That was a great thread on the phenomenon known as the marshmallow zone a few weeks back! I think it was Jeff (Unright Tooner) who started it, and it spurred a lot of good comments.

Now that I've switched to impact tuning completely (CyberHammer on verticals, tap method on grands), I've noticed that there really is no marshmallow zone. Sometimes there's a slight amount of twisting that needs to be undone, but it's far less than what I used to perceive using a smooth pull method. I think sometimes we tend to make things harder than they need to be. With unisons, I think the least amount of manipulation that gets the pin and string where it needs to be is best.

Loren:

I do not think any hammer technique can make the MZ disappear. It is still there, but the technique is using it’s characteristics to an advantage.

For instance, a particular piano that I tune regularly had more spring in the pin than rendering friction. It was difficult to tune stably. So I decided to increase the rendering friction by screwing in the pressure bar. For some reason I thought it would be better to leave the top octave and a half where it was. The result was that a smooth pull worked very, very well. Little overshoot or banging was necessary, except for the top octave and a half. I will be screwing down that part of the pressure bar next time. In fact I am trying to convince myself to do this with more pianos.

The point that I am making is that the size and shape of the MZ was changed so that a particular hammer technique matched it and made it a non-issue. That is what Ken Burton’s great book Different Strokes is all about.

I have no reason to doubt that you are accurately reporting your experience with impact hammer technique. I see little self aggrandizement in your posts. I have not tried an impact hammer, but others have and reported different results.

Here’s how I am looking at stability lately. The difference between the ideal position of the pin and the actual position of the pin (in cents change of pitch) must be less than the amount that the bearing friction can hold between its speaking and non-speaking parts (in cents change of pitch) when played loudly or when there is a reasonable change in the environment in order for a string to be stable. Now if the smallest amount that a particular hammer technique can move a pin is larger than this amount, then you will not be able to always have a stable string on the first try. But even if a particular hammer technique can move the pin in small enough increments, the tuner must also be able to discern when the pin is close enough to the ideal position and then render the string if necessary. If he cannot discern this, then another hammer technique would be better, even if it may take multiple tries.

And let’s not forget what the expectation for stability, or for just tuning accuracy, is. Perhaps (and I really don’t know, I am not exposed to other’s tunings very much) a certain hammer technique always works for a particular tuner because it meets their expectation for stability, not that it is always the most stable technique in a given circumstance.

Myself, I am very picky about my temperament; probably to a fault. I find that I must make every note in the temperament very stable in order to have the RBI progression and the color of the SBIs to my liking. I have to use various hammer techniques to do this. Another tuner may feel that progressive CM3s is all that is required (and they may very well be right) and have a hammer technique that works regardless of the piano.

Anyway, I am very glad you are bringing up this subject again. Maybe there will be some tips come out of it about tuning the seventh octave.

What about the sixth octave? I think that section is even worse!

OK, then what tips do you have for the sixth octave? I'll just transpose up an octave!

I think the biggest culprit in stability (or lack of it) is flagpoling, and that any technique that minimizes it will lead to more stable tunings. For me, anyway, impacting or tapping/bumping provides the advantage of recoil, pretty much straightening the pin quickly and stably. Some very gentle nudges back and forth without a change in pitch tell me the pin and string are set.

My marshmellow zone was reduced dramatically when I got a Fujan.

See, that's my point. If different levers and techniques change the perceived marshmallow zone, then that tells me that it's more a perception than anything.

When I put a coat on to go outside, I do not percieve the cold as much. It does not make the outside temperature warmer because I don't percieve it as much.

No, but it does diminish the relevance of the cold and its effects.

*edit* Good snow tires on your car don't reduce the amount of snow and ice on the road, but in the same way, they reduce the effects of both and their relevance on your ability to control the car.

Car tires may be a good analogy. All season radials are good all-around tires. But there can be better choices for specific uses, like off-road or racing. Likewise there can be better hammer techniques than impact for specific uses, like difficult pinblocks or ultimate stability.

This is like deja vu all over again, especially with Keith involved.

OK, let’s examine flagpoling. Flagpoling exists because the plane that the torque is applied is not in the same plane as the resistance to the torque.

Now please explain how impacting “minimizes” flagpoling itself, not just its effects.

If I find myself frustrated with any “Voodoo Engineering” on this subject as I was a while ago and which precipitated me leaving this Forum for a time, I will try to consider it to be an “Indulgent Mystery” and just exit the Topic.

On a vertical, you don't have the leverage necessary to flagpole the pin. You're holding the lever virtually at the pin.

Loren:

The force is still assymetrical. It does not matter where you hold the lever. The lever only has one handle.

See 'ya Monday....

Jeff, depending on where you grab the handle, you can gain more or less leverage. You know that. If this were not true, extension levers would be useless.

