Belton Digi-Log Digital Reverb to Replace Solid State Driven Tank

[Krwkka]

I just ordered one of the Belton Digi-Log reverb modules to attempt to sub it in for a 4EB2C1B Accutronics tank on a combo amp.

Has anybody tried this? Any tips?

I did see this thread here, but has anybody actually accomplished this?

Digital Reverb in Tweed Princeton
_

[NorCalTuna]

I'm literally doing that right now on a dr clone ish thing. I'll update this in a few days-

[Krwkka]

I would greatly appreciate that!

[jrfrond]

I built a stompbox around the Belton Brick, and it sounds pretty darn good. Not EXACTLY like a spring, but close enough.

Typically, only the medium decay model is available (which is perfectly fine), but it comes in long and short decay as well. It's not hard to implement. The most crucial thing is that you need a regulated +5V 1A power supply, using something like an LM7805 IC (the Brick is a current hog!). If your amp has an unused heater or bias winding, no problem doing this. Otherwise, you'll need a separate, small internal power transformer.

[Krwkka]

The supply shouldn't be a problem. What I'm a bit concerned about is how to simply balance out the existing spring reverb circuit to accommodate the Digi-Log. The Belton can only handle 1.5 volts at the input, and is unity gain. I don't know what the signal voltage output is from the existing 4558 circuit, but even if it's <1.5 volts, I think I'm going to have to attenuate the output before it goes into the return side of the 4558 or I'll fry something!

[jrfrond]

I'm thinking voltage divider and diode clamping at the input.

[NorCalTuna]

Well, my build is ringing ears, literally! It's very tight and loud, and the belton box works. ish. I powered it with an external wall wart kind of thing for the time being, and added a 22 and .1 uf cap at the power terminal to smooth things out a bit more. I used fender's scheme mainly for genealogical reasons: ie, to make my work graft easily for others. Also, I used a .001 cap for the input instead of the 500pf. Thats probably a mistake... I replaced the 1m grid load resistor with a 1ma pot, which is like a dwell, and works to let you get good reverb drive at low volumes while not overloading it (or clipping at the clamper- unless you like tube screamer distortion. It could be interesting, as reverbs do distort...). Also, i put a 10k pot as a varistor the input as a trim. It's mainly redundant and may go away.

I ran the output to another 10k pot- which turned out to be mislabeled, alas- as a divider and back to the fender stage- verbatim. 220k grid load, .003, 100kL etc. Here's the rub- this thing makes a lot more signal than a tank. I'm not exactally sure what a gain of 0 is, which is what the sheet says, but I'm guessing it's unity. It's pretty hot, but it may be as simple as getting the rest of the circuit right so that it's 10k pot, which is it's output load, works correctly. At low volumes you can get more reverb than dry signal. It's kinda interesting, but...

What needs debugging is the hiss. Sounds like a crappy, crappy old pedal. How much hiss? BTW test speaker ev12l... According to my iphone db meter, 30 db. it's raising the noise floor of my amp- my room is 36 db, amp idling is 43db (1m)- over a thousand times. It's not input noise as when I pull the input it's just as bad. I'm going to try shunting hf to ground with a 47pf cap- right over the grid load- to see what that does.

(bench break) (tacks 47pf 500v sm cap in)

no effect. May have actually made it 1db louder but this could be measuring error. Next try- low pass filter from the output of the box before the 10k pot. 10kr, .022 cap- bench break- HUGE improvement, although hiss is still there. About 10 db drop, and the hf isn't really missed over the hiss. I'm not sure how much was there to begin with. Also, stage has way more gain than it needs, so I'm going to put a resistor prior to the 100k pot, making it the bottom of a vd, and lift the cathode bypass cap. I'm also going to try to shunt more hf off over the plate resistor.

Amp is awesome and if you fiddle, you can even bring up a good reverb tone. The dwell knob works well. I also think I should pull the .001 and go back to the 500 or even smaller so I can keep eliminating hf from the other end of the circuit and still have it available. Dinner time. Post dinner fiddle update follows!

[Steve Conner]

What needs debugging is the hiss. Sounds like a crappy, crappy old pedal. How much hiss? BTW test speaker ev12l... According to my iphone db meter, 30 db. it's raising the noise floor of my amp- my room is 36 db, amp idling is 43db (1m)- over a thousand times.

I've had a similar experience with digital reverb. The Fender circuit subjects the reverb to a very wide dynamic range, because it's after the channel volume control, and there is no master volume. A spring reverb can handle it easily, but a digital one can't. If you tune it to handle full volume without clipping the ADC, then when you play quietly, there is lots of hiss and grunge.

You also said the reverb is excessive, though. Taming that will lower the noise floor in proportion.

[Krwkka]

Tuna: These are the exact problems I was anticipating.

I am assuming you have a Fender Something Reverb type clone? The Digi-Log is unity gain: a spring reverb is not. The spring unit produces a very small voltage at its transducer. Consequently, you either need to pad down the output of the Digi-Log more or, alternatively, lower the gain of the recovery stage. A simple thing to try is to remove the cathode cap from the reverb return.

I'm guessing that a large part of the hiss (assuming again that you are inserting the Digi-Log into a Fender reverb circuit) would be from the impedance mismatch between the output of the reverb transformer and the input to the Digi-Log. If so, I believe you need to completely bypass the transformer. It might also help to replace the 12at7 driver with a 12ax7 and rewire the socket so that you're only using one side of the 12ax7.

Does this make sense?

[Steve Conner]

I believe you need to completely bypass the transformer.
Does this make sense?

I don't think it's going to like 150V worth of signal from a tube's plate. Could be wrong, though. There should be a datasheet that explains how to hook it up and what signal levels are required.

[Krwkka]

?? 150 Volts ?? Are you talking about signal or DC?

[Steve Conner]

Signal. That's what a 12AX7 stage puts out when goosed, and I seem to remember the Fender designs drive it quite hard, to the point of clipping when the amp is cranked.

[Krwkka]

Hmmm.

So, Tuna, where are you getting your input to the Digi-Verb?

[NorCalTuna]

There's no 150v as there's no tube or transformer. Steve's right, the box just can't deal with the wide range in input. Lifting the bypass cap on the return stage was the first move of a long night. The amp is about to sprout a master volume.

[Krwkka]

That's sad. I had high hopes for this unit, but if it can't handle the input range of normal playing ---