Jump to content


Photo

Leakage Testing

Leakage Test Experiment

  • Please log in to reply
65 replies to this topic

#55 Chris Ghent

Chris Ghent

    Heavyweight Boxer

  • Members
  • PipPipPipPipPip
  • 1055 posts
  • Gender:Male
  • Location:Blue Mountains NSW

Posted 20 June 2014 - 08:34 AM

In order to avoid the "glue hardens the felt" issue I use a thicker felt. It gets hard to the same depth but there is still some felt unglued to act as a shock absorber. Having said that, some Wheatstones have extremely thin pads but are still quiet. They may have softer leather than I use.

Re bedding in, I usually set the springs arbitrarily strong on first assembly and then adjust them to pressure a week later.

#56 Terry McGee

Terry McGee

    Chatty concertinist

  • Members
  • PipPipPipPip
  • 356 posts
  • Gender:Male
  • Location:Malua Bay, NSW Australia

Posted 20 June 2014 - 06:27 PM

With the traditional leather/felt/card composite pads, I'd be interested in experimenting with different types and thicknesses of leather and felt, and different glues.
 

 

I'm happy to try that, probably as a second level, after we've explored the pros and cons of radically different approaches.  I may be limited in access to materials.

 

I bought some commercially made pads and the glue used to hold the layers together was very weak - several of them fell apart before or during installation. I tried gluing them back together with hide glue and found that although it stuck well, it soaked into the felt and made it very stiff, which caused those pads to not seal properly.

 

Yes, I tried the same - gluing a pad back together but using contact cement - and have invented the circular paving slab.  I think this highlights one of the challenges in making traditional pads.  The glue has to be just strong enough to just not let the pad fall apart.  Achieving that balance will take some experimentation.

 

It also tells us something about the subsequent attachment of pad to shaft.  The pad I tried to revive was an air-release button pad.  It had a very short shaft, which was set at an angle a good way from parallel to the board.  Consequently, the pad suffered being dragged across the board and the hole at the instant of closing.  That no doubt was the cause of its failure - too many sideways drags and the three materials finally disassociated.  I would imagine the best angle for a shaft at rest is horizontal? (It won't matter if the pad is non parallel to the hole when open.)

 

I was still thinking about possible materials as I went to bed last night.  Bad idea - I ended up not sleeping soundly!  It did turn up the idea of trying polythene as a pad cover - it has some nice characteristics - airtight, smooth, soft and flat - but raises the issue again - how would you hold it to whatever comes next?

 

Terry


Edited by Terry McGee, 20 June 2014 - 06:51 PM.


#57 Terry McGee

Terry McGee

    Chatty concertinist

  • Members
  • PipPipPipPip
  • 356 posts
  • Gender:Male
  • Location:Malua Bay, NSW Australia

Posted 20 June 2014 - 06:31 PM

I've also read that new traditional pads have a 'bedding in' period (a few days?) over which the seal gradually improves as the leather and felt conform to the shape of the hole.

 

Yes, I've already noticed this - if I take a measurement and then turn the rig off, when I come back later, the leakage has dropped, sometimes very significantly.  So I plan to do a "longitudinal study" to see how long that actually takes.  Probably only on materials that offer good promise initially, but certainly on leather.

 

Terry



#58 Terry McGee

Terry McGee

    Chatty concertinist

  • Members
  • PipPipPipPip
  • 356 posts
  • Gender:Male
  • Location:Malua Bay, NSW Australia

Posted 20 June 2014 - 06:49 PM

In order to avoid the "glue hardens the felt" issue I use a thicker felt. It gets hard to the same depth but there is still some felt unglued to act as a shock absorber. Having said that, some Wheatstones have extremely thin pads but are still quiet. They may have softer leather than I use.

 

It would still be nice to find a way to secure the leather without using glue at all.  Which I guess may be the clarinet or sax pad.  The ultrathin leather is stretched over the felt and card, and glued only to the back of the card.  I'll include such a pad in the tests to tell us how well they work in this application.

 

Re bedding in, I usually set the springs arbitrarily strong on first assembly and then adjust them to pressure a week later.

