Hoffman AB763 Twin issues (cont from 2012)

[proaudioguy]

I finally have some time off and I'm back to working on my amp.  SO I got some NOS Black plate RCA 6L6GCs for my twin and upped the Bias from 10mA to about 50mA and of course the Trem doesn't work, or it's so weak it's useless.  It's quite strong at a low bias.  I am frustrated.  I tried the LED trick.  I did a couple of other things.  I'm starting to understand why fender changed to the bug, which I have here in the bin.  Does anyone have a parts layout using the Hoffman layout with the BUG?  I'm going to try to work it out but if someone already did it then I'd just as soon take the short cut.


The LED looks really cool.  Can I just attach the bug where the LED is and ground out the signal somewhere?

[HotBluePlates]

... upped the Bias from 10mA to about 50mA and of course the Trem doesn't work, or it's so weak it's useless.  It's quite strong at a low bias.  ...

With your present bias, measure the actual voltage at pin 5 of the output tubes.

Now set the tubes to idle very cold, 10mA or less. Measure the voltage at pin 5 of the output tubes.

What's the difference in voltages? Your trem circuit will have to output a peak-to-peak signal about double this difference to be able to easily turn off the tubes.

This is the problem: tubes are easy to turn on but hard to turn off. Full-on requires 0v at the grid, but for 6L6's and 450v+ at the screen, full-off may require over -56v (screen voltage/Mu G1-G2). You don't need to be a data sheet whiz to see this, just look at how closely the grid lines are crammed together at low tube current on a 6L6 data sheet.

And the turn-off voltage is not linear down at low current. Once you get down to some low current, it takes many more volts to get a smaller current reduction.

All this together explain why you get deep trem when bias cold or when playing softly; the trem doesn't have to overcome much. Once you bias hot and/or play loudly, bias-vary trem has a hard time keeping up.

Can I just attach the bug where the LED is and ground out the signal somewhere?

No, the LED is (should be) attached to the trem oscillator cathode, in place of the original resistor (and bypass cap, if present). In the blackface Twin Reverb, there is an Intensity pot to ground just before the 220kΩ channel mix resistor for the Vibrato channel, and the LDR side of the optoisolator connects to wiper of that Intensity pot. The neon bulb side of the optoisolator connects to the plate output of the buffer triode after the trem oscillator.

[proaudioguy]

That makes sense.  So how do I get a much higher potential negative DC voltage into this thing?  If I bias cold the tubes sound like doo doo.

[HotBluePlates]

Make the measurements first. I have no idea what amount of voltage you need until we have some numbers.

What if it's not even possible to get enough voltage swing with the trem tubes and supply voltage you have? We would determine that by having the hard data to look at.

[proaudioguy]

Well that's exactly what I'm thinking.  That it's not possible, but I'll make the measurements in the morning.


In the mean time I pulled out my large schematic and layout for the standard TR, and got the parts together.  I'll probably load the board anyway while I await your reply in the afternoon or evening.

[sluckey]

The LED looks really cool.  Can I just attach the bug where the LED is and ground out the signal somewhere?

You can do something similar using the Vactrol VTL5C1 optoisolator and a 50K-RA pot available from Hoffman. I used the VTL5C1 to replace the optoisolator in my Sunn trem circuit so I think the LDR part will work well in the Fender circuit. This can easily be installed on the Hoffman AB763 board. And if you don't like it, easy to remove.

See the attached pdf...
EDIT... Removed attachment

[proaudioguy]

The LED looks really cool.  Can I just attach the bug where the LED is and ground out the signal somewhere?

You can do something similar using the Vactrol VTL5C1 optoisolator and a 50K-RA pot available from Hoffman. I used the VTL5C1 to replace the optoisolator in my Sunn trem circuit so I think the LDR part will work well in the Fender circuit. This can easily be installed on the Hoffman AB763 board. And if you don't like it, easy to remove.

See the attached pdf...

Thanks.  I just tried this.  It didn't work.  Just pops when I hit the switch.  I assume the rounder larger component is the bulb and the flatter circular component on it's side is the photocell resister.  I can measure no resistance through it so I may have a bad one.


I had to rework the Bias, which I did.  It's pretty much stock now.  I turned the voltage all the way up.  I'll re-bias after I'm through getting this going.


Thanks for trying.

[sluckey]

Thanks.  I just tried this.  It didn't work.  Just pops when I hit the switch.  I assume the rounder larger component is the bulb and the flatter circular component on it's side is the photocell resister.  I can measure no resistance through it so I may have a bad one.


