Stereo Memory Man – more clock madness

This one is a bit convoluted but kind of cool. Well, nerdy-cool.


I’ve owned a few Electro Harmonix pedals over the years, including a couple of Stereo Memory Man with Hazarai™s (to be referred to as the SMMH hereafter). This makes it one of the only bits of gear I actually cared enough to replace after it broke (to be fair, I broke it myself). I don’t actually use it much as a straight-ahead delay;  it’s the looper functionality where it really shines.

Here are a few of the crucial features which make the looping section unique:

  • Varispeed. There are plenty of loop pedals which enable you to play the loop at half or double speed, but on the SMMH this is via a knob, so you can go anywhere inbetween. Doesn’t seem impressive on paper but you can get all sorts of weird pitching effects. It’s like a tape machine with a speed control; i.e loads of fun.
  • Feedback control. Probably the most awkward to explain, but this controls how much of the existing loop remains when you overdub. At halfway, the volume of the existing loop is halved, so you can create slowly evolving patterns where the existing layers of sound subtly fade out as you add more. Trippy.
  • The footswitch. This will divide people but I am used to the fact that the SMMH uses a ‘soft’ switch for the tap tempo and overdub control. It also works in a unique fashion- you hold the switch for as long as you want the loop to last, and the loop starts playing when you lift off. This might be totally counter-intuitive to those used to other pedals; the majority have clicky footswitches which you tap once to start recording, and tap again to stop recording and start playback. The SMMH works for me because it is exactly like using the sustain pedal on a piano. More on this later.
  • The tap tempo still works in loop mode, and speeds up or slows down the current loop without changing the pitch. You could either use this to subtly change the speed if you’re trying to keep time with other musicians, or you can go to extremes and get some weird grainy time-stretching effects.

Basically the one thing that the looper lacks is the ability to sync to anything; recently I stumbled across a mod on the internet which adds just that.

Nav’s blog is a trove of interesting stuff, and there are 3 posts about the various modifications all of which are explained in excellent detail. I chose to do two of them- adding the clock output and an external footswitch socket, so that the pedal can remain on a desktop and still be foot-operated. The switch I use is a piano-style sustain pedal which works exactly as well as you’d think.

Here’s a pic of the mods (a bit messy but you get the idea). What’s really cool is that you’re basically deriving the clock from the tempo LED, which is something I’d never have thought to do.

So having tapped the clock, I naturally wanted to turn it into midi clock to drive the rest of my gear. Here’s where I ran into a couple of issues; the clock that you get is effectively 1 pulse per quarter note, or 1ppqn* (see appendix). So I tried it out with my Sonic Potions LXR, which has a clock input section on the trigger expansion board. It worked, but sadly there’s a bug in the LXR firmware which causes skipping, which quickly began to annoy me. So instead I decided to go back to the Volca Keys, which I’ve found to be a good way of converting analogue clocks to midi. The glaring problem with the Volca is that the clock resolution doesn’t match – it expects 2ppqn which means that the Volca, and its subsequent midi clock output, run at half speed. Still with me?

In the intervening period, I got hold of an Arturia Keystep, which does support 1ppqn clocking. Except I found it to be a bit fussier than the Korg in terms of what clock signals it works with. Driving it directly from the SMMH directly was a no-go, after a lot of failed attempts. So my final set up is as follows:

SMMH clock out –> Korg Volca ‘sync in’ –> Korg Volca Midi out (clock)—> all other midi gear

With most of the stuff I own, there’s usually an internal resolution, so for example I can run the sequencers of some of my gear at ‘double’ speed to correct for the ‘slow’ clock. You can similarly change the quantise resolution on the LXR to compensate, but it’s a bit of a faff when programming. Hopefully the LXR firmware will get an update and then I’ll be able to dispense with the Volca and shorten the chain a bit.

That’s probably enough about the technical side- what is it like in use? Well it’s a weird process, in that it involves going about things backwards. Usually if I’m trying to make a loop, I’d probably start with the drums or a click track then layer things up on top. With this system, you start with the loop layers then hope that the clock lines up over the top! The key to success is playing in time with yourself, and trying to count the beats as you play (again, see appendix). After a bit of practice it does get a lot easier, and the beauty of it is that you can wipe the loop and record something else if you’re not happy with it. Then reverse the whole thing, change the pitch, do some time stretching and the beat will stay in time.

Let’s face it- this is certainly a niche use for this pedal, but if you’re into modulars and voltage control, there are any number of further applications. Finally- don’t attempt these mods unless you’re confident that you can do them without wrecking the pedal – the pcb is so firmly wedged in the enclosure that you’ll probably have to drill holes with it still in place.


