1-tube transformerless reverb breadboard-fest

[pullshocks]

A long time ago, I bought a reverb tank off Craigslist that turned out to be a high impedence type (xFBxxxxx), not usable for the common Fender-style transformer driven circuits.  About 7 years ago forum member 2Deaf provided a reverb circuit design for an FB tank in a 1-tube transformerless configuration using a 12DW7 tube.  I cut and pasted his driver circuit into a Princeton Reverb breadboard build. I posted about that project here: https://el34world.com/Forum/index.php?topic=20975.msg222095#msg222095 and here: 2deaf's 12dw7 transformerless reverb driver on the breadboard (el34world.com)    It worked OK, with enough reverb for my style, but I didn’t do any more with it.

Until now. 
I have my breadboard set up with a simplified Tweed Bluezmeister Clean Only Reverb (TBCOR) amp ready take another look at 1-tube cap drive reverb circuits. 
Stage 1 is to replace the transformer driven reverb driver circuit with 2Deaf’s transformerless driver,  similar to what I did previously with the  Princeton Reverb circuit.  I note that whereas the Princeton reverb has a 3.3Meg reverb isolation resistor, the TBCOR’s “1-tube reverb” uses a 150K.  My understanding is the return circuit has been tweaked to allow the lower isolation resistor value.
In the future I hope to try some other driver and return circuits.  After looking at a bunch of Ampeg schematics, I see that the GU-12 and GSR112 are 1-tube reverbs, and I hope their driver could be adapted to the 12DW7.  Since these Ampegs use a 2.2 meg reverb isolation resistors, I don’t know how that is going to work out. 


If I had any sense I would just order  a reverb transformer and low impedence tank and use the well known transformer driven 1-tube reverb circuit.  But  down the rabbit hole I go…..who knows I may even try 6BM8, 6DX8 based cap drive reverb circuits.
 

[roarshock]

I'm a fan of the capacitor coupled reverb and wish you the best of luck on your quest.  Guitar.com has a snazzy circuit that they lifted from Swart and posted to the web a while back. I've used it with significant success. There are plenty of great examples out there! I have a Magnatone 440 that has 6BM8 reverb and it goes way cavernous. Benson Amps uses a 12WD7 cap-coupled reverb, and those fetch a high price and plenty of praise.

[pullshocks]

Thanks Roarshock.  I tested the amp without reverb and it worked great, so and I am hooking up the reverb components per the schematic above.


I would love to hear more about your Magnatone.  The only 440 schematic I could find  used a 6dr7 for the reverb driver.


I am familar with the Guitar Magaizine article you mentioned. Did you follow the article exactly? And funny you should mention Benson, I had the Monarch reverb on my breadboard before this.

[tdvt]

  I note that whereas the Princeton reverb has a 3.3Meg reverb isolation resistor, the TBCOR’s “1-tube reverb” uses a 150K.  My understanding is the return circuit has been tweaked to allow the lower isolation resistor value.

In the future I hope to try some other driver and return circuits.  After looking at a bunch of Ampeg schematics, I see that the GU-12 and GSR112 are 1-tube reverbs, and I hope their driver could be adapted to the 12DW7.  Since these Ampegs use a 2.2 meg reverb isolation resistors, I don’t know how that is going to work out.

I built a little Champ circuit but added full 2-tube Fender Reverb (2-tube mostly for the mixer/extra gain stage) & used a 2.2M isolation resistor (w/10pF) with great success. In fact, more reverb than I can use.

[roarshock]

Pullshocks-
My total mistake! The Magnatone 440 does use a 6DR7, I should have checked before typing. 
As for the Swart/ guitar.com circuit, I've built it a few times. The first time or two I used the tone and dwell, but have since added other features to my amps that forced me to remove them. I didn't use them that much, but if I was adding the circuit to a completed amp like the article suggests, it's not a bad idea. It sounds great in my AST clone!  As for the 6BM8 reverb, check out this example and schematic https://el34world.com/Forum/index.php?topic=26945.0, if you haven't already.

[pullshocks]

Results so far:


With 150K isolation resistor, the verb was barely audible with the knob on 10.  Slight improvement with 470K.  2 Meg works a lot better, and it did not seem to have any bad effect on tone that I could notice. 


