Tombola’s DIY spring reverb driver circuit

I’ve been fiddling with spring reverb for a while, probably because of reading Gabriel Roth’s famous “shitty is pretty” article in the late lamented Big Daddy magazine, where he talked about his way of using spring reverb on his throwback funk.

Ages ago I built a little box for this +12v Quasar spring kit, and it sounded ok, if rather hissy. I’ve been using it as my (clanky) main reverb for the last five years or so.

Since then the modular synth thing has gone generally batshit, and Tom of Music Thing recently came up with a spring reverb driver module in 4HP as a smaller alternative to the 8HP Doepfer A-199.

Here’s Muffwiggler user Dego’s stripboard TL074 version, with the additional link to ground on one of the +opamp inputs, which it needs to work. Took me a few months to work that out, aaaaargh:


I had some panels for this project in the batch I ordered from Razorlab, this time infilled with emerald green acrylic paint. Makes me want to drink Pimms and eat strawberries and cream. Maximum recommend for Razorlab by the way, although you might have to go via Ponoko to get it done.


So here’s my version of the module in, erm, 8HP. So I could’ve just bought the Doepfer version then. Anyway. Ta-daaaa.


I added a feedback circuit, with a CGS DC mixer and a switched ext feedback socket – it’s normalled to feed the reverb only signal back into the input, level controlled by the feedback knob. If something is plugged into the ext feedback socket, it mixes that into the input instead. I also added a 1k drive pot instead of the 660R resistor. This helps a lot when switching in different spring tanks via the (handy) phono jacks on the front – most of my motley collection of tanks mostly seem to need a higher drive than the specified 8EB2C1B.

I looked at putting the spring tank connections on a single socket, having missed out on putting an separate output for reverb only, but because the sleeve outputs on the circuit can’t be tied together (they’re not grounded), we’d need a 4-pole socket. Because of the size of the holes in my panel, I tried for one of these.


It turned it out it was nearly impossible to solder wires to the terminals with my clumsily massive iron without melting the plastic dividers. So I gave up and put a stereo jack on the output; one side mix, the other side reverb only.

Round the back, oh dear.


…mmmm. Well it works. My home-made shoddy PCB bracket made out of aluminium L-channel looks a bit better with the blood cleaned off it, at least. I’ve run shielded wires between the board and the sockets for the spring reverb tray in an attempt to try and minimise any hum picked up inside the case.

Because of the additional feedback circuit, added at the eleventh hour, it’s pretty deep and just about fits inside my Doepfer LC9 case. I’d usually favour stripboard but I get the impression that pointboard would produce a more compact layout, probably handy if space is tight. Either way if the sound of springs is your thing I recommend not waiting for someone to knock up a PCB, just have a go.

Here’s my pile of springs, mostly ancient trays presumably yanked out of Hammond organs.


And finally some short demos with the specified 3 spring 8EB2C1B tank, just to give an idea of what it sounds like.

This is an early one, fiddling with an MS20, sequenced from the A-155 through the reverb with the driver circuit running at +/-15v. Murky.

Some more beats, the sound source being Marc Bareille’s Polivoks filter clone with the resonance up, being fed through a couple of CGS30 bandpass filters, including some fiddling with feedback, and pushing the drive.

Another loop from the Polivoks/bandpass filter setup.

This is an 8 step sequence from the A155 into an ASys RS95 oscillator. The distortion is the spring being pushed too hard when turning
up the drive. Again at the end I’m playing the feedback, trying to get it ring without whining.

Update – Oct 2013 : Tom Whitwell has designed a PCB for a new version of his spring reverb; all the information over at Music Thing and the inevitable thread over at Muffwiggler.

Oct 2017 : The original schematic seems to have gone walkabout from the thread at Muffwiggler – here it is: tombola's spring reverb circuit


  1. 8th September 2013Simon Smith says:

    Hi, could you tell me what values for the pots you used? I would love to build this. I only have a small practise amp but I do have a scrounged reverb tank. Thanks

  2. 8th September 2013ua726 says:

    I think I pretty much stuck to the values that Tom specified; 100k for blend/mix, 50k for the drive/recovery, 1k for the extra drive pot I added, and 20k instead of 25k for the EQ/tone pot just because I didn’t have a 25k pot to hand. All linear as far as I can remember.