Try it. Using your traditional lever, place it on a pin. Grasp the top if the lever and try flagpoling the pin. Now grasp where the head meets the pin and try the same thing. Are you saying there will be no difference?

I've suspected for some time that my tuner was also a Jedi.

Loren:

Two things come to mind when you “choke” up on the handle of a traditional tuning hammer that would have little significance when considering an impact hammer.

First, the amount of bend in the handle will be less. Those that prefer a stiff lever see it as a way to avoid this problem. I don’t think it makes much difference because the pin twists without actually moving much more than the lever bends. Second, since the handle of the hammer is at a greater than 90 degree angle to the pin, when you “choke up” the plane of the force and the plane of the resistance are a bit closer together and, yes, would reduce the flagpoling effect a bit. That is a reason that I am considering tuning the high treble of grands with the left hand so that an extension is not needed in order to reduce the flagpoling.

But neither of these effects should pertain to an impact hammer. The force still is applied by the center of the mass of the hammer, regardless of where you hold it, unless you are “muscling” it and not using it as an impact hammer. But in all fairness I suppose it would be impossible to not “muscle” an impact hammer a little bit.

Here is a link to a PTG Pianotech post about hammer technique that I have browsed through but have not studied yet. I think you will appreciate it:

http://ptg.org/pipermail/pianotech/2011-January/033943.html

Jeff, I'm just asking that you try it. Put the lever on the pin of an upright, with the handle at 12:00. Grasp at the top of the handle and try pulling down (as in bending the pin downward toward the floor). Now grasp it at the head and try the same thing. Is there a different in leverage and your ability to bend the pin?

Loren:

Of course there is a difference in leverage. And for the same amount of force, there will be less flagpoling. But when the force is increased to the point that the pin moves, the amount of flagpoling will be nearly the same [Edit: When actually tuning, not just flagpoling a pin as in your example]. Any difference, as I pointed out, is not due to the length of the lever, but the difference between the planes of force and resistance.

But aren’t you changing the subject? I thought this was about impact hammers.

But to deliberately change the subject, can I ask you to try something? This is not to prove a point, but answer a legitimate question. I have wondered just how small the average change in pitch is for each pin movement that an impact hammer makes. I am not sure the best way to do this. Maybe if you give a string a movement up, measure the pitch, then move it up with, say, ten more small impacts. (Now I mean impacts that you know have moved the pin. I image that a pin could be tapped without actually moving and I would hope that the difference can be felt.) Then measure the change in pitch and divide by ten. I realize that this could be different for different pinblocks and in different parts of the scale and the phase of the moon (did you see it this morning!) but just to get an idea.

Milli-cents. (is that a new word?) I can impact with the CyberHammer in the sub-cent range.

I didn't change the subject. You asked me to explain how using an impact lever lessens the chances of flagpoling over a traditional lever, so I did (using a traditional lever method as my example). And I still maintain that because of where the lever is grasped, there is less chance of flagpoling because the leverage required to do it is, by your admission, decreased! The point being, the pin is rotated with less possibility of bend.

Lol, looks like we're dug in pretty hard here....I have a feeling we're going to have to agree to disagree.

But isn't the idea of using an impact lever that the momentum of the weight in the handle moves the pin? If so, then what difference does it make where the handle is grasped? If not, then what is the idea of using an impact hammer?

There is no bending before breaking the static friction of the pin. As there would be if you were tuning lefty with lever at 11:00 or righty with lever at 1:00.

OK, why do you think there is no bending, because of a bend-o-meter or the results? Obviously it is the results (as you perceive them...) Again, that does not mean that the pin does not bend, as in flagpoling. Putting on a coat does not make it warmer outside.

It is too bad that we cannot get beyond this and discuss what an impact hammer actually does.

Jeff, Using a traditional llever, you can FEEL the pin bend and HEAR the pitch of the string change before the pin turns. You don't hear/ feel that?

If the pin suddenly rotates all at once, you neither hear nor feel it. Because you are not pulling in a lateral plane.

OK, you neither feel the pin bend not hear the pitch change before the pin turns with an impact hammer. That does not mean it rotates all at once (how could it, it is still flexible). I have no idea why you think a lateral pull (whatever you think this means) does not exist just because an impact wrench is used.

Maybe this link can help us get on the same page: http://en.wikipedia.org/wiki/Impact_force

I think the confusion is about 2 kinds of flagpoling.
An upright impact hammer eliminates one of them, the grand impact hammer I've seen removes both types.

Imagine a normal hammer on the pin at 12 o' clock. If you
push the hammer away from you the pin will not turn, but just bend upwards (flagpoling 1). If you push the hammer (hold it near the top) in the plane of the pinblock, say to the right, the pin will turn clockwise due to the torque but also bend to the right because of the force (flagpoling 2).

