I know this is a pretty simple adjustment. I have a tool to do it but not sure the best way. Would you bend the spring in the middle or at the sharp end?
soybean
Adjusting the Spring Tension.
Hello soybean, it depends a little on what you want to get and how you want it. This simple question is slightly more interesting then it might seem to be in the very first place. I think I have been bending the springs all over, that would be, depending on how much you want to change the tension, at the very end, (almost) at the post in the middle (of the spring) and all the way in between. I think, for a good spring action, it is best if you have the spring evenly bent, that would be in a soft arc when unattached.
regards
regards
In an ideal world you'd remove the key to gain access to the spring and then use a pair of round nosed pliers to grip the spring at the pillar and gently slide them along the length of the spring while twisting the jaws slightly towards the direction you wish to bend the spring.
This creates a gentle curve in the spring which helps to distribute the tension.
In the real world you probably only have a spring hook, and no desire to take any keys off if it can be avoided...so...
The spring is unhooked from its cradle ( or post ) and the springhook is placed a quarter of the way in from the tip of the spring. The springhook is then gently pulled in the direction of the tension.
You should observe the horizontal alignment of the spring at rest - at a minimum it needs to come out of the pillar and line up with the notch on the spring cradle, for better results the tip should be set so that at rest it's slightly higher. This helps prevent the spring popping out of the cradle when the key is open, as the notch will rise relative to the centre of the key barrel when the key is operated.
It might not be possible to do this with ease ( due to other keys getting in the way ) so you might find you have to tension the spring and then resort to pulling the tip up through the nearest gap in the keywork. This isn't ideal, and is potentially risky in terms of spring breakage - but you may have no alternative.
In certain situations you may have to put a kink in the spring ( rather than a gentle bend ). This is best avoided if at all possible, but can provide very good results in the right conditions. However, it's very much more likely that the spring will break.
You can bend the spring closer to the pillar, and this will have rather more dramatic results with regard to the spring tension - but it also focuses the tension nearer to the pillar, which increases the risk that the spring will break....either during the tensioning process or during use.
If the spring is relatively new then you probably stand a good chance of getting away with it - if it's quite old then the risk of having it snap increases significantly.
If you find you can't quite get enough tension at the quarter-in point, move the springhook a little towards the pillar and try again. Repeat as necessary, moving the springhook closer to the pillar each time.
Spring bending can be something of a knack in that after you've done it enough times on enough different springs you get a feel for how each spring is likely to react.
I can nearly always tell if a spring will react well to being pulled out near the tip, or will take a bend nearer the pillar - or is likely to snap if round nosed pliers are not used.
Regards,
This creates a gentle curve in the spring which helps to distribute the tension.
In the real world you probably only have a spring hook, and no desire to take any keys off if it can be avoided...so...
The spring is unhooked from its cradle ( or post ) and the springhook is placed a quarter of the way in from the tip of the spring. The springhook is then gently pulled in the direction of the tension.
You should observe the horizontal alignment of the spring at rest - at a minimum it needs to come out of the pillar and line up with the notch on the spring cradle, for better results the tip should be set so that at rest it's slightly higher. This helps prevent the spring popping out of the cradle when the key is open, as the notch will rise relative to the centre of the key barrel when the key is operated.
It might not be possible to do this with ease ( due to other keys getting in the way ) so you might find you have to tension the spring and then resort to pulling the tip up through the nearest gap in the keywork. This isn't ideal, and is potentially risky in terms of spring breakage - but you may have no alternative.
In certain situations you may have to put a kink in the spring ( rather than a gentle bend ). This is best avoided if at all possible, but can provide very good results in the right conditions. However, it's very much more likely that the spring will break.
You can bend the spring closer to the pillar, and this will have rather more dramatic results with regard to the spring tension - but it also focuses the tension nearer to the pillar, which increases the risk that the spring will break....either during the tensioning process or during use.
If the spring is relatively new then you probably stand a good chance of getting away with it - if it's quite old then the risk of having it snap increases significantly.
If you find you can't quite get enough tension at the quarter-in point, move the springhook a little towards the pillar and try again. Repeat as necessary, moving the springhook closer to the pillar each time.
