Jump to content

Łukasz Martynowicz

Members
  • Posts

    489
  • Joined

  • Last visited

Posts posted by Łukasz Martynowicz

  1. 28 minutes ago, SIMON GABRIELOW said:

    I am interested to hear 3D printing is used in concertina making; but I hope that the skills of hand and eye, and of making by natural instinct, and the joy of creating from scratch will not be forgotten too!

     

    Don't worry, as useful as 3d printing is, you can't print the classic aesthetics of concertinas and it comes with it's own can of worms. But it is the technology for "Ford model T" of concertina world for sure.

  2. 2 hours ago, Jake Middleton-Metcalfe said:

    I had the chance recently to compare two instruments of the same maker, same design and same materials except the ends were wooden on one and metal on the other. Honestly you would be hard pressed to tell the difference in sound. This sort of flies in the face of what I took to be true but that is what I noticed, it is very very rare to have two instruments identical in every aspect apart from the material of the ends in the same place but I can guarantee these instruments were both identical apart from that. At a guess I would have said that the amount of gap in the fretwork is more important than the material of the fretwork but that is a guess and has not been scientifically studied in any way.

     

    I have made a recording I will upload it at some point and you can make up your own mind.

     

    When I was experimenting with baffles on my big Hayden, it didn't matter at all if the baffle was from hard plastic or way softer wooden filament, despite those two materials having vastly different sound properties. The only variable that mattered was the geometry of the baffle. 

  3. 27 minutes ago, Dashy said:

    I have! I want to modify the model a little to be able to mount it to the pegboard before I really start playing around with it, so I stuck with the regular handrests for working with the mockup so far. After I test it a little I might try and recreate the functional geometry in CADQuery to parametrize it for more experiments.

     

    Out of curiosity, what did you use to make it? That swept geometry is super flowy!

     

    Blender, I'm a graphic designer by trade, so that was the most straightforward approach for me. It is awfully backwards for CAD applications due to poor support of direct measurements, but then subdivision modeling is pretty powerful.

  4. 40 minutes ago, Dashy said:

    Alright, progress!

     

    So, working off of the range I'm targeting (E2-B4/G3-E6) I removed all the sharps and flats and left only the white keys in that range, then went through a collection of songs I would like to be able to play and added the keys I needed (or preferred) to mime them on my mockup. I came up with a 63-key minimal layout:

    image.png.9ff561d9ae10d6d7afd758629142b3ee.png

     

    Kind'a weird how that F#4 on the left never came up... but either way, this is the minimum for comfortably playing my set of test songs. There are a couple keys I could drop and still play the songs, but I would need some weird stretching or fingering shifts that didn't really come naturally.

     

    Now, this layout isn't quite chromatic, so I filled in the missing notes like so:

    image.png.775ea98de4378fc539b2142f441d5d3a.png

     

    D#6 on the right and Bb4 on the left can be a little bit of a stretch to reach, but my reasoning is that since they didn't come up in my test songs, thus probably won't come up as frequently as the other keys, they're less important to optimize for --it's important to me to have them, but I don't need to compromise anything else to have them in a better spot.

     

    This layout has one fewer left-hand reed and four more right-hand reeds than the 65K @Pistachio Dreamer showed earlier in this thread, which given what I've learned about concertina size constraints sounds like a good L/R balance... to me, at least.

     

    Here's the current mockup:

    20220130_142816.thumb.jpg.3822e2f783caa731655d6be3c3eb49ce.jpg20220130_142801.thumb.jpg.f98631ef9ab7c347d71eca8ae82d1df3.jpg

     

    I printed out another pegboard and some peg-mounted handrests, then stuck the two pegboards together with two pegs to get a very thin concertina. Black and white correspond to the colors of a piano keyboard, and blue is the extra notes I used to fill in the minimal layout.

     

    So now I have a keyboard layout that I arrived at systematically and physically tested with a variety of songs. Feeling pretty good about this right now!

     

    Side note, the more I play with the Hayden keyboard the more I appreciate it. It's far more intuitive than it has any right to be.

     

    Have you looked at the antlers yet?

  5. I don't think that the choice of wood would have a big enough impact to decide on the difference between those instruments when you also change dimensions and thus chamber geometry and fretwork geometry.

