Reed voicing / profiling / tuning etc.

[4to5to6]

Spent yesterday building new reed tongues, re-profiling damaged tongues, tuning and voicing as well as experimenting with different valve thicknesses.  The biggest problem was the tongue profiles… the tip and the root were ground down on the worse ones leaving a thicker hump in the middle which made the reed unresponsive and harsh sounding.  Reed tongues were often ground on an angle or notches left that had to be smoothed out or the tongue replaced.  Once this was removed leaving a flat or slightly concave tongue, the responsiveness came back and harshness went away.  Still another day or two to make it perfect while playing it in for a few hours in between tunings but I’m really encouraged that this is actually going to work out just fine.

 

Definitely a challenge… it is so much easier if the reeds are untouched so you can just match the existing profile.

 

i know some use Demels to tune and remember coming across this suggestion in a 70s magazine.  I personally would highly discourage the practice unless you are very confident in how you approach it. I was once told that 40 percent of concertinas have been butchered in one way or another.  You can’t beat a flat file or a very low profile half round on small reeds IMHO… no edge notches or gouges, easy to keep flat so you don’t file on an angle or taking off an edge, easy to rapidly remove material or go slow.  Nice even swooping profiles.  File across at a 45 degree angle so there are no stress lines that could start a crack, etc.

 

The weight of the valves has a large impact on the volume and responsiveness of the reed as well.  I had to leave a few valves off on the highest reeds or leave part of the hole slightly uncovered even though this could cause air inefficiencies or inconsistent chamber sizes affecting harsh harmonic cancellation and the fine tuning due to variable chamber size but found no way around it.  Maybe there is an ultra thin leather I need to find to make these smallest valves out of.  A thicker valve on the lowest notes balanced overbearing bass reeds although slightly decreased responsiveness.  I’m still learning on this one as well so the jury is still out.

 

The key ingredient is patience.  If I started feeling rushed, I had to just walk away and work on something else.

 

Thanks for everyone’s inspiration and advice.  I’m now confident this is another hurdle soon to be overcome. 😊 

.

Edited April 23 by 4to5to6

[Chris Ghent]

23 hours ago, 4to5to6 said:

The key ingredient is patience. 

Every concertina related lesson I ever learned had the word patience in it somewhere…

[4to5to6]

49 minutes ago, Chris Ghent said:

Every concertina related lesson I ever learned had the word patience in it somewhere…

 

Thanks Chris.  Isn’t that the truth!

.

[Clive Thorne]

For optically measuring precise distances these days I would suggest have a digital microsope, with cross hairs on the screen. Then mount the object on a precisely moveable table, either with micrometer type adjustment, or precision feed back. Line up on one side of the gap, set some sort of zero, line up the other side and read the value.

Probably too expensive for an ameteur, but possibly not for someone who was making his living at it.

If you already had a cnc machine then mount your camera on the head, and Bob is your uncle.

 

Of course other methods exist.

 

Precise measuring may be interesting, but I'm sure that once you've done a few you get to know what looks right without the kerfuffel.

[Greg Jowaisas]

Quote

 Maybe there is an ultra thin leather I need to find to make these smallest valves out of.  A thicker valve on the lowest notes balanced overbearing bass reeds although slightly decreased responsiveness.  I’m still learning on this one as well so the jury is still out.

The "great" Malcolm Clapp told me that valving was the black art of setting up a concertina and an important aspect of performance often overlooked.

 

I once had a white kangaroo skin that yielded exquisite small valves.  Mine was used up years ago but left a lasting impression.  You might keep an eye out.

[Chris Ghent]

On 4/25/2025 at 1:49 AM, Clive Thorne said:

For optically measuring precise distances these days I would suggest have a digital microsope, with cross hairs on the screen. Then mount the object on a precisely moveable table, either with micrometer type adjustment, or precision feed back. Line up on one side of the gap, set some sort of zero, line up the other side and read the value.

Probably too expensive for an ameteur, but possibly not for someone who was making his living at it.

If you already had a cnc machine then mount your camera on the head, and Bob is your uncle.

 

Of course other methods exist.

 

Precise measuring may be interesting, but I'm sure that once you've done a few you get to know what looks right without the kerfuffel.

I initially used the xy table on my cnc mill, I clamped perspex to the table, put my reed on the perspex and placed my microscope above the reed. It was not easy and I acquired a bunch of wildly different readings. One issue was the crosshairs exceeded the size of the gap.  Every reading became a judgement call. 

[Chris Ghent]

On 4/25/2025 at 3:46 AM, Greg Jowaisas said:

The "great" Malcolm Clapp told me that valving was the black art of setting up a concertina and an important aspect of performance often overlooked.

 

I once had a white kangaroo skin that yielded exquisite small valves.  Mine was used up years ago but left a lasting impression.  You might keep an eye out.

Agree with “the great”. Lovely to see you poke your head back above the parapet. 

[Clive Thorne]

10 hours ago, Chris Ghent said:

I initially used the xy table on my cnc mill, I clamped perspex to the table, put my reed on the perspex and placed my microscope above the reed. It was not easy and I acquired a bunch of wildly different readings. One issue was the crosshairs exceeded the size of the gap.  Every reading became a judgement call. 

