low power delay SECTION for Shockman

The Missing Shockman Feature

For a while now, I've had it in the back of my mind to eventually add a delay section to the Shockman. It really begs for it, and I want all my bread-and-butter effects together in one gadget. I first thought of somehow duplicating the ECHO feature of a Rockman X100. The problem is that I don't particularly like the way it's done. It's really complicated, and I find that the end result doesn't justify the large amount of work and parts needed. It's not even what I'd call an Echo (hey hey hey...), but actually an analog reverb that uses a long out-of-production 6-tap BBD chip (MN3011) and a whole bunch of op amps and stuff. It sits on its own PCB that's half of the X100, and doesn't even sound that good by 2022 standards.

I looked into alternatives that had to be fairly simple to implement, sound decent, and fit into my very stingy power budget (single digit milliamps from a 9V battery). There's nothing out there that fits the bill! FV-1 was a contender, but it draws something like 55mA from 3.3V. That's 180 milliwatts. I don't have 180mW to spare, man! The whole Shockman -- buffer, compressor, distortion, cab sim, chorus, mixer and headphone amp -- uses about 3 times less. There was no way I was going to degrade its battery life that much.

PT2399 was another option. It's less power hungry than FV-1, at around 25mA from 5V (125mW). That's still twice as much power as a whole Shockman, though. Is it worth it, for just the delay? Hmmm... But what happens if I undervolt it, like a laptop CPU, to save some battery? The datasheet says minimum 4.5V. Will it still work at 3.3V?

Turns out it works fine even at 3V, and its current draw drops dramatically as you lower the voltage. How fortunate! From about 20-25mA at 5V, it went to as little as 8mA at 3V. That's 24mW. Now we're talking! I can swing that.

But wait. My power supply is 9V, not 3. If I drop it down to 3V using a linear regulator, I'll still be drawing 8mA from 9V, which is 72mW. Very inefficient. The linear regulator is out, and a high efficiency switching regulator is in. MAX17550 is about 90% efficient turning 9V into 3, even with a light load of less than 10mA.

Problem solved! SMDelay was born, and it typically draws less than 3mA from 9V. It sounds similar to an analog tape delay, adjustable between about 35ms and 500ms (with a 50k pot), and is barely larger than a full size guitar pot. Longer delays could be achieved with a larger pot value, but then it sounds really lo-fi.

Hopefully, at some point in the future, I'll combine Shockman and SMDelay on the same board. For now, though, I'm having a real blast with it on its own little PCB, alongside Shockman.

-JC, 2022-03-03

Here's the YouTube demo:


Here's the schematic. It's not necesarily any better or worse than other PT2399 schematics you'll find. Note that there's no dry signal mixed into the output, since this is taken care of elsewhere within Shockman. FBSW1-2 goes to a feedback on/off switch. I like the ability to quickly switch the delay to a single repeat (no feedback).

The key part is the powering scheme on the left side, which greatly reduces the current draw from 9V.