KVR Audio

AdmiralQuality · Post by **AdmiralQuality** » Thu Dec 22, 2005 10:21 am

duncanparsons wrote:...but don't go breaking stuff - I don't think I could handle that level of responsibilty!

Heh, don't worry, like I said I had already been intending to try it some day. I'd also really like to have the option of using the spring as a pure wet FX send, instead of always having to mix the tube-amp signal with it.

OK...need sleep... will experiment tomorrow.

deleted · Post by **deleted** » Thu Dec 22, 2005 10:49 am

DELETED

AdmiralQuality · Post by **AdmiralQuality** » Fri Dec 23, 2005 7:04 am

All done... it worked like a charm. You CAN safely drive a spring reverb tank from line level signals!

http://admiralquality.dynu.com/vstdev/S ... .mp3?dpfxb
Original dry test signal. 1. Test tones with soft squarish waveform. 2. Same with Sawtooth waveform. 3. LM4-MK II Drumming (MIDI controller is Roland SPD-20) 4: Strat.

http://admiralquality.dynu.com/vstdev/S ... .mp3?dpfxb
Using a Fender Twin Reverb through a THD Hotplate as an external FX loop. Reverb set to 10 on the amp. Notice near end how switching on the Fender's "Vibrato" (really a tremolo) modulates the wet reverb volume as well as the dry signal. (Not all amp sims can do this.) Some knocking of the amp at the end of the track so you can hear the crashing noise. Using the THD Hotplate to get a line out from the amp (AFTER the power tubes!) works great (there's a lot of hum but my amp needs a tune-up). This technique sounds almost completely convincing when run through a cabinet sim but I didn't do that for this test, you're hearing the raw output of the Hotplate here. My favorite "amp-sim" weighs 75 lbs, but at least the neighbours can't hear it... 85 Watts of total silence.

http://admiralquality.dynu.com/vstdev/S ... .mp3?dpfxb
Using the Twin's Accutronics tank as a stand alone fx loop with an A.R.T. Tube-MP as preamp/driver and another as an output trimmer (neither are really needed, it would have worked straight into my audio interface but the MPs have a hot output which helps.) The Tube-MP driving the spring input is set to it's maximum output level. The spring reverb responded fine at this volume but I felt like it could handle an even hotter input level before it would start to distort the springs or input transducer. Spring reverbs also sound good when overdriven slightly, but it seems I'd require a somewhat hotter preamp than I have to get it there. Still, it's good to know regular line level signals WILL drive this reverb! The output level should also be compatible with most line-ins. At the end of this track I poke at the strings with a screwdriver, first knocking the center connectors, then either spring end, then banging around the outside of the tank, then at the suspended inner tank.

More pictures at:
http://admiralquality.dynu.com/vstdev/S ... .htm?dpfxb

Hope this helps Duncan. I learned quite a few things tonight doing this and it was a lot of fun. I'll definitely be using the spring reverb unit stand-alone as an external effect again really soon, it worked better than I had hoped it would!

duncanparsons · Post by **duncanparsons** » Fri Dec 23, 2005 9:42 am

ta for that AQ, I've had a listen and a look. The unit I had used 3 springs, much shorter than what you have here. I seem to recall the longest spring was about 6 inches, the others we slightly shorter, but in stretched to the same length. The shortest was probably about 5.5 inches. I had two Farfisas as well, back in the day. They both had spring line units as well, so hearing your keys stuff brought back the odd memory! Again, they both had shorter springs, maybe 8" at the most.

Seeing your springs and listening, I can here why there have been calls for more tails.. The units I've had exposure to produced quite a dense sound, not too dissimilar to the plug. I do need to allow settings for a more drawn out sound...

I'm glad you enjoyed your play

, and I really appreciate you sharing the samples!

ATB
DSP

AdmiralQuality · Post by **AdmiralQuality** » Fri Dec 23, 2005 5:06 pm

Just measured and the tank in my Twin has 14" springs (well, they're stretched to 14" anyway, I can't remove them as they seem to be soldered or welded to the transducers) The tank dimensions are 16 3/4" x 3".

FYI, in the closeup picture of the springs you'll see the closer set are slightly wider and thicker guage wire than the set in the rear. And each set is made of 2 slightly different guage/thicknesses joined in the middle... this seems to be to supress oscillation of the fundamental frequency that a single spring would tend to resonate at.

The thinner set are also under noticably less tension than the thicker ones, and you actually get a higher pitched sound from the thicker ones as they're strung tighter. (This certainly isn't the only arrangement, I know Accutronics make other models with more or less springs and different guages.)

I used to have a spring verb in an old Traynor PA and it was probably smaller lke the ones you describe. But this is a guitar reverb and ya, they're designed for that long sustain for the lush sound.

I was actually quite surprised at how NON springy this sounds in a lot of cases. It seems to only do the SPROINNNNG thing when loud, low frequencies are input. Midrange stuff actually sounds almost smooth, even plate like.

The other thing that surprised me was how, even in the 100% wet version, it doesn't sound 100% wet. Some of the "dry" signal is conducting itself across those springs.. at least it doesn't have the mushy response with slow attack and buildup that I'm used to in a wet reverb signal.

Another thing that's interesting is how the springs seem to stabilize when they get a long steady tone. I was expecting them to "swell" during the longer tones... kind of have the effect build up and change during those tones... but was surprised by how static it sounds up until the point where the note was released. Only then do the springs start to sound really reverb-like. Interesting.

