I am a beginning player and have bought 2-3 new harps in the last 6 months, because of not getting the notes out of the harmonica (after a few weeks of use). With new ones it is OK, no problems.
Sometimes it was 1 draw, sometimes 6,7 blow and/or draw. Did my harp went "flat"? Is it because I am unexperienced in how hard one need to blow/draw? i did learn to bend on those harps, but mostly 4,2 and 3 draw, but those reeds stayed fine. Although I think I a not putting to much strength on my blows. how many harp went wrong in your beginning years?
Last Edited by johan d on Feb 15, 2017 6:07 AM
If you get NO sound, the reed slots need cleaning; if note is flat, you did that with technique (probably too much force) -- I've heard this is very hard (impossible?) to fix -- well you can replace the reed if you have the skills or work with someone who does -- others here will say more about that. I learned years ago that the little plastic rectangular thingies that cause an alarm to go off if you leave a store with something you didn't pay for, have very thin metal strips that can be used to clean up the reed and reed slots just by moving the strip up and down, back and forth between reed and slot -- nice to have harps with cover plates that attach with screws to make this easy.
Last Edited by TetonJohn on Feb 15, 2017 7:32 AM
To begin with johan_d, an important question to ask is what kind of harps are you playing? If you are playing a cheap harmonica like a Hot Metal or a Blues Bender or any number of other Made in China junk/toy harmonicas, that's your problem. It could be that there is nothing wrong with your playing and if you invest in a Marine Band or Special 20 or any one of several other quality harps, they will last a lot longer. There is a reason why the Hohner Marine Band became the best selling harmonica in the world with blues player. Most people beleive it is because of the unique tone it produces compared to other harps. That's not really why. It is because the reeds were strong enough to stand up to the additional stress created by bending them. If you do alerady have high quality instruments, then we need to look at how you are playing them. ---------- Tom Halchak www.BlueMoonHarmonicas.com
When I first started playing, I had a few harps go flat in a couple of days use. Typically this is a fatigue failure of the reed. A crack has started propagating which lowers the spring rate of the reed and lowers the pitch.
Frequency = square root(spring rate/mass)
If I would file the end of the reed to reduce the mass, I could bring it into tune but it would go flat again in minutes. A few round of this and the reed would break. Once a crack has started, there is nothing that can be done except to replace the cracked reed. As I have learned to play with lower breath force, these failures are much rarer. Once you learn how to use finesse and resonance, rather than brute force, to get the tone and volume you want, your reeds will last a lot longer.
I use Lee Oskar and Special 20's. So I guess ordering new reed plates would be my only solution... Sound logic that over time I learn to play more controlled and reeds will last longer. Thx all
There was a lot of discussion a while about why reeds go bad - everyone agreed it was metal fatigue. That's caused by lots of flexing - frequency and amplitude are the culprits.
Also that somehow bending the notes stressed the reeds more. But also (as I remember) that the reeds involved in the double-reed bend vibrated with a smaller amplitude. So that doesn't quite square.
Here is another example where someone has reported that reeds not involved with bending notes have failed.
What do you think is happening? Do the reeds involved in bending notes fail more or less than the others? In what way does bending a note 'stress' a reed? If invoking metal fatigue in your explanation I think you have to explain why they are vibrating faster or with a bigger amplitude. Or maybe I have a simplistic understanding of metal fatigue. You people doing manufacturer and working with materials maybe know.
I've never broken a reed, so I tried to break one on a cheap harp by repeatedly bending it's 4D down as far as I could. It certainly 'felt' like I was stressing it. But I gave up after a few days as it showed no signs of weakening.
Last Edited by MindTheGap on Feb 16, 2017 1:31 AM
I did an analysis of a brass cantilever beam the size of an A harmonica reed a while ago. This assumes the peak stress in the reed is the value that would be achieved with a constant pressure difference between the top and bottom of the reed. This shows that an A reed that flexed back and forth at its natural frequency to a shape that a .7 PSI air pressure (around half the value I found for pressure a trumpet player develops in the mouthpiece of the trumpet.) would bend it to, would fail in about 6 hours of continuous playing. I know there are a lot of assumptions here and many may not be valid, but the 6 hours to failure did seem reasonable. I don't know how much typical music you can play before racking up 6 hours on the A reed.