    I tapped pin 7 of the TL074 for the 100% wet output for the feedback.

  3. 16th June 2014snaper says:


    Could you confirm that the above poste layout works? I mean without any change?

    Thank you in advance!


  4. 16th June 2014ua726 says:

    Hey Gábor

    I built that layout ages ago, and it worked. If you want a PCB then Thonk sell Tom’s latest version of his spring reverb.


  5. 29th July 2014trash23 says:

    yes it works! thanks!


  6. 29th July 2014ua726 says:

    glad to hear it!

  7. 19th November 2014dr knob says:

    hi, i do not understand what it’s wiper. Maybe connect pins 2 and 3 for poti?

  8. 21st November 2014ua726 says:

    The middle pin of the three.

  9. 23rd November 2014David says:

    Hey, thanks for publishing this. As someone with almost no electronics experience, I managed to get the main circuit together pretty easily.

    I’d like to add the feedback circuit as well, but, like I said, with no electronics experience I have no idea how to do that. Do I need a second TL074? What would the circuit itself look like?

  10. 25th November 2014ua726 says:

    @david Glad you got it working!

    Cleverly I didn’t make a schematic for the additional feedback stuff, but it’s mostly just taking the reverb only out from pin 7 of the TL074, and mixing with the audio input through a CGS DC mixer (, just a dual op amp – like a TL072 – and a couple of resistors), The output of the TL072 would then be fed into the track marked audio input above.

    The feedback input to the TL072 mixer should go through a pot to attenuate it, I think I used one with an audio taper – it took a bit of experimenting to get something that gave control over the right level of feedback, it can be a bit touchy.

    Dunno if any of that makes any sense, given a bit of time I could work out exactly what I did and do a revised layout if it’ll help, but I’d recommend you have a go on a breadboard in the meantime.

  11. 25th November 2014David says:

    Thanks so much for the reply. I think I mostly understand it; I get how frustrating it can be for newbies to post simple questions, so I’ve got a textbook here that I’m reading to help get me up to speed. Right now I’m waiting on some parts to arrive (I decided to also attempt that Polivoks filter clone you linked to) and once they do I’ll give it a shot. So thanks again!

  12. 26th November 2014ua726 says:

    It’s nice to get comments tbh, makes a change! I’m very bad at asking for help, I tend to stare at something for months, like the System 100 VCO I eventually got sorted.

    The Polivoks is great, but I was just starting off with building things at the time and never quite got the resistor changes perfect for 12v, so a couple of the pots have a zero point at about 7 on the dial, which is mildly annoying but it works ok.

  13. 27th November 2014dr knob says:

    Thanks ua726 it works perfectly!
    There are slight shrill whistle, i hope to disappear by shielding the cables. Excuse me if i make gramatical mistakes, i’m french.

  14. 27th November 2014ua726 says:

    excellent! Moving the spring reverb tray around might help a bit too, I have mine balanced on top of the case.

  15. 7th December 2014David says:

    Hmm okay, so it didn’t actually work. I had it all wired up and assumed everything was okay, but when I plugged it in, SNAP! and smoke. I’m not sure what happened, since I can’t see any shorts and none of the components look damaged. I think I might’ve blown the opamp. I’m going to try again, but I wanted to ask: what do those white circles on your diagram represent? I ignored them the first time since they were unlabelled, but now I’m wondering if I missed something.

  16. 7th December 2014David says:

    Ah, I know what the problem is. In my ignorance I didn’t realize this circuit needed a bipolar power supply (I didn’t know such a thing existed until now). I’m not sure how to adapt my regular 12v power supply to make this circuit work.

  17. 7th December 2014ua726 says:

    Yep, you need a +/-12v power supply. +/-15v should work for this as well, although 12v is the most popular with Eurorack DIYers.

    The cheapest/quickest way of just being able to hear something might be to wire up a pair of 9v batteries in a bipolar fashion (explained It might not sound as good though, and that’s not really a permanent solution.