If you try to turn the pin you can eliminate flagpoling 1 by being very careful to apply force only in the plane of the pinblock, but not flagpoling 2.

With an upright impact hammer you throw the handle left-right, and the moment it hits it will apply an impulsive torque AND an impulsive force (left to right). The latter still causes flagpoling 2, but flagpoling 1 is eliminated by construction.

The grand impact hammer looks like a 'T' and when you "throw" it, it has only angular momentum, but no linear momentum and the impulse is pure torque.

Harpsichord tuning hammers look like a 'T'; when you turn the pin you apply opposite forces to both sides of the upper line of the 'T', resulting in no force, just torque and hence no flagpoling (if operated correctly).

I think Jeff was talking about flagpoling 2 and Loren about flagpoling 1, hence the confusion.

Kees

Thanks Kees, I think you nailed it!

Well, Loren did use flagpoling 1 as and example, but I don’t see where this type of flagpoling would occur with proper hammer technique, unless it was done intentionally. I sometimes do if the rendering is extremely poor. And when the hammer is in line with the string, flagpoling is not an issue, anyway.

Jeff, why don't you design an electrical impact hammer? Something small with a flywheel. You just put it on the pin and press the button. Optionally plug it in the USB of your laptop to let your ETD control it; hit the key ...... tap tap taptap tap - done. Anyone can do it! Sell it at Wallmart for $39.99. Mass produce it and miniaturize it and put one on every pin on the next generation of auto-tuning Chinese piano's.

If it works you should be able to retire on it. Probably won't make you very popular here though.

Kees

I do not think it would work well enough to suit me. But why don't you try it?

So the question remains....if one can set a tuning pin easily and confidently and without guesswork (regardless of technique or kind of lever), does the MZ still exist? I still say it's a perception thing, and that it's for that reason that for some it's big, for others, it's small, and for still others, it's non-existent.

With the grand CyberHammer that Kees used as an example, flagpoling 1 & 2 are both eliminated because the throw is pure spin.

Ah, then with the upright CyberHammer you do understand that flagpoling still exists?

Here's how I look at it. If the hammer technique matches the MZ, then the combined effects cancel each other and it is not perceived. Of course it is still there, and another hammer technique will prove that it did not disappear.

But do we really want it to not be perceived? If we do not know what it is like, how can we know the best technique to use? That is something that bothers me about one-size-fits-all hammer techniques. How do you know that after three test blows that the fourth one won't change the pitch?

The only analogy I can think of right now is what I often do when I am not sure what the road conditions are really like. Starting at a slow speed, I will give it the gas to see if the wheels spin, then brake hard to see if I skid. Then I know how fast I dare go.

With the vertical one, flagpoling 2 in Kees example exists, but 1 is eliminated, yes. What I was concerned with was 1, and that's the one I was stickling on since it influences tuning. It would be the same flagpoling you would get pulling on a lever on a grand if the lever were set to say, 2:00 (you'd be pulling the pin back toward you, equivalent of pushing up on the vertical.

Loren:

Maybe I am mistaken, but it seems that you think it is inevitable, when using a traditional hammer, that the plane of applied force will not be parallel to the pinblock. That the tuner will always apply force somewhat towards themselves in an upright and somewhat towards the ceiling in a grand. Is this what you think?

No, not at all. I just think the chance of doing it is greater and easier to do. BUT...on an upright, unless the hammer is at 12:00, a clockwise pull from say 2:00 is going to also pull the pin downward toward the floor. On a grand, pulling toward you from 2:00 will certainly also bend the pin toward you. No lifting toward the ceiling is necessary. And these are factors that I feel come into play of a so called marshmallow zone.

If when tuning a grand, a tip extension is used and the handle is over the riser and parallel with the strings, then the front/back bending of the pin wouldn't come into play.

*edit* In all these examples, I'm referring to what Kees called flatpoling 2, which occurs even when pushing or pulling in the plane of the pinblock due to the force of the push.

I'm more of a theoretician than an electrician.

(Of course I'm serious.)

Kees

Loren:

OK then! I just needed to make sure before I went further.

It should be obvious that the greater the pin torque, the greater the flagpoling. Keeping (for now) with the idea that it is preferable to have an imperceptible MZ, consider two different pianos with the same rendering friction, but with different pin torque. A different position for the hammer could be chosen so that the flagpoling makes the MZ less perceptible. This is the idea put forward in this link that I posted earlier: http://ptg.org/pipermail/pianotech/2011-January/033943.html

But what if there is no hammer position that makes the MZ manageable. Couldn’t we just add additional (or subtract excessive) flagpoling by making the plane of force nonparallel to the pinblock? Of course we can! That is, we can with a conventional lever. It might not be so easy with an impact hammer. But we can still choose our hammer position with an impact hammer. It may even be possible to choose a position that would be too awkward with a conventional hammer.

