Messages

Hey all,

several years ago I built the old version of the Cave Dweller and really liked it. I decided to try the newer version but the hiss that comes with the repeats even on delay-time settings around noon makes it kinda unusable for me. I mean, I know that the PT2399 introduces a lot of noise because of the analog-digital-analog conversion especially at higher delay settings but it's just too much in this case. I tried tons of different PT2399 ICs I had laying around and also pulled the PT out of my very good sounding Multiplex Jr build but it was always about the same amount of noise accompanying the repeats. I tried increasing R9 to 1k5 to cut more of the higher frequencies and it helped a little bit but it's still not really usable for me.

So, do you guys have any suggestions what I could try to cut more of the noise? I know we don't have the possibility of "quasi-companding" here as we don't have a dedicated output-buffer but couldn't we achieve something similar with the built-in opamp? Like reducing the gain on the input-buffer and increasing the gain on the output-buffer after heavier filtering of the delay repeats? Something like this?
Any ideas would be nice. I assume I can't expect a sound like from the Multiplex or the Deep Blue or whatever but at least it should be usable also when the delay pot is set around noon.

Ah, and one more thing I was wondering about: Why isn't the Mix-pot set up like usual as a voltage divider between the dry and the wet signal? This way we could get 100% wet signal and 0% wet signal and all in between. But as it is it just allows for 50/50 as a maximum.

Thanks a lot in advance! And thanks for the great work all you guys here are doing for the DIY-community :-)

cheers, iefes

Hey Jon, thanks for your fast reply.
I didn't put the LED directly into the circuit but tapped the signal from where I marked and fed it into the circuit on the right in the schematics (marked with "A" in both diagrams). So there shouldn't be a voltage loss, is this right?

Thanks!

Hey all,

I built the Afterlife compressor (old version of PCB without the potentiometer sub-board) and I quite like the result. I added the threshhold knob suggested by jon and an indicator LED which lights up when the compressor is affecting the signal.
It is working at the moment but I am not sure if I've choosen the right position in the circuit to add the indicator circuitry. As you can see in the schematic provided below I tap the signal from behind the threshhold pot/pin7 of the IC ("A"). I used another value for R1 (68k) than indicated in the schematic but the indicator LED lights up nicely when the signal gets compressed. It is also sensitive to the position of the Comp and Threshhold knobs which makes sense.

I am just wondering if you think there might be a better position to place the indicator-circuitry. At the moment the LED lights up also on the lowest settings but with these settings it fades quite quickly. It also flashes shortly when switched on from bypass. Do you have any ideas if this could be avoided?

So, yea no real issues here, just curious to learn! :-)
Any suggestions and/or explanations are very welcome.

cheers

I think the reason that the hard-clipping diodes in the feedback loop didn't have much of an effect is because I already some clipping diodes incorporated into the active filtering section of my circuit. It was overkill. It might be worth breadboarding in your case though.

As far as plans for the custom build delay, I just have them rattling in my head right now. Two PT2399-based delays, each with their own time, feedback, and mix controls, some sort of switching to place them in series or parallel, switching to move the output of the feedback loop between the two delays, and send/return buffers on each delay's wet signal. My friend likes doing wild stuff with effect patching, so I think he will appreciate the extra jacks. I am thinking about tweaking some cap values on the PT2399 so that one delay line is more tape-like, and the other has more of an analog voicing. I need to finish up another project so I can clear out my breaboard and start planning!

Ah, yes this makes sense. So I will try it. I ordered some PCBs and will start with populating just the first Delay unit and socket all the values which might be worth playing around with. When I decided which values sound the best to me I can finish populating the whole thing.

Your ideas sound quite interesting as well! I also thought about voicing some of the delays differently, so I will see what I can achieve with Jons tone control. For the effects-loop you will include an actual buffer for each Delay? Is this necessary?