I had to rework the Bias, which I did.  It's pretty much stock now.  I turned the voltage all the way up.  I'll re-bias after I'm through getting this going.


Thanks for trying.

Did you try this circuit with the VTL5C1 optoisolator I suggested? IT ABSOLUTELY WILL NOT WORK WITH THE FENDER NEON BULB ROACH.

[proaudioguy]

Nope,...I guess I misunderstood you when you said "so I think the LDR part will work well in the Fender circuit."  I took that to mean the 'LDR" that fender used would work like your part did.  I hope I didn't destroy the neon bulb.  Uhg.  I should have just done what I said in the first place.

I may have a vactrol somewhere for spare parts for the Super Champ and the Concert (which may not be similar to what you are describing), but other than that, this is all I have on hand.  I don't want to wait for a week to get 1 part.  I have all the parts to make it like the Fender circuit including an eyelet board if it comes to that.  I could put the Pink LED back and pull the opto apart and put the LED up against the LDS Resister, but then if it doesn't work I'll have to order another one.

[sluckey]

I took that to mean the 'LDR" that fender used would work like your part did.  I hope I didn't destroy the neon bulb

That's exactly what I meant. But that's only half the equation. You already proved that a LED will blink. That's the other half of the equation. The VTL5C1 uses an LED for the blinker. LEDs operate on low voltage but the neon (NE-2) bulb in the Fender roach needs 80 to 90 volts to fire. You didn't hurt it.

I may have a vactrol somewhere for spare parts

They're not all the same. I experimented with 3 different part numbers and found the VTL5C1 to be well suited for trem operation. The others had too slow a response time for the LDR portion.

I don't want to wait for a week to get 1 part.

You've been waiting 2 years. What's another few days? Don't forget you need a pot too.

I could put the Pink LED back and pull the opto apart and put the LED up against the LDS Resister, but then if it doesn't work I'll have to order another one.

That's worth trying. In fact you could run to RadioShack and get their variety pack of photocells and not cut open your roach. Be sure to do this in the dark or at least cover up the LED/LDR.  If it doesn't work out satisfactorily just put the roach back together with heat shrink tubing.

[proaudioguy]

Yes, well it's not that I waited for 2 years,..it's just that I don't get a lot of time to work on it and I have a couple of days right now.


I tore apart all the components front eh board and I'm currently mounting the fender values to the AB763 fiber board.  I have the pot.  I have everything as far as I know unless I've lost some parts.  I originally ordered all the AB763 TWIN parts, then ordered all the substituted parts, as well as the parts I wanted to be low tolerance.  Part of the reason I did this is because Mr Hoffman does not provide lists of part numbers for "kits" and it's very difficult for a novice builder to know which .01 cap to get to fit the layout.  I spent many hours just trying to figure how if a given physical size would fit on the board or not.  If I had a way to select a list of parts for the entire project, then delete or substitute some parts here and there I would have just bought all of it here, since he pulls them all individually anyway.  I prefer his caps because they are smaller.  I did not know that at the time.  Some of these orange drops are huge.


I am currently trying to figure out how to test the bug to see if it's good or not, but yea, glad it didn't break.

[proaudioguy]

Uhg,..missing 1 part, the 10meg resistor....


I guess I'll see if radio shack has one tomorrow.  The rest of the board is assembled and ready to go in.  Hopefully that will work.


I am shielding the wire that goes from the roach to the intensity and from the intensity to the 220k mixer resister.  Does that make sense?  Seems to me the signal would be on that long wire.


So to continue the discussion on the next topic for a bit.


What is the most appropriate way to put the normal channel into the reverb/ trem channel without any issues?  Should I be moving these 220K resisters to that location should I decide to do that?


I read in Weber's book which I bought back around 94 or 95 when I got this amp to remove the 220K resisters at the PI (assuming one channel was not being used) to get better tone through the amp.  Thoughts?


Any suggestions on tone stack mods to the Normal channel to get it into some other flavors?  It would be good to have both channels at the same polarity and use an ABY switch so I can choose between clean fender or something with more meat or be able to blend that clean with FX on the amp.  I finally found a pedal I like for distortion with this amp.  I'm not a distortion pedal fan but it sounds pretty decent.


Is there any benefit to lifting the tone stack entirely on the normal channel and putting in a post V1B level control in a spare hole on the back of the amp, to get a gainy distortion tone without blowing my ears out?

[proaudioguy]

I went to the garage and found the 10M and got the Tremolo working.  The only noise I heard so far was when I leaned over the amp with my guitar volume open to reach the controls on the other side.  I could hear the on off switching.  The actual sound of the tremolo was quite smooth.


