Łukasz Martynowicz

For me, the main problem with polyphonic aftertouch on concertina/bandoneon is not "is it musically useful", because it obviously is once you stop thinking about MIDI concertina as a simply "silent practice" imitation of "the real thing" and start thinking about it as a versatile MIDI controler with bellows driven expression. The problem is the cumulation of functions assigned to the same finger force vector. IIRC Striso utilises lateral button movement for additional expression. Simple "just press harder for aftertouch" also changes bellows pressure and with just 3mm of typical concertina button travel even velocity sensitivity conflicts with my acoustic concertina playstyle. It is different with accordions, where buttons act in perpendicular direction, so you can utilise "just push harder" even on an acoustic box. See blues accordions with note bending feature.

There are two "flavours" of the aftertouch. One is monophonic "per channel" and one is polyphonic "per note". Monophonic is very straightforward to implement in MIDI concertinas - either map the pressure readout to aftertouch message for instrument patches where having bellows driven volume sounds out of place or add a simple thumb joystick and map it's position. But the polyphonic aftertouch is way, way more demanding, since you have to use analogue readout for each button.

Not sounding or late sounding notes probably have badly voiced reeds. That is they have the gap between the tongue and the frame either too big or too small. You can adjust it yourself - you will need a reed hook for push reeds, since reeds on Elise are waxed to permanently mounted reedblocks, but the process is relatively easy. As to the bellows „workout” - instead of air button press all the buttons with your palms and work it out this way. Less resistance, less strain on your arms, so more and faster extensions possible. If there is significant spring back from the fully closed position, you can also try to press the bellows better, but this requires disassembly and some tinkering with woodworking clamps.

Pianoteq, Organteq or VCVrack, all rely on manual MIDI mapping. There are some basic presets for those Moddart apps, but nothing hardcoded. In VCV you have to wire everything anyway. I'll map preset based on official MIDI documentation though.

Ah, you mean a sort of a „modifer keys”? Press a natural note and a modifier to make it sharp/flat, so a fully chromatic octave only needs 8 keys instead of 12? You answered yourself - this is a melody only solution. And „joystick” method actually needs more than 88 inputs, since you would have 3 inputs per note. However, this can be solved mechanically with rocker switches and the physical layout could indeed be reduced to just 7 keys per octave (which operate on 21 switches). So good news here - since what I intend will support button repeats, this will be supported, but it’s up to the user to come up with suitable physical instrument.

I only work with software MIDI via USB/BT, so I just use custom mappings. At this point I don’t intend to include classic MIDI connectivity due to limited pins on Arduino, but I may add it later if I decide to switch to Mega/Teensy. Regardless, if I manage to add display and encoder, then all switches/controls will be customizable onboard. Rewriting code for different boards is not really a problem once the fundamental version is done. It pretty much boils down only to changing pin numbers and rewriting MIDI calls for different library (although the one I use currently seems to be available on all platforms). The important core of the code is pretty much platform agnostic, since all those platforms use C/C++ syntax.

You mean a single octave up/down transposition shortcut, or a switch to make each button simultaneously playing two notes an octave (or another interval) apart?

I don’t understand this. Could you please elaborate?