Spring bending can be something of a knack in that after you've done it enough times on enough different springs you get a feel for how each spring is likely to react.
I can nearly always tell if a spring will react well to being pulled out near the tip, or will take a bend nearer the pillar - or is likely to snap if round nosed pliers are not used.
Regards,
Hmm. I disagree with some of that.
1. Do not bend right at the post. It makes a weak point.
2. Do not make any sharp bends. The spring is likely to break there.
3. Make sure your bending does not make the spring bind against any part of the key, nor any other part of the mechanism.
4. After that, the forced applied to the key depends only on the displacement of the point from the spring cradle when the spring is unhooked, not on the actual shape of the spring.
So apart from getting enough bend to achieving the tension you seek, the only other factor is aesthetics, i.e. the spring should finish up looking approximately straight after it is hooked to the cradle.
The bending stress of the metal in any cantilevered spring increases evenly from zero at the tip, to maximum at the mounting. So to finish up with the spring straight, you will have to put no bend at the tip, and gradually increase the bend until the vicinity of the post.
For flat springs, the smoothness of the feel of its action is greatly enhanced if its effective length is as long as possible. To achieve this long effective length, it should contact the key near the mounting screw for as short a distance as possible during operation. So in this case, most of the bend will need to be near the mounting not so much for aesthetic reasons, but for functional reasons.
1. Do not bend right at the post. It makes a weak point.
2. Do not make any sharp bends. The spring is likely to break there.
3. Make sure your bending does not make the spring bind against any part of the key, nor any other part of the mechanism.
4. After that, the forced applied to the key depends only on the displacement of the point from the spring cradle when the spring is unhooked, not on the actual shape of the spring.
So apart from getting enough bend to achieving the tension you seek, the only other factor is aesthetics, i.e. the spring should finish up looking approximately straight after it is hooked to the cradle.
The bending stress of the metal in any cantilevered spring increases evenly from zero at the tip, to maximum at the mounting. So to finish up with the spring straight, you will have to put no bend at the tip, and gradually increase the bend until the vicinity of the post.
For flat springs, the smoothness of the feel of its action is greatly enhanced if its effective length is as long as possible. To achieve this long effective length, it should contact the key near the mounting screw for as short a distance as possible during operation. So in this case, most of the bend will need to be near the mounting not so much for aesthetic reasons, but for functional reasons.
"In certain situations you may have to put a kink in the spring ( rather than a gentle bend ). This is best avoided if at all possible, but can provide very good results in the right conditions. However, it's very much more likely that the spring will break.
You can bend the spring closer to the pillar, and this will have rather more dramatic results with regard to the spring tension - but it also focuses the tension nearer to the pillar, which increases the risk that the spring will break....either during the tensioning process or during use."
That's a good point.
Putting a sharper bend in the spring alters the dynamic of it. A typical example is a spring that's uncharacteristically heavy, either in terms of size or performance. You can bend such springs until the cows come home and never get the required 'snap' out of it - but a judiciously placed kink dramatically alters its response.
I have two such springs on my YAS62, on the side Bb and C trills. They've been in operation now for over two decades.
The proviso is that the springs must be in good condition and of a certain quality. This doesn't always mean the best quality - it relates more to the stiffness of the spring ( which is different from its strength ).
I note too that the old 1920s Rampone bari I recently acquired has a number of kinked springs on it which, if not orginal, are certainly very old.
It's not a practice I would recommend without any caveats, and my last paragraph refers to being able to feel how a spring will respond.
Regards,
I have found Mr. Howard's Haynes Manual to be a invaluable tool in the subject. I recently adjusted the key action on my alto and it made all the difference.
B
B
Everybody now, the spring might very well break. If it gets in your finger or hand you´d be very lucky at least compared to the degree of unluckiness if it gets into your eye. :crybaby:
One of my worst spring-related injuries was getting my middle finger caught between the opposing springs of the low F and Aux.F keys - I was caught on a 'barb'. That one hurt like hell, and it took me a minute or two to steel myself enough get myself out of it.
I also had a spring go under a nail, and the fingertip subsequently blew up like a foul, yellow balloon - had to have a minor op on it.
Square-ended springs are the worst, they might not slip into the flesh as easily as needle-tipped springs, but they do a lot of damage.