     

    Couple of months ago I have printed a light, travel duet. It was a "fast and dirty" design, engineered in two weeks to last me two weeks of vacation only. A thing to know about 3d printed concertinas is that they are as light as it gets, because they are mostly made of air - prints are hollow inside, only 15% of total volume is material, and the material itself is very rigid, so good at reflecting the sound. Without the endplates mounted, there was virtually no difference between the reed when on a tuning bench and when mounted in the box. The whole sound shaping was done through the geometry of the fretwork and varied between very dry and sharp/bright with completely open fretwork to deeper but muted tone with full cassotto endplates. I also experimented extensively with baffle geometry on my big Hayden as the initial sound with my very open fretwork was ear piercingly sharp. No matter what I did in both those boxes I couldn't change it enough to overcome the fundamental difference of the reed frames material - the big one has brass plates, the travel one has aluminum plates, and this difference is audible on the tuning bench and stays nearly the same up until fretwork design stage. What is even more important is that even 1mm of chamber depth can have way more influence on the final tone than the material of the box does.
     

    • Like 1
  6. I play duet, but I had the same problem - RH melody alone was easy, LH oom-pahs alone were easy, playing octaves was easy, but playing both hands together with non-symmetrical hands movements was impossible. My trick was to stop treating hands separately - instead learn to play very slowly with both hands together, so that your muscle memory encompasses hand-to-hand-to-bellows interactions and timings and then gradually increase speed only when you are not making mistakes at a current speed.

  7. 36 minutes ago, Dashy said:

      

    I was actually thinking more about moving a couple buttons to the flat side (or vice versa), I don't really want to change the spacing or button size (buttons are 6mm diameter, for the record).

     

    I think I put the button field and the handrest a full row too close together, and laterally I think the button field is a whole column too far in the thumb direction, at least on the left. The handrest is also way too shallow.

    Trying to mime a few tunes, it's not too bad -it's just a bit of a stretch between the left and right extremes. I think I'll design an adjustable mockup to play around with now, but it's looking good so far!

     

    I can see how the antler grip helps here -keeping my wrist stationary it can be quite uncomfortable to reach some buttons, but keeping my thumb stationary I can reach the far right with my index finger and the far left with my pinky!

     

    The changes I wanna try now for the next mockup: Move Ab2 and Eb3 to G#2 and D#3 on the left, and move A#5 to Bb5 on the right. That will make it narrower, so a little more compact.

     

    I've seen "Inventor" around here, but I must admit I didn't know those numbers were from him! I got them from the spec sheet given to me by @Pistachio Dreamer.

     

    Ach, I misunderstood.

    My current array goes from Bbs/Ebs to D#/G#, but with antlers I could extend it to span from Abs/Ebs to A#s/Fs without sacrificing the ability to finger all available triads with same fingers efficiently and add two additional rows on each side, giving me the maximum reachable layout going from C2 up to E7.

  8. 4 minutes ago, Dashy said:

    Finished the button generation and handrest, the script is set up so you just feed the list of notes and dimensions and stuff to it in a JSON and it builds it! The alignment is a little wonky because I don't have those dimensions available (and the plate the buttons are on is a just a big rectangle because the version of CADQuery I'm stuck with right now doesn't have 2D convex hull working), but nevertheless I'm really glad I printed these out.

     

    With 9mm vertical spacing and 16mm horizontal spacing these button layouts are a bit wide, I think it should be fine but I do want to see if I can make the button field a little more compact before settling on anything. Probably just gonna print out the button field alone for future mockups just to save filament and time, 'cause now I know I'm fine with the size of the box.

     

    Maybe I should design an adjustable mockup jig so you don't have to print out a whole new thing for each layout, just add and remove buttons on the fly... adjustable handrest would be good too... I wonder if anyone besides me would use it.

    Anyway, pictures of my new mockups!

    20220125_144336.thumb.jpg.34ac5348750ee6066809f6e10b9cbf25.jpg

     

    Closer spacing will in turn require smaller buttons, which in turn make the 'tina less comfortable to play and lever routing even harder to solve than it normally is due to less space between button stems. With the above print I already see a huge problem with lever routing, especially LH sharps side. For me, the smallest comfortable buttons are 5mm diameter, with 6mm ones being significantly better, and at 6.5 mm I start to hit adjacent buttons (on the standard 9/16mm spaced array). This translates to a minimum of 6mm distance between buttons, and less than 8mm gain by squeezing your array to the minimum. Put a hand strap on this mockup, and then try to play some tunes with a lot of high sharps and sharp chords in it to see where you can even reach comfortably and then see if this gain is even enough to fit everything within a playable area.

