Jump to content


Photo

Profiling

reed tongues

  • Please log in to reply
133 replies to this topic

#19 Chris Ghent

Chris Ghent

    Heavyweight Boxer

  • Members
  • PipPipPipPipPip
  • 1066 posts
  • Gender:Male
  • Location:Blue Mountains NSW

Posted 06 April 2014 - 08:32 PM

One last piece of info. Here is a graph given to me by the late great and missed Richard Morse which has been uploaded before. The first pic is raw data, the second smoothed. In the first you can see the variation that comes from hand made reeds. 

Attached Thumbnails

  • 3Dprofiles.gif

Edited by Chris Ghent, 06 April 2014 - 08:34 PM.


#20 Terry McGee

Terry McGee

    Chatty concertinist

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

Posted 06 April 2014 - 11:52 PM

Great stuff, Chris.  Calls for a Tit for Tat* reprisal:

 

Lachenal reed profile 3D.GIF

 

*Historical note:  Sophie Tucker, the "Last of the Red Hot Mommas", referred to being a believer in "tit for tat".  Which, she concluded, meant she was owed rather a lot of "tat".



#21 Terry McGee

Terry McGee

    Chatty concertinist

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

Posted 07 April 2014 - 12:34 AM

And here's something to consider.  Should the length axis be marked in proportions, such as Root to Tip, as we've done so far, and as the Morse graph Chris showed (percent) did, or should it be in mm?  That would show variation in the lengths of the reeds as well as seeing the variations in thickness.  If you give me the lengths of the reeds in the data so far, I can illustrate what it might look like.

 

Do we happen to know in what thicknesses the reed steel was available, back in the day? 

 

Terry



#22 alex_holden

alex_holden

    Heavyweight Boxer

  • Members
  • PipPipPipPipPip
  • 585 posts
  • Gender:Male
  • Location:Lancashire

Posted 07 April 2014 - 02:48 AM

One last piece of info. Here is a graph given to me by the late great and missed Richard Morse which has been uploaded before. The first pic is raw data, the second smoothed. In the first you can see the variation that comes from hand made reeds.


That's brilliant, Chris. I don't suppose you have a higher resolution version of this graph; I can't quite make out the labels?

#23 David Hornett

David Hornett

    Chatty concertinist

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

Posted 07 April 2014 - 03:32 AM

Latest sizes for your graph (I won't bother with any more until I get my new pin micrometer from England, about a week or so, and then i will attack the Jones, but they have been filed, by me. Thank you for the hint re thin cotton, fantastic idea.

 

Low C Heel (area under clamp plate 84   33  30  34  61 82 free length 35.54

Low C# Heel                                      84   40  31  26  52 83                   35.54

 

Low D                                                84   40  31  30  45  82                   35.54

 

Low D#                                              84   40  31  28  49  73                  35.54

Low E                                                84   36  27  33  41  46                  35.54

 

A# below middle C 37  24  25  43  48

Middle C  28  25  25  34  35

D above middle C  36  25  23  33  41

E above C  23  21 22  29  32

A high C .20, 19 16 15 15



#24 Terry McGee

Terry McGee

    Chatty concertinist

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

Posted 07 April 2014 - 05:29 AM

OK, here are two versions of it, both with their merits.  Incidentally, I've assumed, until told otherwise, that all the keys are based on 0.84mm stock material.  If I don't put a value in, the software assumes zero, which is a very unlikely value for the clamped bit!

 

Here's the 3D graph.  It's strengths are in visualising the whole story, but hard to read any values off accurately....

 

Lachenal Duet 8816 Reed Profile 3D V3.GIF

 

Here's the Line graph.  Easy to follow individual reed data sets.....

 

Lachenal Duet 8816 Reed Profile Line V3.GIF

 

And here's an XY graph showing the reeds with lengths.  (I've temporarily assumed the higher pitched ones are half the length of the low reeds.)

 

Lachenal Duet 8816 Reed Profile XY with lengths V3.GIF

 

It could get pretty messy if you add a full set of reeds.  

 

Terry


Edited by Terry McGee, 07 April 2014 - 05:58 AM.


#25 Chris Ghent

Chris Ghent

    Heavyweight Boxer

  • Members
  • PipPipPipPipPip
  • 1066 posts
  • Gender:Male
  • Location:Blue Mountains NSW

Posted 07 April 2014 - 05:57 AM

 

One last piece of info. Here is a graph given to me by the late great and missed Richard Morse which has been uploaded before. The first pic is raw data, the second smoothed. In the first you can see the variation that comes from hand made reeds.


That's brilliant, Chris. I don't suppose you have a higher resolution version of this graph; I can't quite make out the labels?

 

This is not the one Rich sent me, I lost that in the great computer crash of 2thousand and something. I picked it up again from somewhere online, I think here, but my recollection is it was always this resolution.  It never mattered to me what the figures were, the best thing about it is, it allows you to explain the shape of reeds to people very easily. If it is the meaning of the axes you need I can tell you them, roots to the left, tips to the right, bass to the back, treble to the front. The lengths are shown as visually the same, represented as a percentage.  