Regards,
I also had a spring go under a nail, and the fingertip subsequently blew up like a foul, yellow balloon - had to have a minor op on it.
Square-ended springs are the worst, they might not slip into the flesh as easily as needle-tipped springs, but they do a lot of damage.
Regards,
Thanks for the creating indelible, awful pictures in my mind! I never want to do a repad job again!
Thanks everyone for the good advice. I now have a good idea of the various ways to do this task. Most of my horns are recent ones, so I'm not too worried about breakage.
The kink has little or nothing to do with how much tension in total is put into a spring. If you did the experiment and mounted the wires horizontally and placed equal weights on the ends of the wires you would find that they both require the same weight to deflect the spring the same distance ( or near enough as makes little difference ).
It's all about the dynamics of the spring - acceleration, progression etc.
The total amount of energy stored ( and released ) doesn't change - what changes is the way in which it is stored and released. Bear in mind too that the kink adds an area of increased stiffness to the spring ( which by definition will change its response ) and this in turn changes the leverage.
It doesn't require reams of data to explain it - a simple practical test on an instrument demonstrates that there is a definite effect and that in certain situations it produces a key response which cannot be duplicated with a curved spring.
It's a fascinating subject, to be sure.
I'm glad the thumb is healing up nicely, it's always something of a worry that an infection will set in after a puncture wound, which reminds me to check with the doc that I'm up to date on my tetanus jab!
Regards,
It's all about the dynamics of the spring - acceleration, progression etc.
The total amount of energy stored ( and released ) doesn't change - what changes is the way in which it is stored and released. Bear in mind too that the kink adds an area of increased stiffness to the spring ( which by definition will change its response ) and this in turn changes the leverage.
It doesn't require reams of data to explain it - a simple practical test on an instrument demonstrates that there is a definite effect and that in certain situations it produces a key response which cannot be duplicated with a curved spring.
It's a fascinating subject, to be sure.
I'm glad the thumb is healing up nicely, it's always something of a worry that an infection will set in after a puncture wound, which reminds me to check with the doc that I'm up to date on my tetanus jab!
Regards,
"what changes is the way in which it is stored and released. Bear in mind too that the kink adds an area of increased stiffness to the spring ( which by definition will change its response ) and this in turn changes the leverage."
Sorry, I can make no sense of all that at all.
"It doesn't require reams of data to explain it - a simple practical test on an instrument demonstrates that there is a definite effect and that in certain situations it produces a key response which cannot be duplicated with a curved spring."
Nor that, as I hoped I had explained. Just curve it further, then back to the displacement you would have put into a kinked spring, and it will behave just the same as the kinked spring. That is exactly what my experiment demonstrated.
Sorry, I can make no sense of all that at all.
"It doesn't require reams of data to explain it - a simple practical test on an instrument demonstrates that there is a definite effect and that in certain situations it produces a key response which cannot be duplicated with a curved spring."
Nor that, as I hoped I had explained. Just curve it further, then back to the displacement you would have put into a kinked spring, and it will behave just the same as the kinked spring. That is exactly what my experiment demonstrated.
Not to change the subject but I was reading Gordon's post #18 and got frightened by the phrase "Attached Thumbnails".
I sent the details of the kinked spring setup to a colleague who very kindly forwarded it to a relative who's something of an expert in the field ( a professor with interests in the field of mathematics, fracture mechanics, plastic deformation etc. ). He also plays the sax and clarinet, so took particular interest in the idea. The heart of his response was as follows:
"Yes, a kink in a reed instrument spring will change its characteristics.
It will alter the effective length of the spring."
"The portion of the spring in the kink will barely flex, compared to the rest
of the spring. If the kink is close to one end the effective length of the
spring will be, more or less, the longer portion. It should not matter
which end. This will have the effect of shortening the spring, hence
making it stiffer. I'm ignoring any taper of the spring here."
Comment was made on the kink exceeding the elastic range of the material, putting it into the plastic zone - which has an overall effect on the stiffness of the spring.
Note that stiffness and strength are not the same.
These comments tie in with my practical tests and observations.
Regards,
"Yes, a kink in a reed instrument spring will change its characteristics.
It will alter the effective length of the spring."