Here is my problem.  The Bias voltage is now from -63 to about -47.  I can't get it low enough.  All the values in the circuit are stock.  Any ideas?

[sluckey]

The Bias voltage is now from -63 to about -47.  I can't get it low enough.  All the values in the circuit are stock.  Any ideas?

Increase the 470Ω resistor (the one that the PT bias tap connects to) to about 1000Ω. Does that lower the range where you want it?

[proaudioguy]

[size=78%]Before I do that, though, why would it be off so far?  Is it because the B+ voltage is so low?  I'm trying to learn here.  [/size]


Is there a way to figure out why the B+ is low?


Thanks


I'm going to pull the power tubes and measure all the pins and post.  I'm suspicious that this may have been the problem all along.  Not the end of the world, but a good learning experience.

[sluckey]

Before I do that, though, why would it be off so far?

You have a range of "-63 to about -47". I don't consider that being off so far for a quad of 6L6s. Would be nice to be able to go from 40 to 60.

So, do I understand correctly where you are right now? You replaced the Hoffman board with an original style Fender AB763 board. Tremolo works satisfactorily with the roach. Your only issue is you want the bias voltage to go lower than -47.

Did I get that right?

[proaudioguy]

Not quite.  I only replaced the tremolo section.  The rest is still the hoffman board.  In fact the turrets for the bias vary term are all in place.  All the under board wires are still there etc.  I could easily reinstate bias vary trem.  If there was an easy way to post pictures here like Facebook drag and drop I could.


Power tubes out...


Bias can be varied from 46-63


Pin 1/8 0
3 419
4 417 [size=78%](this seems to change by a few volts)[/size]
5/6 with resistor between are -050 to -051


Power tubes in
Pin1/8 0
Pin 3 415 (bias at 63Volts)
Pin 4 412 (this seems to change by a few volts)
Pin 5 -050
Pin 6 -050


Where would I look for a large voltage drop?  Everything in my power supply is pretty beefy.  More so than the original even.

[proaudioguy]

Transformer measures 322 at each end and 52 at the bias supply.  Output of the rectifier measures only 421.  I( suppose I need to pull the cap can off and see what the resister values are down there?


Maybe I'm dropping a lot of voltage?  Or could it be my extra beefy diodes?
They are the 1N540x (I think 8) 3Amp diodes.  I would not think they would drop the voltage but what do I know?

[sluckey]

Where would I look for a large voltage drop?  Everything in my power supply is pretty beefy.  More so than the original even.

There's nothing to drop the plate voltage. The OT center tap is connected directly to the output of the rectifiers. You would typically see about 450-460v on the plates of a TR, but the amp will be very happy at 415v too. Your voltage is only 9% lower than the 460v shown on the AB673 schematic.

How much AC voltage do you measure right on the PT high voltage secondary?

[sluckey]

Ah, you must be reading my mind. You gave me the answer before I asked the question. 

Your PT is supplying a lower voltage than the original PT. That's why your B+ is slightly lower. You would need to replace the PT to increase the B+. I would not be concerned.

[proaudioguy]

This PT is original.  It's got plenty of rust, but there are signs of the original black paint too.  I'd say it was in fair to poor condition.  I did nothing to it as I didn't want to make it worse.


Does the power scale, IOW if the AC at the wall increases, does that increase the Bias as well keeping the tubes happy?


The 1W (or maybe they are 2W) resisters in the cap can are 220k.  The caps are black with gold print TAD.   I think the 70s were beefed up to 100 iirc.  I can't find my notes.  I have a huge stack of paper for this build somewhere.

[sluckey]

Does the power scale, IOW if the AC at the wall increases, does that increase the Bias as well keeping the tubes happy?

Yes. Can't say how happy they would be though. Filaments and B+ will increase also.

[proaudioguy]

I now have 1513Ohms where the 470 used to be and I can get it to 35mA (35mV across the 1 Ohm resisters)  I can go to 2K or even 2500 but it seems like an extreme case.  Should I lower the value of the resister on the bias pot to ground?

[sluckey]

I now have 1513Ohms where the 470 used to be and I can get it to 35mA (35mV across the 1 Ohm resisters)  I can go to 2K or even 2500 but it seems like an extreme case.  Should I lower the value of the resister on the bias pot to ground?

Going to 2K is fine. You could even go up to 5K. Increasing this 470Ω resistor mainly lowers the entire bias voltage adjustment range.

