Cap-driven one-tube reverb

[Diverted]

Hi all,

Attached is the schematic for the Benson Monarch Reverb's one tube, cap-driven transformerless reverb. I like this reverb, it sounds great and I am interested in using this circuit in a single channel AB763 circuit with tremolo and fender reverb removed.
My question is, will this work with the Fender TMB tone stack, if I set up B+ at the same voltages Benson uses? I'm probably going to give it a try but wondering if something obvious eludes me (likely, as per usual). Thanks for the input (schematic shows both gain stages but I cropped it just before the phase inverter).

[passaloutre]

What type of tube are you using for the reverb? I don't see it on the schematic.

[Diverted]

Oh sorry, I didn’t notice it wasn’t on there. 12DW7/7247.

[roarshock]

I like that reverb too. I've built similar ones a few times. If you connect the reverb like Benson, with the tone controls parallel to the reverb circuit and them combining before the second triode, you should have no problem. The reverb might sound even bigger due to the lossy nature of the TMB.  I'm not sure if you're concerned with the lower voltages impacting the tone, but I made a test TMB tone circuit with lower B+ and it worked great. Happy building!

[Diverted]

Thank you Roar, that's the info I was lookiing for. I thought the reverb might be a little excessive with a TMB circuit, but I'm sure I can tweak it if so.

One thing you wrote that I'm curious about are the lower voltages.
In the schematic shown here which I built (it sounds great) preamp voltages are indeed way lower than in the AB763 I plan to build.

Should I just add a new filtering node with a dropping resistor that will get me down to the voltages the reverb circuit I built has? Apart from the impact of the TMB lossy circuit, this is the main thing I'm wondering about before I try to throw it all together.

Thank you!

[tubenit]

In ARCHIVES of frequently addressed topics under "One Tube Reverb" in the first post is some information on different one tube reverb voltages that may be useful to your thinking.


https://el34world.com/Forum/index.php?topic=7957.0


You can also check out a Gibson 17RVT which had a one tube reverb and look at it's voltages which are low


Gibson_ga17rvt.pdf


With respect, Tubenit

[roarshock]

Diverted,
Here's two  schematics of amps I built using a similar reverb circuit (although it's not a CF driver) a while back with voltages included. both have plenty of verb for me and do have a separate node for the reverb.

[Diverted]

Thank you Roar!
These are great! I think I will probably end up sticking with the one I know, which I like, if I can make the voltages work. But I will definitely save these and try them out. I love different reverbs.
In the meantime, can you provide a little more guidance?

So as built, the Benson Monarch Reverb has five nodes:
B+ (OT):  357
B+1 (screens): 340
B+2 (PI) 307
B+3 (pre): 285
B+4 (reverb): 275

The Benson B+4 drops from 275 down to about 270 at the reverb first plate via a 4.7K plate load resistor from the B+4 node.

Whereas the circuit I plan to put it is at about 400VDC at the preamp filter node, which drops down to 275V at the first plate via a 100K plate load resistor. So am I right in assuming that I could just use a small value dropping resistor via B+3 and 4, and then a larger plate load resistor, say 100K, to get reverb plate down to about 270V?

My other question is the reverb first screen, which in the Monarch is supplied with DC by first preamp plate voltage (203V), through a 220K resistor, giving DC voltage at the reverb screen of 198.
In my amp, I'll need to drop the 275 or so preamp first plate voltage down to that 198 number. I imagine I'll have to use a higher resistor to get down to 198. Or a voltage divider?

Any thoughts or observations on my numbers? Either way, thank you for your help so far!

[passaloutre]

The difference between the Benson circuit presented here and all the others is that the Benson reverb driver is a DC coupled cathode follower. It is not adding gain or modifying the signal (at least not intentionally), it's being used to drive the low-impedance input of the reverb tank from the signal at the plate of the first preamp stage. Your chief concerns here will be to keep the plate voltage below the 330 VDC limit of the 12DW7, and keep the cathode less than 200V from the heater. The 4k7 here is not a plate load but a voltage dropper for the reverb node (you can tell this is a supply node because it has a filter cap). The cathode follower has no plate load, it is directly supplied by this node, and the reverb recovery stage is tied to the node with its own 470k plate load.

In other words: I think you're overthinking it. Get the node right (or at least get it under 300 or so) and build as Benson did.

I would also read up on Merlin's DCCF page:

https://www.valvewizard.co.uk/dccf.html

[roarshock]

Tubenit: Thanks for keeping knowledge that's already available flowing. I've scoured that thread a few times.