    I made a simple stripboard +/-12v power supply a while back, it’s OK – it was based on the MFOS power supply.

    Since most of the things I’ve been building recently have been designed for 14v or 15v, I’ve moved to using an Oakley PSU which isn’t the cheapest option, but it’s great for testing. Generally if something is shorted, a pair of lights on the board go out and I jump for the off switch. The output voltage is adjustable within a certain range so it may be possible to run it at 12v but I’m not sure – it might be worth asking in the Oakley Muffwiggler forum if it’s possible to modify the design for the lower voltage.

    If you’re in the US then there are generally piles of PowerOne power supplies going cheap on eBay (like this), but being partly open, it seems too easy to electrocute yourself. The Oakley/MFOS/Frequency Central PSUs all use enclosed wall-wart type transformers, which seems a lot safer.

    The white circles on Dego’s diagram must be for mounting holes, but I wouldn’t put them there because you’ll cut the tracks to the EQ pot. I should just check you realise that the red squares down the middle of the TL074 are cuts in the track.

    Before you plug anything else in, I’d check that you aren’t getting any shorts with a continuity tester in a digital voltmeter. I do this now before I plug a newly built module in, checking that the power supply lines aren’t shorted, that the power pins on all ICs are connected to the correct lines, and that adjacent tracks aren’t shorted (unless they should be). In the past I’ve missed shorts with a visual check, only picking them up when checking with the continuity tester.

    phew hth

  18. 8th December 2014David says:

    Thank you so much for the reply. It’s extremely helpful. I guess now I just need to figure out how far I want to take this and in what direction. It’s funny that the two initial projects I’ve chosen require different voltages, so I’m not quite sure where to start. Part of me is thinking that the best idea now would be to just order two PCBs from Frequency Central and build both 12 and 15 volt versions (or maybe an Oakley 15v and an FC 12v). I suppose I should also do some research into modulars to find out where I’ll most likely end up. In the meantime I really do appreciate your help. It’s hard to break into a scene like this, where everyone seems to know so much, but posts like yours certainly help 🙂

  19. 10th December 2014ua726 says:

    If I was just starting out, I’d just probably get a kit 12v Eurorack style power supply. There’s definitely a wider selection of PCBs and projects available at the lower voltage. FWIW, my Polivoks filter is still running happily in my (12v) Doepfer case despite the aforementioned wonkiness.

    Also check out the Muffwiggler DIY and Eurorack forums, the electro-music DIY forum, and the Analogue Heaven and Synth-DIY mailing list archives, it’s all in there, it’s an entire world. There’s a massive list of links here –

  20. 28th January 2015David says:

    So it’s been a few months, but I ended up getting a used Power One supply and rigging it up for +/-12v, and it seems to work really well. That let me get back to the reverb circuit, which I rebuilt from scratch, and — holy mother of God — it actually sorta worked! Now that it’s at least partially working, I have some more specific questions.

    The overall level seems to be pretty low. I’m using an Acutronics 4ab3c1b (from a Twin Reverb), so it’s got 8ohm input impedance and 2250 ohm output. Anything I can do to bring the signal up? I’ve been testing it with an MS20 mini, and I can only get usable levels with the headphone out at over half volume with my audio interface preamp at like 3/4.

    Secondly, I haven’t been able to get the feedback circuit working. The feedback pot doesn’t seem to do anything at all. I took pin 7 of the TL074 (after the 1n cap) and sent that to terminal 2 on the feedback pot. Then I took the wiper terminal and sent that to the input of the DC mixer, and I grounded terminal 1 of the feedback pot as per Ken Stone’s schematic. Finally I ran the output of the TL072 (pin 7) into the audio input line on the main circuit, and I put it just after the input jack. If that all sounds good, then I’ll go in with the multimeter and make sure everything’s fine, but I wanted to make sure I had the basic signal path right first.

  21. 28th January 2015ua726 says:

    @david, glad it’s kinda-sorta working at least 🙂

    When you say overall level, do you mean the mix of the original + reverb, or just the reverb?