But since you had 46-63V range with the 470Ω it may be better to play with the resistor between the pot and ground. Changing that resistor will mainly affect the low end of your adjustment range and still leave the high end mostly unaffected. So lower the value of that resistor. I would not go below 10K. I'd just pull the output tubes or leave the amp in STBY mode while doing this. If you can get a range of 35-60v I think you will be happy. I'm guessing 12K to 15K may be the magic number.

[proaudioguy]

I now have 1513Ohms where the 470 used to be and I can get it to 35mA (35mV across the 1 Ohm resisters)  I can go to 2K or even 2500 but it seems like an extreme case.  Should I lower the value of the resister on the bias pot to ground?

Going to 2K is fine. You could even go up to 5K. Increasing this 470Ω resistor mainly lowers the entire bias voltage adjustment range.

But since you had 46-63V range with the 470Ω it may be better to play with the resistor between the pot and ground. Changing that resistor will mainly affect the low end of your adjustment range and still leave the high end mostly unaffected. So lower the value of that resistor. I would not go below 10K. I'd just pull the output tubes or leave the amp in STBY mode while doing this. If you can get a range of 35-60v I think you will be happy. I'm guessing 12K to 15K may be the magic number.

OK cool,..while waiting for reply I put in a 4700 Ohm and that got the low range to 26mA.  I think that is too high and I don't have an appropriate value in between like say 3300.


So I'm going to put the 470 Ohm back and try to add something in parallel to the 27K to get that down in the 15k range.  I am just going clip them on because if that fails it won't fry the tubes where as I soldered the other resisters in because I didn't want to chance losing the connection while testing (no bias voltage).

[sluckey]

OK cool,..while waiting for reply I put in a 4700 Ohm and that got the low range to 26mA.

Do you mean -26V? What was the upper end?

[proaudioguy]

No I meant 26mA.


SO I put the 470 Ohm back in and ended up with a 14k resister on the bias pot.  That did the trick.  I was able to get it biased.


The term was clicking and rather than go through the process of eliminating that, I just pulled it out because it was tone sucking in the worst way.  SO now I have no trem.  I pull V5.  The only thing wired to that socket is the heater.


I experimented with lifting the 22K resister that mixed the Normal channel before the PI.  It increased the gain substantially not he Vib channel but didn't seem to affect the tone too much.  Put it back.


I rolled tubes in the reverb recovery until I no longer had massive microphonics but still had good gain.  It's not particularly great sounding but it is good sounding.  It's a mid 90s sovtek, with the short plates.  These things do not keep their printing so I'm not sure which one it is.


I'd love to use one of about 5 other tubes but they are all microphonic on the reverb return.  Does the reverb return in the Hoffman design have a massive about of gain?  I also feel like I'm driving the reverb very hard and wonder if the tone would improve if I lowered the gain on that side.  It does have a 12AT7 in that spot.  I have tried 3 different tanks and they all sound overdriven to me.  The reverbs in my other amps sound smoother in general.  I ave the original tank, a brand spanking new MOD 3 spring long tank, and a 20 year old mojotone 3 spring short tank.  If none of them sound right then I'd say it's not right.  I'm looking for that vintage Deluxe Reverb/ Super Reverb/ Twin Reverb reverb sound.  Not looking for any "improvements" just the original.  I've been told the transformer can cause a lot of issues so perhaps I should try that.

[HotBluePlates]

The term was clicking and rather than go through the process of eliminating that, I just pulled it out because it was tone sucking in the worst way.  SO now I have no trem.  I pull V5.  The only thing wired to that socket is the heater.

It would have been nice to measure the trem oscillator's output at either the oscillator plate or the buffer cathode to know what the peak-to-peak voltage is. A true-RMS meter could give you an indication to start with (you'd need to do some math and assume a perfect sine wave to get the peak-to-peak voltage), or an o'scope could show you exactly.

That info could have been compared to the range of bias voltages needed to go from idle to full-off. Which is where I was going in trying to understand what is/isn't happening in the original trem circuit.

BTW: the Hoffman AB763 oscillator matches the 6G16 Vibroverb, not the blackface amps. But that's okay, because it was copying the 6G16's bias-vary trem.

I rolled tubes in the reverb recovery until I no longer had massive microphonics ... I'd love to use one of about 5 other tubes but they are all microphonic on the reverb return.  Does the reverb return in the Hoffman design have a massive about of gain? 

No, it exactly matches the 6G16 Vibroverb, except that the mixing of the reverb/dry matches the AB763. The Fender AB763 uses an 820Ω cathode resistor on the reverb return, where the 6G16/Hoffman AB763 uses a 1.5kΩ. I don't imagine that would make much difference, but you can try a resistor swap if you like.

