Adding a 1-tube reverb to a 5E3 circuit

[markmalin]

Hi guys
I'm adding Tubenit's 1-tube reverb to a Hoffman 5E3 with an added master volume.  The 1-tube circuit will be per this post:
http://el34world.com/Forum/index.php?topic=7957.0
I understand I will need DC blocking caps on the input and output of the reverb circuit, but I have a question on the B+ supply for the reverb's 12AX7.  Can I just use [C] in the circuit below, or do I need a separate section in the power supply?  And if a separate section, 16uf/450 V cap and what value resistor would be a good place to start?

Thanks
Mark

[tdvt]

I just posted nearly the same question in the "1-tube" thread regarding a 5E2 Princeton circuit. From most of the schematics I've looked at the 4th node was created using a 1K resistor +cap.

I am also wondering about the voltage differential between the reverb tube section.

We'll see..... 

[tubenit]

This is probably what I would personally try first.  I'd use anywhere from a 2.2k to 10k in the B+ rail

With resepect, Tubenit

[Lauri]

Here's another way to add reverb to 5E3 circuit.
http://tube.ghr.fi/5e3reverb.png

Signal to reverb driver tube is taken from the output of the first tube on bright channel and the normal channel is used as a reverb recovery stage.

[markmalin]

This is probably what I would personally try first.  I'd use anywhere from a 2.2k to 10k in the B+ rail

With resepect, Tubenit

Thanks a bunch, Tubenit.  I'll order a 16uf and a few power resistors and pick them up from CE tomorrow and give this a try.  I appreciate the input :)

[markmalin]

Lauri,
Thanks so much for the link!  I'm going to go with the 1-tube because I already have it pretty much in place, but maybe I'll try that one on the next 5E3.
Mark

Here's another way to add reverb to 5E3 circuit.
http://tube.ghr.fi/5e3reverb.png

Signal to reverb driver tube is taken from the output of the first tube on bright channel and the normal channel is used as a reverb recovery stage.

[markmalin]

I finished adding the reverb per Tubenit's suggestions above.  Result:
- It's very faint.  In fact, almost imperceptible
- I replaced the 1M pot with a 2M I had (per your note, Tubenit) but pretty much the same.
- If you turn the Reverb control up to 8-10 there is a faint fuzzy distortion 

I measured some voltages and here's what I came up with.  Any suggestions or thoughts would sure be appreciated.  The amp itself sounds so sweet, and reverb would really put it over the edge :)


(given the schematic Tubenit drafted below)


- before adding the reverb here are the plate voltages
V1A 148
V1B 142
V2A 148.6
V2B 188
6V6 plates: 354
Power supply string:
A 354
B 309
C 238
(no D as it's not in the circuit originally)


(somewhat puzzling they are so low, as previously I swear I was reading 405 at the 6V6 plates, but whatever)


After installing the reverb circuit per the schematic Tubenit drew up

V1A 140.6
V1B 134
V2A 140.6
V2B 177.2
Reverb tube A (did not measure, but we have "D" below)
Reverb tube B 138
6V6 plates: 353
Power supply string:
A 356
B 292
C 220
D 249


So then I left the reverb in circuit, but pulled the reverb tube


V1A 147
V1B 143
V2A 148
V2B 187
Reverb tube A n/a (tube was pulled so didnt' measure)
Reverb tube B n/a ("")
6V6 plates: 353
Power supply string:
A 356
B 310
C 239
D 309


As always, thanks for the support, guys!


Mark

[kagliostro]

Give a look to the Vox Cambridge Reverb



the resistor in parallel with the reverb circuit is way more larger (R14 - 3.3M) and bypassed by a cap (C8 - 10pf)

Franco

[Fresh_Start]

A listening amp or scope would help isolate the problem.


What happens if you turn the master volume up all the way and bypass that 330k resistor between the Reverb pot and the return?

I'm also wondering if the pre-PI master volume combined with the reverb circuit is messing up the bias of the cathodyne phase inverter.  It shouldn't be if you have caps isolating the reverb send and return from V2-B's grid. However, it would be great to have the voltage reading from V2's pin 7 with and without the reverb circuit in place (master volume up Max both times).