What I did start to notice was the decay seemed pretty short.  It is an old tank that does not have code markings to indicate whether it is supposed to be short, medium or long decay. But I suspect it is short decay.  It is a 2 spring tank, 17 inch length


I unplugged the xFBxxxx tank and swapped in a 8EB2C1B  a short format 3 spring medium decay tank I had for the Benson Monarch experiment.  It gave a much better reverb sound.  Despite being the "wrong" impedence it seems to work ok.  This did not surprise me as myself and other Monarch builders found the F and E tanks both worked in the Monarch.  However there is a similar driver circuit that is supposed to be tweaked for the EB tank, plate resisistor 18K instead of 20K, and the resistor going into the tank is reduced from 2.7k to  1.5K.  I don't have an 18K 5 watt on hand, but I can try changing the other one.


I wonder if long decay and/or long spring tank would be even better.  In any event, I will not be using the old xFB tank any more. 

[pullshocks]

After some more playing time, the difference with the 8EB2C1B tank is really striking.  Don't know if this is due to impedence matching issues, or the condition of the older xFBxxxx tank.  The difference is enough that I will go back to a smaller reverb isolation resistors.



"E" tanks seem to have more options as to size, 3-spring, and decay.  "F" seems to have fewer options available.  All the more reason to tweak RVR2 and RVR5 to the supposedly more "E"-friendly values.


I remeasured the power supply voltages now that I have the 12DW7 in place.  small drop at B, big drops at C, D, and E.  As I searched the web for info about cap-driven reverb, there were several mentions of high current draw, so not really a surprise.  But I wonder if the power supply in my SoLoWatt will be able to handle it.


I'm pretty happy with how this is shaping up.  I like the reverb tone.   I may experiment with RC3 and RC6 but I think it has the brightness about right.  It seems  to mostly decay smoothly, although when hitting a loud note and then quickly muting the string, the remaining echo sounds slightly mechanical.  But I think that is kind of inherent in spring reverb.

[tubenit]

I would suggest a 1M reverb dwell. I have found dwell pots to be incredibly useful shaping the reverb tone I want. It can be inside the chassis if needed.


Then I would try a 33k into the grid instead of the 220k.  I'd remove the cap to ground.
And I'd remove the 220k to ground IF that is present in your circuit.


Not saying you "should" do that, it's simply what my approach would be.


With respect, Tubenit

[pullshocks]

Thanks Tubenit.  I appreciate your suggestions.


The driver design came from 2Deaf.  He has not been active on the forum for a few years.



I have only a vague understanding of dwell controls.  not quite sure how to wire it up but I can probably find some examples.


I will definitely try the smaller grid resistor, and reduce or eliiminate the 390 pF capacitor to ground.


As far as the 220 resistor to ground I do not have it wired in.  I found a note about it.  It is supposed to be a pull down resistor to keep the 1 uF capacitor from charging up if no tank is connected.  At some point I'll hook it up and see if it has an effect on the sound

[Williamblake]

I have had good results cap-driving E-Type tanks with an EF80.

[pullshocks]

I’d be interested to see a schematic of your EF80 reverb implementation, if you’d be willing to share.   

[Williamblake]

I have posted it in another thread already but here it comes again. The reverb unit has two output stages to drive high and low impedance tanks alike and this of course not only lets me drive different type reverb tanks but lets me compare the output stages as well so i am quite confident to state it does work really good. Reverb ftw!

[pullshocks]

Thanks Williamblake.  That is really something.

[echuta13]

Glad to see this thread!  Haven't done a 12DW7 reverb but I'd like to (or a 6BM8 if the 12DW7 doesn't make me happy).

Here's an article on a dwell control for a Fender reverb that should be applicable to this one as well:

https://guitar.com/features/opinion-analysis/amp-faq-how-to-add-dwell-control-fender-style-reverb-circuit/
Looking forward to hearing more about your reverb adventures.   

[pullshocks]

I'm jealous of folks that have table top size breadboards.  Mine is just a kludged up 17"x8" amp chassis.  There was nowhere to mount a dwell control as suggested above.  So time for kludge on top of kludge. 

Added a couple switches so I can A/B some of the tweaks suggested above


Can't wait to try my new dwell control........

[tdvt]

Can't wait to try my new dwell control........

I have included or added them to every amp I have built that has reverb, as well as a couple I didn't build.


As Tubenit points out, you can really shape the reverb, playing one control against the other.


In the standard circuit, I just replace the 1M to ground at the driver grid with a 1M pot, haven't fooled with other values. I will be interested in your findings.


I find "Dwell" a pretty vague & unfortunate term, but it's the tradition. I have recently started labeling the controls Reverb "send" & "return" as it is a clearer picture as to their functions.

[tubenit]

FYI,  in reply #6, it says midboost omitted.  Actually the midboost is "always on" in practice in how you're showing this on the schematic. You omitted "normal" which would've been maybe around 330p.  You might find more reverb tone by using a 330p or 390p where the .002 cap is? The reverb is more pronounced on the treble frequencies, in my experience.