Hopefully we can wrap up this whole flagpoling thing and get into what the impacting actually does.

Oh, btw, I got curious about what I asked you to try and tried it myself using a jerk style on a friendly pinblock. I counted ten “ticks” that were felt more than heard and changed the pitch about 4 cents according to my Korg. I was able to set a real good temperament with this hammer technique (I would not use it on all pianos) but usually had to take more than one try to get just the pitch I wanted.

Haha,

well, I've been using an RCT rather extensively for the last half year or so, and an interesting thing is that my stability went bad when I relied on the visual tuning aid. I realized that I do a lot of almost subconscious nudging in the upper octaves when I tune aurally, almost like body reflexes, reacting to the sound. When I went by meter, those nudges where not there, and I didn't set the pins nearly as well.

So I started to try to observe myself working aurally, and that wasn't easy. But I realized some good things. But this is a long story, so instead of hijacking your marshmallow zone I might get back to it in a dedicated stability thread.

I'm not arguing any flagpoling or "voodoo engineering" . I told what I thought were the reasons a Fujan imparted more control over the pin. I can prove what I said with vector analysis which I passed at UCSB. And that weren't no hick college.

The difference now is I don't care because when I wrote that, I didn't own a Fujan. I have been tuning for over 12 years now and I can tell you that the marshmellow zone you talk about, that I experienced and wondered why I couldn't fine tune easily, completely disappeared in the first month of owning the Fujan AND my tuning times went down by 1/4. It has paid for itself. Thank you, case closed. Getting the pin to settle where I want it to is not a problem. I don't need any extra friction from the pressure bar. Besides, that only delays the equalization of the string tension long enough for you to get out the door. When you come back two weeks later after some heavy playing and check your work, then you find out how good your stabiity is. Try it sometime with one of your nicer clients.
Ohh, I have owned a student lever, a wonderwand, and a Shaff extension lever and now I also have a Hale which I got in an old doctors bag that had the exact same stuff in it as my doctors bag. I can't say the old guy was copying me, he died 6 years ago

Keith:

Here is how I check my stability. At the next tuning I see how consistently the strings with the shortest non-speaking length in each unison are out of tune when compared to the strings with the longest speaking lengths.

The vector diagrams in your Voodoo Engineering Topic made no sense. I learned vector diagrams from Charles L. Sauerbier's book, real world stuff.

It's not my Topic, but I am really interested in what you are talking about. I do not think there is nearly the amount of analyzing of hammer technique as there is in temperament sequences. Let me share some observations that may entice you to share yours.

When tuning an RBI directly in the temperament I do not hear the slight changes in pitch that I can when tuning an SBI. I think these slight changes are the key to stability, whether tuners realize it or not. On a particular piano I followed behind what I am sure is an ETD tuner because I know something about the store that provided the tuning. The piano had dropped in pitch a bit, the unisons were out a bit as would be expected, our temperaments were identical as far as I could tell, and there was little difference in the stretch except somewhat in the ends of the piano. What I thought was interesting is that the left string was out of tune in a less consistent manner than the middle and right strings. I think that the left strings were tuned to the ETD and then the other two aurally and the difference could be noted in the stability.

But there is something else that I have been thinking about. First consider the temperament octave. The strings are a decent length, the hz to cent ratio is low and yet setting a string stably to within ½ cent is often a chore. Now go up a couple of octaves. The strings are shorter so the same amount that the pin is turned produces a greater change in pitch. The hz to cents ratio is greater so that any beating in a unison is more noticeable. I hate to admit it, but I am sometimes thankful for false beats in the upper treble to hide problems with difficult pinblocks and string rendering.

Which brings me to what I have been focusing on lately: string rendering. Do we really want strings that render very, very well? The better that they render, the more accurately the pin must be placed. And some pinblocks are terribly jumpy and some pins are terribly springy. And when the environment changes a little bit do we want the strings to immediately render and have the unisons go out which is inevitable because the nonspeaking lengths are different?

It’s nice to think that we can position all pins within, say, 0.5 cent of the ideal position but I don’t think that is the reality of the situation even if it seems that way. I think the reality is we have a personal window, especially in the upper treble, where the pin is close enough and then we render the string to get the right pitch.

I expect there will now be some replies from those that have a hammer technique that makes the MZ imperceptible; that their technique causes the pin to be in the ideal position, the string is properly rendered, and the torque in the pin is in balance with the string tension. Anticipating this, my question is: How can you know that this is the case unless you can perceive the MZ? How do you know that the pin could not be in a better position? But because string rendering friction keeps the string at pitch, the way you judge your hammer technique is simply that the piano stayed in tune, not that the pin and string was in the center of a zone that you do not perceive. But the reason it stayed in tune was that the rendering friction was great enough for the hammer technique that was used.