Diverted: I copy circuits and parts of circuits and don't always understand every part of why they work or what's important. I have seen a trend with cap coupled reverb drivers. They usually have smaller plate load resistors. If I was in your spot, I'd drop the B+ at the power rail and leave the plate load resistor stock. You already have higher B+ than any schematic I've seen for a monarch, so you might want to lower the voltage earlier in the power smoothing section. I built a monarch with pitch shifting vibrato, ala Magnatone and had 6K6s in the power section. I think I built it in too small of a chassis and could never get the hum and hiss to go away. My challenge while building it was to keep the B+ high enough. I think I started with 300 on the plates of the 6K6s. Anyway, I'm a sucker for cap coupled reverb stuff. Build on and keep us all informed!

[Diverted]

Will do!
I've actually started putting in the circuit now and expect I will have to do a lot of tweaking. Thanks for the help so far.

[Diverted]

Well I just finished wiring in the reverb circuit. Good news: IT WORKS! However needs some tweaking.
So far I have only powered it up to about 110VAC as some of the numbers are high. So, a few questions:

1. I used a 47K dropping resistor between preamp filter cap node and new reverb node. That brings preamp B+ supply down to about 240 at the reverb filter cap. I expect once I give it 120VAC it'll end up right around 270-290. So that's looking good. However would I be better off using a lower value dropping resistor and putting a plate load resistor between the filter cap and reverb plate 1?

2. The reverb screen grid 1, which gets preamp AC signal and dc voltage from preamp plate 1 via a 220K resistor. I upped that to 470K. At 110 VDC I'm getting around 220VDC at the screen. So I expect that would climb to 250 or so at full wall voltage. Monarch runs about 200 on the screen 1.
What can I do to reduce this voltage down to about 200? Voltage divider? Other solutions?

Other than that it sounds good and it's clean and quiet.
The reverb effect is not as strong as it is on the Monarch, but on the monarch I built the reverb pot is 1M. I used a 500K in this one as I was thinking the reverb might be louder with the more lossy TMB tone stack and I was trying to accommodate that. I may go up to a 1M.

Any suggestions or observations are appreciated! I don't have anywhere near the skill most of the folks on this site have ... I just like to put stuff together with my limited knowledge, and appreciate all the help so far.

[passaloutre]

A plate load resistor is only going to eat up signal that would otherwise go into the reverb tank.

The grid voltage is going to follow the plate of V1. I don’t think changing grid stoppers is giving you the control you want.

[Diverted]

Thanks, figured it was going to be a challenge.
So then, I'm looking at increasing value of dropping resistor between B+2 (screens) and B+3 I guess?

Currently the power rail resistors are choke>4.7K>4.7K. Maybe start off trying something like choke>10K>10K?

Or adjust reverb cathode resistance to work with higher screen voltage? Not sure how I would go about doing that, but trying to consider all the things that could be changed.

[Diverted]

Hi all,

I've been playing around with values and am getting there. A few questions but first:

1. Swapped out 1K>4.7K dropping resistors feeding PI and preamp for 4.7K>10K.
2. I put in a 27K dropping resistor to supply reverb plate 1 voltage at the new reverb filter node.

Running the amp at 122VDC, voltages:
On reverb plate, now getting 267
On reverb grid, now getting 211
On reverb cathode, now getting 223

The reverb sounds excellent. However ...

This cathode voltage exceeds the heater to cathode voltage referenced by passaloutre (thank you for that info). How about elevating the heaters 20VDC or so? It's a fixed bias amp and I've never elevated heaters on this kind of amp before. But I am not using the 50V bias tap (I tapped into one of the HV secondary outer legs instead). Would it be an option to rectify that 50VDC feed, possibly use a voltage divider to get it down to around 20VDC or so, and elevate the heaters that way?

Thank you!

[passaloutre]

There’s another thread in this forum right now about elevating heaters using the bias tap. Sounds like a good idea to me, but maybe easier to just drop the reverb supply a little bit more…

The middle electrode in a triode is called the grid, not the screen, by the way. Maybe in other languages both words translate the same?

[Diverted]

Gah. I know that ... I have been at this for hours now and my brain is a bit mush. Sorry. I will be more careful.

I may want to try the elevated heater first. I had the plate voltage down to around 230-240 and the reverb effect suffered. It really came to life when I dropped down from 47K to 33K and finally to a 27K dropping resistor to the reverb filter node. It sounds fantastic with reverb plate voltage around 267. I have room and don't mind putting in a heater elevation circuit ... I'll look for that thread. Thanks again.