Interesting. Were the cross hairs physical markings, or electronic/software generated ones on the computer screen?

When I was working for a laser manufacturer we always used electronic/software cross hairs, which could be as narrow as one pixel, irrespective of the magnification.

[4to5to6]

Reed voicing is going great… this is turning out to be a very fine instrument.  Very well balanced.  Great tone.  I bought it as a project to pass it on but it may just end up a keeper.

 

One thing I learned about making reed tongues (I first did it wrong of course) is to always thin down the tip first then work backwards creating a nice sweeping curve towards the root. Then bring the note down to pitch by working on the root rather than trying to bring it up to pitch by filing at the tip.  I tried it the other way at first but then ended up twisting the tongue as the root was too thin to support the tip filing.  I can do a reed tongue about 4 times faster by doing the tip first.
 

I’ve noticed that the reed raises in pitch slightly once the steel work-hardens… the pitch also slightly changes with voicing the tongue curve and gap… that said, the reed pitch drops a few cents mounted versus on the tuning bellows due to chamber resonance (which is a good thing cancelling harsh harmonics)...  I’ve now decided to tune to the exact pitch at first, then play the reed in for a few hours, then do the final touch up tuning later once the reed is stable.  The third final tuning after this is done by ear, counting the interval beats to tune to the temperament.

 

I use a variety of files… some comparatively course that really dig in to remove a lot of metal quickly, then medium, then fine.  I’m also experimenting with Eze-Lap diamond abrasive hone files that use abrasives versus teeth.  Strangely enough, a few tongue clamps screws have been loose and the reed tongue has shifted.  I've had to reset multiple reed tongues in this instrument... I can't understand what was done to it.  The reed profiles are all over he place.  What a challenge... my apprenticeship in reed tongue profiling and setting and voicing!  I can drop a 48 key treble down to A440 in 5 hours and do the final tuning and tweaking in about another 5 total but have spent probably 30 hours on this one so far.  It's getting closer and closer and I'm getting fussier and fussier... much. much better all the time... but I just went over it after playing it for about two hours and 14 keys still need tweaking.  I still need to do the final tuning after all this and it was at A440 when I started.

I’m throwing all this all out there in hopes of some good feedback... I’m open to correction.  I am figuring this out as I go.  I di reverse engineer about 20 high end instruments before i started doing reed studies... handling about 2000 reeds... this helped.

Lots to learn still… I used to tune pianos many years ago so understand temperaments and hear the beats… My wife has near perfect pitch but my main thing is tone.  Concertinas need to be expressive and have great dynamics and balance so the bass doesn’t drown out the higher notes but there is a certain perfect tone that I personally love and go for… my 1927 amboyna TT that Steve Dickinson did has it… it’s a tone that is clear, not harsh at all, smooth and silky but still has a bit of grit when you dig in to bring you forward in the mix.  It must be fast and snappy but still be able to do those long mournful notes.  I’m experimenting with varying the reed profile to bring out this tone.  Lots and lots to learn still.

 

Advice is welcome.

.

Edited April 27 by 4to5to6

[fred v]

Thanks for your notes on voicing. I've played with this a bit with loud notes. I have stiffened up the valve. My hearing is quite bad in the high frequencies so I often wonder if it is the reed or my hearing.

 

[4to5to6]

Regarding a 6x zoom reed gap inspection camera… I just started another thread to try and find one to be more consistent in setting Reed tongues.  I am trying to find a 6X digital camera that is Apple iPad compatible (comes with an app).

 

 

[Duncan Luddite]

Thanks everyone for the above discussion, and information. Found all this to be encouraging, and as a B reed, on the draw, on my Jabez Austin EC has recently gone from in tune to 25 cents low (and was whisper thin for the last quarter of it's length), I had a shot at making a new brass reed.

The dropping reed was nickel, so not an original as all except 4 or 5 are brass and were made by Nickolds (with the open sided screw slots in the clamping bar). I carefully undid the brass holding screws and flipped over the shoe onto some hard, fine, sheet brass I have and ran a scribe along the two sides of the throat, the resulting lines I then continued out with a ruler to make a longer, tapered strip. This was cut just outside the lines, held in a small hand vice and filed until it slipped into the throat (the remains of the scribed lines uppermost so that they would be removed in the tuning process).

Clamping bar reinstalled, and it gave me a nice E below B !

Lots of abrading later I now have an in tune B.

When I originally got the Austin, all the brass reeds had a layer of verdigris on them and, once cleaned (at a clockmaking friend's workshop), nearly all of them needed re-tuning (mid. 1800's style, though at A440), so that part of it I was familiar with; although, as this was a reed I had made, I didn't have to be so concerned with the prospect of ruining it, and as such was able to 'push the boundaries' of what I knew in doing the tuning. So I've done some learning

In the photo the right most item is the dropping nickel reed, the middle my new home-made reed fitted, and the left is it's partner, 'push', reed.

Thanks again for this discussion.