A "Zone-0-Meter" might come in handy here. Got one?

*edit* Sorry Jeff, couldn't resist! :p

I think this MZ zone and flagpoling is not caused by the pin twisting or bending, but by the wood of the pinblock deforming.

A possible explanation for no MZ when using impact hammers is that wood has very high viscocity so it will not deform under a wrench impulse, but will under s steady wrench.

However I don't know if wood viscocity is actually so high.

Another possible explanation I can think of is that when you steady pull and play a note you hear the sound while pulling. Obviously. But with an impact hammer you listen after the impact, so you never get to "hear" the MZ zone if it's still present.

Kees

You know what, Kees, great point, re listening after impact instead of while turning. And without listening while turning, any perceptible MZ disappears.

Halleluiah! We are finally getting into what an impact hammer actually does.

I think it does a number of things by applying force over a very short time. And not having actually used one, I get more curious as read posts from those that do. Like is there a difference between swinging a long slow arc and swinging a short fast arc? I would think there is.

Anyway, I do have experience using an impact type technique with a conventional hammer like when using a jerk style.

The first difference I notice from a smooth pull is that the twisting torque that is generated is released at the same time as the pin’s friction breaks free. This results in the pin moving a much smaller amount than starting at the same place with a smooth pull.

The other thing that I am pretty sure happens is that the “tick” that is felt or sometimes heard is a separate impact in its own right. This tick is a sound pulse that I think travels down the nonspeaking length of the string and helps the string render at the friction points.

Now this is not to say that an impact hammer will always leave the string and pin in the best state for stability. I do not believe that “one size fits all”. What I do wonder is if there can be an adjustment made by the amount of swing and the force applied, as I mentioned earlier.

But what will always bother me about the technique is the lack of feedback. If you don’t know what the MZ is like, how can you know where in the MZ the string and pin are? Unless, of course you have a “Zone-o-Meter”

I'm steadily improving with my CyberHammer.

What I notice regarding swing distance and force is they are variable and independent. That is, I can vary how hard I flick the lever, and I can vary the length of the arc it travels through until impact; both play into the overall technique. Coming from smooth-pull originally, I sympathize with your comment about the lack of feedback. But CyberHammer impact tuning is a different skill, with its own set of input and feedback mechanisms.

Starting with a relatively small arc and throw force, I gradually increase both, to gauge how a pin will react. I carefully write "react," and not "feel," because you don't feel the pin at all, of course. But by rapidly changing the two, you get enough information to proceed.

And now that I've written that, it occurs to me there is another piece of feedback with a CyberHammer, it's the "rebound response." As I described above, when I rapidly tap, tap, tap with the CyberHammer, varying the arc and force, I'm continually feeling how the hammer bounces off the impact. There is a change in the feel depending on how far the pin does or does not move.

Hi Jim,

It is hard to explain the feel of the CyberHammer. Like you said, there is a whole different feedback that you learn to sense.

Does anyone know how the friction compares between (1) the pressure bar/agraffe and (2) the V-bar?

After all there 4 segments of string that can have different tensions.

Kees

If the pin does not move the coefficient of restitution (physics term for how much bouncing back there is) will be a lot larger than when the pin moves. If you throw a rubber ball against a concrete wall it will bounce back, but if you throw it against a marshmallow wall it won't. This is what you are feeling.

Kees

Rather: if you throw a hard ball against a wooden wall with fixed planks, it will bounce back, but if you throw it against a loose-standing wooden plank, the ball will transfer part of its energy into movement of the plank, and so won't bounce back as much.

Also a good analogy, but in this case the ball's kinetic energy goes into kinetic energy of the plank (you are describing an elastic collision), whereas in our case it goes into heat (due to the friction) which you could see as a deformation (i.e. an inelastic collision).

More (or at least equally) interesting I think is if we can bypass the marshmallow zone with impact hammers. If you pull up smoothly, first the wood around the pin gets deformed and when enough stress had built up there will be a slip at the pin-wood interface. But now the wood is twisted in the wrong direction, which has to be corrected. It remains to be seen if with an impact you can avoid this wood deformation altogether and end up with a shifted (rotated) pin without having deformed (twisted) the wood around the pin.

I have a hunch that the answer is "somewhat" depending on the detailed material characteristics of the piano in question.

Kees

This might very well be the most stupid question of this thread, but I'll go ahead anyway:

I use a Schaff and a Fujan. The Schaff for its weight, on uprights, and sometimes on grands when it feels natural to use a tool on the pin, if you know what I mean.

The Fujan is more like an extension of my body, like manipulating the pin with my fingers. I can bend if I want to (like with the Schaff) I can step up/or down in micro-increments, really controlling the pin in the base of the block, and I can hit it fast in an impact manner.