*Appendix: Is it really 1ppqn? Well, sort of. Basically the pedal generates a tempo, always in 4/4, based on the length of time you hold the ‘loop’ footswitch for. It takes some getting used to, but after a while it feels almost as though the pedal ‘detects’ the timing from your playing, although in reality it’s probably just dividing the length of the loop by 4 (or 8, or 16). The fact that the pedal calculates a tempo in the first place was very surprising to me, but it makes sense when you consider the tap-tempo functionality. Either way – this is a seriously over-engineered pedal; and is a testament to the skill of designer David Cockerell, who is a legend in the world of electronic music. Here’s an interview with him if you don’t believe me:






Ikea Pedalboard

I’ve moved into a more permanent home-studio space (hence the huge gap between blog posts). As such I’m trying to organise the room a bit better, and one thing that especially needed a solution was the mess of guitar pedals on the floor under my desk.

My bandmate uses a Pedaltrain Classic Jr which looked the ideal size for my purposes- (typically I use about 4 or 5 pedals) but I began to wonder if there was a cheaper alternative. Googling ‘DIY Pedalboard’, I discovered that IKEA sell shelves which are easily converted; the current shelf-of-choice for guitarists seems to be the HEJNE. As it happens, there’s an IKEA up the road from me so I popped in with the intention of getting a set of HEJNE shelves. However I discovered something much better*…. the KNAGGLIG


*Well okay- this worked out well for me, I can’t actually speak about how good the HEJNE shelves are as a pedalboard because I didn’t buy them in the end.

Here’s what’s good about the KNAGGLIG:

  1. The size. It’s actually a bit shallower front-to-back than a Pedaltrain Classic Jr, but there’s enough room to fit a decent collection of pedals. Crucially, there’s enough of a gap between the the wooden slats to run cables through.
  2. The ease of conversion. You only need to make two cuts with a saw to turn the KNAGGLIG into a pedalboard. You can even use the provided screws to hold the thing together.
  3. The price. It’s £9. Granted, that’s £4 more than the HEJNE shelves but you can make two identical pedalboards out of one box.


IMG_0224The wood just after cutting- the side of the box (r) is used as the main pedal platform. The bottom of the box (l) has the slats too close together to run cables through.

IMG_0225 This is basically as far as the construction goes – screws are included.

Couple of issues- if you’ve got a cheap bulky unbranded power brick like I have, it won’t fit underneath the board. Most decent ones from Voodoo Labs or T-Rex should be fine though, especially if you add rubber feet to your creation. Also I initially ran out of velcro tape before I fully covered the board, so best to buy two rolls.  And it goes without saying that a home-made pine pedalboard won’t stand up to gigs as well as a purpose-built metal one….I’m not planning on any heavy duty action with this but it’s fine for keeping things a bit more organised at home.

IKEA also make a smaller, cheaper version of the KNAGGLIG for £5; I had a look at this in store and decided it wasn’t big enough, but it would probably work if you only want to run a couple of pedals, or if all your pedals are really tiny. If you need more room on the board, shelves seem to be the way to go. Have fun.


Silent Running: Trumpet technology

OK so no DIY this time- instead we’re going deep into the brain- with a bit of technology thrown in.

From the age of about 10 I was a cornet player in a couple of youth brass bands; later incorporating the trumpet and joining orchestras and jazz ensembles. By my teens I usually had four or five band practises a week (and no social life). A side effect of playing all of the time was that my lip was usually in pretty good shape. This lasted until I went to University, where a number of ‘distractions’ meant that I was playing less trumpet, and losing some of the stamina I’d built up for pressing a bit of metal on my lip for hours on end.

The thing I’ve experienced with the trumpet is that it’s a fairly lonely instrument to play unless you’re in a band. Being monophonic, it’s limited to a single melody line rather than the chords that can be formed on a piano or guitar. It’s also a loud instrument- and this was to prove a stumbling block when I moved to London and into a succession of small flats.