"The portion of the spring in the kink will barely flex, compared to the rest
of the spring. If the kink is close to one end the effective length of the
spring will be, more or less, the longer portion. It should not matter
which end. This will have the effect of shortening the spring, hence
making it stiffer. I'm ignoring any taper of the spring here."
Comment was made on the kink exceeding the elastic range of the material, putting it into the plastic zone - which has an overall effect on the stiffness of the spring.
Note that stiffness and strength are not the same.
These comments tie in with my practical tests and observations.
Regards,
Body tubes are brass, so have nothing to do with a discussion on steel.
I totally disagree with your expert, Stephen.
1. He is saying that if you put a kink half way along a spring, the spring effectively becomes equivalent in stiffness to a spring of half the length. If a spring is made half the length that it was before, it has a most noticeable and disagreeable stiffness constant, that the player may interpret in layman's terms, as very resistant. The force needed to counter it will dramatically increase during its travel. That is not what happens.
2. If what he says is correct, and a spring with a kink half way along its length effectively reduces its length to half, then it would have to behave like a spring of half the length, with only half the spring moving. However you will note that half way along the spring, there is still a displacement during operation, that is approx half the displacement of the tip. The spring is still behaving more or less as a spring as long as it was before.
I would like to see his spring design calculations to support his statement. It just does not make sense. Perhaps he could rethink his expertise. You are welcome to show him this post, and i hope you presented my previous one (s).
"Comment was made on the kink exceeding the elastic range of the material"
Yes, that is true for the perhaps 1 or 2 mm at which the kink occurs. However note that in my experiment details, as I explained, I separated out this quite separate parameter, by exceeding the plastic range for both my kinked spring and my curved spring. Did you tell him that? (And that is what you are seemingly not doing.)
You are talking about the increased tension you can get in a spring by exceeding the elastic range, a long way, then returning the spring to its cradle. The same can be done over the entire length. It is not the kink that makes your spring stiffer , but the fact that you have bothered to significantly exceed that limit for at least part of the spring. Well exceeding it over the entire length would actually produce the stiffest spring possible. That would be a spring with a kink, so to speak, over its entire length. (And according to your expert's statement, that would make the spring equivalent in length to a spring with no length. Obviously this is incorrect, so the "expert's" statement is incorrect.
Time for a re-analysis!
I totally disagree with your expert, Stephen.
1. He is saying that if you put a kink half way along a spring, the spring effectively becomes equivalent in stiffness to a spring of half the length. If a spring is made half the length that it was before, it has a most noticeable and disagreeable stiffness constant, that the player may interpret in layman's terms, as very resistant. The force needed to counter it will dramatically increase during its travel. That is not what happens.
2. If what he says is correct, and a spring with a kink half way along its length effectively reduces its length to half, then it would have to behave like a spring of half the length, with only half the spring moving. However you will note that half way along the spring, there is still a displacement during operation, that is approx half the displacement of the tip. The spring is still behaving more or less as a spring as long as it was before.
I would like to see his spring design calculations to support his statement. It just does not make sense. Perhaps he could rethink his expertise. You are welcome to show him this post, and i hope you presented my previous one (s).
"Comment was made on the kink exceeding the elastic range of the material"
Yes, that is true for the perhaps 1 or 2 mm at which the kink occurs. However note that in my experiment details, as I explained, I separated out this quite separate parameter, by exceeding the plastic range for both my kinked spring and my curved spring. Did you tell him that? (And that is what you are seemingly not doing.)
You are talking about the increased tension you can get in a spring by exceeding the elastic range, a long way, then returning the spring to its cradle. The same can be done over the entire length. It is not the kink that makes your spring stiffer , but the fact that you have bothered to significantly exceed that limit for at least part of the spring. Well exceeding it over the entire length would actually produce the stiffest spring possible. That would be a spring with a kink, so to speak, over its entire length. (And according to your expert's statement, that would make the spring equivalent in length to a spring with no length. Obviously this is incorrect, so the "expert's" statement is incorrect.
Time for a re-analysis!
What he said was that by putting a kink *near either end* of a spring effectively shortens its length, with the longest portion being the effective length.
He didn't say that a kink dead centre would halve the effective length, and I strongly suspect this is because it's not a convenient linear arrangement.