My benefit from the Fujan is in what I mentioned in the middle of the paragraph above - the micro-increment feel. Like bending, but micro-steps instead. When I use that, I know where I come from, where I'm heading, and where the tension is at. Practically no MZ problem at all (just like Keith, if I remember correctly, stated earlier).

If I can mimic all three methods, why would I go for a hammer specialized in realizing one of the extremes? I've seen Bill B. work too, he uses his hammer for all three methods mentioned - bend, nudge, impact. To me, it seems like a rigid hammer with a fairly long shaft indeed might be able to do all three of them - if your hand technique is good enough.

Comments?

PS Jeff: thanks for your interest in what I wrote earlier, I will soon start a thread about my experiences of stability in aural tuning compared to ETD tuning.

Patrick, I'll make a few comments, as a learner in the CyberHammer style. I was never a "jerk" tuner, so I can only go by what I observe, and that is the half-dozen or so people I have been able to observe using the jerk style do it with a mix of smooth pull, jerking, and otherwise manipulating the lever. So, as you were saying, I think that's definitely a toolbox approach, and more power to those who use it.

With the CyberHammer, it's different -- it's designed for impact tuning only, and as a result you end up learning a different set of skills. Yes, you could still use it in the other modes of operation, but its design does not encourage it, particularly that 45-degree play being very awkward in switching directions while smooth pulling. When I feel I need to go back to my smooth pull technique, I pull out my Fujan. But many who have perfected the CyberHammer technique have encouraged me to keep at it, because just as with the finer details of smooth-pulling, so too there are mysteries to be decoded by practicing straight-impact technique. I don't think of it as being at one end of a spectrum of techniques, but a unique one on its own because of the position, and the manner in which the energy is applied, by twisting, which allows rapid and fine-grained switching of directions and variation of force.

I've noticed that my tunings so far are almost (maybe not quite) as stable as my smooth-pull technique, and I have been working at it for only a half-year or so. I also observe that I am improving more quickly than my smooth-pull technique did when I was just starting out. I got a tip on technique from Nate Reyburn at the last national convention where he held a CyberHammer mini-technical, and that was to try and make my last, small impact be in the sharp direction to attempt to leave a slight bit higher tension in the non-speaking length near the pin. The equivalent of what someone smooth-pulling might do by slightly flagpoling or twisting in the counter direction as their last motion. So I've been working on that.

Anyway, it's been interesting, and I am having fun learning to do it. If you find it not-fun and/or frustrating, then maybe it's not for you. For me new techniques and tools are somewhat of a "new toy." And, oh yes, I think I've mentioned before that the tuning position for small uprights is soooo comfy! My back and shoulder have never felt better after multiple vertical tunings.

Still not tuning enough grands to justify a Grand Cyberhammer purchase though

Jim, those grand CyberHammers sure look sweet. Expensive, though! But imagine, with its perfectly balanced design, the only force being applied to the pin would be pure torque with no bend. That would take some getting used to in itself!

The reason I brought up the zone-o-meter is, there really is no way to measure the scope of a piano's so-called marshmallow zone....or even its presence. We've already agreed that different techniques can eliminate it while others can exaggerate it. So it's a subjective thing directly related to the tuner's perspective.

Earlier in the thread, the comment was made that putting a jacket on doesn't make it warmer outside. True! But in the case of weather, we have a thermostat which measures the temperature. So while one person may be warm and the other cold, we can point to the thermostat and say that regardless of whether you are warm or cold, it's 30 degrees out. We can't do that with something like a marshmallow zone. This thread is working on the assumption that it exists, yet I'm not convinced. It seems to me that "the zone" is directly related to a tuner's technique and ability. If one can go from piano to piano, setting pins quickly and easily, confidently leaving a pin and string in place with no fudging around afterward and instead going to the next pin, does a marshmallow zone exist?

Hey, that can be the piano technician's version of "If a tree falls in the forest, and no one is around to hear it, does it make a sound?"

I do not see this as a subjective thing at all. The one part of the MZ, being the twist of the pin, has been well established. There have been demonstrations of tuning pins in a pin block with a pointer at each end. The torque left in the pin can then be observed.

And I routinely check what a piano’s MZ is like, and if I had an ETD I could measure it. I will bump a note low of pitch and then use a smooth pull to bring it to pitch without any overshoot and the hammer in a position to avoid flagpoling in a direction that will affect the pitch. Then I will apply a very small amount of reverse torque and bang on the key. The amount that the note is now low in pitch (In my case beats, but with an ETD it could be in cents) is one half the size of the entire MZ. If the MZ is small I will usually tune using a smooth pull with an overshoot about ½ an MZ wide. If it is large I may use a jerk style or I may use a hammer technique that reduces the size of the apparent MZ.

