That sounds interesting. Talking of dimples, I imagine covering the entire interior of the mpc with small 2mm diameter dimples like the surface of a golf ball. I wonder if it would behave like a large chamber because of its volume, or a smaller chamber (dimples on golf balls actually reduce surface resistance), or otherwise...?

But how about the accuracy of such 3D printing? If it cannot guarantee accurate facing and rails, the "printout" will need some finishing and that could introduce enough error that throws the entire test results away.

 

That sounds interesting. Talking of dimples, I imagine covering the entire interior of the mpc with small 2mm diameter dimples like the surface of a golf ball. I wonder if it would behave like a large chamber because of its volume, or a smaller chamber (dimples on golf balls actually reduce surface resistance), or otherwise...?

But how about the accuracy of such 3D printing? If it cannot guarantee accurate facing and rails, the "printout" will need some finishing and that could introduce enough error that throws the entire test results away.

interesting thought, I believe the dimples do reduce resistance but also enlarges the surface area..

the accuracy is .25 mm that means that "sharp" edges will have a fillet with a radius of 0.0049213 inches, and the overall texture will be quite rough on the feel
it`s comparable to the picture below.

I`ll als the workshop manager what finishing options they have dont know for shure if whe have that possibility.

on the other hand I could draw the mpc a bit bigger for some hand or machine surfacing..

 

Do you have a way to image a mouthpiece and export a cloud array of points? If so, you could start with a real mouthpiece of known character and modify it to learn what effect it has.

 

@Dr G--

I have the possibility to do a 3d measurement.. that allmost follow the procedure you are prescribing only you are drawing in 3d as you are measuring..

good idea! I`ll ask the guy who manages the workshop today!

 

0.25mm is quite large an error for a mpc facing...but at least you can draw a perfect radial facing curve for start.

On the other hand...can we make a mpc with different beak shape.. eg. more "lip -friendly" / polygonal / cylindrical..? diagonal mpc? ....just some thoughts.

 

It can certainly be done, and the only limit is your imagination and drawing ability. I used 3d printing to prototype most stages of my new tenor mouthpiece. At .25mm resolution you can probably get a playable facing printed. I tried a couple of different machines, SLS (laser sintering) with plastic was marginally better than FDM (fused deposition). IIRC the Z increment on both was .1mm which I didn't expect to produce a playable facing but it did. I haven't found any plastic in these systems suitable for production (i.e., strong enough, pretty enough, and cheap enough), but for prototyping I found it useful for a while.

This is a perfect application for this technology IMO. Sax mouthpieces are awkward shapes to machine (especially large chambered ones) and so are great candidates for using this kind of manufacturing. You can make shapes that are difficult or even impossible to machine.

I'm having trouble visualizing what chamber shape you mean by 3 5mm dimples, but if you can draw it and have access to the printer, might as well try it. Good luck with your experiments.

 

It can certainly be done, and the only limit is your imagination and drawing ability. I used 3d printing to prototype most stages of my new tenor mouthpiece. At .25mm resolution you can probably get a playable facing printed. I tried a couple of different machines, SLS (laser sintering) with plastic was marginally better than FDM (fused deposition). IIRC the Z increment on both was .1mm which I didn't expect to produce a playable facing but it did. I haven't found any plastic in these systems suitable for production (i.e., strong enough, pretty enough, and cheap enough), but for prototyping I found it useful for a while.

This is a perfect application for this technology IMO. Sax mouthpieces are awkward shapes to machine (especially large chambered ones) and so are great candidates for using this kind of manufacturing. You can make shapes that are difficult or even impossible to machine.

I'm having trouble visualizing what chamber shape you mean by 3 5mm dimples, but if you can draw it and have access to the printer, might as well try it. Good luck with your experiments.

thank you for your comment mrfy. With dimples i mean; in stead of making the round chamber part, as in your picture, the chamber is only bigger than the bore on 3 places which you could indeed see as 3 larger golfball dimples.. sorry I dont speak/type english fluently enough to explain better, so i drew an example. these are oval dimples they were fastest and easiest to draw..

 

The whole subject is very interesting and keep us posted with the results.