There's clearly something going on which you have not accounted for, and my experiment with the wire visibly bears this out - where both halves of the spring can be seen to curve under stress
As regards the 'expert', his many publications in the field of stress mechanics and his somewhat lengthy CV lead me to believe, quite reasonably, that when he says something is so, there's more than a very good chance that he's right on the button.
He also made some interesting comments regarding perpendicular and parallel force, which I'll experiment on and perhaps incorporate into my future spring-setting technique if I find any noticeable benefit.
I have again forwarded your latest comments, if he cares to respond I'll most certainly copy them here ( I'd like to see the maths too, though I suspect it's a case of "Here's the answer, now go find out why" ) - but until a more authoritative counter-claim is made I feel I'm more than happy to agree to disagree.
Regards,
He didn't say that a kink dead centre would halve the effective length, and I strongly suspect this is because it's not a convenient linear arrangement.
There's clearly something going on which you have not accounted for, and my experiment with the wire visibly bears this out - where both halves of the spring can be seen to curve under stress
As regards the 'expert', his many publications in the field of stress mechanics and his somewhat lengthy CV lead me to believe, quite reasonably, that when he says something is so, there's more than a very good chance that he's right on the button.
He also made some interesting comments regarding perpendicular and parallel force, which I'll experiment on and perhaps incorporate into my future spring-setting technique if I find any noticeable benefit.
I have again forwarded your latest comments, if he cares to respond I'll most certainly copy them here ( I'd like to see the maths too, though I suspect it's a case of "Here's the answer, now go find out why" ) - but until a more authoritative counter-claim is made I feel I'm more than happy to agree to disagree.
Regards,
Ive kind of gotten lost in the debate, so I drew a picture, are we debating that there is no difference between these items regarding spring strength measured at the end of the spring, or something else entirely
For me Mathematically of course there will be a difference, stress / strain / point of load force / etc there are a lot of factors involved, however in our purpose for musical instruments I think it is really irrelevant, I know if I want the key to be a bit snappier I put a kink closer to the tip, for general smooth movement I distribute the radius
For me Mathematically of course there will be a difference, stress / strain / point of load force / etc there are a lot of factors involved, however in our purpose for musical instruments I think it is really irrelevant, I know if I want the key to be a bit snappier I put a kink closer to the tip, for general smooth movement I distribute the radius
There's no reason why there has to be a difference in the strength of the spring, it's not what's required anyway - it's simply a way of altering the response. As you've observed, the spring takes on a bit more snap in some circumstances without necessarilly making it any stronger. I dare say you've done the same as I have down the years, and wrestled with springs that just don't quite seem to have the right feel no matter what you do...until you put a kink in.
Regards,
Regards,
No.
Just take out some of the bend in the spring and the keys will be lighter on your fingers.
Alter the spring, do not bend the keys.
About the kink and the spring: In the area of the kink, the spring stops being one, or at least the action is highly altered. If that is close enough to the end of the material, it will effectively shorten the length.
An infinitesimal short kink in the middle of the spring wont have much impact besides being a fantastic point to break due to stress.
The math is pretty linear when it somes to springs with parallel sides. As long as material and dimensions do not change, it is 8th grade math at best.
About the kink and the spring: In the area of the kink, the spring stops being one, or at least the action is highly altered. If that is close enough to the end of the material, it will effectively shorten the length.
An infinitesimal short kink in the middle of the spring wont have much impact besides being a fantastic point to break due to stress.
The math is pretty linear when it somes to springs with parallel sides. As long as material and dimensions do not change, it is 8th grade math at best.
Stephen, you are just not addressing the issue I raised.
With your test, in order to get these equal deflections, as far as I can gather:
1. You put a sharp kink in one spring. The rest of the spring was left in its original, relatively stress-free state. The approx 1 mm where the kink is likely to become work hardened (especially with stainless steel or copper alloy), so yes, the spring is effectively shortened by say 1 mm. Only. I concede that, and the result will be a very slightly stiffer spring, but with a greater likelihood to break. However this increased stiffness in only significant if the kink is in the mounted end. After all, there is very little bending force on the other end, and it deflects and undeflects very little during operation compared with the mounted end, so stiffening that end makes makes little difference.