 

 

Now, you'd imagine that the repeatability of the concertina keying system wouldn't be enough to land the pad in the same place every time.  But I guess that isn't critical - if the pad lands slightly off the hole, the thin crescent of bulging leather cops most of the restoring force, and quickly submits.  And it's not like a flute, where the temporary leakage would render the instrument unplayable.  At worst, there would just be a little extra leakage until the bulge flattens.

 

Terry


Edited by Terry McGee, 20 June 2014 - 06:51 PM.


#59 David Hornett

David Hornett

    Chatty concertinist

  • Members
  • PipPipPipPip
  • 186 posts
  • Gender:Male
  • Location:Hobart

Posted 20 June 2014 - 07:31 PM

Terry,

In your considerations I hope you have paid attention to the gussets. I have made about ten seven fold, and one eight fold bellows over the last couple of years from chrome tanned kangaroo skin (Parker leather www.packerleather.com, QLD: fantastic people to deal with) -- I wondered one day if I could do it then got carried away -- seems to happen to me occassionally. Anyway the bellows work beautifully, tight, no sag, no spring -- sit perfectly flat -- but when one kept the pressure on they would close, and if falling under their own weight would open in about 60 seconds. Pads! ... , it is always pads! Well so I thought. After wearing out my pad press, pads with felts, pads without felts, pads on card, pads on MDF, plastic pads (the leather pealed off, odd that) and pads leathers attached with Quick Grip, feltless straight 1mm soft leather pads (the best for clarity of sound, with no noticeable deterioration in seal) I got smart, Gussets?? Now being lazy I had used .025 - .03 mm skived (Skin on) Parker Kangaroo skin gloving leather -- folds beautifully, much tougher than other leathers, elasticity way up there, so I opened a set of fitted bellows, got some warm soapy water and  .... hundreds of little bubbles covered the gussets. (I could not stare in wonderment too long, water and rabbit skin glue, but I stared enough before running for the paper towelling and the blow heater) The wife says, "Well, if you had asked a doctor, not that you can ever believe I'm one, rather than fooling around in that dungeon of yours, she would have told you leather breathes!!" Terry, leather breathes.

 

Solution: I sprayed the bellows with satin clear. Those bellows do not leak, you could sit on them at the paschal feast if the springs could stand the pressure, and they are 8 fold, lots of places to leak   -- but, other than the semi shine, the ester very slightly forms little ridges not on the gussets, but on the outer leathers due to stretching from opening and shutting -- detracts a little from the cosmetic effect. I have been wondering if shellac would be better. Anyway, I put up with the seepage now, it still takes unsealed bellows a minute to close (eg the bellows on the Jeffries) and that's not too bad in my book. 



#60 Terry McGee

Terry McGee

    Chatty concertinist

  • Members
  • PipPipPipPip
  • 356 posts
  • Gender:Male
  • Location:Malua Bay, NSW Australia

Posted 20 June 2014 - 08:00 PM

Very interesting, David.  I'm aware that the bellows of my main concertina close faster than I'd like, and I suspect it's generalised seepage rather than any holes.  So that question is definitely on my do list.  

 

Now when you say you sprayed the bellows with satin clear - are we talking estapol or something similar?  Wow, courageous!  I might have thought it would flake off when dry, but I suppose it soaks in and fills the pores first?

 

I'd be interested in hearing other ideas for doping leather to ensure airtightness, while maintaining flexibility.  Ones I've considered so far include: Neatsfoot oil, Dubbin, silicone rubber (thinned in benzene).  Any others?

 

Terry



#61 David Hornett

David Hornett

    Chatty concertinist

  • Members
  • PipPipPipPip
  • 186 posts
  • Gender:Male
  • Location:Hobart

Posted 20 June 2014 - 09:46 PM

Terry,

 

Cabothane Clear Bunnings..

 

I tried the oil and dubbin, but of course if anything goes wrong you can't clue to it and it discolors the papers. You can't glue to Cabothane either I suppose, I have never tried, but then when all the bellows are coated who needs to. Anyone tried diluted shellac as a sealer? I too am interested in what else to seal the gussets with, before they are glued into place, so the whole instrument does not need to be coated.