Respectfully,


Chip

[markmalin]

the Pre-PI MV in this circuit is a little wonky IMHO...0-2 on the MV is a big jump in volume, then not much until about 7.  I'm considering putting in a post PI MV, but that's another story... 


In any case, I've tried the reverb with the MV at 10 and no difference in sound.  However, I'll try what you said and about bypassing the 330k with the MV on 10.

A listening amp or scope would help isolate the problem.

What happens if you turn the master volume up all the way and bypass that 330k resistor between the Reverb pot and the return?

I'm also wondering if the pre-PI master volume combined with the reverb circuit is messing up the bias of the cathodyne phase inverter.  It shouldn't be if you have caps isolating the reverb send and return from V2-B's grid. However, it would be great to have the voltage reading from V2's pin 7 with and without the reverb circuit in place (master volume up Max both times).

Respectfully,

Chip

[markmalin]

I can play around with the 220k mixer resister per the Vox.


One thing - in general do the voltages I'm seeing with the reverb in circuit seem to be "reasonable" values?  If I look at the Vox schematic Kagliostro posted, they do seem reasonable.  Thoughts?

[2deaf]

The VOX has has a PI that is acting like a mixing stage with plenty of gain for EL84's.  You have a PI with less than unity gain, so the VOX scenario will not work.

The MV will work a lot better and the reverb may even work if you take the 1M pot. and the 220K resistor out of the cathodyne PI and isolate it with caps.

BTW, do you get thunder when you kick the tank? 

[markmalin]

The VOX has has a PI that is acting like a mixing stage with plenty of gain for EL84's.  You have a PI with less than unity gain, so the VOX scenario will not work.

The MV will work a lot better and the reverb may even work if you take the 1M pot. and the 220K resistor out of the cathodyne PI and isolate it with caps.

BTW, do you get thunder when you kick the tank?

So are you suggesting a post Phase Inverter master volume?  I guess I'm not following what you mean by "the MV will work a lot better....if you take the 1M pot and 220k resistor out of the catho. PI and isolate it with caps..."  Sorry, just confused that's all. 

[2deaf]

Like this

[markmalin]

Like this

Interesting.  So this DC isolates the MV and mixing resistor...and also leaves in the original circuit's 1M resistor. 

[Fresh_Start]

Mark - 


Please tell me a little more about how that pre-PI master volume works (or doesn't as the case may be).  It's my understanding that most of the effect should be at the bottom of the dial IOW 1-3 because the 1 meg pot is "bootstrapped" rather than being grounded. Are you saying that there is some action in the 7-10 range?

Like this

That should "work".  However, you've now doubled up on .022uf coupling caps over 1 meg resistors. Two identical hi pass filters. If you go this route, I would try a .1uf for the first coupling cap. Either way, there are multiple voltage dividers to be careful of.


Hope that helps,
Chip

[markmalin]

Chip, it's like the Master is really touchy from 0-3 then the net volume gain levels out from 3-7'ish.  What I mean is the first little bit of turn on the MV jumps the volume a lot, then it levels out (less steep of a curve).  Then about 8 there is an additional boost.  That may be typical of a 5E3 with an MV, but it just seems strange.  Like the two MV and the other Vol interact...


Along with that, the first volume also behaves like that.  I suspect this is typical of a 5E3 (it's been a while since I've played one), but the net effect of the MV and original Volume acting this way is, the 0-3 on both are a very big change.  For bedroom volume you end up with both at about 1-2.  Dont' get me wrong, it sounds awesome, but I can picture someone saying "can you make the Master more gradual?

So here's something - I wired the MV up per 2deaf's idea, but did not put in the reverb yet (so I left out the 220k mixer, but the rest is per his schematic).  Substantial difference in the MV behavior:

- now the MV is very nice and gradual (like you'd expect an MV to behave)
- however, unless you have the MV on 10, if you turn the Vol up past about 3 you get overdrive...and when you crank the Vol to 8 or 10 and turn down the MV (say to 5) it's shredsville...which I dont' care for
- the tone doesn't seem as sweet, but that may be in my head

So in some respects I like the way the MV behaves now, but I don't like the severe overdrive when i turn up the initial volume, and it seems to have lost just a touch of it's charm, unless you have the MV on 10 and the Vol down low (I like playing cleanish blues)

Help me understand why this is so different wired up this way.  (Chip or 2Deaf) Why is the MV more gradual, but the volume overdrives so much.  I understand the idea of the  multiple hi-pass filters, and perhaps that's why there's a faint touch of "harshness" (the non brite channel sounds pretty good, actually). But I'm not entirely sure I like it this way.  (all reverb aside...)