But what I find very interesting is that the MZ is larger when the nonspeaking segment is smaller. This is an indication of how elastic the strings are. And as I have posted in another Topic, the separate components of pin twist and string rendering can be deciphered by noting how far the pin twists before the pitch changes and the static friction is overcome.

I do like the tree in the forest analogy. Here is another application of it: “If a tree falls in the forest and no one could hear it because their walkman was too loud, did it make a noise?”

[Edit:] Better yet, "If a logger cuts down a tree and cannot hear it fall over the racket of the chain saw, did it make a sound?"

Are you saying the pin actually shears? I thought for all practical purposes the pin can be considered perfectly rigid, and any "marshmallow" effect (and flagpoling) comes only from deformations of the wood. That seems also Owen Jorgensen's opinion ("Tuning", p13-14).

Kees

or the version a lecturer once proposed: "If a husband goes around talking in the forest, and his wife is not around to hear him, is he still wrong?" :-D

Unright, The only way you will see it is if you are here in person. Then I can put the two different "real world wrenches" that I based the vector stuff on, in your hand, then you will see.

First off, somebody said my stuff was wrong because of the length of the pin,angle , blah, blah. They obviously have no knowledge of the most BASIC of mathematical princples.
The equation we are trying to prove or disprove is,
Schaff tuning lever = Fujan Tuning lever
Let's put the pin factors in the equation. We'll call angle, length, type of pin block, P1 and P2. Ohhhh, wait, we are using the same pin for both tuning hammers. So just P1 is needed. We have,
Schaff TL + P1 = Fujan TL + P1.
Geeee, now basic math says if we have something that is equal on both sides of the equation, we can remove it from the equation. The proof is simple, you can do anything to an equation as long as you do the same to both sides.
Let's subtract P1 from both sides of the equation... Get it?This is not voodoo mathematics to say that the tuning pin has no bearing on this equation.
Like wise, we determine the angle of the head. Yes we can buy a 5* angle and a 15* head. If you you were analyzing the difference of the two TL's, would you get two different angles?My suggestion was to make the vector analysis EASY and make the angle 90* or 0*, which ever direction you are looking at it from.
Now my vector stuff only shows the force applied to the head of the two different levers IS ROTATED BY 90*, That is all I was saying. It's not that hard to understand when you get the other stuff out of the way.
Back to the difference between theory and practice. I have used both these style wrenches extensively in the plumbing pipe fitting trades. Large valve handles can be either these type of wrenches or levers. I have both type in my tool box. I saw the difference immediatly when handed the Fujan.

The best way for you to understand is to have me hand you the wrenches. I have been pulling on wrenches for 48 years, longer if you count my erector set. The proof is in the practice. If I had never used anything but a Fujan, I wouldn't have a clue of what you are talking about... !!!!!

Only if the equation is linear and I don't think it is since the arithmetic sum of a pin and tuning lever is not really math.

More seriously can you point me to this vector analysis of yours? I'd like to read it and didn't find it on this site.

Kees

Try this for yourself: Place your hammer so that it is in line with the string and flagpoling will not affect the change in pitch. Move the lever without breaking the friction between the pin and the pinblock. Unless the pin torque is very low, the pitch will change. Obviously the pin is twisting. On very tight pinblocks this twist can be maintained without breaking the “foot” of the pin loose, and can be used to advantage in fine tuning.

Have you read Different Strokes by Ken Burton?

Here is a link to the Topic: http://www.pianoworld.com/forum/ubbthrea...ite_id/1#import

Please respond to that Topic and do not bring “The Cliffs of Insanity” here. Otherwise, I will just bail out. I am not even going to read Keith's last post.

Nothing you describe indicates the pin itself is twisting (i.e. shearing). It is likely just the wood near the interface that is deforming and the pin is just rotating.

Kees

Fair enough. I was thinking after my last post that what I described could also be explained by the wood acting elastically. But the sudden reduction in torque that is needed to twist the pin accompanied by a “tick” indicates to me that some sort of static friction is overcome after the pin is twisted so far. This, along with the experiment that I mentioned with the pointers, indicate the pin twists before the friction between the bottom of the pin is released by the pinblock. I am sure there are tables or equations available indicating the amount of twist various steels will develop under different torques.

The tick is just a slip at the pin-wood interface. This does not indicate a twist in the pin itself, rather the reverse.

Can you reference this alleged experiment with the pointers? It would have to use some very sophisticated pointers as any twist in the pin will be so small as to not be visible to the naked eye.

The tables of equations you mention do not seem to support the idea that the steel can actually deform, but I don't have the exact numbers.

Think of all the types of marshmallow zones you have encountered. What made these piano's so different? Did they use different pins or was the wood of the pinblock different?

