5F1 Champ w/ Capacitor Coupled One-Tube 6BM8 Reverb (Eric Barbour/Ampeg-style)

[caps8419]

Thank you, since I'm adapting this to a 6U8A I'll mess with these values and the existing voltage divider I have implemented now. Trying to eliminate components so we'll see what happens!

[dchang0]

It helps to think of the reverb circuit separately from the passive mixing components. I treated the reverb circuit as a black box when I was working out the problems with my passive mixer (with help from some people from this forum).

Certainly the reverb circuit was developed in a vacuum long before attempting to splice it into a 5F1.

Keep in mind that I was on a very strict budget for space. I could not add or use a lot of components or wiring due to the tight confines of my 5F1 chassis, and that includes forgoing something as simple as a voltage divider. You will probably want to choose something far better than what I have here; the flip side is that this schematic is the bare minimum to get the reverb spliced into the amp.

[afm_380]

At the risk of being chastised for defacing a vintage silver face dual showman, here's the board that I built with this design (although this pic was taken prior to wiring it up). I had tried multiple attempts to get the tremolo circuit to work correctly, then I ran across this thread. So ... I pulled the 12ax7 out and rewired the socket and circuit to a 6bm8. Now this dual showman has a nice reverb and gets the playtime it deserves instead of gathering dust ...

[dchang0]

That is AWESOME--may I ask what make and model of caps you used? They are far smaller than the Orange Drop 716Ps I used...

[dchang0]

g1 caught a typo in post #87 above. Item 2) should say "500pF" and not "500uF".

[j, claudius]

Hi dchang0 I am going to attempt this build. I have found an antique tank 1FB2C1B. Wondering if I need to tweak the impedance or power from your design. Do you have those characteristic curves for the 6BM8 Pentode-as-triode? That original doc as googled seems to come up with dead links. Thanks and I will let you know how it goes! Thanks for your design and information.

[dchang0]

Very cool!

Sure, here are the curves of the pentode-as-triode that I used:



I just realized that the printed hardcopy that I drew the actual line on is not in my files. I must've thrown it away without scanning it and storing it as a PDF. I'd have to reverse-engineer the line I chose based on the values of the components in the schematic. (Actually, looking back at the other thread, I described in words where I drew the line.)

[dchang0]

Turns out I did scan it into a PDF file.

Here is the chart I used to design the original 6BM8 single-tube reverb. Check out the lines in pencil. I ended up choosing the one labeled in pencil "5K" on the y-axis.

[dchang0]

Necropost. I still have this amp and use it as my primary practice and recording amp.

Over time, I've tweaked some of the resistors to clean up the reverb sound and adjust the wet/dry mix. This is Rev. 1.0m, and it sounds the best to my ears.

The components in red were changed (from Rev. 1.0l) and the one orange component was added.

a) 6BM8 driver side strapped-pentode plate resistor increased from 5K ohms to 10K ohms to improve impedance matching with the tank's input transducer for higher frequencies. Overall result is a cleaner, more transparent/airy reverb. 5K is probably still more optimal; 10K results in a pretty oblique line relative to the curves.

Side note: I recently read this article published in 2012 and realized that I probably should have gone with an 8E tank (lower 800 ohm input impedance) instead of the 8F tank (1925 ohm input impedance) so as to go for the constant current design. Someday I might drop-in an 8EB3C1B tank without changing anything else to see how it sounds. I could probably go back to 5K if I went with an 8EB3C1B tank.

http://www.channelroadamps.com/articles/reverb_driver/

b) 6BM8 driver side grid leak resistor went up from 470K to 1M.

c) 6BM8 recovery side grid leak resistor went down from 1M to 470K. I might try as low as 220K (typical Fender 6A14 value). When it was 1M I was getting feedback from the reverb tank pickup transducer due to its close proximity to the speaker (everything is stuffed inside a tiny 5F1 tweed champ combo cabinet--the reverb tank is less than 1 inch from the 10in. speaker's rim). Feedback no longer occurs even if both knobs are maxed.

d) The main mixing resistor went up from 100K to 180K and the resistor off the Reverb pot's leg went down from 33K to 27K to adjust the wet/dry mix. I arrived at these two values using gradual trial and error. They can be changed to suit your preference for how the Reverb knob behaves. I tried as high as 1M for the main mixing resistor (keeping all other values the same), but that was way more wet signal and way too little dry signal than I wanted. I also tried as low as 10K for the main mixing resistor, but the wet signal was so anemic I couldn't get a decent slapback with a hard, muted strum. Also, these resistors were previously carbon comp, and they developed some crackling noise over time (like a vinyl record), so I replaced these with higher-wattage-than-necessary metal film resistors to get back to clean.

e) The one orange-colored symbol is a resistor I added per the channelroadamps.com article mentioned above (they refer to it as R6). It is for the rare case when the reverb tank might be disconnected. The value doesn't matter so long as it's a lot higher than the impedance of the input transducer of the reverb tank. I chose 180K because I had a spare. Anything above 50K is fine. I put mine off-board but still inside the chassis, across the legs of the amp's jack that goes to the reverb tank's RCA input jack.


I also tried various capacitors in place of the 500pF at the main mixing resistor including no capacitor at all and went back to the 500pF. The above tweaks made more of a difference. I might go back now and focus on tweaking just the 500pF and .47uF capacitors to see what tonal range it has and try to add more midrange to the wet signal. pietro_moog mentioned a long time ago that the 500pF cap forms a low-pass filter with some other components; it might be worth figuring this out.


Schematic is attached.

[dchang0]

Turns out I did scan it into a PDF file.

Here is the chart I used to design the original 6BM8 single-tube reverb. Check out the lines in pencil. I ended up choosing the one labeled in pencil "5K" on the y-axis.

Re-attaching this scanned image that the forum software lost, especially since I did eventually go with a lower/flatter line (10K ohm instead of 5K). I had planned on testing values in between such as 6.2K ohm and 7.5K, but I just went for 10K and was pleased enough that I didn't bother trying the others.

[Helmholtz]

Sorry, I didn't read the whole thread.
My question is, why triode mode, as pentode would provide much better current drive?

BTW, the DC loadline doesn't tell the whole story.
What matters with signal is the AC loadline, which is much steeper because of lower impedance (around 2k@1kHz).

[dchang0]

Sorry, I didn't read the whole thread.
My question is, why triode mode, as pentode would provide much better current drive?

BTW, the DC loadline doesn't tell the whole story.
What matters with signal is the AC loadline, which is much steeper because of lower impedance (around 2k@1kHz).

I chose to strap the pentode as a triode mainly because of a desire to reduce total number of components due to the tiny space available inside a 5F1 chassis. If I were building in a larger chassis, I would likely have gone the pentode route, as it wasn't that much more to do (one 22uF 450V capacitor and one 30k 10W resistor, both too large for my 5F1).

So yeah, this circuit could be a lot better. I invite anyone else to modify it into pentode mode.

And thanks for the wisdom re: AC loadline. That might be the graph someone asked for that I couldn't find. There is a high chance I threw it away in a pile of other papers.