I'd recommend including an fx loop to process the wet signal. If you have a Darkside build, try that out in the loop for tape saturation. Also IMHO don't overdue the tape like filtering or the delay will wind up too dark. Have fun.

Thanks JC103! I thought about including an fx loop but I know me and I know that I don't like bothering with external effects too much. I have included the LP-filter which is in the Multiplex and in the ZPSDX and will give this a go.
I did a quick research on the darkside. it looks interesting , but a bit too much for this purpose. If I decide to put in an FX-Loop I will try my Germanium-COT50 in it, this could give some sweet saturation. I will let you know :-)

Thanks Chris!
The mode you mentioned is just a consequence of the other modes I wanted to realize. I don't know if it will be useful, but I am keen to try it.
You are probably right about the "busyness" of three parallel delays. You will have to be very precise when dialing in each delay-time to get a proper rythmic pattern. When it gets too noisy, you can just turn down the volume of one or two of the delays. This Volume-knob on each delay really gives a lot of opportunities.

I also thought about voltage-mirroring the delay-pots like you did in the Sagan but as I also wanted independent delays, it would have been to complicated with switching.

I think I won't bother too much about the saturation, maybe just try to incorporate something like the low pass filter of the ZPSDX. In your build doc you mentioned that the two hard-clipping diodes in the feedback-path won't make a difference in sound, so I was wondering if it is worth it to incorporate them...?

I'd be happy to have a look on your files if it was okay for you. And I'd also be interested in your approach for the custom-build for your friend, if you wouldn't mind sharing your ideas.

Thanks!

Thank you, so I should try to locate each cap as close to pin 1 as possible? Seems to make sense.

Hey there!

Inspired by all the great PT2399 layouts out there, I started working on my own circuit, including different ideas I liked the most.
I decided to use three PT2399s.

My goals with this project are:
- Three independant delays which can be combined (parallel and serial operation), inspired by the Red Witch Titan
- long delay times due to three Chips in series, inspired by the Sagan, Abductor, Zero Point SDX etc.
- seperate Time, Feedback and Volume Controls for each delay
- Switchable feedback-paths (seperate for each delay or one for all the three)
- reproduction of the rythmic patterns provided by the "Multiplex"
- Preamp section of Jons "Hamlet" (18V)
- Filtering section of Jons "Hamlet" which may give the opportunity to tune each delay a little different if desired

Below you can see the circuit I came up with so far. I got most of the values from the Hamlet but I also learned a lot, looking at the Multiplex, Saban, Abductor 2, Zero Point SDX, Dirtbaby, EchoBlue etc. (Thanks to all those guys!) I included Jons idea of putting a LED between pin 7 and GND. I also included the "Dirt"-control on one of the Delays for some nice rythmic noise. I think the rest of the circuit is pretty self-explanatory but feel free to ask :-)

I would like to ask you experienced people here to have a look at the circuit and tell me if you see any things that seem not to make sense or if any other ideas pop into your mind. I included pads for modulation, in case I want to use external modulation on it. I also included one pad for +18V to power one external effect by this charge pump.

I also have a few questions I would like to point out here:
- Why has every chip a 100n cap from pin 1 to GND in most circuits? Couldn't you just put one of them from +5V to GND to get the same filtering? Or do you need a higher capacity with each chip (because the Cs are all in parallel)?
- How could I include a simple (low parts) clipping circuit in the long feedback path so that the repeats get saturated more and more with every repeat? Like in the Zero Point SDX but without adding a complete active filtering-stage? Thought about one transistor-stage with diode-clipping but would appreciate other ideas.
- Is a capacitor from the feedback path to ground necessary (C39,38,53)? I saw this in some layouts but not in all.

I am going to socket a lot of parts, to find what suits me best. In all the different layouts I used to come up with this, values vary a lot. I am looking for a more tape-like feeling of the delay rather than analog.

Thanks a lot in advance!

I am going to let you know if I come up with any new questions or something :-)

cheers, iefes