 

If anyone wants the thicknesses of the reeds can I recommend two things, firstly as this thread shows it is easy to get detailed info yourself, while it takes time if you can't spend that much time then you won't get a concertina made, and in gathering the info you will learn much more.  You will need to prepare to be confounded by the roughness of the data. Secondly, though this graph is from a Jeffries, it is better not to measure Jeffries concertinas as a starting point. While I cannot speak for everyone, I can't think of a maker I know who makes anglos with only 9 reed frame sizes. I do know a few who started off intending to make 9, myself included,  but they all expanded their frame sizes to as many as possible because it makes good reed making considerably easier.

 

The thing is, Jeffries best reed makers were astounding good. They could fit into a single size of reed frame not only the wider range of pitches needed if you only have 9 where the rivals have up to 19, they could also do it for other keys of concertina. My Af/Ef has the same size reed frames as a C/G in the same positions, it is just the pitches that are different. However the reeds they made, even though they are acceptable when in a fabled concertina, in fact are part of the charm, would not be acceptable in a good modern concertina. Talking of good Jeffries reeds, that is to say ignoring about 25% of Jeffries concertinas, the clearances are often OK, but they are too pitch unstable. There are number of people today making much better reeds.



#26 alex_holden

alex_holden

    Heavyweight Boxer

  • Members
  • PipPipPipPipPip
  • 585 posts
  • Gender:Male
  • Location:Lancashire

Posted 07 April 2014 - 06:11 AM

Thanks for that Chris, it's very useful.

#27 Terry McGee

Terry McGee

    Chatty concertinist

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

Posted 07 April 2014 - 06:12 AM

Interesting point to note in the 5 longest reeds.  Note reeds 1 to 4 pretty much return to the full stock value of thickness at the tip, but reed 5, low E, ends up with a much thinner tip:

 

Lachenal Duet 8816 Reed Profile XY with lengths V3.GIF

 

How can there be such a jump in one semitone, you ask, when the first four are also separated by only one semitone.  Look where the minimum thickness of Low E is - at around 15mm from the heel, compared with most of the others at around 21mm (keep in mind these are guesswork figures at this stage, but they are all following the same assumptions).  So Low E's flexing point is lower down the reed, leaving more of the reed to be weight, not flexure.

 

Low C# conversely is equally thinned and therefore flexible, but further up the reed.  So it's tip is the same thickness as that of low C.  

 

Terry



#28 Chris Ghent

Chris Ghent

    Heavyweight Boxer

  • Members
  • PipPipPipPipPip
  • 1066 posts
  • Gender:Male
  • Location:Blue Mountains NSW

Posted 07 April 2014 - 07:02 AM

These look like the normal variations you get when assessing a set of hand made reeds. When I measure a set I discard the ones well outside the average for the area.

 

When you are filing it is good to reach the thinnest part first but with a thou (sorry I think thou in reed thickness, its easier than .72 to think 28 thou, plus my only micrometer is in thou) in hand. Then work back (talking reeds above about 350 hz here) to the thickest part watching the curve as I go. The thou in hand  is so you can adjust the curve if you have to. 

 

If you are doing a low reed like the one you mention above and you thin the belly too much you can get it back to tune by thinning the tip. You have to decide how much you want to deviate from the ones around it, you are trying to make a set not just one reed.  Your deviation above looks like one of those where the maker has over thinned the belly. The thing you don't know, because the sampling is too coarse, is whether he really blew it and had one tiny thin spot which would make a really weak reed or whether the thin part extends widely in which case it would be weak but perhaps less so.

 

If I was sitting at my own kitchen table in Victorian London desperately trying to get a set of reeds finished and back to the factory for Mrs Lachenal by the end of the day I would just leave that one in too.



#29 Terry McGee

Terry McGee

    Chatty concertinist

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

Posted 07 April 2014 - 07:47 AM

Chris, earlier up you mentioned Jeffries using 9 frame sizes, too few in your view.  I've assumed in all this that the frames were cast, which would make sense having a limited range of sizes.  But I've never thought to look when I've had a concertina open.  Were they cast, or made in some other way?

 

Terry



#30 Terry McGee

Terry McGee

    Chatty concertinist

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

Posted 07 April 2014 - 08:02 AM

And another question.  The Lachenal I've been looking at has remanent signs of deep pitting on parts of the reeds, which generally have been polished up.  Clearly the reeds were very rusty until recently.  Now, cleaning off most of deep pitting on both sides will require removing quite a bit of metal.  Then tuning the instrument will mean removing more.  Obviously, the profiling can never again be the same as the maker intended.  The reeds will be more flexible (thinner in the waist) but also lighter (thinner at the tip).  What would we expect that to do?  Make the instrument play on less bellow pressure, but run into problems at higher bellow pressures?  

 

Terry



#31 alex_holden

alex_holden

    Heavyweight Boxer

  • Members
  • PipPipPipPipPip
  • 585 posts
  • Gender:Male
  • Location:Lancashire

Posted 07 April 2014 - 08:06 AM

Chris, earlier up you mentioned Jeffries using 9 frame sizes, too few in your view.  I've assumed in all this that the frames were cast, which would make sense having a limited range of sizes.  But I've never thought to look when I've had a concertina open.  Were they cast, or made in some other way?