...  I also feel like I'm driving the reverb very hard and wonder if the tone would improve if I lowered the gain on that side.  It does have a 12AT7 in that spot.  I have tried 3 different tanks and they all sound overdriven to me.  The reverbs in my other amps sound smoother in general.  I ave the original tank, a brand spanking new MOD 3 spring long tank, and a 20 year old mojotone 3 spring short tank.  If none of them sound right then I'd say it's not right. ...

I've only ever had reverb in original blackface/silverface amps, and with an Accutronics 6-spring tank in a reissue 6G15 Reverb unit. So I don't know how any of the modern tanks compare to any of those.

I know the '67 Super Reverb I used to own always seemed to have way too much reverb, the circuit had hiss if you turned the Reverb knob up at all, and was silly-cavernous at 4. I always kept the reverb turned down to 3 or below on that amp & let it be what it was; I never tied to "fix it". I only mention that to say your mental image of what it should sound like may be different from the reality of what the old amps actually were.

Regardless, there is a solution: look at the grid resistor for the 12AT7. It's a 1MΩ to ground. Temporarily swap a 1MΩ pot in place of that resistor, wired like a typical volume control. Now you exactly match the Dwell control on the Fender outboard reverb unit. Fiddle the pot until you find the amount of drive into the tank you want. Turn off the amp, and carefully measure the resistance from each outer lug to the wiper (which is feeding the 12AT7 grid). Replace the single 1MΩ resistor with 2 resistors matching your measured values to create a "Dwell Voltage Divider" going into the 12AT7.

[proaudioguy]

Thanks I'll try that. I had a 93 concert and the reverb was perfect.  The tone was a bit steril but the reverb was perfect.  It was a very good gigging amp even though I never gigged with it.  Yes about 3 is about right, agreed.  And yes I like it to be able to get surfy.
The issue is it's got a lot of tones rather than being diffuse and these tones are dissonant. The recovery tube has so much gain that if I tap the chassis of the amp with a guitar pick or finger nail with the reverb turned up I can hear it through the speaker.   If I tap the tube it's very loud. I had to go through about 8 different tubes to find one that minimized this.  If the reverb is not placed just so and the send and return cables routed just so, I get feedback just by turning up the control. No guitar even plugged in.  Just crank the reverb knob and it feeds back.

Wouldn't putting an 820 on the cathode increase the gain?

I consider all of this a learning process and time wasted on it is time well spent.  I have learned a ton.  I know now how to make my amp sound as bad as a reissue.   I'd like to know how to mix the normal and Vib channels before the reverb rather than at the  PI.  I'd like to put a master volume on the back in a spare hole.  I would not mind playing with the V5 as another gain stage and using the speed and intensity pots to control it.

[proaudioguy]

The 3.3 meg resister. Is the dry signal passing through that?  Is the 10p cap there to let some highs through?  Or is all this so the reverb return can feed back to the reverb send?  Can the gain of the recovery triode be reduced by using a larger resister on pin 1 to ground?

[sluckey]

Wouldn't putting an 820 on the cathode increase the gain?

Theoretically yes. But Fender used a 820Ω resistor because that resistor is shared by another triode. Doug used a separate 1500Ω on each triode.

The 3.3 meg resister. Is the dry signal passing through that?  Is the 10p cap there to let some highs through?

Yes and yes.

Or is all this so the reverb return can feed back to the reverb send?

No. The 3.3M is the mixing resistor for the dry signal. A 470K is the mixing resistor for the wet signal. The reason such a large value was chosen for the dry signal is because the dry signal is huge in relation to the wet signal.

Can the gain of the recovery triode be reduced by using a larger resister on pin 1 to ground?

Don't do that! Reduce the 100K plate resistor. Another way to reduce the gain would be to replace the 12AX7 with a 12DW7. That would give you the equivalent of a 12AU7 for the recovery triode.

[HotBluePlates]

... The issue is it's got a lot of tones rather than being diffuse and these tones are dissonant. ...

Welcome to Fender reverb with the stock 2-3 spring long-delay tank. Any outboard recording effects unit gives better reverb unless its digital reverb is horribly grainy, and unless you need a spring-reverb sound (for surf, or a particular sound).

I knew I didn't particularly like the reverb in most Fender amps. I did enjoy the reverb in my old Princeton Reverb, but I played it at low volume around the house, which fed a proportionately small signal to the reverb, which made it seem not so boomy.