I still had the occasional trumpet gig, but these would usually crop up on average once every couple of years. This would result in me attempting a bit of practice, usually with a mute in, and feeling disgusted at how difficult it was to play even simple pieces. The problem was: in my head, I could still play! The musical knowledge and the muscle memory of how the valves worked was still there, but my lip was disobeying me. Coupled with this was a feeling that I was making a horrible racket and disturbing people around me. It’s very easy to feel self-conscious and embarrassed when playing out loud- I suffer from this, so always use headphones when practising keyboards or anything electronic in the house. Fortunately I can now do the same with the trumpet- as the solution was to get a gadget!*


This isn’t a review of the new Silent Brass system as such. It’s not that it doesn’t sound good (it does), or isn’t easy to use (it is); it’s more about the concept and what it does for your brain. The ability to completely blast horrible sounding, split, out of tune notes in the knowledge that no-one can hear you is amazing. Short of a sound-proofed room, or a cabin in a remote location, this is the best solution I can think of for rebuilding brass-playing confidence. If they made one for vocal training, I’d buy it- although I’m not sure that the resulting gimp mask/space helmet would be as practical, or even safe to use. On the subject of practicality- there’s an older Silent Brass model, which is bigger, heavier, and doesn’t sound as good (based on me playing one in the late nineties, YMMV.). So it’s not ‘new’ technology as such- the idea was as good in 1995 as it is now. The best thing is that after a few weeks silent practice, I realised that I haven’t completely lost the ability to play. I’m a former ex-trumpet player. And as a result, I’ve got more gigs coming up.

Flowers of Hell play Truck Festival on Saturday 18th July.

Touring in September.

Thanks to CDM for this article from Feb 2014.

*the solution isn’t always to get a gadget

DIY Time Part 3: Kits

If you say “home-made synthesizer” to someone, the mental image is probably a rat’s nest of wires making cranky mad-scientist noises then breaking down in a puff of smoke. Whilst this is cool in itself, it’s not a fair reflection of what the music DIY scene is today. A synth from a kit can sound as good as- and cost far less than- the equivalent ready-made unit, and you don’t need any serious electronic knowledge to have a go. So far in 2015 I’ve managed to make two units which not only worked first time but have proved to be musically useful; proper synthesizers rather than novelty noise-makers.

A bit of background: In 2013 I bought my first kit- a drum machine from Standuino in the Czech Republic, now operating as Bastl Instruments.


It was a bit of a problematic build for me (for the sake of brevity I won’t go into detail about it) and with hindsight it was probably a case of me trying to run before I could walk. So I went right back to basics and built the fuzzbox you see next to it. (this time from Pedal Parts UK). Something easy like a guitar pedal is definitely recommended as a starting point; if you make mistakes, they’re generally easy and cheap to fix! Despite some wonky wiring, the fuzz continues to work, and is in regular use.

I wanted a bigger challenge for the next project and decided my “skills” had improved enough to try a synth. I was already familiar with the French company Mutable Instruments through owning a Shruthi-1, however I’d bought mine pre-built, and had always felt a bit guilty that I hadn’t soldered one myself. At the time, the company were scaling back their DIY line (although the classic Shruthi kit is still available today) and had a sale on unpopulated boards. This would mean sourcing all the remaining components, a further step into full electronic geek-dom. In actual fact, I found the self-sourcing quite fun, mainly due to the fact that it meant multiple parcels turning up (let’s face it, getting a parcel is always exciting). There were certain small inconsistencies, like the fact that I am still waiting for a knob cap to arrive 4 months after finishing the project – this sort of thing is par for the course, and a good argument for ordering a full kit wherever possible. The only moment of doubt came when I powered on the finished device and the screen was blank- simple solution: turn down the contrast trimmer.


So with the Shruthi complete and sounding great, I started looking into new projects which I wouldn’t have dared to attempt previously. I ended up getting the boards and parts for the Sonic Potions LXR Digital Drum Synth. I took more photographs of the build this time around, so you can see the progress below.



20150320_145143 20150320_155144

I finished the whole build in a single day, which was much faster than the previous kit: the extra soldering experience probably had a lot to do with it (not to mention longer daylight hours!). No contrast difficulties this time around, as it uses a fancy OLED screen, but I did have an awful moment when I first connected it to a mixer and it produced a horrible static-interference type noise. Turned out the power supply wasn’t giving it enough amperage – when I located the right supply the digital noise instantly became pristine digital drums!

Julian Schmidt from Sonic Potions says “A machine you’ve built is a machine you love!” I agree- and would add that building kits helps to de-mystify what’s going on under the hood, so “a machine you’ve built is a machine you have more chance of fixing if it goes wrong!” There’s also the fact that both Sonic Potions and Mutable Instruments designs are Open Source, which means if you want to hack the source code or check out the schematics, everything is freely available. I think this concept is extremely forward-thinking, not to mention generous. Having said that, I’ve got work to do before I can hack any firmware myself!

