I got this unidentified spring reverb tank with what looks like a driver. It’s not an Accutronics tank and there is no indication where it comes from, but a little detective work showed that it might have been pulled out of an old Ace Tone (the precursor of Roland) powered mixer (like this one).
The circuit, which has seen better days, is transistor-based with three 2N5089 (I think it’s the low-noise version of the classic 2N5088) and two 2N5855. Rather than trying to build a driver for a spring tank with unknown impedances (probably futile), I’d like to see whether the driver at hand works. I therefore need to:
Sort out a way to power it
Figure out the location of the input and output
There is a reading on the tank itself that says “6V”. Do you think that I might be okay with switching supply at 6V or should I have a regulator for the job (like
L7806 or LM317)? How much current do these spring reverb tanks typically draw?
I also tried to line up the back and front images of the circuit board to trace the circuit to locate the input and output. I have uploaded a file here where you can play with the transparency between the images (the images do not line up perfectly, but you get the idea…) The black and red wires of the 4-pin Molex connector are power and ground, just like the wires soldered on the back of the board (no idea why). So I guess the other two are the input and output of the circuit. One of them connects to a 100k trimmer and then to a 100nF capacitor, while the other to a 56nF capacitor. I obviously need to dig further, but on the first look, the former is probably the input?
Nice find. Elliott Sound has some really useful information on spring reverbs. Here’s a link: Care and Feeding of Spring Reverbs
In order to know the current that your tank draws, you’ll need to work out its impedance. Is there any code at all printed on the tank?
It looks like it should be pretty simple to draw a schematic from your PCB – there are only a handful of parts, and the traces are all very clear – which would make it much easier to say with some degree of certainty what’s going on regarding which leads are input and which output.
For testing purposes, a switching supply should be alright, but in the long run, you’ll need a regulator. Spring reverbs are prone to picking up noise. Good supply and in particular, good grounding are critical in keeping that to a minimum.
I’ve read the Rod Elliott article, but without any markings on the tank whatsoever, I have no idea of its input impedance. Since it’s not an Accutronics tank, I cannot even guess the model by measuring the input resistance. This leaves me to try the actual circuit, which is probably easier than building one of my own and hoping it will work with this particular tank. All the parts are through-hole, and I don’t see anything exotic in terms of components, so I am confident that I can repair it if needed and make it work. I guess I will have to bite the bullet and do all the tracing before I try anything though!
Once you’ve done that, it should be obvious which side is the input and which the output (I think that your hunch is probably correct – it makes sense that the trim pot is there to adjust the input level, but better to know for sure), and we’ll also be able to make a more educated guess as to what level and impedance the circuit is expecting.
I ended up storing the tank and driver away. A few weeks ago it felt like it was time to revisit this mystery.
For some reason, I had ignored that the Rod Elliott article mentioned a push-pull driver when I first looked around this circuit, and when I looked again this time around, it hit me. The two TO-106 package transistors are a complementary NPN/PNP pair in a push-pull driver, so the right part of the circuit board is the driver and the left part is the return. From that, it was easy to figure out which wire was the input and which was the output.
The next step was to determine the supply. The tank had a “6V” stamped on it, but I thought that sounded quite low given the supply of spring driver push-pull circuits. I remember trying image search to identify the source of the tank and its driver a year ago, to no avail, but this time around when I put the top of the driver in Google Lens and indicated that it is a spring reverb, I got a breakthrough. The AI model training must have gotten much better because I got pointed to this blog post where someone had pulled an identical driver on a larger tank from a 1970s Viscount organ. The post had the relevant schematic from the service manual which specified a 23V supply for the driver. With a little bit of searching I found that the same person had posted a question at the SOS forum back in 2015. To cut the long story short, those who know more about transistor drivers concluded that the suboptimal design of the driver meant that the supply is not particularly critical, and that something above 18V should work fine.
Armed with this knowledge and with a 19V laptop “brick” supply, I gave it a quick test, and it worked! Not as impressive sounding as the larger tank in the video at the end of the forum discussion, but still very workable. Another quick test indicated that the driver draws something like 30mA (maybe 40 when pushed harder). Not too bad!
Now I will need some mix/feedback circuit around it to make it into a module and I think that the Doepfer A-199 traced by @Sandelinos might be a good start.
Sometimes it’s better when you leave something aside, even if you revisit it some 21 months later!
Thanks for writing this up, I’ll come back to this when I build mine.
I found a couple of rusty reverb tanks at a radio swapmeet in April 2025, but have so far only fiddled briefly: I connected the outputs to my TAPE-HEAD AMPLIFIER module and got satisfying explosion noises when I abused the springs.
One of them has fairly high input impediance; I tried driving that with a Eurorack +/-5V signal and it surprised me by reverbing quietly, but I expect there’ll be a lot more to doing it properly.
A point to watch is that my tanks have input grounded to the metalwork rather than floating, so that will be relevant to driving them.
