Eric Barbour's 6BM8 One-Tube Reverb Imagined...

[überfuzz]

What size are the coupling/high pass caps you're using?

[dchang0]

Do you mean physical size or capacitance? I assume you mean capacitance...

All the coupling caps are 715 Orange Drops, 0.022uF, EXCEPT for the 0.470uF cap coupled directly to the reverb tank.

All my high-pass filters are tiny silver mica 500pF.

Check the other thread (5F1 with 6BM8 reverb) for the entire circuit that I built and tested.

[pietro_moog]

i'm thinking about this reverb.
is it good sounding? and reliable? is it far from fender?
can the .470uF cap be something like .1uF ?
what's the best spring can? i can't find this type

[dchang0]

Great sounding, yes. It's a lot like a Fender Princeton Reverb with its famous "surf tone," though a little bit thinner in the midrange and fuller in the bottom. This scoop in the midrange may be a consequence of the smaller tank I am using--a type 8 short tank with medium delay rather than the typical type 4 long tank with long decay. It might also be the consequence of an impedance mismatch, because this circuit is originally intended for the highest input impedance 8FB tanks, whereas I am running the next-highest 8EB tank (the 8EB2C1B for Blues Jr., which was all I could find but are plentiful and cheap--you can find them at tubesandmore.com and tubedepot.com).

Still, as it stands, the reverb sounds great--dark and woody. I get lots of "slap" if I palm-mute the strings and strum. This was a design goal when I started out--I wanted full-on Fender surf-style reverb but in as few tubes and components as possible. I went to all the trouble of figuring out how to triode-strap the 6BM8's pentode in order to get more than enough current to drive the tank at maximum just to get that slap sound.

I am currently running this circuit in a 5F1 tweed Champ build, whose fully-tested schematic (rev. 1.0i) is in another thread on here. That amp is fully capable of going all-out surf reverb. I run the reverb at around 5/10, but it is responsive through the full range all the way up to 10/10, which is much better than the reverb knob on a Princeton Reverb which seems to top out at around 3/10. My only complaint is that I didn't get the passive mixer down just right so that there's a slight dip the dry signal volume at 9/10, but that's not noticeable unless I point it out to the player. The guitarists I've shown this amp to have all had huge grins on their faces after playing it--they've never heard a tweed Champ with a lush reverb like this! People have liked it so much that I am being hired to install reverbs in other amps.

The final schematic (rev 2.0f) shown above AND the final schematic shown in the 5F1 with 6BM8 reverb thread (1.0i) are not truly current. I did some further, now-forgotten tweaks to the passive mixer to get the reverb's surf capabilities to come out. However, I did not modify the driver and recovery sections, so these schematics are both accurate in that regard. You CAN use the final schematic to full surf-effect if you simply modify the passive mixer section to suit your application.

Yes, you can change the .470uF cap to whatever you want such as 0.1uF or 0.22uF.

[tubeswell]

Here's a 6BM8 reverb from one of the guys at the dumble forum. I've never tried it but it looks interesting. Not sure how good the stage's Rk and Rg2 values are for it tho'.

There's also a 2 x pentode self split PP driver for 6BM8s originally posted by Steve Conner* (IIRC) a couple of years back, which looks downright provocative.

(* apologies to Steve if it wasn't you)

[dchang0]

That other 6BM8 reverb schematic reminds me to remind anyone considering building my design: I worked under very limited design contraints: "get Fender-like surf-style reverb in as few components as possible." Part of this was due to practical matters: I was fitting reverb into the tiny 5F1 chassis. But part of this was also due to trying to imagine a "purely exemplar circuit" along the lines of what Eric Barbour would've done for Svetlana.

Certainly I stripped out many components that could serve to improve the tone in order to achieve this goal. For instance, Ampeg's cap-coupled designs nearly always use an additional lone 22K resistor parallel to the reverb tank's pickup to properly match the load. I'm sure I'd get better, lusher results if I had included this and done the math to figure out the proper value for my design. (In all likelihood, 22K is close enough--the output impedance for the 8EB tank is nearly the same as that in the 4FB.)

