Tube Amps / Music Electronics
For current discussions, please visit Music Electronics Forum. New: view Recent Searches.
New: visit Schematic Hell!
The sunn still shines online!

Listen to great tunes streaming live right now!

ampage archive

Vintage threads from the first ten years

Search for:  Mode:  

Increase the decay time on spring reverb?

10/10/2000 1:16 AM
Increase the decay time on spring reverb?
I've got an old Hammond reverb tank pulled from a tube GE stereo console. Was able to modify an ampeg style reverb driver/recovery circuit to work with the impedances, but the thing's got a short decay. any way to mod the innards to increase the decay?  
10/10/2000 1:46 AM
Joe Gagan

I'm guessing, try to put more springs in there.[?] Also, a friend of mine ran several tanks in series or parallel, he said it sounded great, but I didn't hear it.  
10/10/2000 3:24 AM

really can't put more springs in there, mine's a 2 spring unit. tried messing with the spring tension, either lessening it or tightening it, just made the decay shorter. i'm thinking probably less stiff springs, also don't know if the stationary magnet pieces can be modified.....I've read that longer springs mean longer decay, wondering if there's some hotrodding to be done here.  
10/10/2000 1:45 PM
paul perry

This may be blasphemy, but 2 things to try are:  
1. a compressor  
2. running a digital delay in series with it.  
I'd try the digital first, but try both ways.
10/10/2000 1:58 PM
Mark Hammer

Technically, there are two ways to increase decay time: the springs themselves, and the circuitry.  
Much like guitar strings themselves, spring reverb characteristics depend on both length and tension. Shorter, tighter springs transmit physical energy more efficiently, and accomplish it in less time because it has farther to go.  
Spring reverb pans, such as those available from Accutronics, will vary in a number of characteristics, often available in the same physical package:  
- the input and output impedance of the transducers  
- the number of springs (more springs generally equals a more realistic reverb "wash", with less focus at any given delay)  
- the length of springs (longer = more delay)  
As far as I know, you may be limited in terms of the length of the springs in whatever reverb pan might conceivably match the physical and impedance requirements of your amp; not tightly so, but somewhat. Certainly the simplest solution might be to buy a replacement pan with more, longer springs, screw it in place and plug in the connecting cables (they often connect with simple RCA plugs/jacks); assuming the in/out impedances are appropriate.  
The second route is via circuitry. Although there are numerous differences between spring and bucket brigade devices, one of the things they have in common is the capacity to recirculate some of the output back to the input, so that it can be "re-delayed". On 3-knob reverbs, this is often accomplished with a control labelled "dwell". Now just *where* it gets tapped from and fed back to will depend on the specific amp, but it IS feasible to increase the *apparent* decay of a spring reverb by recirculating the reverb signal back through the springs. Common sense suggests that this requires careful tone shaping to do right, but you can do that by ear and the use of prudent choices in capacitor values.  
Bear in mind that just as cranking up the recirculation in a bucket brigade or even digital device set for a delay of 40msec will NOT sound the same as more recirculation with a delay of 300msec, the sonic pleasingness of increased "dwell" with a spring reverb will probably increase with longer delay and more springs. Two springs with a short delay will probably sound box-ey if one increases the recirculation. More springs will defocus it.  
You would probably run into more expertise on the matter in the amp forum. Best of luck.
10/12/2000 2:05 AM

Hi Mark,  
How you been? Thanx for the thoughtful in-depth reply. Coupla observations, and a few resultant questions....I've messed with several tube reverb circuits, found I really prefer the Ampeg capacitor-coupled circuit tone wise to the more common SE tranny coupled.  
Ran across this EL-84 cap coupled reverb circuit in an old Magnavox stereo console, was intrigued by the design. Unit used a Hammond tank (2 springs, long tank)with high impedance and short decay.Couldn't adapt it to work well with guitar, way too much in the way of microphonics. Kevin O' Connor has a nice section in one of his books describing that this is a real inefficient approach, so I went back to my old habit of breaking things to fix them to learn how they work....learned a bit about impedance matching of the tanks, etc. I've got several tanks that were pulled from SS reverb circuits (high impedance also, one has medium decay/2 spring and the other has medium decay/3 spring/short tank.) Both of these sound great with the modified Ampeg circuit.  
I was hoping to modify the Hammond from short to long decay to use in another project. I tried lengthening the springs, (tacked on additional springs of similiar characteristics) and shortened the springs by changing the hook-up point. In both cases, didn't really change the decay time much. that's what got me to thinking about the stationary magnets.....  
As far as changing circuit parameters, I was under the impression that "dwell" is really a drive control, determining how much signal voltage is presented to drive the pan. Up the signal voltage too high, and you go from increased decay to mud which happened when I messed with that circuit parameter, even with significant tone shaping).  
I'd like to learn more about the BB type approach, still waiting on "Guitar Gadgets" to arrive. Is there info there, or can you suggest other sources?  
10/12/2000 3:39 PM
Mark Hammer

I'm okay and back at work. Thanks for asking.  
Just looked in my Pittman book at the Fender reverb unit schem, and you are right. The "dwell" control IS a drive or sensitivity type of control, with no additional controls providing any recirculation. On the other hand, there is little reason why you *couldn't* recirculate some of the post-spring signal, as long as you tap it from the right place, and bring it back to the right place. I would recommend lopping off some of the low end, though, just to take the slap out of the transients. I honestly can't see springs taking kindly to too much bass.  
Don't know that I could suggest other *technical* resources, but I do have an interesting article from Polyphony, back in 1981 or so, on why spring reverb will never die. (The name Craig O'Donnell comes to mind as author, but I can't be certain of that from the office.) The author enumerates the interesting characteristics of springs that result from it being a mechanical device whose physics change depending on signal characteristics (which likely won't be duplicated by DSP units for at least a few years, although I say this without knowing what Lexicon has up their sleeve). What becomes clear from the article is that desirable reverb tone probably has as much to do with tone and gain shaping of the pre-spring and post-spring signal as it does with the actual springs themselves. This also jives with the degree of desirability that 3-knob reverbs have, since they accomplish some of that shaping, albeit crudely.  
From another angle, although you can't string springs end to end (since they will sag from the weight), there is no reason why spring units couldn't be *electronically* cascaded to extend delay time, although one should bear in mind that this will likely NOT sound the same as one single long spring unit, since the springs have to start vibrating to transmit the signal and two cascaded units provide two startup points. You also have the frequency-limiting of two sets of transducers. Worth thinking about, though.
   Page 1 of 2 Next> Last Page>>