I share MindTheGap's skepticism that bending directly reduces reed life. I do not dismiss the observations of many harmonica players and technicians that bending too hard correlates with reed failure. My hypothesis is that as you bend, the pitch reaches a floor that is near the natural frequency of the lower pitched reed in the pair. As you attempt to bend past this floor, you cease to get a pitch change and the natural response is to apply more breath force to attempt to get the pitch to go lower. I suspect it is the increased breath force, whether initiated by hitting the floor, of the bend or by ineffective bending technique, that raises the stress in the read and causes early failure. To test this hypothesis, one might instrument a harp to measure pressure to see if there is a pressure increase a a player tries to bend past the floor. One might also build a constant pressure device that could play the note consistently at the lowest possible bent pitch and compare the hours to failure to an unbent note at the same pressure. It would also be interesting to look at the difference between a novice and a pro as a note is bent. I suspect that a pro has very little pressure change as a note is bent while a novice will increase pressure greatly as the bend is attempted. Has anyone heard of any studies done with a pressure transducer in the reed chamber?
This is a model of a reed using finite element analysis (FEA) to show where the highest stress point is as the reed vibrates. The red color is where the stress is the highest.
This is a photo of a reed that has failed fully. Not that the position of the crack is in the high stress area predicted by the FEA.
Last Edited by STME58 on Feb 16, 2017 8:42 AM
Likewise I'm not dismissing the observation that bending breaks reeds, but I don't see how an increased breath force/bad technique is actually applying stress to the reed.
I would have thought that GOOD bending technique would stress the reed more, as you'd get more reed amplitude = nice loud, resonant sound = more metal fatigue.
Last Edited by MindTheGap on Feb 16, 2017 9:43 AM
I get what you are saying about maximum stress occurring at resonance, and that bending necessarily moves the frequency away from the resonance of the single reed, thus likely reducing the amplitude of the vibrations. Due to Hooke's law, the stress in the reed is directly proportional to haw far it is flexed. If there were a way to see the reed as it vibrates, you could calculate the stress in the reed from the shape it bends in to. The analysis of a reed as a cantilever beam is pretty straight forward. You can see from the analysis I posted previously that the larger the pressure difference from one side of the reed to the other, the further the reed bends. Any vibrations that occur are likley to bee from the starting point of the reed as deflected by pressure. The larger the breath force, the larger the excursion of the reed as it vibrates.
This is only a small part of the problem though. We have multiple reeds and a column of air and other things that are also vibrating. The volume of the sound you hear is not due to the amplitude of the reed vibrations, but the amplitude of the pressure wave generated in part by the reed vibrations. If you set you oral cavity such that you get a Helmholtz vibration set up at the resonant frequency if the reed, it is quite conceivable that the amplitude of the resulting pressure wave though the reed with a small excursion of the reed, could be larger than the amplitude achieved without the Helmholtz vibration and a larger reed excursion. The problem is complex and interesting. In other words it is possible that a player could have a lower volume and yet stress the reeds more than a player using another technique.
Last Edited by STME58 on Feb 16, 2017 11:35 AM
I think I'd need some convincing that a large resonance in the mouth doesn't give rise to a big deflection of the reed. My experience with a dan moi (or however it's spelt) where you can see the reed, is that when you get resonance you get a louder sound and the reed goes wild. It's coupled.
