Sunday, 20 April 2025

Synthesizer Build part-68: VC DELAY by BMC.

This is BMC 83 the voltage controlled delay using the PT2399 chip. This is a eurorack friendly project.

Dispite the fact I built over 68 projects I never build a digital delay or reverb, except for project 11 but that was a ready made effects unit. This project takes care of that. It can deliver delays of up to 4 seconds, but not in high quality. The PT2399 wasn't made for such long delay times but shorter times sound really good and with the long times you get some cool distortion, sort of a bitcrush effect.
This was quite an easy project to build. You can find the original article on the Barton Musical Circuits website. There are audio demonstrations on that website so you can hear what the delay sounds like. I also made a demo video myself which you can find at the bottom of this article.
This circuit will work fine on both a dual 12V or a dual 15V powersupply.


The finished delay module

Here's the schematic I used to make my layout from. I changed the opamp numbering to match that of the layout.


I didn't use the 10 Ω resistors in the powerrails as shown on the schematic. But if you have problems with hum you can include them. You could put a 10 Ω resistor from K-3 to I-3 and then lead the red wirebridges from there and the purple wirebridge could be replaced with a 10 Ω resistor for the negative voltage rails.
The diode and 1M resistor in combination with the 100nF cap and the top transistor with collector to pin 6 of the PT2399 make up an anti latch-up circuit that presents a high impedance to pin 6 in the first 400mSec after you switch on which gives the internal oscillator time to warm up and prevents the chip from latching up and crashing which can happen if the resistance between pin 6 and ground is less than 2K at start-up. After start-up this resistance can be much lower but not a straight short to ground. In this module the resistance is then controlled by the second transistor which is opened up by the time control potmeter or external CV input. This resistance controls the delay time.
So there are voltage controls with level potmeters for the delay time and the return amount and the module has an audio output that outputs just the delayed signal and a mixed audio output which mixes the original signal in with the delayed signal controlled by the 'Mix' potmeter. There's also a tone control potmeter which also influences the return time I noticed (see demo video below)
The delay time range goes from 60 milliseconds to 4 seconds but like I mentioned earlier the audio fidelity drops quite a bit with longer delay times, mostly at times longer than 2 seconds but that doesn't have to be a bad thing. It has quite a cool distortion effect. With the longest delay times you do get some clicks and artifacts mixed in the audio but it's not much. The delay times are controlled by the two transistors forming a voltage controlled current sink. The 47 Ω resistor at the emitter of the bottom transistor determins the shortest delay time while the 330K in parallel from the collector to ground determins the maximum delay time.
In my own build I did notice quite some dead space at the beginning (ccw side) of the 'Time' potmeter but lowering the value of the 47 Ω resistor didn't do anything. 
I urge you to download the PDF accompanying the original project. It has a comprehensive description of how the circuit works and what all the components do. 
Here's a block diagram of how the delay works. This is also from the PDF that comes with the build instructions on the BMC website


Audio in 2 is the Return input and it has the Direct Output normalled to the socket switch. So if you take the direct output into an external effect module and take the output from that module and connect it to the return input you can have an external effects loop going, creating all sorts of possibilities. You can, for instance, lead the direct output into a lowpass filter and have the VCF out connected back to the return input.

HOW TO PATCH UP THE MODULE:
To get the best out of this module you need to make a synthesizer voice in your modular synthesizer where this delay sits behind the VCA at the end of the signal chain. You can also patch it up so that the delay sits inbetween two VCA's and have the second VCA opened by an ADSR with a slow Release time. That way you get more control over the Delay time, but it's not necessary. The minimum Delay time is 60mSeconds so it won't be able to create flanging or chorus effects. But you can mix in the effect with the clean signal by using the Mix control and the Mix output.

LAYOUTS:
Here are the layouts I made for this project. They are verified as always. I used them to build my module. This was almost another hole in one. I made one little mistake. I had all four non inverting inputs of the TL074 grounded only the last opamp with the direct output must not be grounded. Once I corrected that the circuit sprung to life. Pins 5 and 10 of the TL074 are connected through the strip underneath the chip. The 'Tone' control potmeter has pin 1 not connected. It's important to wire it the way you see in the layout or it won't work properly.
Wiring:


Stripboard only:


Cuts and wirebridges. You know the drill, mark the cuts on the component side using this guide and then stick a pin through the marked holes and mark them again on the copper side. Then cut the marked positions with a hand held 6- or 7mm dril bit.
Don't forget to cut position P-8 underneath the ground wirebridge.


Here is the Bill of Materials. 
It might be a good idea to use a logarithmic 100K potmeter (A100K) for the return potmeter. A lot of changes happen quite early in the throw of that potmeter. However I used a linear 100K myself and that works fine too. But the log type would be more convenient. You could use other value potmeters for all but the Tone Control. That has to be a 10K linear potmeter. The other potmeters are just voltage dividers in this circuit.


PICTURES:
Here are some pictures from the build proces:
I left out the two short wirebridges that connect all 3 ground strips at the eurorack connector together. Instead I soldered them together with some extra solder bridging the gaps.


Stripboard all wired up for testing. I normally only wire things up when I have the panel ready so I can keep the wires as short as possible but with this module I had to be sure first that everything worked. Anyway, it made mounting the board behind the panel easier coz no need for soldering and I was able to stuff all the wiring underneath the stripboard out of harms way.


This is the panel with the waterslide paper applied ready to receive a final thick coat of clear lacquer. The panel is 14hp wide (7CM). the width I normally use because it allows me to mount the stripboard flat behind the panel keeping the depth to a minimum.


Here's the panel design I made in Photoshop just in case you want to use it. It's in A-4 format 300pix/Inch resolution.

Module on the test bench:


The rear of the module. It's 3.8 cm deep so it will fit any Eurorack case.


VIDEO DEMO:
Here's a little demo I recorded showing the module in action.



Here's an interesting look at the inside workings of the PT2399 chip: --- click here ---

Okay, that's it for this one. Hope you like it.

If you have any questions or remarks about this project please put them in the comments below. Remember comments are moderated so they don't appear straightaway. Only after I read them.
You can also post questions on the special FaceBook group for this website.

If you like this content and would like to support these projects and the upkeep of the website which is presented to you free of costs, you can buy me a coffee. There's a button for that underneath the main menu if you're on a PC or Mac. You can also use this PAYPAL.ME link to donate and cut out the middle man. All donations go straight back into new projects. Thank you!

No comments:

Post a Comment

Note: comments are moderated and do not appear straightaway. Your first comment is not allowed to contain any links.