 

Got the idea to spray them because the bellows papers, which are by far the most time consuming things to attatch when making bellows, I used to spray with Cabothane one side only (the back was left unsprayed so the glue would adhere). When doing up my Lachenal and Jeffries I had found some papers displayed on the net, copied them, enlarged to size, got the colours right and then printed off sheets. Because the print paper was just good quality photo paper, but thin, I gave them a dose of Cabothane, stiffened them up and kept them clean when in place. The papers I designed I also did the same to, but there is the distraction of a slight gloss. On the Jeffries I kept them all in the natural, no coating, they look great but will discolor over time. The Cabothane does not noticeably discolor, but really it is the sealing of the gussets before they are placed, with a sealer that can be glued, that I am interested in, that's the only mystery I have left about good bellows making.

 

David



#62 Ann-p

Ann-p

    Chatty concertinist

  • Members
  • PipPipPipPip
  • 139 posts
  • Gender:Female
  • Location:Worcester, England

Posted 22 June 2014 - 05:10 AM

To solve the lack of wine bottle corks, you could also switch to fizzy! Have to look carefully as really cheapo has plastic corks!



#63 Dana Johnson

Dana Johnson

    Heavyweight Boxer

  • Members
  • PipPipPipPipPip
  • 662 posts

Posted 06 July 2014 - 10:00 AM

I'm of the general opinion that controlling leakage is a good idea. Certainly, tests of different leathers / thicknesses for gussets would be useful for those not set in their ways or considering what to get for a first try. But like pads, leather for gussets has to meet more than one requirement. I don't care much for the idea of sealing naturally porous leather so much as using less porous varieties. Most critters leather comes from are covered in hair or wool or some other fur. A few have sweat or oil glands that go deep into the secondary layers of skin. Hair follicles can go deep and cause pronounced porosity as in pig skin which is renown for it's breathability. Gussets must flex probably more than any other part of the bellows, resist cracking and be flexible enough not to cause undo resistance. On some leathers, achieving these parameters requires splitting it to a degree that exposes these natural pores. Other leathers are naturally thinner to begin with and need little splitting and are flexible enough without the excess porosity.
This brings up the other issue of how tight do things need to be? How tight is Too tight and is there any such thing. While leakage enough to adversely effect response should be avoided, I find that the few ultimately tight concertinas I've played are like playing rocks. As much as the little felt bumpers under the buttons cushion the button's stop and keep playing from becoming painful, a very small amount of give in the bellows accomplishes the same thing, making playing more pleasant. I expect there is a crossover point where our ability to reverse bellows direction is slower than the reed's response to the speed of pressure change in the bellows (due to leaks in pads or gussets), where further tightness only serves to make the instrument feel "hard", and adds nothing to the music. Some leakage can even be helpful with reeds that have their set optimized for response over volume, that tend to choke under rapid bellows changes. ( especially the slower lower reeds ), allowing them to begin to start before the pressure is high enough to hold them closed.
If, as I expect there are diminishing returns to ever increased tightness as there are in the difference between SPF 50 vs SPF 75 sunblock, before you go to measures heretofore deemed unnecessary by our predecessors, you might consider stopping using bellows drop as more than a coarse diagnostic tool. ( especially since it isn't particularly useful for detecting weak springs which can allow leakage on the press). At some point leakage influence drops below the influence of "set" or valve weight/ springiness/ condition, loose reed shoe mounting, or poor reed pan to corner block contact, all of which might need to be addressed before playing differences due to non pathological leakage causes like slightly porous leather could even be noticed. It might be nice to have a rating chart for the different gusset leathers, with relative values for flex, durability, tightness, availability? Some historical input would undoubtably be useful to avoid reinventing the wheel. Leather has also changed according to the needs of the market over the years, mostly the needs of concertina makers have not been at the top of the list. What commonly available leathers ( list sources please!) are equivalents of what used to be used?
Dana

#64 Terry McGee

Terry McGee

    Chatty concertinist

  • Members
  • PipPipPipPip
  • 356 posts
  • Gender:Male
  • Location:Malua Bay, NSW Australia

Posted 06 July 2014 - 09:28 PM

Heh heh, I have this mental image of a concertina with ornately turned brass pressure release valves on the top of each side.  Pressure on the left and vacuum on the right?  You open the valves just enough to take the edge off the otherwise-perfect airtightness.  Or maybe they open themselves according to a preset spring pressure?  Sorry, I'll get real now....