  9. 6 hours ago, resistor said:

    I'm not going to claim to really understand how your "antlers" system works, but I was curious why you're choosing to immobilize the thumb in a downwards-pointing position? It seems like holding the thumb in a bent position like that for an extended period would be painful. Would it be possible to achieve the same effect with the thumb pointed straight, but still in a thimble?

     

    There is a very good reason for this orientation, but first I want to stress one thing - this whole setup is aimed at relaxing the hand as much as possible, so no, there is no pain involved, at all.

     

    The main concept behind this handle design is based on the same principle as rock climbing chock techniques - because of how the thimble, anvil and handlebar parts of this system interact, the thumb is locked in all planes but one (plus the leeway described below) and in all axis of rotation but two. Because of this, the hand is effectively attached to the concertina in a very "glove like" fashion, with all positioning absolute and all lateral thumb movements directly transferring to the bellows movement. You mount/dismount your hand by a move that does not naturally happen when playing but when locked in the hand simply rests, hanging from the thimble in the same position you hook the thumb on e.g. backpack straps when you want to rest your arms, or on a jacket button etc...

     

    For the handle to act as described above, it must be rigid. Any elasticity reduces the absolute nature of finger positioning. Now the main reason why the thumb has to be oriented this way is that in order to reach various areas on the keyboard in finger configurations required by Hayden chord shapes, the thumb must rotate a bit. The thimble is oriented exactly along this axis of rotation. A rigid handle like this, if made with the thumb straight wants to literally break your thumb as you try to reach extremes of the array and even if it didn't you'll loose the hook-and-lock aspect of the handle. During prototyping this I gradually expanded the concept from a simple, english style thumb "strap" (only made from rigid filament) and then iterated it about 20 times to the final shape you see here. There is a short video on the previous page of this thread when I present how the lock aspect of this handle works and how the movement is transferred onto the bellows, and in the final seconds od the video you can see the unlock move.

     

    Now, the common confusion I got about this thumb position is that it is a 90 degree bend. It isn't - it varies between close and far rows fingering and the thimble is not "snug fit" vertically, you can move the distal part of the thumb enough for all movements required. In the "ready" position of the thimble there are two forces acting, one upward acting near the joint and one downward acting at the tip of the nail. Then when playing those forces vary, both in direction and strength, so there is very little fatigue. After a 45 minute session my hand is more relaxed than after playing the same time with a hand strap setup on my CC Elise. This is mostly because my handle design does the gripping for you.

     

     

  10. 10 minutes ago, dabbler said:

    Very nice linkage solution.  I wonder what the lifespan of the Bowden cables is.  I suppose they could be made replaceable if it is an issue.  I'm also impressed with the finish on the handles.  Is that stained wood-fill PLA?

    -George

     

    Those are PTFE, which from I can gather has an unlimited shelf life. With this curvature I suppose there won't be any serious wear on those. But if any part of the link breaks, it is simply push fit into the board with a little notch cut on the button side so it does not slide back in. To completely replace the whole assembly you just need an exacto knife and a piece of tube for the Bowden and a pair of pliers and a proper gauge spring wire for the actuator. I have actually replaced those a few times when I was working on them, since I made them too short the first time around and the resulting curvature was too tight. It took about 2 minutes to replace them.

     

  11. 29 minutes ago, Don Taylor said:

    Brilliant!  Simple and ingenious, using Bowden cables like this should be taken up by other makers.

    Thanks! It was one of those "how have I not thought about it sooner" moments. I don't mind others taking up on this, as I don't see myself in a concertina building career anytime soon. Maaaybee some 3d printed 46 button Haydens if I ever design a travel box for myself.

  12. 34 minutes ago, Dashy said:

    Another consideration is the mechanical properties of the material. He says he uses PLA, which is generally fine, maybe a bit heavy but plenty stiff -just don't leave it in a hot car!

     

    He uses PLA for the box and reedpan and CF infused filament for all the moving parts.

    • Like 1
  13. 11 minutes ago, Dashy said:

    I actually came across it earlier today, it's really cool!

     

    I am curious about the acoustics, it sounds fine through video but acoustics is very hard. My brother and I actually have experience playing around with different materials and whatnot for some pretty involved 3D-printed headphones, PLA generally doesn't sound very good. Our best success was with PETG, and we haven't tried it yet but from my experience with other prints I think CF-filled materials are even better. Another material that comes to mind is woodfill (generally PLA-based), I'd love to try that with something like a concertina...

     

    Another consideration is the mechanical properties of the material. He says he uses PLA, which is generally fine, maybe a bit heavy but plenty stiff -just don't leave it in a hot car!

     

    As I wrote, Ed spent a whole year meticulously fine tuning the sound of it. Before he started making concertinas, he was an accordion tuner for more than a decade, so you can trust his expertise. But you don't have to rely on trust alone, he got some of the best UK players to "test drive" those. Also, the material of the concertina plays only a secondary role in the acoustics of it, sound reflections are the key.

    Now, on the unrelated topic, I have updated my thread with some pictures of the link mechanism: 

    As to sharing my handle design, my ultimate goal is indeed to popularise this design, so I don't say no, but you will have to be patient just for a little bit longer. At this moment those are calibrated for my hand only and the proper calibration is key to this design. I'll have to come up with some sort of adjustable solution first before I can offer it to the public.

    • Like 1
  14. Now on the second photo above one other feature can be seen - my air pad box. This is why it looks this way - my air lever. Because my "antlers" immobilise the thumb, I needed something different than a traditional air button. This can be operated by any of the four fingers and as such can be operated mid-phrase to close/open bellows faster. Though I only use it in one tune, because this beast have enormous air supply. Which points to a fun fact - with the bellows perfectly airtight, this box can play the lowest note, F2, continuously for 30 seconds. But because DIX reeds with the valves I use are so responsive and the bellows cross section is so large on this box, I actually had to introduce a controlled leak between the bellows and the reedpan to get a proper dynamic range for reeds above C4, because they were pretty much on-off with the slightest of the bellows move.

    The last update for now is my lap support. This ensures, that the LH side hand positioning stays absolute, as it eliminates any wobble around the vertical axis due to eccentricity of forces acting on the endplates on draw and anchores the LH side of the bellows. By the nature of my bellows (it is very rigid in both axis perpendicular to the bellows travel) it also stabilises the RH side, so this whole thing feels much more like playing on a static instrument like a piano rather than a concertina with a hand strap. It is a click-in mount, so there is only a little peg on the concertina itself. 

    air lever.jpg

    lap support.jpg

  15. Hopefully, this is my penultimate post before completion, as the box is fully playable for a few months now and only bellows papers left to do. Playability is in fact the reason why it is not done yet - I prefer playing on her in my precious little time I can spare for concertina.

     

    So, to the point. In another thread I've mentioned, that I've solved the links problem. My first prototype link all those years ago involved traditional levers, but it has proven impossible to route direct levers through the array and routing them around required so many intermediaries, that there was just too much loss of travel and response. And then occurred to me, that I could go through the inside of the box for a direct route and this is how I did it. Those are for Ebs, and if I knew this will be my solution I would arrange a few reeds differently to accommodate Abs also (at this point I could readily make one Ab link, but I don't see the musical point of just one). Bowden setup has all the advantages I can think of - it is smooth, PTFE tube+spring wire have virtually no friction, so no additional spring is needed; button travel is exactly the same, because lever arm length is the sam for both buttons; and with my button-to-lever connection design, the "donor" button does not move when the receiving button is pressed; and last but not least, there are no meaningful button force differences, so linked buttons work pretty much exactly the same as the rest of the array. With careful planning of the reed placement, I could get full range of enharmonic duplicates this way. I guess this is a "to do" for my future full piano range box, somewhere in the next two decades :D

    link bowdens.jpg

    links and air pad box.jpg

    • Like 4
  16. 18 minutes ago, David Barnert said:

    Have you seen this?

     

     

     

    I can wholeheartedly recommend Ed as a maker. I was a close witness to his journey from the early prototype to production models and as an accordion player himself, he is very focused on the sound of his designs as much as he is on the ergonomy of them. He is currently working on duets, Crane to be precise and is open to designing other systems as well, but not one-and-done projects, so it's more probable to get a standard 46 button Hayden from him than a 66 button one.