Finally I should add that my success in building these kits has everything to do with the clear step-by-step instructions provided by the manufacturers. It’s basically grown-up Airfix, but with the distinct advantage of not having to wait for the glue to dry.

Next time: CASES.

DIY Time part 2: Click to Clock & Nitrotracker

For a period of time in around 2007/2008, before the iphone got fully into its stride, the Nintendo DS was the platform for portable music apps. I bought a DS Lite specifically to take to work and make tunes in my lunch break (a 24 year old with a kid’s game console, although not ‘cool’, was just about socially acceptable at the time). Synth giant Korg even got involved in the scene with their DS-10: an excellent handheld recreation of the MS-20. However, the pick of the software for me was Nitrotracker, written by a German coder called Tobias Weyand and distributed for free.

So what is Nitrotracker? It’s a sample-based tracker that takes its cues from Fasttracker II – in other words, an old-school music sequencer. On classic trackers the note entry was done via your computer keyboard: here you use a stylus and a mini piano keyboard on the DS’s touchscreen. Believe me, once you are used to it, the workflow is insanely fast.

Here’s the kind of thing I could knock up in a lunchbreak:

The main problem that I encountered was how to integrate my Nitrotracker compositions with the rest of my studio. First off, when the tempo marker in Nitrotracker reads “140”, this doesn’t necessarily equal 140 beats-per-minute. So if I recorded some nice blippy patterns from Nitrotracker into my computer, and wanted to add a drum machine over the top…..well, let’s say it was more ‘miss’ than ‘hit’.

I tried several solutions to this problem. First off- the vaunted MIDI functionality of Nitrotracker itself. I bought a DSerial interface, but found the latency to be too high for drum hits, and experienced a lot of stuck notes. Then I tried an audio-to-MIDI clock converter in the form of the RedSound Soundbite Micro. This was better, but my propensity for wiggy arrangements meant that it often dropped the beat. Bottom line – the Soundbite is a good unit if you have a solid 4/4 kick drum. The solution only presented itself much later, when I came into possession of a Korg Volca Keys.

The Volca range have 3.5mm ‘sync’ inputs and outputs. The specs from Korg state that these send and receive a 5V clock pulse- the sort of voltage-based signal usually associated with analogue sequencers and modular equipment (of which I have none). However, Korg also promote an app called “SyncKontrol“. Here’s where it gets interesting (no, really!)….


The SyncKontrol app runs on any iOS device, so iphone, ipad etc, all of which have a standard stereo 3.5mm headphone jack, but nothing in the way of CV/Gate outs. I concluded that the sync input threshold of the Volcas must be low enough that an audio signal could trigger them. I downloaded the (free) app, and recorded about 30 seconds of the signal from the headphone output of my ipad into my computer*. I saved a single ‘pulse’ from the resulting recording, and imported the WAV file into Nitrotracker. Here it is:


Next up, I created a pattern which played back the pulse at a rate of 8 notes per ‘bar’. If you want to get technical, the Volcas sync to 2ppqn, or “2 pulses per quarter note”. On connecting the headphone output from my DS to the Volca, the tempo knob started flashing. Sync was established. But how to get music and sync from the DS simultaneously?

The solution was simpler than I thought- split the signal coming from the DS (I’m working on a custom cable for this, but in the meantime a £1 headphone splitter does the job).


Next, I hard-panned the sync pulse to one side of the stereo mix, and everything else to the other side. I ended up with a mono signal from Nitrotracker, which I found to be a decent compromise for two reasons; 1) the panning in Nitrotracker is pretty rudimentary anyway- no automation or anything, so not much to miss. 2) mono output means it’s easier to patch it through fun guitar effects like distortion, phasers etc.

So the Volca is synced to the DS, but it doesn’t end there. Korg have been generous in making the Volcas extremely easy to hack– so I added a MIDI output (god, more MIDI???), in a quick-and-dirty fashion:


The Volca Keys will now happily pump out midi clock in time with the incoming clock from Nitrotracker. Party time!

Here’s the first test, with the sync pulse audible at the beginning for reference. Not the finished article musically, but you get the idea:

OK- so not that many people will have a burning desire to plumb a discontinued Nintendo handheld running homebrew software into a musical environment! However, the Volca element has some excellent mileage in terms of live performance. You could record backing tracks to one side of a stereo mix, with a sync pulse on the other side, and gig with an mp3 player, a Volca or two plus MIDI equipment and keep it all locked rigidly in tempo. In this article, Dan Deacon describes a similar set up, but simply sends his click track to a drummer of the human variety. It’s not only the Volcas that do this- the new Teenage Engineering PO series also have 3.5mm sync inputs. And if you’ve got an expensive modular synth to play with, I’m sure the applications are infinite!