Recently I started exploring keyboard layouts beyond concertinas to build virtual organs around. I tried classic piano (I still hate it), then converted one to Janko (much better, but octave span is still a problem), then bought 70 years old CBA (and fell in love with the B-system layout)... and down the rabbit hole I got fed up, that there are no viable MIDI options for basically anything other than classic piano. They are either flawed, non-existent, no longer in production or simply stupidly expensive. This includes of course MIDI concertinas of any flavour. So I decided to dust my old MIDI project. Right out of the gate I was very happy to discover, that some current iterations of Arduino support USB MIDI out of the box, and that one of them is physically compatible with my old circuitry. Software however needs a rewrite, since the old version relied on a software driver and there are off-the shelf libraries available now. So, a question to the concertina hive mind: What features would you personally want in a MIDI box? What controls beyond buttons and pressure sensor would you find use for? To clarify, this project is only for MIDI controller, not a standalone digital concertina. No onboard synthesizer, no speakers, USB (and perhaps BL) MIDI only (no MIDI connectors), and my question is not about physical instrument this will go into. The main goal is to provide a flexible, open platform built around off-the-shelf circuitry and components (so no custom PCBs), that can be used to build MIDI free reed instruments, either from scratch or as conversions. Some things that are currently implemented: Hardware: - basic keyboard with 64 buttons using multiplexers - 2x pressure sensor - four bi-stable utility switches - one potentiometer Software - basic MIDI communication for the hardware above - button mapping customisable on the fly (and persistent), without the need of hardcoding it. Currently this procedure requires another MIDI keyboard as an input. - four selectable pressure response curves - unisonoric and bisonoric modes (this is old code and IIRC bisonoric mode had some issues; not yet tested with the current setup) - piano mode (pressure controls velocity instead of volume) - transposition knob - since there are two pressure sensor inputs (initially I could only source uni-directional sensors), the other one is used as a wind input for additional MIDI CC My current "to do" and "this might be an interesting idea" list: - expand basic keyboard capability to 2x64 notes with independent channel setting via I2C bus - drop multiplexers in favour of full matrix, so that links can be used freely (for expanding Hayden layout sideways into enharmonic duplicates and to accomodate CBA duplicate rows) - implement keyboard matrices in a way, that disconnects them from physical layout of switches (intermediary PCB) to allow any sort of physical layout - expand bi-stable switches pool to accommodate organ and accordion registration requirements (stops, combinations etc) for multiple manuals - add rotary encoder and display so that I can add an "onboard" button mapping procedure and allow for menu-driven settings and wider customizability of available controllers - add one or two thumb joysticks or other suitable controls for more MIDI CC inputs - add drums mode and perhaps sequencing/looping capability for finger-drumming backing track (this one only if I decide to upgrade to more powerful board later on) - add accordion standard basses modes - add expression/sustain pedal connectivity - add "strum" mode, where pressed notes are arpeggiated up/down/fast/slow depending on pressure direction and value [obviously, not everything on this list concerns MIDI concertinas :D] Currently I work around Arduino Leonardo, as this was a straightforward modernisation of my old circuitry, but it is a rather limited board, so an upgrade to Mega or even Teensy 4.1 at some point may happen... As a side goal, I'm also testing some ideas for 3d printed switch suitable for free reed MIDI, with a form factor that will fit Hayden spacing requirements and durable enough for musical instrument. If it'll work, conversions of any kind of concertinas will be possible without the need of lever action. But that may prove to be a quest for a holly Grail...

I don't know, if he's a proffesional, but he is damn good at playinng English concertina with fan-like bellows technique. Additionally, I believe he uses a tape to bind the bottom edges together. Why? Because if you look closely, this way he can play with eight fingers instead of just six, or even only four in case of the Wheatsone intended technique. So, as you can see, this absolutely makes a ton of sense. But I agree, that this is a dumb idea for an Anglo player to do so.

Jackie uses accordion style bellows, that is inherently springy. It will ease a bit with time, but some of the springiness will always be present. On an English you are not restricted by note availability on push/pull, so just use shorter bellows phrasing. If you look at some top players out there, many of them use very short, fan like bellows articulation. Pretty much any higher tier concertina you’ll upgrade to in the future will have proper concertina bellows, which has little springiness, and the top ones have pretty much linear response up to nearly full extension.

About harmonikas.cz reeds. The lower the class of the reed, the less air efficient and responsive reeds are - in terms of concertina building, this translates the most to how long single bellows direction phrases you will be able to play. So if you go for mechanika reeds, make the bellows with more folds. They also have less rich sound due to softer tongue steel used and have lesser tuning stability. Mechanika class is perfectly fine for a first time build, as you won't have to worry about destroying a top quality reed worth a quarter of the full mechanika set However, given that Tipo A Mano are only 50 euro for the set, I would highly recommend going straight for this class, as you don't really need classic accordion reeds of higher quality in a concertina. DIX reeds are a different breed and you pay premium for the brass frame/shoe sound - way more like traditional concertinas (or bandoneon if you go for the zinc plate option). They are pricy, but worth every cent.