With respect, Tubenit

[pullshocks]

FYI,  in reply #6, it says midboost omitted.  Actually the midboost is "always on" in practice in how you're showing this on the schematic. You omitted "normal" which would've been maybe around 330p.  You might find more reverb tone by using a 330p or 390p where the .002 cap is? The reverb is more pronounced on the treble frequencies, in my experience.



With respect, Tubenit

Well, I guess I'll have to hit the tone stack calculator and review my understanding of the mid-boost function.  Thank you.  For now I can revise the  schematic to say "switch omitted  "


I'm glad you pointed it out so I can get that squared away  before changing any of the caps in the reverb circuit.


In other news, I did try the "dwell control" briefly.  I used a linear 1 meg pot.  Not sure if most would use audio taper.  In any event, turning the dwell down about half way did seem to maintain a sense of spaciousness but with less of a "poing" effect.  Which I like.


Thanks again for your interest in this topic.

[Williamblake]

According to this paper
https://www.pure.ed.ac.uk/ws/portalfiles/portal/12456049/Spring_Reverbation_A_Physical_Perspective.pdf
which is way out of my depth the drip results from the difference in time it takes for high and low frequencies to travel through the reverb spring.
What i do have experienced is that an adjustable high pass on the drive side shapes the the boing/drip just as much as a volume (dwell) control does but leaves the ability to drive the tank reasonably hard so you can still achieve a good output signal alas without drip if you wish so. And vice versa.

[pullshocks]

Lots to report.
I should begin with a disclaimer.  My hearing is pretty messed up.  I use Signia  hearing aids set up by a very knowledgeable PhD audiologist.   They have a “live music” program, that is supposed to be the most natural, least processed setting.  None the less, the sound quality I perceive is not necessarily the same as others would perceive.  I may miss details others would hear, or may not be able to hear differences in an A/B test that others can detect.  And conversely there may be times something sounds harsh to me that would not bother others.

1  Resistor RVR5 (going into the tank)
 RVR 5 changed from  2.7K to 1.5K.  This is based on a 12dw7 reverb schematic by D. Tinnhopher  for an 8EB3C1B tank . Copy attached.  I do not recall where I downloaded it from  .  Note that it calls for B+400 volts, and I have more like 260. 
 I was not able to tell any difference with this resistor change.

2Mid-boost circuit
 I wired in the mid boost switch .  I did not have a 390pF cap available.  Subbed in 270 pF.
 I am not able to tell any difference in reverb performance in either switch position.
 One other preamp note—I do not have the bright cap wired in.  I don’t think it is needed.

Dwell control
 I used a  1 meg linear.  I do notice a difference  turning down the dwell.  It gives a reverb sound I like better, by reducing the  “poing” sound .

RVR1 (Grid resistor on V2A)
 Tubenit suggested to  try 33K in place of 220K.  I have this on an A/B switch. The  33K  gives a somewhat stronger reverb signal, and I “think” it sounds a little brighter.  Not sure if I like it better, but it is noticeable for me.

RC3 (390pF cap to ground at V2A grid)
 Tubenit suggested  to try the circuit without this cap.  I now have it on a switch with choice of 390pF, no cap, or 200pf (on-off-on switch)
 Reverb is noticeably brighter without this cap.

Thoughts about reverb isolation resistor
 I still have 2 meg in this position, with no bypass cap.  I notice when I turn the reverb knob all the way down, the dry signal is not as bright as the signal with reverb.  I will try reducing the resistor and/or adding a bypass cap.

Thoughts on this project so far
 Tonight I felt like I was getting a reverb sound I like.  More tweaking to come, but I like it.  One thing I realized is I was playing without a pick tonight, that may be why the reverb sounded smoother—less percussive attack on the strings.

Closing thoughts
 So far I have only played at one volume setting, and only with the Eminence lil’ Buddy speaker, which is not known for its brightness. 

[shooter]

kludge on top of kludge.

been kludge'n since '79 

[pullshocks]

Have made a couple changes.  Happy with the results so far. 


Not directly related to the reverb, but I switched V1 to a 12AY7.   Also tried 5751, but like the 12AY7 best.  If any Tweed Bluezmeister builders are reading this, I'd be interested in what you liked for V1.  I had not tried the TBM circuit before this project.  I'm really loving it and I've got to say a think it and the reverb go great together.


Changed the reverb isolation resistor to 1 Meg.  Better highs in the dry signal, less total reverb.  Ahhh, the tradeoffs in life.  The next lower value reisistor I have on hand is 470K. 


Still experimenting witht the 500pF-220K-390pF network going into V2A.  Having these components on switches really helps.