Finally, I’m aware that this is not exactly new technology- we can trace the technique back to 1977 when an engineer called Robbie Wedel synced a tape machine to a modular Moog. The result?

*If you don’t own an iOS device, IN THEORY you could connect the sync output of your Volca to a mixer and record the output as audio: UPDATE 03/03/15 – I have since tested this out and the volume was fine into a standard line input.

DIY Time Part 1: Arduino USB Midi Converter

In the last couple of years I’ve become aware of the huge boom in DIY music equipment. The internet provides a massive (and free!) repository of interesting and inspiring projects to try. A great many of these projects are built on the Arduino platform, an open-source board which will do pretty much whatever you program it to. Not being a coder by any stretch of the imagination, I had wondered whether or not I’d have the skills to make anything useful using an Arduino. It turned out my fears were unnecessary- this project can be realised by a total novice.


I was looking to get a small midi controller, but found myself frustrated by the fact that most of the smaller, cheaper ones were USB only. Fair enough – there are issues with the amount of space to fit a MIDI port on a small controller, and they are marketed mainly at computer musicians. However, a bit of googling led me to this discussion on a V-Guitar forum, which suggested that using a USB Host shield and an Arduino could translate USB Midi to a ‘regular’ 5-pin Midi output, as well as providing power to the controller itself. So I took the plunge and ordered the two boards. Total cost was around £21, so even if it didn’t work, I hadn’t spent a bomb.


Now- important disclaimer: I am not the author of any of the code here, the credit must firstly go to Collin Cunningham for writing the code (way back in 2010). Yuuichi Akagawa’s updates on Github made it possible to use multiple devices via USB Hub (more on that later), and “Spider” over on the V-Guitar forums provided the invaluable step by step guide to making it work. This is the beauty of open-source and the internet, and really the point of this post: anyone can get the info and do-it-themselves!

The picture below shows what I ended up with (click for larger version)


As per Spider’s instructions, the TX port of the Arduino shield connects up to the Midi out socket, with a 220 ohm resistor inbetween, along with the ground and 5V pins. The small piece of breadboard holds a 6N137 optocoupler for the Midi input. This, it later turned out, was unnecessary. My advice would be to stick to a one-way conversion- USB Midi in to hardware midi out, but more on that later.


So far so good. While I was at home over Christmas, my Dad offered to case the thing up (having a much better selection of tools, not to mention a lot more practical skill!). The resulting device is not much bigger than two packs of playing cards.





Here you can see it at home in a portable synthesizer rig I constructed over the Christmas break. This is completely standalone from any sort of computer, and runs off just two plug sockets. The little Akai keyboard (MPK Mini mk1) is great for this setup, as it has different presets which allow you to switch between controlling the two synthesizers- each of which is set to a different midi channel. In the example shown, the Mutable Instruments Midipal provides a midi clock for both instruments, but I did try it out as a channel splitter, which enabled the entire thing to be played as a 4 voice synth (albeit with a very cramped keyboard!).


So the converter works, and works well. It runs quite happily on USB bus power, any generic USB plug adaptor that you might get with a mobile phone, or a 9v DC supply. However there are a couple of things that it currently won’t do.

You can’t currently convert midi from your computer to hardware midi using this device. The way I see it- it isn’t designed for that. It acts in place of a computer in hosting the USB device, and is therefore more suitable for stand-alone operations.

I failed to get the hardware midi in port to work at all. The biderctional_converter (sic) example specified in Spider’s guide would not compile, no matter what I tried. This may just be an issue with my particular computer, and I’m hopeful that at some point in the future a modified code will appear. The beauty of the Arduino platform is that it’s so easy to change the program running on the chip, so this isn’t a dealbreaker. I now think of the midi in port as ‘for future expansion’!

I did however test running multiple USB inputs via a generic (unpowered) USB hub. This worked fine with two USB devices, and I couldn’t make it “choke” by sending multiple CCs. Adding more than two devices may be possible, but I was unable to test it out.


Thanks to the generosity of technically-minded folks such as those mentioned above, it’s easier than ever to get into DIY projects like the one detailed here. For such a minimal amount of money, I’m really pleased with this little black box. The fact that it runs on an open source platform means that it’s endlessly customisable (with the right code, of course). For a first time toe-dipper in the world of Arduino, it’s extremely encouraging. In the words of Alec Baldwin: “Go and do likewise, gents*!”

*and ladies, Alec.