As a result, when I bought a reissue '63 Reverb, I also bought a "6-spring tank" from Accutronics (when they still were Accutronics...). This has 3 delay lines, each made of 2 springs, and all individual springs are different lengths. That helps with getting a more diffuse reverb. Additionally, I got a "Medium Delay" tank instead of the stock long delay. That reduces the apparent dwell, and along with an actual dwell control allows for a more subtle reverb.

... The recovery tube has so much gain that if I tap the chassis of the amp with a guitar pick or finger nail with the reverb turned up I can hear it through the speaker.   If I tap the tube it's very loud. I had to go through about 8 different tubes to find one that minimized this.  If the reverb is not placed just so and the send and return cables routed just so, I get feedback just by turning up the control. No guitar even plugged in.  Just crank the reverb knob and it feeds back.

Well, yes, that's microphonics. Don't know if it's due to the tube, the tank not being isolated in a bag/etc, microphonic part in the amp or what...

...   I'd like to know how to mix the normal and Vib channels before the reverb rather than at the  PI. ...

Does this really mean "I want reverb on both channels"? If so, the answer is not changing how they are mixed, so much as routing the Normal channel through the reverb circuit. In general, it takes 1 wire to make this change.

[sluckey]

There is another easy way to reduce the gain of the reverb recovery tube. Don't know why it didn't come to me earlier today. Just remove the cathode bypass cap. Big difference.

[proaudioguy]

I didn't work on the amp today.  I had a recording session and spent time editing vocals and midi tracks as well as roughing in the mix. Just have to add guitar drums and bass.

Yes I want reverb on both channels.  Do I move the two 220k resisters to just before the 3meg?
If not then why do I need them there?

I will try to reduce the drive using the 3 meg I have.  I don't have. Spare 1 meg pot.   I'll just put one leg on the connection from the 500pf to the tube and the wiper to ground and start at 1meg and just gradually reduce the resistance to ground.  If that cleans up the sound of the sound of the reverb then I'll try reducing the gain of the recovery if for no other reason than to get rid if the microphonics.

Yea the tubes are microphonic but not more than most.  When the reverb is turned off there is no perceivable microphonics from any of those tubes and the signal is passing through the other triode.

[sluckey]

Yes I want reverb on both channels.  Do I move the two 220k resisters to just before the 3meg?
If not then why do I need them there?

No. You move the wire that connects the normal channel to that 220K mixing resistor to just before the 3.3m. See attachment.

I will try to reduce the drive using the 3 meg I have.  I don't have. Spare 1 meg pot.   I'll just put one leg on the connection from the 500pf to the tube and the wiper to ground and start at 1meg and just gradually reduce the resistance to ground.

I would connect one leg to the 500pf and connect the other leg to ground. Then connect the wiper to the tube grid. This is called a DWELL control. Many people do this mod to tame the reverb.

Removing the cathode cap on the recovery tube is the dead nuts easiest way to reduce recovery gain.

[birt]

if all the tubes in that position are microphonic i would suspect it's maybe something else in the circuit rather then the tube. tap those components...

[proaudioguy]

Ha ha,..that looks like an old Torres article.  One of the main reasons I ask you guys how to do this is because that guy is a total hacker.  I spent plenty of money on his Super Texan kit for this vary amp in the 90s and it was worthless.  I found the instructions to be written in the poorest grammar I ever saw in print.  I called them for help and they were not helpful.  I told them the instructions read as if a child had written them.  He repeats himself a lot.  Biggest mistake I ever made.  I should have replaced the noisy resisters, and just given it a cap job and kept all the mojo locked in.  It had the greatest smell when I fired it up!  After that Torres mess, I just tore it all apart.  I didn't have the internet back then.  I was just a do it yourself hobbyist, that had several complex solid state audio devices under my belt and wanted to simply tweak my amp to do more than what it did.  Wish I still had those two 16 Ohm greenbacks (one was sort of tan) it was loaded with.


Brit, I have tapped everything.  All the noise I hear from tapping is around that tube and tapping the tube is loudest.  Turn down the verb and it's clean, no tapping sound coming through.  Half of that tube is always active regardless of reverb level.

[proaudioguy]

Yes I want reverb on both channels.  Do I move the two 220k resisters to just before the 3meg?
If not then why do I need them there?

No. You move the wire that connects the normal channel to that 220K mixing resistor to just before the 3.3m. See attachment.

I don't understand how I can have the output of 2 tubes coming together without causing problems.  And if they can (solid state can't do this), then why do the 220K resisters exist at all?


Thanks

[sluckey]

I don't understand how I can have the output of 2 tubes coming together without causing problems.  And if they can (solid state can't do this), then why do the 220K resisters exist at all?