Last Edited by MindTheGap on Feb 16, 2017 10:55 AM
I need some convincing too. I am just presenting hypotheses, and methods for testing them, to try and explain why novice players break reeds often and pros tend not to. I break reeds far less now, than when I first started, and I think the acoustic volume I am able to get out of the harp is higher now than when I was a novice. Both this experience, and my understanding of how a harmonica makes sound (more like a siren than a guitar string), indicate that there are other factors in how loud the harp sounds than just how hard you blow. My understanding of mechanics of materials leads me to believe that stress in the reed is closely related to how hard you blow (or suck). Of course if you are blowing hard and have an excitation component at the natural frequency of the reed, you will get a larger reed excursion than if that frequency component is away from the reeds natural frequency. If you are at resonance, you could reduce the input energy to keep the excursions lower. This is what I expect a skilled player does.
It would be cool to get videos like this along with sound pressure level measurements to see if the volume of sound can be somewhat independent of reed excursion.
Last Edited by STME58 on Feb 16, 2017 12:00 PM
58, I don't understand why you use a beam as a spring (reed) bends at the same rate. the tip bends the same distance as the middle which is the same distance as 3/4 of the down the spring that is why it brakes at the place where it does't move.
I see a lot of bad reeds. Reeds in various stages of failure. Sometimes I see reeds which are flat, not cracked, still have some springiness but I can see a definite crease in the reed. I tend to take it as a sign the reed has been attacked 'too' forcefully and physically 'bent' (as distinct from playing a bent note). I'm not sure why it would still bounce back in that case though, as I'd expect the spring to be compromised. I've never seen this in a steel reed, always brass. Anyway, I suppose I've looked at 600 broken reeds and I probably have seen this 30 times. mostly reeds are visibly cracked.
I guess when I'm repositioning reeds I am taking care to not crease them. These (assumed) wild attacks which crease reeds could be simply resulting in a slight change to the offset which remains playable but has changed the pitch of the reed and created a weakness which will soon progress to failure. Practically, I just replace the reed
Right, that's a least a viable explanation for the failure - wild attacks. I like the name too. Has anyone experienced it happening? I'm going to try it - blow into an old harp as hard as I can. Or draw, I guess as it's the draw bends in the frame.
Last Edited by MindTheGap on Feb 16, 2017 3:30 PM
Short answer, when we are learning harp we often attack it with too much force, with fatigues the metal faster, leading to reed failure.
Long answer, ironically, from my experience, better harps can actually be easier to blow out. They are more airtight and use thicker reeds which worry faster (but sound better) than cheap reeds. (Think how many times you can fold a penny in half and unfold it... doesn't work well, but you can fold a sheet of aluminum foil and then unfold it). (Engineers, maybe you can explain that part better.) Airtight also means when you blow harder even more air gets to the reed, so it gets hit with more force.
It gets less common as you learn how to relax and play gently. (Or you increase your harmonica budget!) Save broken harmonicas. Even if you don't want to bother replacing individual reeds having a spare comb when a comb cracks saves money, and if you have two C harps and the 4 blow goes on one and the 5 draw goes on one you can use the remaining good reed plates. They are also good to practice minor tweaks on, like opening the backs up or changing the gapping.
I had a cheap Piedmont. It didn't sound great but even trying to blow as hard as I possibly could it was hard to blow it out, but that was because it was leaky and most of my air was being wasted.
MX, I am not sure I understand your question, but I will try to explain the analysis a bit. The equation for the shape a cantilever beam takes with a constant pressure applied to one side can be found in many engineering texts and is in the middle of the analysis under the heading "Deflection along beam" and starts with y(x):= just below that is a graph created by using this equation. This is the shape the beam would take with a higher pressure on one side than on the other. This is a curve along the whole length of the beam.
The colored image of the reed is from a model I made using dimensions of a Seydel reed and subjected to a Finite element analysis. This modal analysis shows the first mode of vibration. I trust the analysis because the first mode frequency calculated by the analysis matched the sounding frequency of the reed. This is a slightly different curve that what the beam would take under constant pressure. The actual shape the reed takes when played is probably in between these two.
Surprisingly the reeds don't typically break right at the root.Both the broken reed I photographed, and the FEA modal analysis indicate the failure point is a ways out from the root.
