Tape delay Simulators

[sjaltenb]

Yes yes Ok I thought about this last night and I finally got on track.

I think that covers the basic controls+

-A Swell mode switch would simply be a switch that bypasses all the head output SPSTs and turns them all "on"
-We can easily add LED's to each SPST switch with DPDTs (not a biggie, just sayin)
-Outputs could easily be routed from each head to a Mono/Panning circuit for semi-stereo operation
-Should we overlook the possibility of using a full or low voltage Tube preamp with a tiny bit of gain for the pre-amp section? There are plenty of designs, so I think not!
-Internal Trimmers inline with the main control pots could be used to set the range of each delay head time, right? Once setup, 12:000 on the Front-Panel pots Should be equal to the original Binson times.
-Modulation is always a possibility! But what type of circuit would be appropriate? It could easily be added to each head or individually.

I think this would all fit nicely in a 2U rack space...1U could probably work but would be crammed.

[jacobyjd]

The modulation I was suggesting wouldn't be isolated to any individual heads--it would be more to simulate actual tape wear for the whole unit. Basically an envelope-controlled vibrato--louder notes would cause more warble in the delayed signal. Probably after teh feedback section--I wouldn't want to put individual envelope setups on each feedback loop and delay output.

pushbutton switches would be cool, too

I probably wouldn't use a stereo/pan feature, but it's a great idea. I would try to put mine in a larger enclosure and keep it on the floor--but that's just my style

[slacker]

-spst switches for each delay 'head's feedback

The Echorec just took feedback from one head, that's what the block diagram is supposed to show, so I was thinking a rotary switch to choose which one. Having the possibility of feeding them all back at once is cool though and would give a lot more flexibility.

[sjaltenb]

ah ha! That makes more sense....

[jacobyjd]

-spst switches for each delay 'head's feedback

The Echorec just took feedback from one head, that's what the block diagram is supposed to show, so I was thinking a rotary switch to choose which one. Having the possibility of feeding them all back at once is cool though and would give a lot more flexibility.

Yeah, I guess I'm not really looking at this in terms of 'how closely can we emulate

• '...it's more of an issue of how to make a delay that functions in a similar fashion to a particular vintage tape delay, using the aspects that were favorable (i.e. using 2399s as 'heads', adding envelope-controlled pitch modulation for 'tape-warp', appropriate LPFs, and adding a compressor before the delay circuits), and optimizing any aspects that were unfavorable (i.e. lack of options that could be considered 'cool' in my mind  )--with the end goal of making something better/less temperamental.
  

[sjaltenb]

well It sounds like we have a group of interested people here.

So lets look at what needs to be done:

In my mind, given all the circuitry, etc, we could use external power supplies providing the necessary 5 or 9v, or whatever, to each part. Thus eliminating part of the Rebote circuit.

Also, because we will be splitting the circuit 5 ways, and eventually mixing it externally, we can eliminate the buffer and mixer part of the circuit.

So what we are left with is the need to create a layout for JUST the delay circuit (or even better 4 on a single PCB) with the addition of:

-Delay time limiting trimmers to create the proper intervals
-a Single feedback controller
-The compression and envelope filters

Volume will be controlled with the external mixer, and the dry signal shall be removed from the delay. We could use the simple GGG mixer to sum both the outputs and the feedback parts. A simple Jfet buffer/splitter- maybe jack ormans- to split the signal 5 ways.

Do we need to create a splitter with 4in/2out for each head to send the signal either out or to the feedback? or simply T off it.

Let me know what I can do to help. If someone shows me the proper parts of the PT circuit to create a layout for, I can do that.

[jacobyjd]

by 'external' power supplies, you mean external to the rebote circuit, correct?

and yes, I think we could use any good DIY mixer to get the signals where we want.

As for feedback, how does the Rebote handle that currently? I don't really see a need to deviate from it--if it uses a splitter, then use a splitter. If not, then just T it.

Also, I think the whole thing should be on 1 board--I'm never much for multiple boards if you can't help it, so input/feedback summer, splitter, compressor, modulation, delay circuits, output mixer should be an entire design. If I ever get around to building this, I'll probably make my own, simply because I'll want to optimize it for my use.

I do think it's a fantastic idea though. The more we can optimize how it's constructed the better it will be when it's finished.

[sjaltenb]

I hear ya there.

I guess if anyone has some suggestions on what type of compressors/pitch mod, etc would be appropriate that would be great.

Other than that I think I am going to simply start a project in Eagle and see how it flows from there. I already have the power supplies, splitters and mixers in there so I can copy those schematics in and go from there.

I think you are right though, with the amount of offboard wiring that will be taking place anyway, it seems logical to put as much as we can on a single PCB, even if it is simply sectioned off.

[grapefruit]

Yes, one, or maybe two or more PCB's connected with IDC ribbon cable would be the way to go, with all components PCB mounted.

Do you really need switches for the feedback and output switching. You can just turn down the volume controls, or is this more of a performance thing??