There's an article by Geoffrey Crabb somewhere that talks about stamping out the rough shapes with a fly press before hand-filing.

#32 Terry McGee

Terry McGee

    Chatty concertinist

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

Posted 07 April 2014 - 08:14 AM

Ah, that makes sense.  I could imagine the many downsides of casting.  And stamping requires custom punches and dies, so still good reason to keep the number of sizes down to a reasonable minimum.  Thanks Alex.



#33 Chris Ghent

Chris Ghent

    Heavyweight Boxer

  • Members
  • PipPipPipPipPip
  • 1066 posts
  • Gender:Male
  • Location:Blue Mountains NSW

Posted 07 April 2014 - 08:31 AM

All of the big factories had diesets and fly presses, there is a reference somewhere (Tommy Williams?) which talks of 17? fly presses in a row at Lachenal's. In those days this was high technology. It was probably a two step process, one die cut the outside shape and oerhaps roughly cut the window, the next process broached the window and created the relief behind the reed aperture, and perhaps flowed the metal outwards creating the bevel in the dovetail.

Early revival makers used fly presses also, recent makers are more cnc mill orientated, and the most high tech, Wally Carroll, is probably using wire EDM. It is very expensive to get a set of diesets made, if you fronted an engineer here for a set of 9 today I would estimate you would be up for $30,000. The same restrictions would have been extant in 1890, probably accounting for Charles Jeffries only having access to 9 sizes of reed.

Diesets are surprisingly accurate; well setup there would be little need for supplementary filing. They are very predictable, the end result is the same each time, meaning the reed fitting would also be predictable. An anecdote; Richard Evans showed me a Jeffries in which every reed tongue of a specific size ie. one of the 9 sizes, had a tiny nick in one corner of the tip about 1.5 thou in size. We looked at the frames under a microscope and there was a tiny piece of metal left by the stamping process in that same corner, must have been a defect in the die and reproduced each time exactly the same. What interested me, the reed maker knew exactly what that obstruction looked like even though it was truly tiny. Did they call the microscopist I wonder. Was there a fellow with a horse and cart who would come around, on the side written George G Fallows Esq, Microscopy by appointment or somesuch, who would show them the defect so they could plan around it.

Funnily enough there is a rooster in NSW who is setting up to cast frames, I have seen an early example and the result is surprisingly good. Cheap too.

#34 Chris Ghent

Chris Ghent

    Heavyweight Boxer

  • Members
  • PipPipPipPipPip
  • 1066 posts
  • Gender:Male
  • Location:Blue Mountains NSW

Posted 07 April 2014 - 08:36 AM

And another question.  The Lachenal I've been looking at has remanent signs of deep pitting on parts of the reeds, which generally have been polished up.  Clearly the reeds were very rusty until recently.  Now, cleaning off most of deep pitting on both sides will require removing quite a bit of metal.  Then tuning the instrument will mean removing more.  Obviously, the profiling can never again be the same as the maker intended.  The reeds will be more flexible (thinner in the waist) but also lighter (thinner at the tip).  What would we expect that to do?  Make the instrument play on less bellow pressure, but run into problems at higher bellow pressures?  
 
Terry

I think I would not remove the pitting. Just fix the actual problems, that is, if it is out of tune, tune it. They don't have to look good to work well. Thinning the reeds could change the playing characteristics but it is all relative. How much would you be removing?

#35 alex_holden

alex_holden

    Heavyweight Boxer

  • Members
  • PipPipPipPipPip
  • 585 posts
  • Gender:Male
  • Location:Lancashire

Posted 07 April 2014 - 12:00 PM

Funnily enough there is a rooster in NSW who is setting up to cast frames, I have seen an early example and the result is surprisingly good. Cheap too.


A rooster?

#36 David Hornett

David Hornett

    Chatty concertinist

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

Posted 07 April 2014 - 04:22 PM

Hi all, give me a few days and I'II give you the readings for all 60 reeds, heel, length and thickness. But one thing I do suggest is that a close look at these rather pristine reeds, no rust, just nice clean reeds is I suspect the profiling was done over some type of form, the reed held bent over a male profile, the file passed across in a series of long strikes and then the reed mounted and tuned. I suspect this because the file marks are uniformedly approx 45 degrees over each reed (a guess), and no file mark interferes with another, all consistent with the reed being held firm, and longitudinally tensioned, and a number of file passes. When the required centre thickness was reached the reed could be lifted from the form, it would automatically flatten, inserted in the shoe, trimmed to length, and then touched up. Each reed for the required note would be consistent (a form for each) and with a bit of experience (doing the same thing every working day of your life) very little final tuning would be required. The real skill would have been in making the forms. That's my guess The file marks are deep, direct firm and non interfering, the same certainly cannot be said when I do my little bit of tuning every now and then.

 

That's my guess for this instrument.

 

Have a great day.






0 user(s) are reading this topic

0 members, 0 guests, 0 anonymous users