics and will now take after pics and post them after the testing. 
 
 


 

[SILVERGUN]

It's soooooo loud he runs the MV on 2. And it's soooooo bright he sets the treb. on 2 or 3 and bass on ~8 and mid on 10. And he faces the speaker cab, 15", towards the wall.   

I have to say...this sounds like a good candidate for a simple power resistor attenuator.

That will kill highs and shave serious db'eez

[Willabe]

I have to say...this sounds like a good candidate for a simple power resistor attenuator.

That will kill highs and shave serious db'eez

Simple power resistor attenuator!!!!!        

Oh yee of little faith! 

[Willabe]

I'd experiment with replacing the 33K that's on the wiper of the balance pot with a 25K pot and 22K resistor connected in series between the wiper and ground.

Ok, adjustable -bias installed. I used a NOS PEC mil spec locking pot, 25K with 22K R to ground. It's working good.

All measurements with variac set to 120acv;

At max -bias; NOS 6L6GC's are running ~ 60% dissipation.

-bias pot swing; -46.5 ~ -55.6 

6L6GC plate; 439dcv.

I'll post more voltages later and post pics, gotta go eat.

[Willabe]

.....locking pot, 25K with 22K R to ground.

All measurements with variac set to 120acv;

At max -bias; NOS 6L6GC's are running ~ 60% dissipation.

-bias pot swing; -46.5 ~ -55.6 

You nailed it Sluckey, but, can this be adjusted for a little wider voltage swing? 

(Schemo link is in OP.)

[shooter]

It can but I'm not a confident fixed bias guy, I'd start by increasing the 22k, no tubes, and see which way that goes

[sluckey]

Put a 50K pot in and see what happens. Pull the output tubes while you are experimenting.

[Willabe]

I'll see if I have a 50K pot. 

[Willabe]

..... I'd start by increasing the 22k, no tubes, and see which way that goes.

I think I have to go smaller, because I have too much -bias dcv. Lowest setting on pot now is just getting the tubes to 60% dissipation.