Mark

[2deaf]

That should "work".  However, you've now doubled up on .022uf coupling caps over 1 meg resistors. Two identical hi pass filters. If you go this route, I would try a .1uf for the first coupling cap.

Two identical high pass filters in series forms a second order high pass filter which will have the same cutoff frequency as either high pass filter alone, about 7Hz in this case.  Should you feel that my calculations are off or that I am using the wrong formulae, by all means use a 0.1uf cap.

[2deaf]

- however, unless you have the MV on 10, if you turn the Vol up past about 3 you get overdrive...and when you crank the Vol to 8 or 10 and turn down the MV (say to 5) it's shredsville...which I dont' care for
- the tone doesn't seem as sweet, but that may be in my head

So in some respects I like the way the MV behaves now, but I don't like the severe overdrive when i turn up the initial volume, and it seems to have lost just a touch of it's charm, unless you have the MV on 10 and the Vol down low (I like playing cleanish blues)

As the amp was originally designed, the 6V6's would have been overdriven before the preamp would.  With the MV you can overdrive the preamp before the 6V6's, although two-stage overdrive shouldn't be all that radical.  The parallel load on the plate of V2A is not significantly different whether the MV is there or not, so I doubt that it makes much difference in the overdrive of the preamp. 

[Fresh_Start]

Mark -

Here is a discussion from the music electronics forum, some of which addresses Rob Robinette's pre-PI master volume for a cathodyne phase inverter:
http://music-electronics-forum.com/t41344/


This was a shorter and more useful discussion here:
http://el34world.com/Forum/index.php?topic=19782.msg206407#msg206407


As far as I can tell. That pre-PI master volume affects the signal going to the power tubes in two different ways. As you reduce the value of the grid return resistor for the cathodyne, you change its bias point. The 1 meg pot also acts as a voltage divider, although not quite the way you or I expected due to the "bootstrap" effect. My guess is that one works at the top of the pot while the other works toward the bottom. Unfortunately, I don't know which is which.


I also cannot explain why having the revised MV at "5" puts you in "shredsville". Try that .1uf cap before the MV. If that helps but not enough, try a .2 or even .3. Also, have you experimented with the Volume pot for the channel you aren't plugged into?  I haven't built my 5E3 yet but the volume pots and tone control are weirdly (and wonderfully) interactive from all that I've read and heard.


Chip


P.S. I was going to put this in a PM but thought someone else might benefit, especially if they come across RobRob's 5E3 mod page.

[markmalin]

Mark -

Here is a discussion from the music electronics forum, some of which addresses Rob Robinette's pre-PI master volume for a cathodyne phase inverter:
http://music-electronics-forum.com/t41344/


This was a shorter and more useful discussion here:
http://el34world.com/Forum/index.php?topic=19782.msg206407#msg206407


As far as I can tell. That pre-PI master volume affects the signal going to the power tubes in two different ways. As you reduce the value of the grid return resistor for the cathodyne, you change its bias point. The 1 meg pot also acts as a voltage divider, although not quite the way you or I expected due to the "bootstrap" effect. My guess is that one works at the top of the pot while the other works toward the bottom. Unfortunately, I don't know which is which.

I also cannot explain why having the revised MV at "5" puts you in "shredsville". Try that .1uf cap before the MV. If that helps but not enough, try a .2 or even .3. Also, have you experimented with the Volume pot for the channel you aren't plugged into?  I haven't built my 5E3 yet but the volume pots and tone control are weirdly (and wonderfully) interactive from all that I've read and heard.

Chip

P.S. I was going to put this in a PM but thought someone else might benefit, especially if they come across RobRob's 5E3 mod page.