As for the compressor, maybe a soft clipper would be better. I have a schematic here for a tape saturation circuit that I got from somewhere but I haven't built it. It depends how clean you want it really. If you set up a soft clipper so that the level to the delay chip never quite gets high enough to get digital distortion that could be cool. I've biased diff amps so they clip really gently too. If I get time on the weekend I might try something out...

Data on the PT2399 is a bit thin. I did a search on here but didn't find too many definite answers. I'm guessing an op amp/transistor based voltage to current converter could be used to control the delay time via voltage control.

Cheers,
Stew.

[sjaltenb]

Well I just had almost the entire schematic drawn out on Eagle then it crashed.

I had the 3 Power supplies ( 5, 9 and 18), the jfet splitter, the output and feedback mixer, and the 4 pt80 circuits



I was making so much headway!!!

Luckily most of these are simple circuits that are Copy/Paste able like the splitter, mixer, etc.

[Serge]

I was just thinking of a tape delay simulator just this weekend! Great minds think alike I guess 
Anyway, I found this post :http://www.diystompboxes.com/smfforum/index.php?topic=25096.0
Check it out, there are some ideas on simulating compression and tape saturation in that post which you may find interesting as well.

[Dimitree]

I guess you can get some ideas even from the "Boss RE-20 Space Echo Delay"

[Ben N]

or Akai Headrush, which also simulates multi-head tape delay with modulation--even gives you five separate output jacks--but no feedback routing options. Anybody see a schematic of one of these?

[peterv999]

Hi,
Just joined this excellent forum and very much interested in this old tread. I'm presently designing a tape simulator using the PT2399 delay chip. On my website www.echotapper.nl I've introduced the inner workings of the vintage tape devices and also offered a VST module that's emulating a handful of vintage gear. In order to design this I have a library of fingerprints of most devices I could have access to so there is sufficient information on that site already. My idea is to give a stompbox design a go as allot of guitarist are not (yet) in using computers. The PT2399 was selected although the max delay of the 6Th head could be around 700mS. There are some idea's in my head to resolve the 700mS issue by using one first head delay and the other 5 in parallel (from that first delay). I have most of the parts in twofold available and a friend of mine will (as usual is the echo projects) help out! I have already sent Slacker a PM with some questions regarding the capabilities of the PT2399 chip.

I'm wondering if there are here some people here that would find this interesting

Peter

p.s. it's fully non commercial but it's to become a winter project  so my retired brains are kept warm during the winter.

[peterv999]

i'm just curious, how sonically different would this be?  you're still using analog or digital delay, you're just changing how the feedback (multiple repeats) are configured, right?  Slightly different controls but the same beast, 4x?

Not trying to shoot it down, i think its interesting.  I'm just thinking outloud, if its not that different sonically, will it be worth the extra cash for 3 extra circuits?

please don't take offense i'm just curious.

the main difference between a modern echo device and the vintage devices is:

- multihead setup is with fixed times but with variable attenuations. sometimes per head or per group controlled so it generates a specific pattern
- multihead tape/disc machines require a shelving filter in the feedback so the returning signal in the echo is to be lowered over time
- freq troughput of the tape/disc echo is with a much smaller bandwidth then the modern version.
- wow & flutter are key components for multihead tape/disc machine emulations

-peter

[wavley]

To add my $.02, I have a couple of space echoes...

The RE-501/SRE-555 Chorus echo uses a NE570N compander and a simple 2 band eq (maybe the big muff tone control would sound good here?)

You might look at the schems for the Morley/Tel-Ral oil can delays, there is a Tel-Ray forum somewhere with all of the schematics.

Are you thinking selectable heads for your delay time control like the Binson or delay time controlled by tape speed and fractions of that tape speed for the different heads like a space echo.

The headrush is a good place to look, it's a great sounding pedal, but it does not ramp up/ramp down pitch like a tape speeding up or down, it clicks.

The rotary switch on space echoes for head selection is cool, but the Korg Stage Echo uses head selection like a binson but in a Roland type machine and I've always found this to be quite nice.

A tube buffer would be nice, I've often thought of adding this to my 501.

Things like the fender tape echo and the echoplex use a head on a slider so you physically move the head to change delay time, while the moving of this head while playing sounds REALLY cool I favor multi-head machines myself.

You might also take a look at the schems for Univox tape delays because they're more bare bones than the other multi-head machines and might help figure out the basics.

[oldschoolanalog]

Worth a look. Some good ideas here:
http://electro-music.com/forum/topic-35276.html

[Nasse]

I have RE-20 and have used it some time, I love how the EQ works

[frequencycentral]

OK Josh, I'm here! Perf layout?

[jacobyjd]

OK Josh, I'm here! Perf layout?

Hey man, I'm not forcing you to do anything :-P I mainly figured you'd wanna check it out--plus, with 50 bazillion toggles at your disposal, why not? :-P

I need to think through all this craziness again before I would start the build. I'll collect my thoughts/notes and get back. If this is something you want to take on, a perf/pcb layout would certainly be welcome

I think I may start breadboarding this monster--at least in a 1-head version first, then add onto it as I confirm findings. I really think it could be amazing when properly executed.