 

I will! havent you got any unconvential / unmachineable ideas lurking around in the back of your mind? (that you are willing to share offcourse)

 

This sounds like a lot of fun. Let's hope it produces some amazing new gear.

 

It is fairly common to take inspiration from aerodynamic designs for mouthpieces. But acoustic science is different from aerodynamics. Putting dimples in a mouthpiece makes as much sense as putting a high baffle on a golf ball. But some baffle designs were inspired by the shape of airplane wings. They can work quite well in spite of the different physics at work.

 

I`ll see if i can find some aerodynamics books, I only had fundamental aerodynamics and fluid dynamics but it`s a start

 

We have use here at our museum, some new 3D scanner to use on small archaoelogical artifact. ypu dont need to put point on the surface like the old technology. We use bigger Leica scanner for the in situ remain for 5 years now. This technology use cloud array of point. they even compatible with SketchUp.

 

More details, please, regarding your scanner(s), Avan. My NextEngine isn't doing the job I need - specifically, scanning reflective surfaces to high resolution.

 

Dr G: How well is your next engine scanner working on the inside of the chamber and shank on your mpc scans? If you have been successful capturing those areas, do you have a specific angle that you are using to place the mpc? Do you mount it on the table? How are you holding it in place as the table turns?
As far as reflective surfaces: at Target you can buy Halloween face paint (tempera paint pens) and with careful brush strokes, you can spread out the washable paint fairly evenly. It probably adds .002-.004 thickness, but captures a much better scan! You can also buy at Target, "Bright hair spray" which is a washable colored hair spray. The colored hair spray paint tends to run more. Black color works best with the next engine scans.

 

We are experimenting with the small hand scanner. We also have some difficulty with high reflective surface with the small scan. The resul I see are good (good enough for a mouthpiece) but some intricated detail where lost. it's like in digital photography. it's hard to keep detail in the white or highlight zone. This is the company we are working with : http://www.graphsynergie.com/, talk to Richard Lapointe for more technical stuff he is THE guy to talk about scanning and cloud, and it is very friendly: rlapointe@graphsynergie.com The result with the big scan are impressive. We can see and mesure the tickness of a crack deep in a stone wall, so it help for the restauration and the survey of the site.

 

one of the greatest potentials of this technology is that any mpc could be scanned and then enlargened for solving the problem of 10 to 15% reduction of casting. A good florida replica with OEM link ligatures compatibility and chamber size, a slant cast in delrin, possibilities are virtually infinite!

 

thats true, only I dont have those scanning technologies at hand, only thing that comes close is on this site http://sax.mpostma.nl/NL/ He is a Dutch refacer and mouthpiece maker.
he has a few pics like this on his site and some X/ray pics

 

^^ that's what I'm talking about!

 

Those x-ray picture are scan, this is one of the render, you can have cloud of point also. If you take a picture you can map it around.

 

I will! havent you got any unconvential / unmachineable ideas lurking around in the back of your mind? (that you are willing to share offcourse)

Of course I have some, but I believe the greatest value for a mouthpiece designer would be the ease of experiment with the 3D printers, especially in few years when they develop them further.

 

interesting stuff. why not put it to the test. scan your own mpc, print it and see how it plays, you can do a one on one comparison with the original. I'm curious of what you'll find.
Suppose it really works fine and becomes as common as a PC in a couple of years, we can all copy our favourite precious florida's , Gaia's ,slants , early babitts, Dukoff stubby's ,Morgans , offset Bergs etc.
I'm afraid it might raise some legal questions though. would it be the end of the high priced mpc ?

 

would it be the end of the high priced mpc ?

I`m afraid not, the manufacturers of 3D printers make the most money on the plastic that is necessary for printing.. The machine I`m going to print with costs around 12000 dollar and one roll of plastic thread costs 400 dollar.. dont know the volume of the plastic thread roll though but not a whole lot.

I`m a student so this "assignment" gives me one freebee, if I want to make more of such projects in the future I'm afraid I`ll have to pay up.

Still cheaper than a company because I dont have to pay for the "hours" programming, printing and after treatment and machine costs + overhead..

