I’ve been building a couple of Sympleseq analogue sequencers from hexinverter and they’re coming on, most of the soldering is done.
Essentially, they’re basic analogue 8 step sequencers based on a 4017 decade counter, offering note on/off/reset per step. There’s an internal clock provided by a 40106 with quite a wider range, and there’s an option for externally clocking, although you’ll have to get quite close to the input voltage to clock it. I’ve not had any success yet in my brief attempt at trying to directly clock it from an MFB Dual LFO.
Here’s the rather lovely Eurorack-sized panel – hats off to Ben at resynthesis.co.uk, he’s done a great job, I’m really happy with this:
This is the logic board, attached by a header on the right hand side to the switch board, it’s a bit wobbly:
Here’s a (mildly chopped up) really quite horrible drum loop I sequenced from the Simpleseq, sound source being a homemade Polivoks filter and a Doepfer noise source mixed together, with a bit of spring reverb and occasional delay from an RE-201.
I was trying to be a bit Human League circa Reproduction, as ever.
Hats off to Mr Gaudreau for getting this out, they are good for the money. Having said that, I’m not keen on the switches, which are very easy to burn up when you’re soldering, although the build manual does give fair warning about this. Not that I read manuals *cough*. To Stacey’s credit he does recognise this and include a couple of extra switches in case you do fuck up. Which I did. But only twice.
Also I think as a function of trying to fit two sequencers into a small space, the switches are pretty small and… not really the best. They seem to be a bad combination of stiff and fragile – I managed to snap one in my fingers when trying to switch it down… it’s also hard to tell when they’re on, off or set to reset.
On the positive side, it’s really good that the sequencers work off a single sided power supply, so you can attach a 9v battery and start bleeping. It’s also not very deep, so it could be fitted in a slim desktop case, which is my current plan.
Because they’re based on 4017 decade counters, there are no extra features like reverse play, or pendulum or CV control over which step is played next. All the attempts I’ve seen so far at linking two 8 step Sympleseqs together to make a 16 step sequencer have been a bit tricky, the most reliable seeming to be switching on each trigger between the two sequencers, rather than switching at the end of 8 steps.
Thinking about it a bit more, it would be fairly easy to replace the two logic boards with an Arduino Nano, and use a pair of shift registers to drive the sixteen steps, with the remaining digital and analogue ins used for selecting playback modes. I say easy, it just takes time.
Hey Steve!Thanks for the blog mention. Very cool to be able to follow along with your experience!I think you hit the nail on the head with the switches. That was a major oversight on my part, and I apologize. My initial encounters with the switches during beta were successful, so I chose to go with those for the final product. I probably did not spend enough time with them to realize that they aren’t the greatest.The unfortunate reality of it is that the cost of a sympleSEQ kit goes up dramatically to have nicer switches — to the tune of about +$25 per kit. If I offer kits in the future, it will only be with the more expensive switches.Another thing I missed, which you also mentioned, is the issue of levels in a standard modular setup. The results of me failing to design for this are two-fold:1.) Gate/CV levels leaving sympleSEQ are much higher than most normal 0-5v signals. To fix this, install a voltage divider (10k and 10k is fine for R1/R2) on the CV output, and an LM7805 5v voltage regulator on the gate output. This will reduce levels to a standard 0-5v (or thereabouts for the CV)2.) Clocking sympleSEQ from standard amplitude clock pulses doesn’t work very well. It needs a really high clock signal. To fix this, use a standard NPN transistor like the 2N3904. Have your clock input connected to the base through a 47k resistor, and have the transistor switching the sympleSEQs supply voltage into the clock input. Now, when you put 5v into your clock input, it will turn the transistor on, allowing your supply voltage to flow into the clock of the sympleSEQ. This essentially makes it activate on a lower voltage :)Hope that helps! Feel free to email if you’d like to ask questions or anything!Also, you may be interested in my next eurorack/DIY project (if you have not seen it yet) — it is a programmable sequence expander/sequential switch that should pair with 2 or more sympleSEQs very happily! A final beta + demo video is in the works over the next weeks, but you may check out the muffwiggler.com forum post detailing the function of it, as well as the indiegogo.com funding campaign if you’d like, here:http://www.muffwiggler.com/forum/viewtopic.php?t=56316&postdays=0&postorder=asc&start=0http://indiegogo.com/seqsquaredI wish you all the best in your creative endeavours!– Hex
Hey Stacy, thanks for stopping by and commenting. I totally understand the reason behind choosing the switches on the Sympleseq. I might be tempted to replace the switches on mine with the better type, but I’ll see how I get on for now. Aaargh desoldering, etc :-)Thanks for the suggestions on the signal levels. I actually quite like that it puts out such high levels by default. Old sequencers like the Roland 182 and the System 100 104 sequencer were switchable between different ranges, so at least the way you’ve done it, we have that option. Also my Roland System 100 101 keyboard seems to need a meaty gate voltage.I’ll give that transistor switch idea a go – sounds a bit simpler than the alternative I had in mind which was to implement a comparator on the input, as on Ken Stone’s gate sequencer.Good luck with the seqSquared 🙂
2 comments