Thanks for the discussion(s), Chip.  Sounds like you're working on the same circuit as I am.  It was the Rob R. mods I was looking at originally for the MV and the thought that it was an easy insert into the circuit.  Interesting how the discussion on music-electronics-form talks about the MV basically being full on right away due to the "bootstrap" of the master volume pot - that's exactly what I'm seeing, so that explains that.  I'm really curious to know what you come up with on yours, and which way you decide to do it.  (you haven't built it yet, right?).  I'd leave it at Rob R's layout if it wasn't for that immediate on/off.  Maybe your adding the 500k resistor would tame it down.

I also cannot explain why having the revised MV at "5" puts you in "shredsville".

It's just the pre-amp overdrives so quickly, like 2deaf mentioned, the original circuit was designed so the 6V6's would overdrive before the preamp.  IMOH it's just too much overdrive wired up this way - I don't find it very pleasing, but I'm a jazz/blues guy.  I was going for a "blues/jazz" amp.  Probably should leave the MV out, but it's nice to have for bedroom volumes.

[markmalin]

- however, unless you have the MV on 10, if you turn the Vol up past about 3 you get overdrive...and when you crank the Vol to 8 or 10 and turn down the MV (say to 5) it's shredsville...which I dont' care for
- the tone doesn't seem as sweet, but that may be in my head

So in some respects I like the way the MV behaves now, but I don't like the severe overdrive when i turn up the initial volume, and it seems to have lost just a touch of it's charm, unless you have the MV on 10 and the Vol down low (I like playing cleanish blues)

As the amp was originally designed, the 6V6's would have been overdriven before the preamp would.  With the MV you can overdrive the preamp before the 6V6's, although two-stage overdrive shouldn't be all that radical.  The parallel load on the plate of V2A is not significantly different whether the MV is there or not, so I doubt that it makes much difference in the overdrive of the preamp.

I sure like that original design :)  it's just nice to have a master for quieter playing (bedroom/livingroom).  I've got to play around with it a little more.

[PRR]

> Two identical high pass filters in series forms a second order high pass filter which will have the same cutoff frequency as either high pass filter alone

Ignoring isolation: two networks each -3dB at 7Hz, in cascade, will be -6dB at 7Hz.

By thumb-rule, -2dB at 14Hz. So the nominal "-3dB" point has shifted up about 2/3 of an octave.

As 7Hz or 14 Hz or 11Hz are all FAR below guitar band, I don't think it is significant.

Two identical C-R networks, directly cascaded (no isolation), gives an even softer result. While probably still not nicking the guitar's balls, I agree: use 0.1uFd just-to-be-sure.

[markmalin]

I'll give a 0.1uF a try then.


I need to play the amp more, but so far I'm still not convinced this is the master volume scheme I want.  I really don't want so much overdrive/distortion - I'd like to have more headroom to be able to play the amp a bit louder with just a little "bite".  Again, I need more time playing it, but I'm not looking for "Marshall" distortion out of the pre-amp section.  I'm admittedly over my head other than maybe trying a Post-PI MV.  I've done that on other Fender circuits.  Any ideas would be appreciated.


short term, though, I need to play it more to see if it's what I'm looking for.

[shooter]

don't want so much overdrive/distortion

If you have one, pop in an AU in place of the AY, might be an easy *fix* for pre distortion

[markmalin]

don't want so much overdrive/distortion

If you have one, pop in an AU in place of the AY, might be an easy *fix* for pre distortion

Thanks, shooter.  I probably have one laying around :)

[tdvt]

I am following your progress as I still hope to include the one-tube reverb in an SE amp I am drawing up.

With my limited, enough-to-be-dangerous level of knowledge, I can see where the boot-strapped MV could (does) have weird unanticipated results & interaction.

2deaf's drawing makes much more sense to me, but it sounds like the amp has lost it's edge.

I have never tried it, but have you thought about power-scaling (VVR) to lower the volume? (but hopefully retain the character of the amp)

[markmalin]

I am following your progress as I still hope to include the one-tube reverb in an SE amp I am drawing up.

With my limited, enough-to-be-dangerous level of knowledge, I can see where the boot-strapped MV could (does) have weird unanticipated results & interaction.