All of you are welcome to take my "foundation" and improve upon it--and please share your results with everyone when you do. Many of you aren't working under the constraints I was and could really get all that the 6BM8 has to offer as a single-tube reverb, which I believe is more than a 12A?7 single tube reverb can offer.

[dchang0]

An update: I increased the 22K grid leak on the reverb recovery side to as high as 470K without trouble. THIS REVERB SOUNDS AMAZING! At 470K, it becomes very obvious that the 6BM8 driver is much stronger than the 1/2 12AX7 single-tube driver designs out there. With the 6BM8 triode-strapped-pentode, you can hear the springs getting twisted about very clearly on strong palm-muted strikes.

Reverb starts in gently from 1/10 to 2/10, then gets strong from 3/10 to 5/10. Goes to Fender surf-ey at around 6/10 to 8/10, and then outer-space reverb-ey at 8/10 to 10/10. With the recovery grid leak at 470K, most people would find the useful range of this circuit to be between 2/10 and 5/10 and anything higher than that to be too much reverb. Drop the 470K to a more reasonable 270K (also tested) to shift this range towards around 5/10 to 8/10. The tonal characteristics aren't different, because the drive side is still the same, but the wet/dry mix seems more reasonable.

I also tested the recovery grid leak at 22K, 68K, and 100K and found that it's much too weak at anything less than 220K. I'll update the schematic with my findings.

Most important, I am finally, finally happy with the way this thing sounds!

Final, final schematic, after over six months of playing/tweaking, 2.0g.

[imaradiostar]

I've lost track- what kind of amp is this in and where is it inserted in the circuit and what value of resistor are you using on the dry side of the reverb blend?

jamie

[dchang0]

Yeah, it's been six months. Even I lost track of the changes I made, so I opened it up again.

It's a 5F1 Champ. See this thread for the exact schematic. 5F1 Champ w/ Capacitor Coupled One-Tube 6BM8 Reverb (Eric Barbour/Ampeg-style)

I am currently running the recovery grid leak at 470K, not 270K as shown in the schematic, but with 470K the levels are far beyond what most people would probably want.

[dchang0]

It seems my schematics were lost in the calamity that occurred some time ago.

Here is a re-post of the last revision of the single-tube 6BM8 schematic, 2.0g:



A reminder: you will likely have to modify this design to get the levels of reverb you prefer. The current design is quite powerful and can overdrive the reverb tank depending on your input signal (especially if you input a lot of midrange and bass).

I also suggest that you NOT use 470nF for the coupling cap--it is kinda overkill, especially in terms of its large physical size and higher price. A typical .022uF coupling cap is probably good enough and far more manageable in size. I only chose 470nF based on a Valve Wizard schematic and did not test to see if it was any better than .022uF.

[Chuck H]

It might be good to mention that the circuit seems suited to the modern Russian 6bm8 currently being made (triode AF=70, pentode power=1.8W). I only mention it because data sheets make it look like an entirely different tube than a vintage/nos 6bm8 (triode AF=20, pentode power=3.5W).

[dchang0]

Ah yes, the intended 6BM8 of this design is the still-in-production and very affordable Electro Harmonix 6BM8.

If anyone happens to modify the circuit for the vintage 6BM8, please post the data sheet you used and resulting schematic here--thanks!

[Niquel]

Hi,
dchang0, I have just added your 6BM8 Eric Barbour's inspired reverb in some sort of a Princeton Reverb with good results.
I've used triode strapped pentode with 282V from EL84 screen-grid supply, 500pF and 220k for pentode grid, 470nF for plate into a 9EB1C1B. I need to experiment a bit more with the recovery stage and maybe lower a bit attenuation to the mixing stage.
Thanks so much for sharing!

[j, claudius]

Attempting this as well

Hi I am attempting this as well... with a head-style, 12W OT, 1FB2C1B tank, a 5E1 circuit (with choke), and I'd like to try out a dwell pot. I've posted on the other thread with the extensive tweaks described - Will update....

https://music-electronics-forum.com/...t=22535&page=3

[dchang0]

Someone PMed me over a year ago and I didn't get the email notifications. I just happened to stumble across the message now.

They were asking me for the schematics in this thread, saying they can't see the attachments, so I'll post the latest up again here in case it was lost in the great image calamity that happened a few years ago.

This is the last version I came up with, 2.0g.