Friday, May 10, 2024


The NTDD is a PT2399 based dual-delay(DD) processor, and analog Dub-Siren circuit. 

There are two PT2399 delay chips in series with independent delay time controls on the X and Y axis of the "TIME" joystick(Left). 
The delay-mix and feedback-amount are controlled together by the "ECHO" knob. The "REGEN X2" switch enables the output from delay-chip 1 to be fed back to its input, as well as to delay-chip 2's input. When disengaged, only delay-chip 2 will have feedback(echo). 
The Echo and Feedback circuits are designed more for experimental exploratory noise, and less for traditional audio processing. 
The delay times can go very low and become very distorted, and the feedback is set very high to allow maximal feedback and self-filtering with the use of auto-leveling. 
An optical compressor is used to prevent feedback runaway, while allowing lower signals to regenerate continuously. 
An internal trimmer resistor can be used to adjust the sensitivity of the compressor.

The "Dub-Siren" consists of a square-wave VCO, gated envelope generator and LFO. 
The envelope generator is active while the "GATE" switch is held, and the "ENV"(envelope) knob determines the Attack and Release time when the Gate switch is pressed and released. 
The three position switch below the Envelope knob determines the Envelope knob's range. In the upper position, the Envelope knob will have a fast attack, and the knob will only change the release time of the envelope. In the center position, the Envelope knob range will be split between Attack and Release. Increasing the Attack time will in-turn decrease the Release time, and vice-versa. In the lower position, the envelope will have a fast Release time, and the ENV knob will only change the Attack time.
 By default, the envelope generator controls the volume of the square-wave VCO, but can also be used to modulate either the Pitch of the VCO or the Rate of the LFO with the "MOD" 3-way switch. In the center position, the MOD switch is "OFF".
The LFO modulates the Pitch of the VCO. 
The LFO can be set to Square-wave or Triangle-wave with the "SQR-TRI" switch. The LFO Depth and Rate are controlled with the X and Y axis of the right joystick. 
The LFO voltage can be offset with the "BIAS" knob. While the LFO depth is set to minimum, the BIAS knob can be used to change the pitch range of the VCO independent of the LFO.

The NTDD also comes equipped with;
 Volume knob(VOL)
Power ON/OFF switch with LED
1/8 inch(3.5mm) input jack
1/4 inch line output jack
5.1mm DC adapter(9V tip-positive)
Built-in speaker
3XAAA battery compartment.

The concept for the NTDD came to life while I was working on my big SUZUKI PK-37 project. 
I had been experimenting with different delay chips, and started to think about what I should do with tall these PT2399's I had laying around. These days I would prefer to use higher quality digital effects processors for my synth builds, so what to do with all these chips? I though it might be nice to have something weird to compliment the NTSH mini noise synth. The PT2399 may not be the best quality delay, but they can do some pretty freaky stuff when you mess with the filtering, feedback and sample-rate time. I thought it might be interesting to have two delay chips in series and have separate delay times and feedback. I had never tried it before, and wasn't sure it would even work, but with a little experimentation, I was able to get a pretty decent signal through both chips, albeit a little alias-y. One thing I really wanted to include was a self leveling circuit in the feedback path. I love being able to crank the feedback, and not have to worry about turning it back down before it runs away and starts clipping really bad. I have experimented a lot with compressors, compander chips, and auto leveling circuits with envelope followers of all kinds. Most of the time I usually end up scrapping them because they are either too strong and oppressive, or too weak and choppy. I built all the usual stuff trying to tame the feedback with envelope followers and filters. I tried OTA's and JFET's, but everything that worked had too big a footprint for the tiny box, and would probably require a lot of fine tuning. in the end I tried out these old cheap vactrols I had bought years ago. They come in handy sometimes, but are usually too inconsistent to rely on. They worked really well with my full-wave rectifier/follower circuit. They worked so well that I didn't even need any rectifier/follower circuit. Instead I just sent the signal to a comparator circuit that drove the vactrol, and there was no warble, and the the response was beautiful! When the input signal passes a certain threshold, the comparator sends out short pulses to the vactrol, which attenuates the input. The vacrols have just the right amount of lag to keep it subtle but effective, and the variation from one vactrol to the next makes no difference in the performance. Perfect! Sometimes things just work out like this, and I think this circuit was just meant to be.


After a few iterations of the dual delay circuit, I kind of wondered if the whole thing offered enough to be a standalone device. I had intended for the NTDD to be a signal processor for the NTSH, and the continuous feedback did give the NTDD some stand-alone quality. I just wasn't happy with it as-is, so I thought of ways I could add to the design. I thought it would be good to have at least some kind of VCO. I really wanted to keep it simple, though. There is really not a whole lot of space to work with, and I knew I would be super limited for controls. I thought maybe a simple dub-siren would pair well with the lo-fi delay. I was able to come up with a simple dub-siren scheme that was both simple but still pretty dynamic. I was able to fit it all in with some extensive designing. While the dub-siren was the inspiration for the circuit, I would have to say that it is more of a drone synth in practice, due to its limited ranges and control. It has a special character of its own, and I'm sure it would sit well in some more experimental dub music. 


I designed and built several variations of the circuit until I had something that I thought would be  a good final product, but when I assembled the full prototype, I just didn't think it was ready yet, so I shelved it for a few more months, and started working on another big project. About a month ago, my long neglected laser cutter literally went to pieces, and the big project had to be put on hold. Once I had dealt with the laser crisis, I thought I should finish up the NTDD before diving back in to the big job. I made a few changes here and there, mostly just bolstering some of the circuitry inside, and putting in the BIAS control knob in place of a dedicated pitch knob, as well as putting in the square-triangle switch in place of a high-low VCO range switch. The tweaks to the design were a huge improvement on the overall layout of the box. The design was finally finished and ready to build.

Over the course of building and assembling six units, I have had to ask myself some hard questions about whether these are worth all the time needed to build them, considering how much I am able to sell them for. For now, I think it is more important to just make interesting new things, and get them out into the world and make people happy. At the very least, I'm doing something way more interesting than working in some awful kitchen again :D
The six I made will be for sale on my Etsy and Reverb shops. Get one while you can!

No comments:

Post a Comment