Solid state certainly can be connected together like this.

[HotBluePlates]

I don't understand how I can have the output of 2 tubes coming together without causing problems.  And if they can (solid state can't do this), then why do the 220K resisters exist at all?

The article probably is from Torres. But that doesn't make it wrong (in this case). I've tried it and it works.

It may not answer the why or how the way you'd like, but accept for now that the 3.3MΩ resistor now performs the function the 220kΩ resistors did before, with respect to enabling the final stage of each channel to develop an output voltage from their coupling caps.

To reiterate, if you just do it, you will have reverb on both channels with no extra parts or issues.

[proaudioguy]

I don't understand how I can have the output of 2 tubes coming together without causing problems.  And if they can (solid state can't do this), then why do the 220K resisters exist at all?

Solid state certainly can be connected together like this.

Well I have worked on many mixers and the idea of taking 2 active outputs into 1 input is very bad on the components.  That's one reason why mixers have resisters on each channel feeding the opamp.  And yea you can take 2 XLR outputs and Y them into one input but it is considered a very bad thing to do.


And the 2 220K resisters tend to support this.


I know I'm all over the place but as long as I pulled that wire out from under the board, which was not at all easy but I managed, I went ahead and tacked it on to the V2 grid to see what would happen.  I figured since there was that .047 cap between that it would be OK.  It actually sounds really really good.  So Now I'm wondering if I should use that 3 Meg pot I have and put a Volume between the V2 output and the reverb input to act as a gain master so I can get the level of the V2 Volume up without it being so blasted loud, thus driving the V2B stage.  I'm assuming I would put the wiper on the 3.3Meg, one side to ground and the other side to the cap coming from V2B.


This 20 minutes is all the time I had to play with it today.  Time for bed.

[proaudioguy]

http://www.rane.com/note109.html

[HotBluePlates]

...  So Now I'm wondering if I should use that 3 Meg pot I have and put a Volume between the V2 output and the reverb input to act as a gain master so I can get the level of the V2 Volume up without it being so blasted loud, thus driving the V2B stage.  I'm assuming I would put the wiper on the 3.3Meg, one side to ground and the other side to the cap coming from V2B. ...

Why make it so hard?

The first 2 stages of both channels are exactly the same. Now that you've moved the Normal channel to connect to the reverb input just like the Vibrato channel, everything after the first 2 stages are exactly the same for both channels. At this point, with each channel's volume at the same setting, each so be equally-loud.

So there's no need to make a "gain master" for V2B; the volume is controlled by the volume control between the 2 triodes for that channel. If you still want to control the dwell, or reverb strength, the right way is still to swap a volume control in place of the 12AT7's 1MΩ grid resistor to turn down hard hard the 12AT7 shakes the springs.

[sluckey]

Well I have worked on many mixers and the idea of taking 2 active outputs into 1 input is very bad on the components.  That's one reason why mixers have resisters on each channel feeding the opamp.  And yea you can take 2 XLR outputs and Y them into one input but it is considered a very bad thing to do.

I totally agree. But there are plenty of other circuit examples of where collectors or emitters of two transistors are coupled together sharing a common load resistor.

[proaudioguy]

...  So Now I'm wondering if I should use that 3 Meg pot I have and put a Volume between the V2 output and the reverb input to act as a gain master so I can get the level of the V2 Volume up without it being so blasted loud, thus driving the V2B stage.  I'm assuming I would put the wiper on the 3.3Meg, one side to ground and the other side to the cap coming from V2B. ...

Why make it so hard?

The first 2 stages of both channels are exactly the same. Now that you've moved the Normal channel to connect to the reverb input just like the Vibrato channel, everything after the first 2 stages are exactly the same for both channels. At this point, with each channel's volume at the same setting, each so be equally-loud.

So there's no need to make a "gain master" for V2B; the volume is controlled by the volume control between the 2 triodes for that channel. If you still want to control the dwell, or reverb strength, the right way is still to swap a volume control in place of the 12AT7's 1MΩ grid resistor to turn down hard hard the 12AT7 shakes the springs.

The Normal channel is FEEDING the vibrato channel tube at the point where the input jacks would normally go.  The idea would be to put a DPDT switch in there to switch the V2 in and out of circuit.  Switch one way and Normal feeds reverb,..switch the other way and normal feeds vibrato and vibrato feeds reverb.  With normal feeding vibrato, the gain is extremely high.  Nice OD tone, but way too freaking loud.  Need something to take the level down.