    The reverb level could probably be tweaked by varying the 660R resistor, try a 1K pot instead. Most of my collection of dodgy old springs can be made to work this way, although watch out for distortion.

    If the overall mix level is low, I imagine you could vary the 100K between pin 8 and 9 of the TL074, maybe make a bit smaller (68k? I haven’t tried it) Have to say that the level on mine is fine running from the send on my Yamaha mixer, with the audio straight back into another channel, gain fairly low.

    Here’s the schematic if that’ll help:

    Your feedback plan almost sounds right, here’s a dodgy diagram I just knocked up, you might need to open the image up in a separate window if it’s too small.

    spring reverb feedback

  22. 28th January 2015David says:

    Oh, I see, I need to put the main input through the mixer too. That’s helpful, thanks! I know that i take the feedback from pin 7 of the TL074, but should I take the sound input right from the input jack itself?

    I should’ve been more clear in my original post — it’s the reverb signal that’s weak. The original signal is quite loud and clear. I have the 1k drive pot in place. I’ve been toying with the recovery resistor, and that seems to help. I’ve even tried removing it altogether and just using the drive pot, and have actually gotten some good results that way. You kinda have to work the blend and drive knobs in tandem, but I like that the drive knob can add a touch of reverb even if the mix is 100% dry. It can distort a little more easily, especially when the tone knob is down, which seems to boost the lows a lot, but it’s workable, although not a setup you would ever find on a retail reverb 😉 I’m not running it through any sends, though. Just the MS20 into the verb into one of the line-in channels on my interface.

    After playing around with it all day, it really does sound quite good, just needs a little more adjustment. Thanks again!

  23. 28th January 2015ua726 says:

    Yeah, the thing I’ve labelled “sound input” on the DC mixer just becomes the input for the spring reverb as a whole, the DC mixer becomes the front end for the spring rev driver. It’s a bit cranky, I’ve found that the blend doesn’t always take the reverb completely out like you say, possibly the resistors need to go a bit lower there.

    This article here covers making a spring reverb driver work for different spring impedances

    …although inevitably the circuit is a bit different, I think that R7 on that circuit roughly corresponds with the R2 in Tombola’s circuit, so it might be worth varying that 2K a bit. Enjoy fiddling, it’s addictive though…

  24. 29th January 2015David says:

    I didn’t quite get the feedback circuit working, but replacing the recovery pot with a 150k fixed resistor makes it work pretty well with my MS20. It actually sounds fantastic now, kinda reminds me of old BBC Radiophonic Workshop stuff, although I’m not sure they had springs back then.

    Anyway, I really appreciate the effort you’ve put into helping me . I definitely wouldn’t have got it together on my own 🙂

  25. 27th April 2015Ed says:

    I’ve just built this and worked first time. I replaced the 660r resistor with a 1K pot and added a feedback circuit. I didn’t think the CGS style mixer stage is really necessary and I just took a line from pin 7 and looped it back to the audio input without any additional opamp buffering via a 47k pot and a 100r in series. That makes it similar to a few guitar pedal feedback mods I have seen. Everything sounds excellent-thanks for a great circuit!

  26. 27th April 2015ua726 says:

    Fantastic, glad to hear it works. I didn’t do nothin’ really, it’s all Tom Whitwell’s circuit and Dego’s layout.

    Agree that the CGS mixer is probably overkill but that’s what I do really 🙂

  27. 8th September 2015Fredrik says:


    I have the driver doepfer a199. I have changed the tank to a long tank with two springs.
    But i feel that i has lost a bit of its power even thug i checked the ohms before i ordered it from tube doctor.

    Now i starting to wonder if i have to build a own driver. Are you driver suitable for my tank do you think?

    My goal is to have the same sound as you can have from the classic grampian 636.
    With that said i mean the snatchy and clatchy sound that you want in dub music.

    Thanks / Fredrik

  28. 8th September 2015Fredrik says:

    is it possible to add a LED for input signal in this schedule?

  29. 9th September 2015ua726 says:

    hello Fredrik

    Heh, I heard someone else’s A199 and thought it sounded great with the small tank.