I`m Using the FDM method (Fused Deposition Modelling). this will work for these protoypes but there are better technologies available for better surface quality.. but these are a Factor 3 to 5 more expensive..

I`m afraid it will take quite some time before we are sitting at home thinking; hmm what mouthpiece should I play today.. Ahh A nice Florida Link should do: CTRL+P and wait.. (Like the idea though)

 

Have you guys seen this yet? I'll be curious to see how well this HP 3D scanner will work.
http://www.hp.com/united-states/campaigns/topshot/index.html

 

It does not look like a 3D scanner to me. It just takes pictures from different angles without moving the object.

 

Koen
I would keep the changes simple.
1) sample changes in the outside diameter of the mpc prints.
2) Change the diameter of the chamber area (slightly larger then the shank inside diameter)
3) See how accurate the facing curves are when you change the tip opening of the printed models.
4) Change around the table angle, in relation to the tip.
5) change around the height of the floor area.

I've messed with adding dimples on a metal mpc. It didn't work too well. It seemed to slow down or disrupt the air flow.

 

I would also use the high dollar plastic material if possible. Something with higher temperature tolerances.

 

Can you draw something like a step baffle but instead of having the one really big step baffle do lots of small step baffles like:
#### \
##### \_
###### \|
####### \
######## \_
######### \|
########## \
########### \_
############ \|
if you orientate that 90 degrees clockwise and ignore the hashes

 

or maybe try a square chamber that morphs into a tri angular chamber then into a circular chamber With a straight baffle that has checkered waves going on. Basically the baffle is straight then you make a chess board design over it and every white square you hollow out a few mm at the middle and continue hollowing the square outward like a golfball dimple.
Or maybe even just rounding off the edges that the stream of air follows so it is like \_/ for a cross section of each one but with it rounded off at the top.

 

One thing to keep in mind---different 3D printers use different materials. Many of them use a ABS compound, which is fairly inert and doesn't pose much of a health risk. But there are some methods of 3D printing that do use some plastics that really aren't very good for you....particularly if it's something you put in your mouth

 

It`s ABS or PC both are safe, i checked it allready, but thanks for your top anyways.

 

"Scanning chambers -- you can make a mold, scan that and then paste it into the scan of the body."
Genius! Next time I'm in Leicestershire, I'm going to buy you a beer!

 

you are all talking about chambers but i`m having the greatest difficulty to reproduce an accurate facing of an Otto Link stm. I'm trying to build a mathematical equasion on these measure points so I can describe it most accurate.

And "just" adding dimples isn`t all to it, A golfball has a certain pattern with different sizes of dimples which contribute to height of the turbulant layer of air, on which laminar air can flow more easily thus making a greater flow between mpc and body..

 

you are all talking about chambers but i`m having the greatest difficulty to reproduce an accurate facing of an Otto Link stm. I'm trying to build a mathematical equasion on these measure points so I can describe it most accurate.

You can draw splines in your CAD program, right? Why not measure a good Link with a glass gauge, then use those data as control points.

 

The equation you seek is, according to Theo Wanne, Ax-squared + Bx + C. The kicker is what A, B and C are. See here: http://www.theowanne.com/resources/ringFacingCurve.php

You might also like a radial curve - use a length and tip opening to calculate the radius of a circle through those points, then use the formula X = SQRT(2*YR - Y-squared) to determine the points on the gauge where your feelers go (X is the gauge measurement, Y is the thickness of the feeler). Mouthpieces with this curve play very smoothly.

If you are using a spreadsheet to try to find the curve formula, you may get better results if you swap X values - for example, if your 0.0015 inch feeler is at, say, 48 (or 24) on the gauge, setting that to 0, and setting the tip opening to 48 (or 24) might give you better results. Spreadsheets do better with curves that start at 0,0...

 

I`m using Matlab ,which is a matrix based program, and i`m going for the theory of a quadratic facing curve but not just ax² +bx +c but which is a quadratic equasion but i was thinking more of an 3rd or 4th or n-th if you will..

the problem is that i dont just want to make "a facing" i want the exact facing, so i can be sure there are no other elements shape en dimension wise so i have one control model (standard STM) and one with a modified chamber..