Regarding reeds for an Anglo, it will border on a miracle to source exactly what is needed from anything other than an Anglo. You not only need specific pairs of notes, you also need coherently sounding set, so mix-matching from different boxes will most probably result in quite chaotic timbre, response and volume. The easiest option would be to cut accordion reeds from two dry tuned M voices of one accordion to singe tongue reeds and build with that. Be aware, that building your own instrument from scratch is vastly different than restoring an old one. There are two layers of free reed inatrument construction. First one you can easily see, understand, copy or design - the material/mechanical layer. It is not harder than any other wood-, metal- and leather- working in similar scale and with Anglos it is also relatively not very time consuming. But… You are pretty much guaranteed to build a poor sounding, unbalanced box the first time you do it, because understanding of the invisible aspect of free reed instruments - the acoustics, can only be learned by experience. My advice here is - build slightly oversized endboxes and bellows, and design the interior in a way that allows repeated rebuilding and refitting of the reedpan. Preferably also re-designing and refitting the fretwork. This will allow you to re-iterate your design and fix inevitable sound-impacting mistakes/bad choices without the need to build multiple whole instruments.

I suspected, that it'll be the answer. Rochelle uses accordion style bellows, which is a folded carboard that has innate springiness to it. It will get a bit less springy with time, but it will never be a linear, no bounce-back experience of traditional concertinna bellows. But, as I wrote earlier - your brain will adapt to this.

What concertina do you have? Because that sounds like you have an accordion style bellows, which is seriously tight.

Two separate questions are being conflated to one in some of the answers above: „what is the useful language to describe the timbre” and „how and by what the actual timbre is created”. The second one is quite „easy”, with the answer being „by everything”:D The first one is a trickier one, because we don’t really perceive the timbre objectively. Not only from person to person, but also between pitches, even within a single instrument. I don’t have absolute pitch hearing. Feed me pure sines and I can only tell you a general octave. Give me some time with a specific instrument, and I’ll tell you the notes, because with enough practice I can recognise the individual character of each note. So, when describing the timbre, IMHO the best way is not to try to take the timbre apart into small pieces like bright, mellow, muddy, round etc… because those are useful only in a relative context: this specific concertina has a mellower tone, than that specific concertina, etc, because you’re merely labeling some audibly different qualities to another person who can also clearly hear them. You might as well say „this concertina’s timbre is more red and the other one has a colder tone” and you will be understood. A better approach is the same, that is used to describe accordion registers: by invoking the general impression of another instrument. I’m currently restoring a 70 years old accordion. It has three voices, two of which have reeds in brass frames, one has aluminum frames. The differences are clearly audible between them - one invokes memory of an organ flute stop in the upper range and leans toward a bassoon in the lower range, and the second one clearly sounds trumpet-like. That is until my wife takes out her trumpet, then it sounds nothing like a trumpet. Same goes for the other voice. I have a virtual organ set up around Organteq software. Flute and bassoon stops are not even the same family of stops and both sound completely different than my accordion. One last example: as with any accordion, I can also mix voices. A mix of the basoon one with a third one, also a „trumpet-like brass” but detuned by 20 cents, commonly named „sax”, has a timbre that can be best described as a creepy circus, and I can bet most of you have now a very adequate impression of this timbre in your heads

The asymmetry of the force you work the bellows is normal, as is the change of the force as the bellows extends towards it’s limit. The bellows will get softer a bit with time. However, it sounds like you are pulling and pushing too hard. Reeds need unintuitively small amount of airflow and pressure to speak. You will learn to adjust to this unconsiously by ear, and the best excercise is to play as quiet as you can manage. Preferably a slow tune played legato, so you have to manage the air perfectly to fit the phrasing.

This happens mostly at the early stages of learning how to play, when you are generally less conscious of what exactly happens. My advice is to play offending passages a lot, lot slower and repeatedly in the proper neighbouring context until you no longer make this mistake.

Looks like he has simply looped and sewn the end of the handstrap to create a fixed diameter loop. It would be easier to suggest a solution if you can provide a picture of your handrests.