[passaloutre]

You have the right idea: rectify the bias tap, divide it down to what you need, and apply it to the heater center tap (or artificial center tap). I usually defer to Merlin for building blocks like this: https://www.valvewizard.co.uk/bias.html

Remember you need a positive voltage, so reverse the diodes and electrolytic caps from those drawings.

Let us know how it works out! I’m definitely intrigued by the cathode driven reverb. Do you mind sharing the rest of the Benson schematic?

[Diverted]

Thanks, and sure — schematic and .diy layout attached. This was sent to me months ago by someone else on a different forum. I also reached out to Chris Benson when I started looking for one. He didn't seem to mind the data being shared around (super nice guy) so I suppose it's no problem.
If you follow the layout and particularly the ground scheme, it's super quiet.
I'll report back once I build the elevator. Many thanks for the help.

PS: In the original schematic (which I changed slightly) R2, the phase inverter balance resistor, is a 10K linear pot. If you don't want to use up a pot for the PI balance, just tack one in temporarily and measure the resistance that gives the least hum without killing the signal. I ended up using a 6.8K resistor in its place.
Also, C3 value not shown on the schematic but is listed as 10uf on the layout. I had an original on my bench, as built from the factory, and while it was covered with electrical tape it was bigger than 10uf. I didn't bother to pull it to get the exact number, just used a 47uf.
Also, reverb pot is shown as 500K on the shematic, but 1M on the .diy, which I changed. The original I had had a 1M.

A few other notes: transformer secondary is 275-0-275. Original was a Mercury Magnetics, I used a Hammond 270EX which gave the right voltages rated @144ma. My voltages are probably slightly higher than original which used GZ34 for rectifier. I didn't have one so went with diodes.
Also the switch is a really cool three-way off-standby-on. I ended up finding the right part ... if you decide to build it with that switch and need a point in the right direction let me know. If I had built it with a tube rectifier I probably would have left it out.
Reverb

Ted

[Diverted]

Just wired in the elevated heater circuit:
Bias supply>18K>diode>56K/100uf>ground. Heater center tap goes to positive side of filter cap.
It works: Getting about 25VDC on the heaters, reverb triode 1 at 269 plate, 210 grid and 220 cathode. So I should be good there, correct?
Only issue is that now the amp has a buzz/noise that was not there before that gets increasingly loud as the volume is turned up, no guitar signal input. Prior, I got no noise turning up the amp with no input source other than standard empty amplification/white noise.
Elevation circuit is grounded to one of the transformer mounting bolts with the HV secondary CT.

EDIT: Just added a second 100uf/100V Nichicon to increase filtering. Little/no change. Might I be better off creating the center tap somewhere else, possibly after the choke to get even more filtering?

Or should I dial back the elevated voltage a bit? In the original the cathode is a bit positive w/respect to the grid. If I subtract the elevation, it's a little bit negative (210 grid/195 cathode). I should add that the noise is not caused by the reverb circuit. Turning it up or down has no effect ... it varies with position of volume pot.

Thank you!

[passaloutre]

Do you have two stages of filtering in the bias supply, like the Merlin circuits? I think multiple filters is better than one big filter.

[Diverted]

No. I pretty much copied the Princeton Reverb circuit, upped the capacitance and changed resistor values. I will have a look at those.
Thanks.

[Diverted]

Something like this?
I'm not really understanding this schematic — not sure what P1 represents.
Also changed descriptions; It was originally negative rectifier at left and grid leaks at right

Or regarding having a second stage could I modify mine as built to add a second stage?
Currently it's xfmr bias lead>18K>200uf/56K>ground, with 6.3V CT to positive of filter cap.

Thank you!

[Diverted]

Or would something like this work? I remembered that Fender Tremolux 5G9 uses two stages of filtering, and photoshopped this out of the schematic.

[passaloutre]

P1 is a potentiometer, and R3 is a protection resistor in case the pot fails. If you don't need it to be adjustable you can just use a fixed resistor divider, which simplifies to the second drawing you showed. Start with 10uf caps and a 10k/10k divider. You don't need the big dropping resistor because it's already a low voltage tap. Like the attachment, but with a fixed R instead of a pot (again, note the diode and cap polarity in your circuit will be opposite of this).