 

I'd also like to know what is perceived as being too tight.  Here's the current perception chart:  

 

Concertina Leakage Perception.gif

 

Does anyone have a concertina that answers Dana's "too-tight" description?  Where does it fit on the chart above?  

 

Now taking on board the criticisms Dana makes above in regard to the drop test, I think we'd come to agreement that we should use a pressure test.  Open the bellows halfway, play a note, noting the amount of pressure that needs, open the bellows fully and apply that pressure without playing a note.  Count the time to close the bellows.  You could then reverse the process, counting the time to open the bellows fully starting at closed.

 

Now, Dana, I wonder if there is a mechanism at work other than leakage that might avoid the "playing a rock" feel, while keeping leakage low?  I'm thinking of flexibility in the gussets as opposed to leakage through them.  As you play staccato (am I right in thinking this is when it will show up most?) the gussets flex in response to the sudden pressure difference, ameliorating the shock.   

 

In which case we have a new parameter to consider and try to measure.  Sigh....

 

Terry


Edited by Terry McGee, 07 July 2014 - 03:42 AM.


#65 Terry McGee

Terry McGee

    Chatty concertinist

  • Members
  • PipPipPipPip
  • 356 posts
  • Gender:Male
  • Location:Malua Bay, NSW Australia

Posted 07 July 2014 - 06:53 PM

Hmmm, more questions.  

 

Am I right in assuming that, in a new and well-made concertina, leakage through the bellows is probably through the gussets?  

 

And that the next most likely would be the top skive and bottom hinge?  Do we have any reason to believe that would be significant?

 

In saying that I'm assuming that leakage is unlikely through the card stock?  (And yes, as an officially registered obsessive, I ran a leakage test on card stock and found it negligible.)

 

Do we have any easy way of estimating the area of a gusset, and therefore the total area of gusset?  Approximating it as a diamond, it appears on my Simpson to have a long axis of about 56mm and a short axis of about 14mm.  

 

Area of a gusset taken as 2 triangles each 28mm x 14mm = 392mm2

 

6 folds and 6 sides means 6 gussets by 7 valleys = 42 gussets.

 

42 gussets each 392mm2 = 16464mm2, or about 165cm2

 

 

So a total area equivalent to about 128mm x 128mm, or 5" square.  An area about as large as the end of your concertina.  That's a lot to leak if leakage is in season.

 

Are they all reasonable assumptions, or can we do better?  

 

Terry



#66 Dana Johnson

Dana Johnson

    Heavyweight Boxer

  • Members
  • PipPipPipPipPip
  • 662 posts

Posted 09 July 2014 - 05:53 PM

I think in practice , gusset area may be reduced depending on how they are applied. If the glue is applied to the card, the full flex area is " leak able" but if the glue is applied around the perimeter of the gusset, some covers the flex area and fully seals it. The hinge area at the valley is very narrow, and I think for the most part sealed with the film of glue. The top runs likewise are coated with glue when they are applied.
Gussets can bulge under pressure, and if too thin will pop in and out when the bellows is reversed quite annoyingly and delay the moment where the change takes effect.
Regarding your chart, even though the curve is trending upward, it's slope isn't near infinite. Since I am not the only one to have the experience of playing an instrument with insufficient give, perhaps that point lies beyond the range you chose to illustrate.
I have seen leakage around action pans, depending on how they are made. Often the action pan seal against the bellows frame and reed pan is not particularly tight. The seal on the leather flap valve hardly counts for any but the active reed pair. Air passing other reeds is free to migrate across chambers to whatever leakage may occur around the edge or through pad seals.
Where have other people found leaks and after fixing them noticed a difference in playing characteristics?

Edited by Dana Johnson, 09 July 2014 - 05:57 PM.






Also tagged with one or more of these keywords: Leakage, Test, Experiment

0 user(s) are reading this topic

0 members, 0 guests, 0 anonymous users