What you're really asking is not which side to support, but which to immobilise. Duets are best played seated and fully supported, bandoneon style, with straps as loose, as you can manage to still controll the instrument properly. Take a look at Didie Sendra (https://www.youtube.com/@Soloduet0703), IMHO the best Hayden player out there. He started with only the right side supported, then experimented with center support and fan-like bellows articulation, but nowadays he supports both ends, with the preference of lifting the left side if the need arises (articulation, fully closing the bellows, etc...). Aidas Rusa (https://www.youtube.com/watch?v=Mb-C_sZFnsQ) also supports his box on both ends, with his right side being fully fixed. I on the other hand (https://www.youtube.com/@desnou/videos), play with the left side fixed, as it makes oom-pahs easier, but rarely ever lift the right side, only for extreme extension or bellows effects like vibrato etc. When I do lift it, the direction of articulation (with the gravity or against the gravity) depends on the side of the keyboard the passage is played, that is, is it in flat keys or sharp keys or the center of the keyboard - whichever way is the most comfortable to play. When I played Elise I also experimented with the neck and shoulder strap (as seen in my two Elise videos). Currently I use a completely custom handling system, that has no strap at all. With the Elise, you can experiment with the thumb strap, just set the straps loosely and put your thumb in the small loop. I found this to be a superior way to handle this instrument, as it allows for way easier use of pinky. You can also see this in my Elise videos. George from "George plays music" YT channel (https://www.youtube.com/watch?v=d_4Bfk1ere4) [edit: the rest of this sentence somehow vanished...: "...also modified his handrests with thumb spacer/holder for easier control." One last tip - remember, that your unused fingers can assist with the bellows movement and stability. Don't be affraid to put fingers down onto the endplate when necessary/convenient. You'll be able to use even looser strap with this kind of finger support.

And I thought I was imagining thing when I recognised some of the characteristic phrasing and expression styles of his.

There are two main reasons for adjusting padhole size: volume adjustment and pitch stability. The problem starts around F3 and increases significantly below C3, when reed size starts to grow rapidly. Large reeds need a very stable and sufficient airflow, otherwise they will be very prone to pitch bend when overdrawing the bellows. This behaviour can be used deliberately, but above certain level of fragility of pitch it becomes impossible to use those low notes, as their behaviour becomes too erratic. Just few days ago I learned, that some people exploit this trait for bassoon register on accordions, by altering how much this register opens up. Now the flipside is, that the bigger the hole, the louder the sound. My big box is balanced for 80dB, which is low for a concertina (CC Elise peaks at more than 100dB when overdrawn, 90-100dB at normal pressure. So for the lower reeds a balance between stability of tone and volume of the lower reeds had to be established carefully, so they don't overpower everything else (F2 and G2 especially).

Either I don't understand your answer, or you didn't understand the question. You do experiment with those parameters every time you design a new box. Padhole aperture vary within a single instrument for the reasons I mentioned above. Perhaps you're thinking too big? We're talking about just milimeter differences in aperture. In my big box I actually made my bass reeds' holes too small at first, then re-bore them too big, so had to print inserts to reduce them again. I also had to remake some levers and even move one reed further away (which required reshuffling two others), because the lever was to short to lift the now bigger pad high enough to achieve full airflow and sound projection.

See my post above - there are many parameters affected by padhole diameter and lift, all at once, differently for different pitch ranges.

That’s why I wrote „lowest tension that still keeps the pads closed under normal push force”. I can get away with lower tension, because my lever-button joint being completely unrestricted does not add to action resistance, and large bellows cross section lowers the pressure. About „higher tension makes closing of the pads faster”. That is true, but at the same time it increases necessary muscle tension and fatigue, resulting in slower play. For me, buttons should resist only so much, that you can „prime” the note. After switching to my big box I had quite long adjustment period, because Elise has 3x higher button force and I was constantly triggering notes too early. I’m also the „lifting finger to play” kind of player, so buttons should activate on simply releasing the extension muscles of the fingers. Regarding pad lift, diameter, airflow and sound projection. This is tricky and gets even trickier for lower reeds, especially on „singing accompaniment”, lower volume instruments. Action chamber height affects tibre and volume, so there may be limitations in how high your lift can be, padhole aperture affects both volume and pitch stability, and you have to balance relative volume of low, mid and high reeds. As a result, it’s a puzzle of compromises, and action’s speed is just one of the possible goals here.