Not sure if this is the source of your new noise, but it should be easy to try it connected and disconnected to the heaters to find out.

For what it's worth, plenty of classic designs have cathode followers above the rated voltage. You may find it's not necessary, or you may find yourself replacing 12DW7 tubes frequently. Are you using new or old tubes?

[Diverted]

Thanks!
I am using NOS USA Tung Sols, I have a few of them.

I went ahead and wired in the Tremolux circuit, I used 82K/56K resistors which is what that circuit had, which puts out -25V or so bias voltage.
I used 100uf/100v caps as everything smaller in value that I have are axials and are rated 500vdc. No room! Elevation at 24v and the noise is gone! But replaced with something that almost sounds like motorboating…. A faint hollow sounding tap tap tap.
Should I try to scrounge up some lower value caps? Thank you!

[passaloutre]

I don't think reducing the cap value is important. I've used plenty of 100ufs when recapping Twin Reverb bias supplies with no issues.

Tap tap tapping sounds more to me like a lead dress issue or possibly a weak solder joint.

[Diverted]

The tap tap tap went away .... and then it came back — when I placed my iphone too close to the preamp tube on my bench I wish they were all that easy to solve.

As it stands now, I think it's getting close (don't want to say all the way there because the gods are always listening).
Amp is noise-free apart from amplification white noise when turned up. At 123VAC from wall, 12DW7 triode 1 currently stands at:
Plate: 271
Grid: 214
Cathode: 226
6.3V elevation: 26, so I'm right at that 200V limit.
I've read exceeding that limit shortens tube life. Hopefully being right up against it doesn't also. If so I can always increase the elevation output voltage I suppose. Would you recommend I drop it down a little more? Perhaps a 120-150K resistor coming off the bias diode? (currently 100K)

Either way, I hope I don't have to come back reporting problems. But right now the amp is stable and sounds great. Reverb is much, much more pronounced than on the Benson Monarch, approaching Fender swamp country turned up — perhaps because the TMB tonestack is more lossy than the singlee tone control in the Benson? I've got the same 1M reverb pot the Benson had in it, may swap it out for a 500K.

A little bit of backstory: I originally built this amp with stock fender reverb but due to the confines I'm working with and probably some layout/placement mistakes I made as a result, it was super noisy with noise even at idle, volume 0. I couldn't tolerate it and could not root it out (it originated on the grid side of the 3.3M mixing resistor) so I removed the board, cut off everything before V3 (reverb return/third gain stage), rewired that V3 socket for the preamp, and put the reverb socket where V1 used to be. Built the reverb circuit on a terminal strip as shown in that layout I posted. It was a hail mary pass and I'm very happy to have made the amp more well-behaved.
Many thanks to you and other posters' invaluable guidance. I'm not as technically minded as I'd like to be, but I'm learning. Stuff like this helps a lot.
Will report back with any other issues that crop up, but hoping they don't. Gotta play around with it this afternoon and see how it behaves after being on and played a while.

Thank you.

[passaloutre]

Awesome dude! Glad it’s working and sounding good! Got any photos or sound clips?

I might have to try out that reverb soon, and the Monarch itself reminds me of my Supro Thunderbolt, which is a long time favorite.

In not sure what the voltage limits of heater elevation are, but Merlin talks about 30-60, so I think you can safely bring it up a bit more to get you further into the safe zone for your 12DW7.

[Diverted]

Thanks, I'm going to increase that elevation a little. Can't hurt going up another 10V.

No vids yet, the amp hasn't sounded good enough before today. Plus I need to stuff everything in the cabinet and finish it up before I do it. But I'll swing back in a couple days provided nothing goes sideways. Here's a few photos. Forgive the disaster of a bench, burnt insulation here and there and crappy look overall — it was all nice and neat before I decided to go nuclear and rip out half the circuit hahaha.

[passaloutre]

Looks great! What’s up with the separate chassis for the PT?

[Diverted]

I wanted to put it in a small cabinet, and the way I built it originally, with reverb transformer on top of chassis, no way it would have all fit. So I dropped the transformer onto its own chassis and ran all the AC secondaries plus wires to on/off switch up through that harness. Cabinet is 20x17x10.5 deep.
I've done the same thing for a two or three friends and it's always worked out well so I thought I'd do it again. This time for whatever reason It was noisy as hell. Weird. Anyway it's no problem because I'd been wondering how the Benson reverb would work in a higher voltage amp. So it ended up being good in the end... thanks again!

[Diverted]

Here's the cab.