    I can’t guarantee that your tank will sound any good with Tombola’s driver circuit as standard, but you should be able to vary the drive to get a reverb sound out of it. Like I said, I bought a job lot of old Hammond organ springs, and they all sound a bit different, some better than others.

    As to whether it sounds like the Grampian 636, also no idea! First I’d heard of it, to be honest. I know Tubby used a Fisher Spacexpander, and I always thought Scratch used a Space Echo, but that was just going from pictures. I haven’t tried to play the kind of scratchy instrumentation they would have used through it, all my fiddling has been with farty-sounding analogue synths.

    I would think that either the A199 or this would be good enough as a starting point, along with a tape echo and plenty of eq fiddling.

    As for the input LED indicator, you might be able to adapt Ken Stone’s LED driver.

  30. 27th December 2016Ted K. says:

    Hey thanks so much for this! I’m like beginner-intermediate to electronics but I just found this bipolar supply that seems almost too good to be true in my search for bipolar supplies:
    It looks to me like it’d totally work and is only 10 bucks, what do you think?

  31. 27th December 2016ua726 says:

    I reckon it’d power the spring reverb circuit mentioned in this post, plus maybe one or two other small-ish circuits – but I don’t really know for sure, ‘cos I’ve never measured how much power the spring reverb circuit draws, and I can’t see any info about how much power the Futurlec PSU could provide.

    The “output power of up to 1A” bit I imagine is about the DC transformer you’d plug into this to power it – I’d imagine the maximum reliable output current from this board would be a lot less.

    The datasheet for the regulators might help give you some idea of how much current it could provide. Doesn’t look like there’s any heatsinking (which would help), and there’s not much room for retrofitting any.

    Futurlec tend to be slow and rubbish at telling you when they’re waiting on stock – I had particular trouble with ordering slide potentiometers from them, despite appearing as “in stock” on their site. You might find similar to this on eBay from other vendors.

    I’ll stick to my big-arse heavyweight linear supplies (Oakley PSU board + Yamaha PA-20) for most stuff – reliable power means one less thing to troubleshoot – but for ten bucks I’d be tempted to give this (or similar) a go for a couple of modules.

    I’ve got a Meanwell switcher around somewhere as sold by Curetronic that I’ve never got round to testing out, they might be worth looking at too – although they’re rather more expensive.

  32. 1st February 2017Ezta says:

    Hi I am about to attempt to build the circuit in the diagram. Have just read the comments and am wondering about power supplies. Will a standard 12v dc work for this? Cheers!

  33. 1st February 2017ua726 says:

    Soz, you need a power supply with a dual output, +/-12V would work.

    For a quick check it might also work with two 9V batteries like this but I haven’t tested it. My DIY 808 bass drum boomed happily when powered from such a set-up.

    If you fancy doing more of this, then buying/building a +/-12V power supply would be a good idea – something like might do the trick.

  34. 2nd February 2017Ezta says:

    Cheers for the response! I will definitely try the battery trick.

    Would any of these do the trick for a more permanent solution:

    Sorry I don’t know much at all about power supplies!

    Cheers 🙂

  35. 2nd February 2017ua726 says:

    This one would do the trick

    and assuming you don’t fancy doing some mains wiring (I wouldn’t recommend it) – couple it with a Yamaha PA-20:

    which is centre tapped 17.5V-0-17.5V.

    You might be able to get away with just using a single phase wall-wart type AC adaptor, have a look at the wiring suggested by Tony Allgood for his PSU, you might be able to adapt it.

  36. 2nd February 2017Ezta says:

    Would this do the trick:

  37. 2nd February 2017ua726 says:

    I’d say so, but mains wiring.

    (Years ago I shocked myself trying to fix my CSQ-100, decided it was a lesson learned)

  38. 2nd February 2017Ezta says:

    Thankyou very much for the info. Really appreciate. Will try and build and let you know the outcome


  39. 2nd September 2017Lukie says:

    Need to say the TLE2074 is a huge improvement on this….loads of noise is gone now 🙂

  40. 22nd October 2017Tombola’s DIY spring reverb driver circuit | The Paupers Electronic Music Studio says:

    […] […]

Write a comment: