Material test on Yamaha P37D

Homepage Forums Melodica construction, repair and maintenance Material test on Yamaha P37D

This topic contains 27 replies, has 4 voices, and was last updated by  Ofir 2 years, 7 months ago.

Viewing 13 posts - 16 through 28 (of 28 total)
  • Author
    Posts
  • #5226

    Lowboy
    Participant

    I think I see another technological hurdle with my design idea. To seal the valve chamber, the linkage between each key and valve would have to be sealed as it enters the valve chamber. That is a lot of seals and complexity and could prevent good key movement.

    But wait! Make the key and valve independent. “Connect” them by way of a rubber diaphragm.

    Lowboy

    #5227

    Lowboy
    Participant

    I have it!

    Use quasi-isometry criterion to make the face of the valves polyhedra in shape, so they impinge on the ellipsoidal shape of the valve tunnel using a congruent, interlocking, but quick-release seal. Simply use procrustes analysis to ensure the rhombial subshapes do not prevent affine transformation of the air flow.

    Lowboy

    #5228

    Alan Brinton
    Participant

    Maybe the reed plates could be designed to slide in and out to make both tuning and replacement easier (a “snap” so to speak).

    Jazzman 1945 has referred to “the principle of the lever” in explaining how to bend notes. Wouldn’t the 1-to-1 relationship actually result in less control?

    #5230

    Lowboy
    Participant

    Can you explain Jazzman’s principle or provide a link to the post? Lowboy

    #5231

    Alan Brinton
    Participant

    I’m not sure how well I can explain this. But think of a seesaw. Depending on where the fulcrum is, more or less movement (or pressure) will be required at one end in relation to the other. If you have a 50 lb. child at each end, the fulcrum should be at the exact center. If your two participants are of unequal weight, then you adjust the fulcrum so that more weight (and movement) is required at one end and less at the other. The key on the melodica is a lever, and the distance you press the key is greater (and the pressure required is less) at the front end than at the back end just over where the key pad opens. So the longer the key, all other things being equal, the finer control you will have over the movement of the key pad and the amount of air released. Consequently, all other things being equal (which of course they usually aren’t), the longer the key, the finer control you will have over the amount of air. In theory at least, it should be easier to bend at the front end of the key and get harder as you slide up toward the back, and it will be impossible to depress the key when you reach the fulcrum. I suspect that even just thinking of the key as a lever makes it easier.

    #5233

    Ofir
    Participant

    Lowboy, I guess that you’re also a mathematician, and you somehow lost me a bit on your solution for the sealing in the over mathematical post.

    To be precise, I’ll use a terminology of “reeds-before-pads” (what’s currently implemented, at least on most of the instruments), as opposed to “pads-before-reeds” (your thesis).

    I see two major shortcomings to pads-before-reeds:
    1. I believe that pads-before-reeds makes the reeds more distant from the musicians lips, hence less control over the reeds. Using the harmonica as an idol design for bending, in the harmonica the reeds are VERY close to the musician lips. So I believe that you’ll lose control rather than gaining more of it.

    2. An air-tight chamber is something that no one would give up on. Pads-before-reeds require an air chamber with moving parts in it, as you need the levers connected to the keyboards to be sealed in a way that air will flow only into the correct reed (via its dedicated tunnel). Moreover, such mechanism will probably result in a pad that blocks air from going into a small chamber (the reed’s tunnel), which is way harder than using pads for blocking air from going out of the chamber (which is generally implemented)

    Sound wise, I don’t think that you gain any benefit from pads-before-reeds. In reeds-before-pads, the sound is mostly extracted from the open pad, which is very close to the reed. In good designs, sound runs roughly 1cm from the reed to the pad, and then goes out condensed so is can be manipulated (if you wish to). In pads-before-reeds, you can make the distance to the reed smaller before manipulation, but it’s not condensed therefore may be harder to manipulate.

    I think that the future is in reeds-before-pads for its clear benefits, but yet sound manipulations can be integrated in several ways. I.e., think of an air dimmer button at the entrance to the air-chamber. Pressing it helps reducing the air flow into the instrument in a controlled way, which may help with bending; think of that as a dedicated bending handle like you have in guitars.

    #5236

    Lowboy
    Participant

    Hi Alan. Okay, I see what you and Jazzman are saying and I agree one hundred percent. My only comment is that the technique of depressing a note ever so slightly to bend it is impractical in my experience. You need such precise control and you must move the key so little that actually executing this technique while playing at speed is very difficult if not impossible. Maybe this technique can work on a very slow song where you are holding down a not for a long time. I also find the bends are very shallow using this technique.

    I combine several techniques together to get deep bends, but that is a topic for another post.

    But now that you have cleared this up, I do question whether a 1-to-1 ratio between the key movement and valve movement would be good. For some reason, it seemed like a good idea to me. It would probably only be a good idea if there was a difference in air flow or sound over the entire range of the valve movement and that valve movement was large. But I suspect as soon as the valve is open 1/8 of an inch, opening further would not make a difference.

    Regards,

    Lowboy

    #5238

    Lowboy
    Participant

    Hi Ofir,

    You seem to keep missing my humor. The following was me having some fun:

    “Use quasi-isometry criterion to make the face of the valves polyhedral in shape, so they impinge on the ellipsoidal shape of the valve tunnel using a congruent, interlocking, but quick-release seal. Simply use procrustes analysis to ensure the rhombial subshapes do not prevent affine transformation of the air flow.”

    Maybe I should have put a smiley face in there.

    In response to your thoughts:

    You said, “I see two major shortcomings to pads-before-reeds: 1. I believe that pads-before-reeds makes the reeds more distant from the musicians lips, hence less control over the reeds. Using the harmonica as an idol design for bending, in the harmonica the reeds are VERY close to the musician lips. So I believe that you’ll lose control rather than gaining more of it.”

    Yes, I noted in a post just few days ago how close the lips of a harmonica player are to the reeds and that this gives them great control over expressiveness. I do believe they bend notes by changing the angle of the air flow over the reeds. On the melodicas I play, in particular the HM-32 with an 8-inch tube, the highest reed is about 26 inches away from my mouth. The lowest reed is about 41 inches away from my mouth. Seems hard to believe, but I just completed the measurements. These long distances result from this model having a long channel that runs the entire length of the melodica before the air even enters into the reed chamber.

    So moving the reeds a couple of more inches from the players mouth should have no appreciable effect.

    Outside of pressure control, I don’t think any melodica design allows the player to influence how the air strikes the reed. The reeds are just too far away from the mouth. On a 37-key Yamaha, I just don’t think you can control how (except for pressure or valve manipulation) the air flows around a reed. The high note reeds must be at least 20 inches away from your mouth on a Yamaha 37-key melodica when using a mouthpiece.

    In regards to your comment: “2. An air-tight chamber is something that no one would give up on. Pads-before-reeds require an air chamber with moving parts in it, as you need the levers connected to the keyboards to be sealed in a way that air will flow only into the correct reed (via its dedicated tunnel).”

    Yes, you must have missed my follow-up post that said: “I think I see another technological hurdle with my design idea. To seal the valve chamber, the linkage between each key and valve would have to be sealed as it enters the valve chamber. That is a lot of seals and complexity and could prevent good key movement. But wait! Make the key and valve independent. “Connect” them by way of a rubber diaphragm.”

    In regards to your comment: “Moreover, such mechanism will probably result in a pad that blocks air from going into a small chamber (the reed’s tunnel), which is way harder than using pads for blocking air from going out of the chamber (which is generally implemented).”

    I don’t see the problem here. That is exactly how it is implemented on a harmonica: small inlet hole.

    “Sound wise, I don’t think that you gain any benefit from pads-before-reeds. In reeds-before-pads, the sound is mostly extracted from the open pad, which is very close to the reed. In good designs, sound runs roughly 1cm from the reed to the pad, and then goes out condensed so is can be manipulated (if you wish to). In pads-before-reeds, you can make the distance to the reed smaller before manipulation, but it’s not condensed therefore may be harder to manipulate.”

    I don’t quite understand what you are saying here. I see a huge advantage–again like a harp–in having the reeds being exposed to the air for further manipulation any way you like. When the reeds are inside the melodica, you are stuck with the sound generated by the design of the melodica. There is little opportunity to modulate the sound unless you have a HM-26/27/32. When the reeds are exposed on the outside of the melodica, you have unlimited opportunity to work with them.

    Regards,

    Lowboy

    #5239

    Ofir
    Participant

    Hi Lowboy,
    Yes, I tend to miss the humor then I get lost with terminology; a smiley will help 🙂

    As for the angle of the airflow hitting the reed, I don’t see how you can change that in either of the design alternatives (p-before-r or r-before-p). This is a benefit you get from BEING the valve.

    As for the rubber and indirect pressure, I read that but I find it non durable as you count on a stretched rubber to maintain its sealing. I believe that this mechanism will break quite fast, which makes me go back to distinguishing between stopping the air from going-into or -out-of the chamber / tunnel.
    Again, in the harmonica, you ARE the valve. Therefore you recreate the sealing yourself as you need.
    Using mechanics for such task is very hard. For the possibly best scenario, I imagine myself a tunnel with a valve inside, where one side you have the small air chamber splitting into the tunnels, and on the other you have the reeds.

    As for the sound manipulation, I’ll try to explain.
    In order to make sound manipulations such as the examples detailed, you must do that after the air has passed the pads and the reeds (the order of things is negligible). Having the reeds last, the sound will go directly from the reeds (let’s say that you don’t need to protect these). Having the pads last, 90% of the sound will go through the pad. I believe that you missed that I wrote that on my original post on material test.
    so either way you have a specific region where you can control the sound using different materials (such as wood, muter, etc.) just by having the cover of the instrument built for that, letting you access to the area.
    I believe that in reeds-before-pads you have a “tube” of sound from the reed to the pad, just like a trumpet (or the closer relative, the Claviola); unlike the trumpet (and like the Claviola), you have a dedicated tube for each note. The fact that the reeds are berried inside the instrument doesn’t mean that the sound behaves like that.

    #5240

    Lowboy
    Participant

    Hi Ofir,

    I agree. Without an overly complex and expensive design/construction model, melodica players will not be able to change the angle of attack of their breadth on the reeds. The reeds are too far away. So we must manipulate the sound after it is made.

    I can think of numerous ways to keep the keys and the valves independent so the valves can reside in an air-tight chamber. Valve design can be very simple: think valve in an overhead-cam, internal combustion engine. You have a valve with stem, spring, and spring keeper.

    Yes, I suppose you can manipulate the sound from each valve in the valve-last model. But there is nothing stopping one from created a cover than has chambers for each of the exposed reeds in a reed-last model either.

    In any case, there are scores of possible designs. Daren will certainly have his hands full developing and testing prototypes that are simple, effective, economical to build, and which provide a high quality sound. The basic overall configuration that provides the most flexibility and expressiveness must be figured out first.

    Regards,

    Lowboy

    #5241

    Lowboy
    Participant

    Hi Alan,

    As I was thinking about the design of the HM-26/27/32s, I realized that in fact the valve control is near the end of the key: about an inch in from the front end of the white keys and right at the very front end of the black keys). The ratio of movement of the key to the movement of the valve approximates 1 to 1.

    Maybe this is why I have trouble bending notes using the “slightly depressed key” technique.

    Regards,

    Tom

    #5242

    Alan Brinton
    Participant

    Fascinating exchange between you guys, Ofir and Lowboy. Maybe someone will try to implement pads-before-reeds and see what actually happens, which occasionally is not what we expect.

    You seem to be morphing into a hat, Tom. Any chance we could get a look at a higher resolution version of your new avatar? It looks promising for the Members’ Artwork page, wherever that is now.

    #5243

    Ofir
    Participant

    Lowboy,
    I was also thinking of an engine-like design, but the big difference is that in engines the control is inside the engine, while in melodicas you must have outside control. This is why no matter what you do, the outside lever must move an inside valve so air will be allowed, and you must connect them.
    Any moving part is subject to many failures, therefore must be considered carefully.

    So basically we agree that reeds last are way more complicate to design (I’m not sure if feasible as relatively low cost and durable design), and should first prove its necessity (which I’m very much not confident of).

Viewing 13 posts - 16 through 28 (of 28 total)

You must be logged in to reply to this topic.