  17. Welcome to the fascinating world of solving Hayden range puzzle :D

    I went through the same process when deciding on what I want to have on my 66 button instrument all those years ago (self built, currently just bellows papers left to make and it'll be finally done, after 8 years). So you already know that overlap is important, but what you don't know is that reeds below C3 are way larger (about twice the area or even larger, I'm talking about a hybrid, so accordion reeds) than those from C3 upwards and require even larger chambers. You have eight of them, I have "just" five, two of which are long scale (so about 150% the size of normal scale). And by "larger chambers" I mean 150-250% of the volume relative to the reed tongue length than reeds from C3 upwards. My instrument is 8 2/3" and I could only make it about 1/3", maybe 1/2" smaller without sacrificing response and timbre. And I use linked buttons for 4 notes, so I'm talking about "only" 62 reeds... This is why duet concertinas have more notes on the RH side than on the LH side. It is LH that dictates the physical size of the instrument, and then you fill RH size with as many useful notes as you can route the levers to. Which leads to the second most important restriction - the problem of routing levers, which is a can of worms in its own respect. My concertina has handles a bit too much to the back, so the bellows works just a bit to off-center, all this because I wanted it to have a RH range from F3 to E6, so had to fit two rows of high reeds in front of the array.


    Now another problem with theorising the layout - concertina is not a piano, you have to reach all those buttons with your wrist pretty much immobilised if you go with a traditional hand strap design. It took me quite a lot of work to come up with a solution that frees the wrist, but it is highly unorthodox. You can see it here: 

    Another problem with sharps heavy array is that it inherently pinky heavy array on the RH side, which adds to the hand straps problem. So before you decide on the layout make physical mockup, including handles, and check if you are even able to reach all those buttons in practical ways.

    Now last but not least, to my knowledge no modern maker except me (and I'm not a professional) uses linked buttons because it is a non-trivial problem on a concertina. So you can forget about all those Abs, Eds and A#s without adding duplicate reeds and more layer routing/box size problems.

    To sum up - with your layout you are already approaching bandoneon sized instrument (because of bass notes), so you might be better with this: https://bandoneon-maker.com/professional-model-c-b-and-russian-b-system-bandonion/

    Or if you stick to the concertina, with this bass side, you can afford all the reeds you want on the RH side, probably another 10-15, especially if you extend towards "dog whistles".

    Ah, and I do use E6 and D#6 in at least one tune each, so I wouldn't drop them.

  18. On 1/21/2022 at 1:57 PM, hjcjones said:

    If I were starting again, I would seriously consider a duet.  However (not having seriously tried to play one) I wonder how intuitive it is to play?  Forming chords on anglo is very straightforward (in the home keys anyway) even with no knowledge of music theory.  On the other hand, I think of chords as shapes rather than notes, and find them on the anglo by poking about until something sounds right, so I'd probably work it out.

     

     

    On 1/21/2022 at 4:42 PM, David Barnert said:

     

    In a word, very. I can only speak about the Hayden duet system. Thirty-six years ago, Rich Morse put one in my hands and out came a tune. I had previously played many other instruments but no squeezeboxes.

     

    It's something that baffled me since my first contact with an Anglo, and then was further emphasised when I switched to Hayden - the claim that chords on an Anglo are straightforward. Generally speaking, they aren't, only those few in home keys are. On the other hand, chords on a Hayden are pretty much more fundamental to the layout than melody is. If you know the shape of the chord class, then you can play all chords of this class. For vast majority of tunes out there, you only have to know how maj an min chord looks like (other triads are just as simple). This is true as long as you don't have to wrap around the keyboard edge. When I got my CC Elise, I've been able to play all sorts of songs from guitar tabs in just couple of hours. This feature enables working on tunes (and understand music construction) from harmony structure towards the melody line instead of harmonising melodies - this is yet another perk of a Hayden layout, that it actually teaches music theory. Not only because it is isomorphic, but most importantly, it groups black and white piano keys together and is structured around diatonic scales and maj/min music theory. This means, that you can directly see which chords belong to any given key and why, which notes play what role in any key etc. It is all there in the button array itself. 

  19. On 1/13/2022 at 4:49 PM, Stefan said:

    There are two small microphones under my palms, one left one right. They go into a guitar multi effect board that has two separate effect strings. So I can use different effects for the left and right side. 


    Thanks! But I do wonder - how are there no feedback loop problems? Sorry to ask so many questions, but I simply love the electrified concertina sound since I first heard your recordings few years ago, and wanted to try this myself, but I have no real experience outside of acoustic and MIDI environments.

  20. 59 minutes ago, Stefan said:

    Yes Geoff, it´s from 1916 and I am still greatful that you gave it to me. It plays and sounds like a dream. And yes, I take good care of it.

    Do you know where Alex West got it from? He is a member here, maybe I can ask him.

    If Concertinas could talk....

     

    I always wanted to ask you - how is it electrified? Is it a pickup, a microphone, a MIDI pickup? And do I guess correctly, that only a single side is electrified and the other is plain acoustic?

×
×
  • Create New...