2deaf's drawing makes much more sense to me, but it sounds like the amp has lost it's edge.

I have never tried it, but have you thought about power-scaling (VVR) to lower the volume? (but hopefully retain the character of the amp)

With my limited, enough-to-be-dangerous level of knowledge,

That makes two of us :P


I once bought a power scaling circuit (VVR), but never implemented it because it would have required lots of re-work.  My goal really isn't to get power tube overdrive at livingroom volumes. It's more about being able to turn the pre-volume up enough just for a little edge (again, thinking more BB King than Santana) while still having enough headroom to play jazz at club volumes or livingroom volumes.  The standard 5E3 circuit does this nicely at club volumes with no master, but not at "honey, I'm on the phone!" volumes.  Anyhow, I figure that kind of ruled out the idea of the VVR, but maybe I'm wrong.  I also am wondering if a post PI MV would do better, but the more I read it seems since the PI on this amp is unity gain it wouldn't make much difference which side of the PI the master is on.


2deaf's drawing actually does work more like you'd think an MV would (nice and smooth - thanks again, 2deaf!), and that makes sense since it's more of a voltage divider (with one side going to ground)... but it's not that it "lost it's edge", it's more that it lost it's headroom.  Unless the MV is nearly in the 8-10 range, turning the pre volume up past about 2-3 you start to get distortion (more so with humbuckers), and by the time you're at 5 and over it becomes a "5E3 Marshall".


Again, I need to play the amp more - this may be a workable solution, but I liked the other MV for the headroom it gave, it just came ON FULL right away.  Like an on/off switch.  I'm understanding now that's because of what it does to the circuit (I'm using the term "understanding" loosely here ;) )


Mark

[kagliostro]

Mark have you ever seen this pdf file ?

https://courses.physics.illinois.edu/phys406/student_projects/fall04/james_milsk_gabe_jacome/james_milsk_gabe_jacome_modified_5e3_deluxe.pdf

Ciao

Franco

[markmalin]

Mark have you ever seen this pdf file ?

https://courses.physics.illinois.edu/phys406/student_projects/fall04/james_milsk_gabe_jacome/james_milsk_gabe_jacome_modified_5e3_deluxe.pdf

Ciao

Franco

no.  grazie mille!

[kagliostro]

Pay attention

the modifies on that file are principally about the use of a 12ax7 as V1 instead of a 12ay7

but however there are interesting things to be read

Franco

[2deaf]

Ignoring isolation: two networks each -3dB at 7Hz, in cascade, will be -6dB at 7Hz.

Yeah.  So I left that detail out for dramatic effect, but it in no way invalidates my assertion.

By thumb-rule, -2dB at 14Hz. So the nominal "-3dB" point has shifted up about 2/3 of an octave.

I don't know the rule or how you applied it, but the -3db point probably moves up more like a whole octave for the second order filter because it is well into the knee portion which is curving to the right on the bode plot and the first order filter is just entering its knee.  Whatever it is, it sounds furly dramatic until you look at what happens with the curves between the -3db points and 82 Hz.  At 82Hz, the width of the line used on the graph begins to become significant in determining the difference in low-cut.

I agree: use 0.1uFd just-to-be-sure.

An amzing conclusion given the premises cited.


I know it is all splitting hairs and I wonder what a fraction of a db has to do with distortion and performance.

[Fresh_Start]

Pay attention

the modifies on that file are principally about the use of a 12ax7 as V1 instead of a 12ay7

but however there are interesting things to be read

Franco

While it is interesting reading, the fundamental premise of that research paper is that the Tweed Deluxe 5E3 needs to be "fixed". 

Seriously, the 12AY7, the "backwards" volume pot configuration, the weirdly interactive volume and tone controls, and the absence of negative feedback are all part of what makes the 5E3 circuit a CLASSIC.  I've been doing a ton of research myself on this circuit and the only thing that might be improved without exorcising the mojo is phase inverter distortion.  A suitable grid stopper should do the trick.

Sure, you can improve the power supply filtering, add a choke, add negative feedback, use 6L6 power tubes, beef up the output transformer and... whatever.  But it's not a Tweed Deluxe anymore, is it?

My old hound dog didn't think that he needed to be "fixed" neither! 