Switched PT from 278-0-278 to 320-0-320.  It brought the reverb driver B+ node from 260 to 310V.  I can say the reverb got discernably stronger, but not dramatically stronger.

[tubenit]

Thanks for the update on this project!  I've become quite a fan of 12AY7's in V1!  Although I still like the 5751 tone also.


Just a thought for you ......................   


I've used a passive FX loop in this amp topology before with good success. Generally speaking, I like delay before reverb but that's just a personal preference and I have no strong feelings about the order.


There is a $17 Kmise delay pedal that I think sounds great in front of reverb in an FX loop. I've had Boss delays, Wampler delay, MXR Carbon Copy, Moen Pretty Dolly delay in passive FX loops and found then all to sound good to me.  The Kmise sounds reasonably good to me compared to the other delay pedals.  (The Moen Pretty Dolly has been my favorite by far)
I have the "time" and "repeat" dial at 9 0clock  & the "mix" dial at noon.  Just slight flavoring to the tone but it's working great for me.


Amazon.com: Kmise Analog Delay Electric Guitar Effects Pedal Mini Single Type DC 9V True Bypass : Musical Instruments


You might try that pedal between guitar and amp ........... and if it sounds good, then install it with a passive FX loop with slight delay.  It makes for a wonderful lush clean sounding reverb/delay tone in my experience.


So, $10 for passive FX loop jacks & $17 for a pedal to improve the reverb spacious tone ............ is a pretty good deal, IMO.


Just a thought ............ NOT saying you "should" do this. 





With respect, Tubenit

[pullshocks]

Thanks Tubenit. 
I appreciate your sharing your experience  adding delay/echo before reverb.  That may be something others reading this thread may want to try.  Especially at the price of the Kmise Delay Pedal!!! 
For now, I just prefer to stick to plain reverb and not add another variable.  As stated above, I like the results so far with the transformerless reverb, and find the amp/reverb combination very immersive or absorbing to play.  But it has its drawbacks for a permanent build, and I am open  to the possibility that a reverb effect box in an active or passive loop may work out better.  .  In fact, all my amps were built with that in mind,  as in the attached pictures.  Unfortunately the passive loop approach hasn’t worked out with either of the 2 reverb effects I have.   I don’t currently have a passive FX loop in the breadboard, but I should probably wire one up and try the passive loop one more time.
This whole quest is starting to remind me of the “Rainbow Passage”
”…There is, according to legend, a boiling pot of gold at one end (of the rainbow). People look, but no one ever finds it. When a man looks for something beyond his reach, his friends say he is looking for the pot of gold at the end of the rainbow.”

[tubeswell]

Guitar.com has a snazzy circuit that they lifted from Swart and posted to the web a while back.

Got a schematic handy? (Or a link?)

[pullshocks]

Guitar.com has a snazzy circuit that they lifted from Swart and posted to the web a while back.

Got a schematic handy? (Or a link?)

I think this is the one he means.
https://guitar.com/guides/diy-workshop/build-tube-spring-reverb-unit-amplifier/

[pullshocks]

Making progress on the transformerless reverb. 


2Deaf sent me an updated reverb driver circuit that increased reverb very noticeably.  He has been a tremendous help on this project.    The driver plate resistor is reduced to 15k and the cathode resistor increased to 750R.


2Deaf also suggested  to change RC6 to 2.2nF from the 500pF used in the “Classic 1tube reverb.”  He says the larger cap works better as far as an LC resonance. The change reduced the. brightness of the reverb.  To restore some of the lost brightness, I will be trying different values of RC3.  As mentioned above I have a switch set up to A/B those caps. 


Years ago, I tried an Ampeg GU-12.  I remember the “Dimension” control being pretty cool.  I did not know  then that it is a capacitor driven reverb, a one-tube capacitor driven reverb at that.  Looking at the GU-12 and other Ampeg schematics, their return circuit uses  470K plate and 3k cathode resistors, a higher gain configuration. 


I decided to try the Ampeg Rp/Rk values in the breadboard, and it did give an additional increase in reverb level, with no drawbacks that I can detect.


Interestingly the 470K/3.3K combination is used in the “Swart-style” reverb in the Guitar.com writeup that is linked above.


I want to do some more tweaking, but I think this is getting there.  I’ll try to get some recordings up in the next week or so.

[kagliostro]

The old italian FBT 500r & 500r2 amps had a Transformerless Reverb on it


(the ecl82 is a 6bm8 tube)





Franco

[pullshocks]

Thanks Kagliostro.  Interesting amp FBT 500 R-2 amplifier (youtube.com)

[pullshocks]

I tried bypassing the reverb isolation resistor with a 10pF cap.  With my hearing loss, I could not tell a difference.


