The board I'm modifying here cost me less than £4 (which is somewhere around the US$5 mark) - but did take a couple of weeks to arrive as it was coming from China to the UK. If I'd wanted it quicker than I could have spent £7 and got it next day. However, the point being - they're easy to get hold of and they're cheap.
The delay itself is a bit lo-fi and does become quite noisy at higher delay times. And on the two copies of the board I've got it sounds like there's an 8-bit / 22khz sampling thing going on so it cripples the top-end - but as I grew up with the Amiga I'm OK with that audio aesthetic - and gives it a really bit-reduction vibe.
Look, it's a delay for less than the price of a pint of beer - it's not going to be up there with an AMS or Eventide for quality.
The information for this comes principally from watching Mike Sisk's Youtube channel, specifically this video and this one. There's a couple of other resources I used and I'll link to them all at the bottom of the article.
Disclaimer: I am absolutely a beginner at this whole electronics / soldering / making stuff malarkey, so most of what is below is be figuring stuff out as I go along, watching Youtube videos and reading things on that there inernet - so take everything I say within that context. Also, I'm no great shakes when it comes to soldering so some of the joints you'll see below are truly gruesome and for that I apologise to you all.
Here's what the board looks like and the bits you'll need to break it out -
Fig. 1 - the basic PT2399-based board and additional components
Bill of materials -
The first thing we'll need to do is desolder the 50K pot that is already attached to the board. How you remove it is up to you - suction thingy, wicking tape, whatever - but I had real trouble getting these pots off the boards I had. The legs were a really tight fit and I ended up breaking them when trying to remove them. Arguably, that's down to my uselessness - and I won't disagree. However, I'd already decided to use my own pots so that my finished delay would have all the same size pots, so it breaking wasn't the end of the world.
Next up we need to desolder the suface-mount resister labelled R27. Here's where it is -
Fig. 2 - where is the R27 resister?
To the left of R27 you'll see three holes labelled S, G and G. These holes need to be cleared out because we'll be wiring a pot to these. This pot will replace the R27 resister and give us control of the delay time via pin 6 of the PT2399 chip.
Here's what the board looks like after the desoldering -
Fig. 3 - a depopulated board
Our next step is to start putting in wires and attaching them to pots. Before we do though, we'll need to solder a 1K resister to the left hand leg of the delay time pot (the B100K). This is is to stop the PT2399 hanging at power-up if the resistance in pin 6 is too low.
Fig. 4 - the delay time pot with a (poorly soldered) 1K pot on the left leg
With the resister connected to the pot we can solder the wires between it and the board. The left leg (with the 1K resister) goes to the point labelled 'S', while the centre leg is attached to the point labelled 'G' (I did this to the point next to 'S' to keep my wiring tidy).
Take the 50K pot you either desoldered off the board earlier, or - if you're like me and broke it - take a new 50K pot and solder wires between the legs and the corresponding holes in the board where the original pot used to sit.
Your wires should look like this - notice the missing R27 resistor -
Fig. 5 - breakout wires from the board and the removed R27 resistor
Out next part involves a bit of precision soldering. It's a little bit fiddly, but I'm no expert with the iron and I managed it. We're going to be soldering a wire to the capacitor C15 on the side next to R21. This wire will be for the feedback control.
Here's where we'll be soldering a wire -
Fig. 6 - where is C15?
Space is tight, so a small tip on your iron and a steady hand will be helpful. When I did mine I decided to come in from the side to maximise the contact area. It also allowed me to use a bit of superglue to stick the casing of the wire to the board for a bit of extra security.
Here's what that looks like when it's soldered in place -
Fig. 7 - soldered in the feedback wire
The other end of this wire goes to the right-hand leg of the A100K pot (which I didn't have any of, so I used a B100K instead - works just as well for my needs).
You should now have three pots soldered in place (though the feedback pot will only have a single connection - keep reading, you won't believe the one weird trick for where it goes). It'll look like this -
Fig. 8 - three pots connected to the board
Our final connection will be to connect the centre leg of the feedback pot to the centre leg of the mix pot. I knew the size of enclosure my delay was going into so I just used a short cable - use whatever length is appropriate for your design. It'll look like this -
Fig. 9 - jumper between the centre legs of the feedback and mix pots (again, apologies for the terrible soldering)
And that's it really, you've broken out the controls so all that's left is to power it and get some audio in and out. The audio connections are through Dupont header connectors. I'd recommend using proper jumper cables for these as I had connection and noise issues when I just soldered the wires from the jack sockets to the pins (which is understandable really). Just grab a female cable, chop of the other end and solder that to your jack socket, works great.
A word about the power connections. On the two that I got the blocks were really bad at making a connection. So much so that I rounded the screw-heads on one trying to get it tight enough to hold the wire. I ended up having to solder directly to the points that protrude through the board - they're quite big so it's not a big deal.
With this way of modifying the boards, the mix does have a control over the feedback - there's a relationship between it and the feedback knob. I'm not sure if that is just because of the shared ground on the two pots, or if that's the design of the board - I didn't check how these work before getting into the modification.
Links:
All work © Darren Shaw 2020 (except where noted)
darren@anexium.com
Back to the index