BTW I understand, from the little there is written about harmonica physics, that the reed vibrates because of vortex shedding NOT from air forcing the reed to swing open, then shut like a sprung door, as I've often read on MBH as the 'common sense' explanation. It's the same thing as wires whistling in the wind. Not sure how that effects your model, but I think the assumption of a constant pressure along the length is a simplification. It may be a valid simplification though - after all that is where the reed breaks!
Nacaron - I'm trying to draw the distinction between two quite different things which you've bundled together there...
1. Metal fatigue is caused by repeated, often small flexing. The kind of flexing that wouldn't pull a reed out of shape by itself.
2. SuperBee is hinting at a force/impulse/attack that is so strong it DOES bend the reed right out of shape.
I would suggest it is more likely that a 'good player' with lots of resonance will make that reed vibrate strongly, although with more efficient use of air. That would mean lots more flexing of the kind that results in metal fatigue. (For the record STME58 has suggested that possibly the 'good player' gets big acoustic volume while the reed doesn't move much.)
Last Edited by MindTheGap on Feb 17, 2017 12:36 AM
So, to be clear, I'm questioning the often-made statement that metal fatigue is the culprit. Or is always the culprit. It that were the case I'd expect the gently-played harp to suddenly fail at some point, whereas from what I read it seems to be about 'incidents' when there's been some hard playing.
I'll try and 'bend' a reed with breath force and report back. I mean properly bend it.
Last Edited by MindTheGap on Feb 17, 2017 12:39 AM
Here's another thought. Metal fatigue is highly enhanced in a corrosive environment. Beginners are often reporting that they get a lot of spit in the harp, and that improves over time as they improve. Less spit = less corrosion = less metal fatigue = fewer reed breaks.
MTG, it may be that the metal fatigue creates weakening that a catastrophic force then uses to finish the reed off.
One thing I have wondered about is the difference between steel and bronze reeds. I'm not an engineer, but from what I've read steel can undergo limitless numbers of bends without suffering from metal fatigue- as long as those bends stay under a certain flexing threshold. Bronze fatigues a small amount with even the slightest flexing. From anecdotal evidence (although it's a fairly large amount of it) hard players seem to blow out Seydels fairly quickly, but for light players they seem to last forever. My theory is that they are staying mostly below that threshold. Maybe bronze can handle more total cycles of stress? That's a question I've raised a few times.
Playing around with wires, bending paperclips, etc, it seems that fatigue occurs faster the farther you flex it, and my understanding is that thicker metal worries faster if you bend it X number of degrees than thin metal does when bent the same number of degrees, and my understanding (which may be wrong) is that it's because the outer layers actually have to move farther to achieve the same degree of bending on the thicker pieces.
As for the opening door analogy, I've used that before, but not so much to describe how the 'door' is being moved but to describe how it being moved creates a specific sound- how a free reed creates sound compared to say, a string being plucked.
nacoran - yes could be a combination of things. Re the fatigue threshold, yes that's right but we don't know what regime we are in with harp reeds.
There's no hard information about any of this, which is why I don't think anyone should say it's definitely this or that - which they do.
Re the 'door' metaphor, I think it's useful in the way you use it there, the danger is people taking a simple idea and extrapolating way out of scope.
Last Edited by MindTheGap on Feb 17, 2017 12:27 PM
This discussion has got me thinking about this again. In doing a bit of searching, I see that Pat Mission has updated his technical links. These look interesting and I will go thought some of them when I get a chance.
My experience with failing reeds seems consistent with a fatigue model. I will first hear that a reed seems just a bit out of tune, it will slowly get flatter until I am compelled to tune it. The first tuning will last a short while, but soon it will be flat again. After the second or third tuning the read will go flat in minutes after a re-tune and I know it is time to replace the reed. Sometimes at this point, just plinking the read will cause it to break off. That was what occurred with the Seydel reed I photographed.