Edit...Let me rephrase,....V1B feeds V2A, V2B drives reverb stage.  Switch the other way and V2 is out of circuit and V1B drives reverb stage.  Just inserting V2 for preamp OD.

[proaudioguy]

I tried the dwell control,...really cool,...I lifted the cathode cap.  Yes reduced the gain tot he point that the microphonic character is gone.  Tap not he chassis near the tube and I hear the ringing.  The only part I have left to try is the transformer.  I was told by my friend that is Fender certified (authorized service center) that the reverb transformer is the root of many reverb issues and he has them in stock so I'm going to borrow one and if that doesn't fix it then I have no place else to go with it.


Thanks for your help.  In the mean time I'd like to explore the gain staging idea.

[HotBluePlates]

The only part I have left to try is the transformer.  I was told by my friend that is Fender certified (authorized service center) that the reverb transformer is the root of many reverb issues and he has them in stock so I'm going to borrow one ...

What exactly is the problem being caused by the reverb transformer? Meaning, what exactly is going to be fixed by using a new reverb tansformer if your your amp has reverb now? I think it will be a waste of time, but if he lets you borrow instead of buy then at least it's not a waste of money.

The Normal channel is FEEDING the vibrato channel tube at the point where the input jacks would normally go.  ...  With normal feeding vibrato, the gain is extremely high.  Nice OD tone, but way too freaking loud.  Need something to take the level down.

Edit...Let me rephrase,....V1B feeds V2A, V2B drives reverb stage. 

You plug a mic into the mic preamp of a channel strip with effects. Then you unplug the mic and plug it into a standalone mic preamp, and out of the that mic preamp and into the mic preamp of the channel strip. Sure there's going to be a volume difference...

You're turning the Fender into a different amp, but maybe not the best way. The AB763 is 2 gain stages + output stage (or 2 gain stages + effects + output stage), with a volume and tone circuit between the 2 gain stages (this impacts the gain structure). Your new setup is 4 gain stages + effects + output stage, which is roughly what the highest gain amps use.

But if you want it to work well, you have to copy some of the high gain amp's approach for building up & knocking down signal. Or you need to copy the London Power Standard preamp, which has 4 total gain stages but allows switching from 2 of those for cleans to all 4 for distortion.

Regardless, that's far afield from "AB763" and "I want reverb on both channels."

[proaudioguy]

This is about my amp which is a learning platform for me.  It will never leave my house.  It's too heavy for me to lift.
It makes no sense for me to have 18 different threads.  If you want to solve a problem, and have it be finished, it's not going to happen.  This is for me to experiment and learn on.  I will make changes and make changes again.  I already had it working as an AB763 and I've already torn a bunch of stuff out.
I think for the high gain thing I need to drive V2B, first then from there go into V2A, rather than the other way around.


Eventually I'm going to take what I learn from this amp and apply it to mods to my Concert.  If I want to go somewhere, I carry the Super Champ.  It's ever so slightly moded, mostly to clean it up.  Got rid of all the fizzys, pops, and humming.


I'm open to suggestions.  I'll check out this london thing.  I never heard of it before now.

[HotBluePlates]

This is about my amp which is a learning platform for me.  ... It makes no sense for me to have 18 different threads.  ...  This is for me to experiment and learn on.  I will make changes and make changes again.  ...

I see, that was the missing link. Since the thread started with "my reverb needs to change" I didn't realize the amp was meant to be a test-case for modding.

[proaudioguy]

This is about my amp which is a learning platform for me.  ... It makes no sense for me to have 18 different threads.  ...  This is for me to experiment and learn on.  I will make changes and make changes again.  ...

I see, that was the missing link. Since the thread started with "my reverb needs to change" I didn't realize the amp was meant to be a test-case for modding.

Yea, I want to get it all "right" then start going to town one thing at a time and learn what changes do what.  When all is said and done, I may use it as a recording amp, but to be honest with a mic a few inches away my Super Champ sounds pretty much the same clean.



So here is a video of my problem with the reverb.
I don't know what to do about it.  Tapping on components makes no real difference.  The whole area in general can be heard a bit but the real problem is the high pitched ring.  It's quite loud int he room  Hopefully you can hear it.  Sometimes if I move the tank around it will just start feeding back.  I've tried 3 different tanks.  Makes little to no difference on the issue.  The reverb sounds different but the overtones sound the same.  If the overtones were just coming from the guitar hitting the reverb tank, that would be fine, but it's absolutely unrelated.  The guitar is not plugged in and the volume is all the way down.


https://www.youtube.com/watch?v=-FA-6MpFsVY&feature=youtu.be