2. With the other spring, you put a gradual curve in it, so the entire spring is in a stresses state, and as soon as you bend it far enough back to hook onto the cradle, it will lose much of that original pre-cradle deflection you gave it.
So you are not only comparing methods of of obtaining deflection. You have introduced another facto of whether or not you are using a spring stressed in an undesirable direction or not.
With my experiment, as I said, I eliminated this factor by curving my spring significantly past the intended deflection, then returning it to a deflection that matched that of the kinked spring. This results in not having those undesirable stresses that you introduced into the curved spring.
This is why my experiment - photo shown - backed up what I predicted.
If you are making comparisons, and using experiments to demonstrate something, you need to eliminate other factors.
With your test, in order to get these equal deflections, as far as I can gather:
1. You put a sharp kink in one spring. The rest of the spring was left in its original, relatively stress-free state. The approx 1 mm where the kink is likely to become work hardened (especially with stainless steel or copper alloy), so yes, the spring is effectively shortened by say 1 mm. Only. I concede that, and the result will be a very slightly stiffer spring, but with a greater likelihood to break. However this increased stiffness in only significant if the kink is in the mounted end. After all, there is very little bending force on the other end, and it deflects and undeflects very little during operation compared with the mounted end, so stiffening that end makes makes little difference.
2. With the other spring, you put a gradual curve in it, so the entire spring is in a stresses state, and as soon as you bend it far enough back to hook onto the cradle, it will lose much of that original pre-cradle deflection you gave it.
So you are not only comparing methods of of obtaining deflection. You have introduced another facto of whether or not you are using a spring stressed in an undesirable direction or not.
With my experiment, as I said, I eliminated this factor by curving my spring significantly past the intended deflection, then returning it to a deflection that matched that of the kinked spring. This results in not having those undesirable stresses that you introduced into the curved spring.
This is why my experiment - photo shown - backed up what I predicted.
If you are making comparisons, and using experiments to demonstrate something, you need to eliminate other factors.
I don't know about you, but I hardly ever look at the springs on my horns unless they aren't working. Unless they're sticking out from behind the rod considerably, the aesthetic issues are insignificant.
"... and an expert in the field of stress mechanics confirms that there is something going on outside the normal state of affairs."
I explained it. You ar introducing a second parameter other than the initial deflection and kinks. You don't appear to want to even consider that explanation.
And no it had nothing to do with the simplicity of eating nuts, and had a lot to do with whether a deflection (with the spring at rest) was with the spring metal - at least the bulk of it - in a stressed state, or a stress-free state. The latter is obtained only by judiciously bending too far and then back to that inital relaxed deflection.
I am quite sure that if I had a direct discussion with your expert we would agree, after very brief discussion of the parameters, on everything, and those things we agreed on would not quite match your perception of things. I'm done with wasting more writing time.
I explained it. You ar introducing a second parameter other than the initial deflection and kinks. You don't appear to want to even consider that explanation.
And no it had nothing to do with the simplicity of eating nuts, and had a lot to do with whether a deflection (with the spring at rest) was with the spring metal - at least the bulk of it - in a stressed state, or a stress-free state. The latter is obtained only by judiciously bending too far and then back to that inital relaxed deflection.
I am quite sure that if I had a direct discussion with your expert we would agree, after very brief discussion of the parameters, on everything, and those things we agreed on would not quite match your perception of things. I'm done with wasting more writing time.
I have to say that although debate on the forum can get a little robust at times, it should never be thought of in terms of winners and losers as this tends to discourage people from posting for fear of ridicule.
I very much enjoy a lively discussion, and often find that positions on both sides benefit from scrutiny.
As such I will not be associated with the obvious comment made above - and I feel sure that Gordon and others will join me in condemning this deplorable attitude.
Regards,
I very much enjoy a lively discussion, and often find that positions on both sides benefit from scrutiny.
As such I will not be associated with the obvious comment made above - and I feel sure that Gordon and others will join me in condemning this deplorable attitude.
Regards,
-
?
-
?
-
?
-
?
-
?
-
?
-
?
-
?
-
?
-
?
-
?
-
?
-
?
-
?
-
?
-
?
-
?
-
?
-
?
-
?