I really agree with the comment by MindTheGap, "There's no hard information about any of this, which is why I don't think anyone should say it's definitely this or that". There may be some recent research that is shedding more light on this, perhaps even in the link I posted above. The University of New South Wales department of Music Acoustics does not have anything on the harmonica, but they have some very interesting studies on other instruments that might be good models for research into harmonica. Unfortunately, I have lost access to high speed cameras, CAD modeling and finite element analysis, so I can't play in this arena right now.
@mlefree Yeah, it probably a good idea not to file a reed at the base--I like using a polishing bit if I am going to lower pitch that way, but at the Hohner factory, they are trying to produce as many harmonicas per hour as possible. I think the Suzuki harps are laser tuned.
Gnarly, how different is removing material with a laser from removing it with a file, from a stress riser point of view? Have you seen anything from Suzuki on why a using a laser to reduce the stiffness of the reed to tune it might cause less of a stress riser than a file? I was planing on modeling some tuning scratches to see the relationship between depth and frequency as well as looking at the difference in longitudinal, transverse and angled tuning scratches. Unfortunately I was laid off and no longer have access to CAD and FEA. Perhaps when I find a new engineering position I will have access to these tools again.
I have not had any conversation with Suzuki Japan regarding this issue--but the word on the street is that our reeds seem to last (of course, YMMV). My personal feeling (and bear in mind that I play a lot of SP20s!) is that it's considerably worse to file a reed than polish it! I still use files (and sanding wands, and I have a couple of tools Wally Peterman made me from hardened steel, one of which is a draw scraper) but try to polish when it is expedient. Greg Jones swears by polishing, so much so that I suspect he is Polish!
Polishing is well know to increase fatigue resistance. It would make sense from a mechanics of materials standpoint to polish the area of the reed that bends the most to a mirror finish until you are at, or below the desired pitch. If you went below, removing some material at the tip would bring it to pith without impacting fatigue resistance as there is very little stress at the tip.
Hi 58, a cantilever beam is not a spring as springs have memory beams don't. "higher pressure on one side than on the other" this is not how a spring moves if the spring is bent down then the top surface is stretched were as the bottom surface is in compression that is why the EQ you used gave you the wrong info.
mx, you are quite correct that when a flat piece of material is bent, the convex side is in tension (stretched) and the concave surface is in compression. Somewhere along the thickness of the bent piece of material is the neutral axis where the material is in neither tension nor compression.
I am puzzled by your statement that a cantilever beam is not a spring, and I wonder if you are using the term spring differently than I do. A spring moves proportionally to the force applied to it. It does not matter if the force is a pressure, a point load, or a moment. If the elastic limit of the material is not exceeded, it will return to its original shape when the load is removed. This is true of coil springs, cantilever beams, leaf springs, and other types of springs.
The equation I used was for a uniformly loaded cantilever beam. This equation is valid no matter how the load is applied. It could be a snow load, or an air pressure load, but if it has a uniform pressure on it, the beam equation is quite accurate at predicting the shape the beam will take and the stress in the beam. If you want to argue that the pressure load on a harmonica reed is not uniform, but varies in time and with the position on the reed, I expect that is correct. the uniform load is just a first approximation and is undoubtedly an oversimplification.
mx - "springs have memory, beams don't." Either you have some deeper understanding or you don't know it about it all. Can you elaborate? Thanks.
STME58's model may be a simplification (well, models always are) but it does seem to give a reasonable result. From observation that's where I'd expect the high-stress region to be - simply because that's where the metal is bending most.
The more interesting thing would be to use it to try to compare failure times with experiment. But that's obviously a big undertaking.
Gap, Structural Steel: used in the fabrication of buildings, bridges and other structures, mainly low carbon steels, made by open hearth process. Spring Steel: steel suitable for manufacture of springs, medium and high carbon plain steels and quaternary alloy steels, made by electric processes mostly. And I agree that doing the experiment is better than the dry lab method above. MX
.JPG)