Respectfully,
Chip

P.S. Yes, I know that I've been actively pursuing tweaks and switches for this circuit.  However, my objective always has been to maintain the original 5E3 when the switches are "off".

[2deaf]

Did those kids graduate?  We didn't have Senior Projects at the University of California when I graduated, but the Cal State colleges did and they were a pretty big deal.  I don't think that little dissertation on an old guitar amp would have cut it.

[markmalin]

I played the amp for about an hour yesterday with 2deaf's MV in circuit.  The behavior of the MV vs pre amp volume is not that bad.  I don't like the sound of the overdrive when the Vol is on 6-10, it's brittle, but I do like that i can back down the pickups on my Les Paul and it sounds pretty nice.  And with the MV on 10, the volume behaves like a normal 5E3.  With a Strat I find it sound pretty nice in that it doesn't overdrive anyware near as much due to the weaker pickups. 


So I decided to wire in the reverb per your idea, 2deaf.  It works for the most part, but is somewhat faint  (was only able to test it quietly about 10:00 pm last night with everyone else asleep).  I need to play around with it some and play the amp some more today.  I think this may be a good starting place, though.  I can get clean jazz tone by backing down the volume on my les paul (or plugging in to the lower gain input), and can get a descent blues overdrive if I keep the Volume down to 2 or 3 and turn the MV way up. 


With the reverb and mixing resistor in circuit I'm getting some hiss now, but that could be due to me re-working the components so much (heating them up), and I didn't pay much attention to parts or outside foil orientation just yet, so I can debug that after I've settled on a layout.  (it's looking pretty squirrelly right now, and I won't guarantee any of my solder joints just yet!)

[Willabe]

Do you know/have you tried using the interactive volume control setting for a clean/scooped mids sound?

Plug into the channel with the tone control, set that channels volume at 12:00, set the other/unused channel on full. Should be clean with scooped mids. Then reverse the volume control settings and see how that sounds. Then do the same plugging into the other channel.

The 12:00/full up settings are the max clean/full distortion sound you can get with a 5E3. But you can play around with the 2 volume control settings for different sounds/tones of your liking. And you can use an A/B box for a 2 channel set up.       

[tdvt]

Do you have this saved?

From the other thread:

[markmalin]

Do you have this saved?

From the other thread:

I do, thanks.  I'm using the values listed (i.e. haven't changed anything from the base parts on the schematic).  After I play it a while I may start playing around with the different values to add more gain, for example.  I did try a 2M pot for the reverb control, but didn't notice much difference - however that was before using 2deaf's master volume layout...

[markmalin]

Thanks, Willabe.  I'll have to mess around with this.  I've never tried connecting the two channels together either.  I forget how they do it, a Y cable maybe?  But now i'm curious to see how this clean scooped mids sounds. :)


Cool idea about using an A/B box.  Thanks!

Do you know/have you tried using the interactive volume control setting for a clean/scooped mids sound?

Plug into the channel with the tone control, set that channels volume at 12:00, set the other/unused channel on full. Should be clean with scooped mids. Then reverse the volume control settings and see how that sounds. Then do the same plugging into the other channel.

The 12:00/full up settings are the max clean/full distortion sound you can get with a 5E3. But you can play around with the 2 volume control settings for different sounds/tones of your liking. And you can use an A/B box for a 2 channel set up.     

[markmalin]

Been playing the amp more, I think this is a livable solution.  A couple peculiarities, though. I'd appreciate any thoughts.

- when I turn the Reverb control past 8 the volume drops.  Could this maybe be a parasitic oscillation, or is that to be expected?
- reverb off = more gain than reverb on.  Can I expect this due to the signal going through another set of filters (caps/resistors)?
- the higher I turn the Tone control, the more gain (which means more overdrive). This is subtle and may be typical of a 5E3.

The addition of the reverb and it's added power supply have become somewhat of a hack on the board.  I had it laid out nicely, but the extra parts going with 2deaf's has added a lot of crossing wires and such.  I need to clean up the lead dress, but as it is now here's what the board looks like.  Please offer any suggestions of obvious lead dress concerns.  I'm embarrassed to show these...but there it is.