Anyway, I must try this on real music some time, not just test tones. Been meaning to try this since I bought the amp, so glad your project got me inspired to finally set it up!

AdmiralQuality · Post by **AdmiralQuality** » Fri Dec 23, 2005 7:15 pm

http://www.accutronicsreverb.com/rt60.htm !!!!

http://www.accutronicsreverb.com/rvbtheop.htm !!!!

Spring Reverberation units use a combination of electromagnetic and mechanical elements to simulate paths of delayed sound. an audio signal drives the COIL of the INPUT TRANSDUCER, which applies a twisting force to miniature cylindrical MAGNETS attached to a set of precision stainless steel TRANSMISSION SPRINGS. The twisting motion travels as a wave impulse down the length of the springs until it is rebounded by the OUTPUT TRANSDUCER, which also uses magnetic components to generate a delayed output signal.

The springs TWIST!!!

And it turns out I could have checked this page to find out the impedances: http://www.accutronicsreverb.com/ioic.htm

duncanparsons · Post by **duncanparsons** » Sat Dec 24, 2005 1:20 am

I was trying to get some of that twist using a gentle chorus. I've updated to Beta3, which is a mix of 1&2's settings, with a couple of new controls.

Conductance is the wrong term, and is basically a damper on a network of 6 allpasses. It works the wrong way round as well! Low should give to more length, high should be more damped..

Also Rectification is kind of the wrong word. There's a halfbridge rectifier in there, with a desaturator. I tried a saturator, but it blew everything up! If you imagine a saturator making the signal more convex, the desaturator makes it more concave, thinning it, which is a pleasing effect. It's linked with the variable rectifier for a laugh. zero is fully desaturated and no rectification, full is normal signal, full halfbridge (with the signal rescaled).

Anyway, it sound better. I need to do some presets. Some settings approach your real one. I'm not going to try to match the tail times. This also needs a notch filter to remove some unwanted resonances, but have a play any way.

Oh yes, and there's a stereo width

Beta3 [See first post and RoughDraftAudio forum]

I'm not going to have much chance to be online for about a week; but do leave any more comments/suggestions

Merry Christmas
DSP

AdmiralQuality · Post by **AdmiralQuality** » Sat Dec 24, 2005 2:29 am

I dont think the twist translates to a chorus effect. Its simply the dimension of motion they chose to propagate the sound from one end of the spring to another (makes sense, helps to eliminate interference from all the lateral and vertical bouncing the springs do. Just walking in the same room as the unit makes them bounce around a LOT.) And there should be nothing like a rectifier effect in there at all (it twists in both directions). Non-linearity sure, and probably some (soft?) clipping, particulary if you want to model how it sounds when overdriven.

The Accutronics site actually mentions the two "delay" times they use for the springs: "34 and 41 milliseconds, chosen for the fullest and most even overlapping of delayed sound." That might be some help. A spring line of course is not the same as a discreet delay line (whether it be composed of digital samples, tape, or a bucket brigade). I suspect different frequencies probably travel through the spring at different rates, but those times should be a good starting point.

Another interesting thing I learned from the Accutronics site is that the input is best driven with the signal filtered through a -6dB/oct HPF due to the increase in input impedance with frequency. This explains why my stand-alone test has a lot more low end than is typical for spring reverb. Next time I'll process the input signal before it goes to the 'verb. (You don't need to model this of course as they cancel out.) The output end is more of a flat response and, like I observed, regular line level pre-amps should handle it just fine.

Don't mean to hijack this thread with too much reality... back to the virtual!

Will try out Beta3...

deleted · Post by **deleted** » Wed Jan 04, 2006 4:45 pm

DELETED

soma · Post by **soma** » Wed Jan 04, 2006 5:32 pm

Sounds pretty good. Two together sounds better. I put one in front of dub box and it sounded very much like an space echo.
I hope you announce the final version on the KVR main page so I'll see it!

sinkmusic · Post by **sinkmusic** » Fri Jan 06, 2006 1:46 am

beta3 has a better sound but crashes Live5. (the sound is getting very very high when turning some knobs, then the cpu rises up to 256%, the sound gets so high on the vumeter than you can't even hear it, and it freezes..).
So i don't know if the reveb is dynamic (the higher the sound is, more "boingy" you have ?)

AdmiralQuality · Post by **AdmiralQuality** » Fri Jan 06, 2006 3:48 am

sinkmusic wrote:beta3 has a better sound but crashes Live5. (the sound is getting very very high when turning some knobs, then the cpu rises up to 256%, the sound gets so high on the vumeter than you can't even hear it, and it freezes..).
So i don't know if the reveb is dynamic (the higher the sound is, more "boingy" you have ?)

Generally my real-world experience is the opposite. It's the low frequency sounds that get the thing boinnnnnging.

duncanparsons · Post by **duncanparsons** » Fri Jan 06, 2006 9:36 am

I have taken on board alot of the feedback here. The project is currently on ice whilst I get some other things sorted (another rather involved delay project intended for extremely public consumption, and a novel synth idea that will have free and payware versions).

When those are done I will return to this, and it will probably be a BetaBugs release

DSP

Lunch Money · Post by **Lunch Money** » Fri Jan 06, 2006 9:43 am

Hey, I'm a choosy cheese chooser, too!

This is a really unique little plug-in. Good work so far, mate.

xoxos · Post by **xoxos** » Fri Jan 06, 2006 3:58 pm

"springline" indeed :p cue right said fred. admit it, you're a mason aren't you.

will try 3, cheers

SpringLine Reverb Model :: Now in the RoughDraftAudio labs!