I have been clawing my way up the learning curve of sound recording with my Presonus Audiobox 96 studio package.  I have a long way to go with my recording technique, but here is a demo of the breadboard amp (schematic above), through an Eminence Lil Buddy speaker.   Guitar is my homebrew T-thinline style with PAF style pickups from Stewmac Thinline project completed. (el34world.com)




 I recorded this with the  Presonus M7 condenser mic close to the speaker cabinet grill cloth.     The recording sounded boomier than what I hear in the room, so I used the software to roll off lows.  I did not use any compression or other processing

https://on.soundcloud.com/eXxLevenXoGFzv7B7

[Carlsoti]

FWIW, I have a Magnatone M10 with a capacitor-input, xformerless reverb that is SOOO deep, though I can't recall the input/output impedances. The downside is that it's a single tube driver, with a transistor recovery circuit.

[pullshocks]

Carlsoti, thanks for posting.  I hope I get to hear an M-10 some time.

[pullshocks]

I now have  1-tube capacitor driven reverb under test in my "5 Watt Imperial" single ended 6V6 build.  As with the previous versions, the circuit design came from Too Deaf.  This version includes an LND150 stage between the reverb isolation resistor and the power amp. 

The Imperial reverb circuit switches some components depending on whether the rhythm or lead channel is used.  If I understand correctly when you switch to the lead channel it reduces the reverb drive signal (like turning down a dwell control) and rolls off lower  frequencies

I'm going to experiment with this

[tubenit]

Thanks for sharing your work!  Appreciate it. 


With respect, Tubenit

[pullshocks]

Ok I wired in a lead/rhythm switch for the signal going to the reverb driver.  It is sort of an approximation of the Tone King circuit shown in reply 32 above.


It makes a difference.  Without the switch, the reverb was a little harsh when pllaying hard on the lead channel.  The new switch takes away that harshness.  There is still just enough reverb there, but it is more balanced.


Right now it is not foot switchable, but I can easily do that with the unused half of the channel switching relay.

[pullshocks]

Slowly migrating the tranformerless reverb circuit from the breadboard into the chassis.  Talk about cramming 10 lbs of s**t into a 5 pound bag....keeping my fingers crossed as to lead dress and cross-talk & interactions.


Taking the opportunity to upgrade the power supply to fully choke filtered.


This chassis has been reworked several times, as you can see from all the adapter plates and leftover holes.  The choke goes in the area circled in red

[DummyLoad]

There is actually a G spec tank - Mostly used in in the newer Carvin amps before they went bust, Carvin used G tanks with a single tube reverb, driven by 1/2 12AX7 with capacitor coupling. No one at the time I was fiddling with the MTS 3200 sold them in the US, Tube Amp Dr. had them in stock, and I ordered four, one for a backup for my MTS 3200, one to experiment with, with the remaining pair for service parts. They seem to come up for sale on Reverb.com occasionally. Might get that deep surfy tone with a single tube reverb CC ckt with a G tank - Give it a try. Ampeg used a 6SN7 in early amps as a one tube CC, later amps use the 6CG7/6FQ7 as the driver and 12AX7 recovery.

9GB2C1B is the type designation - Input Z is 8KΩ; Output Z is 735Ω

--Pete

[pullshocks]

There is actually a G spec tank - Mostly used in in the newer Carvin amps before they went bust, Carvin used G tanks with a single tube reverb, driven by 1/2 12AX7 with capacitor coupling. No one at the time I was fiddling with the MTS 3200 sold them in the US, Tube Amp Dr. had them in stock, and I ordered four, one for a backup for my MTS 3200, one to experiment with, with the remaining pair for service parts. They seem to come up for sale on Reverb.com occasionally. Might get that deep surfy tone with a single tube reverb CC ckt with a G tank - Give it a try. Ampeg used a 6SN7 in early amps as a one tube CC, later amps use the 6CG7/6FQ7 as the driver and 12AX7 recovery.

9GB2C1B is the type designation - Input Z is 8KΩ; Output Z is 735Ω

--Pete

Thanks Pete.
In post #26 above I included the Carvin Vintage 16 circuit with the G tank, along with some other commercial cap drive reverb circuits. AFAIK Carvin used this in several other amps.
I have never heard any of the Carvin amps nor found any good on line demos.

[pullshocks]

Here are a couple of the Carvin schematics.


I was curious about them, but not curious enough to order the G tank from Germany.  Like Dummyload, I was unable to find any vendors in the USA offering the G tank.