Available soon: Bodhizafa analog reverb

OK, here's the latest. I used my function generator and oscilloscope to fine tune the auto-bias and reworked the input booster to squeeze every little bit of headroom out of that power mosfet.

Running it at 9V sounds absolutely awesome, but you're gonna need a good (ahem - OneSpot) 1A power supply.

The two power resistors should be rated for at least 2 watts and soldered elevated off the PCB so heat can dissipate under them. I'm thinking about just including them with the PCB.

Screenshot 2026-05-11 at 5.07.53 PM.png
 
I will put my name down for one of these once you get it rolling. Reverb is one those effects that be finicky to dial in. I have high hopes for this. Will it play nicely with bass as is?
 
anevilspar7an said:
I will put my name down for one of these once you get it rolling. Reverb is one those effects that be finicky to dial in. I have high hopes for this. Will it play nicely with bass as is?
Click to expand...
Bass? Really hard to say. The effect needs a good bit of low cut up front, which could be good or bad for bass, depending on exactly how you want to use it.
 
I'm interested in this. What impedance tanks can you use with it? It would be a bonus if I could use my existing tank (600 ohm input, 2250 ohm output impedance) with it, but it's probably not a dealbreaker. I am a little horrified about what shipping it to Europe might cost these days :F
 
Fama said:
I'm interested in this. What impedance tanks can you use with it? It would be a bonus if I could use my existing tank (600 ohm input, 2250 ohm output impedance) with it, but it's probably not a dealbreaker. I am a little horrified about what shipping it to Europe might cost these days :F
Click to expand...

The tank must be 8 or 10 ohms.
 
With it working pretty well at 9V, I want to try a different type of mosfet to see if we can get another volt or two of headroom out of it - and maybe make it a bit more efficient.

The VG(th) of the IRF510 is 2 to 4 volts and the RDS(on) is 0.54 ohms.

I found a "logic level" Hexfet, the IRLZ44NPBF, with a VG(th) of 1 to 2 volts and RDS(on) of 0.022 ohms.

Another volt or two of headroom would make a big difference on a 9V supply.

Also, the Hexfet has a much larger transconductance factor, which should help it drive the tank more efficiently.

The downside to the Hexfet is a higher input capacitance. It's listed as 1.7n on the data sheet, but that should be no big deal since I'm driving the mosfet with a low-impedance output from an opamp. I'd already added a 470pF from drain to gate to tame the high end a bit going into the tank.
 
When you write up the BOM would it be easy to make this run off of 12V or 18V with some part swaps? I use Cioks for my pedalboards so I would have no issue with clean isolated power.
 
anevilspar7an said:
When you write up the BOM would it be easy to make this run off of 12V or 18V with some part swaps? I use Cioks for my pedalboards so I would have no issue with clean isolated power.
Click to expand...

You won't need to build it any differently to run it at higher voltages. Just make sure you're using 25V capacitors and don't plug in anything over 18V.
 
While I'm waiting for parts, I've got a couple BD139 power audio transistors to play with. I'm going to try making a power buffer with that and the LM317 constant current source in the feedback loop of an opamp.

The power mosfet amplifier setup sounds awesome as-is. It genuinely beats the reverb in my hand-built Deluxe Reverb. Since I've got that in pocket, I'm going to see if I can get the same results with something even a little more efficient.
 
Reverb tanks are just voodoo.

The low-on mosfet showed much more bad distortion and didn't have as much headroom as the irf510. It's output was lower impedance, but hitting a tank "hard" is not the same as it sounding good.

I've tried:
The Elliot Sound circuits with the tank in a feedback loop.
Several chip amps in single-ended and bridge-tied load.
The class A buffer configuration with the IRF510 and BD139.

None of those sounded as good as the Class A mosfet boost.

I've rebuilt it with the IRF510 and fine tuned the bias. With the opamp servo "autobias", there will be no adjustments necessary when you build it. The plan now is to take a day or two to check the layout for defects and send it off to JLCPCB for them to do their magic. I hope the faceplates turn out. I had a lot of fun designing those.

Here it is in its tack soldered glory with a new breadboard sidecar:

IMG_2381.JPG
 
If you happen to know much about reverb tanks I could use some help actually. I have an ampeg vh140c with what seems to be the appropriate spec reverb tank. I've tried it both ways and smacked the tank and both sides sent audio which should mean both half of the tanks are working correctly I think. I've tried both cables going straight from the in and out to check the cables and turning the reverb knob and they both work. By testing it that way it does some phasing stuff to the sound which means the amp is definitely sending and receiving signal.
Yet somehow it doesn't work when I actually play through the tank. :(
It just makes no sense to me. Sorry I don't mean to derail the thread btw I just haven't been able to find any help in other places.

added https://forum.pedalpcb.com/threads/reverb-tank-issue.29345/
 
Last edited:
Vild said:
If you happen to know much about reverb tanks I could use some help actually. I have an ampeg vh140c with what seems to be the appropriate spec reverb tank. I've tried it both ways and smacked the tank and both sides sent audio which should mean both half of the tanks are working correctly I think. I've tried both cables going straight from the in and out to check the cables and turning the reverb knob and they both work. By testing it that way it does some phasing stuff to the sound which means the amp is definitely sending and receiving signal.
Yet somehow it doesn't work when I actually play through the tank. :(
It just makes no sense to me. Sorry I don't mean to derail the thread btw I just haven't been able to find any help in other places.
Click to expand...
Start a thread about it and link it in your post above so anyone who wants to help can reply on your thread and not derail this one even more.
 
jubal81 said:
I've tried:
The Elliot Sound circuits with the tank in a feedback loop.
Several chip amps in single-ended and bridge-tied load.
The class A buffer configuration with the IRF510 and BD139.
Click to expand...
In my personal testing/opinion, the ESP current-drive style (tank in the loop) really should have at least 10x gain before it. Otherwise it won't get driven fully. When driven hard it can sound more "hi-fi" compared to a surf-y style drive using IRF510/9510. Which isn't better or worse, just different.

Anyway, your Bodhizafa design looks awesome. Best of both worlds. Very relevant to my interests. I'll stay tuned.
 
SpringReverbFan1234 said:
In my personal testing/opinion, the ESP current-drive style (tank in the loop) really should have at least 10x gain before it. Otherwise it won't get driven fully. When driven hard it can sound more "hi-fi" compared to a surf-y style drive using IRF510/9510. Which isn't better or worse, just different.

Anyway, your Bodhizafa design looks awesome. Best of both worlds. Very relevant to my interests. I'll stay tuned.
Click to expand...

The tanks definitely react differently. When driving them in a bridge-tied-load configuration, I got a similar response as from the ESP current drive.

I also noticed a substantial difference in the attack and feel between Class A and A/B.
 
You must log in or register to reply here.
Back
Top