Kees

Thanks. No need for me to reply, as it is too wrong to even argue about.

Kees

Doel:

OK, I agree that the tick is ”just a slip at the pin-wood interface.” In other words, static friction is overcome. The question is whether the wood, the metal, or both are twisting or deforming prior to the breaking of the static friction.

I will look for a reference to the experiment with the pointers.

Here is a link to a real world use for the same kind of twist that I am talking about: http://en.wikipedia.org/wiki/Torsion_bar_suspension It certainly does happen with metals, otherwise springs could not exist.

The portion of the MZ that I am talking about is “larger” in pianos with tight tuning pins.

Doel:

There is an explanation of the torsion of the pin, the resulting change of pitch and a reference to the PTG experiment in Reblitz's book. I found it in a cached link on Google, but do not have the book with me to give you the chapter or page.

I don't deny that a steel rod can twist but am not convinced it does more than infinitesimally under tuning circumstances. Of course if experiments show otherwise I will be convinced.

When the tuning pins are tight there is more friction at the metal-wood interface, hence the wood will be "dragged along" longer than otherwise, hence more of that MZ effect.

Kees

I don't have references, but I have witnessed this experiment myself.

Jim Busby, I think, had a jig at one of his classes at the 2008 PTG national in Anaheim with such a setup: pinblock material drilled all the way through; lengths of music wire attached through the bottom hole to the bottom of the pin and bent at a 90 degree angle such that it pointed like a clock perpendicular to the axis of rotation of the pin; another wire attached to the pin above the pinblock.

You could see how much twist was applied via the top pointer before the bottom pointer moved. I didn't take any measurements, but it was easily visible, so a few degrees at least, by my recollection. I have seen this experiment discussed many times in the PTG Journal as well as in online discussions, so I believe it is easily repeatable.

Also, Nate Reyburn tells me they did this experiment with impact levers as well, to verify it is more difficult to leave twist in the pin with impact technique.

I've seen the demonstration that Jeff is talking about...

It uses a scrap of pinblock with one pin installed. On the bottom of the pin, a wire has been welded or soldered on to protrude through the bottom of the pin block and is then bent at a 90 degree angle. The demonstration begins with the lever and wire in line. The pin is rotated with the lever - we could see the lever clearly move without the wire moving. Once the friction of the block was overcome, the wire moved as well. I was surprised to see the difference in angles for the top and bottom of the pin.

Ron Koval

Thanks guys, that settles it. I found the passage in Reblitz which is in the form of a thought experiment mentioning in passing that it has been demonstrated at a PTG convention.

Apologies to the trees for underestimating the strength of wood!

Would there be any benefits to using carbon fiber pins?

Kees

I had thought that there may be some benefit in imbedding high carbon steel rods within tuning pins to make them stiffer. But I think the real solution is better QC when building pianos.

And FWIW I think there is something going on with the wood, also.

I tuned a piano yesterday which was my fist real-world experience with a huge marshmallow zone. The piano was about 10 cents flat overall. In octave 6, with a smooth pull I would overpull 20 cents (measured by ETD). Then the test blows. BANG (10c sharp) BANG (on pitch) BANG (5c flat) BANG (4c flat). There it would stay.

Trying the same thing with a jerking motion made no difference.

Jeff, what is your approach in a situation like this? Do you just figure out you have to overpull 40 cents and risk breaking strings? Or do you do like I do; take your time, and overpull again, a little less this time, and be patient till the string settles?

Kees

Kees
My explanation is:
You had twisted the pin clockwise to a point where it reached a point resisting more torsion. It then partially moved the more bedded foot of the pin. You then let go with the pin in a state of disequilibrium between the pin torsion and the string tension. The front of the pin then wants to unwind because of string tension against the lesser friction at the front of the hole to a point somewhere near equilibrium. The position of the pin's foot governs this position. It is likely not in tune because you had little control on the precision the foot. What you need to do, using smooth pulling, is to ease the pin back until you feel the area of equilibrium, somewhere early in the marshmallow zone, to a point where it is in tune AND where the friction in the hole is able to hold against the string tension and the now more stable pin torsion. If (when) the pitch is still incorrect then repeat with finer movements, but when you go too sharp then the movements are reversed. It takes judgement in feeling the tension and friction through the tuning lever. Impact style techniques are supposed to move the foot move precisely in small increments, but I am not sure about this.

It depends if this type of MZ is due to a very tight pinblock and moderate rendering, or a moderate pinblock with very easy rendering.

If it is the first, I often will jerk the pin well above what would at first seem the ideal position. And then work a twist into the pin without breaking the foot of the pin loose so that a slight torqueing of the pin up and down will show that the top of the pin is in a neutral condition in regard to the pitch. I tune a particular piano like this that can hold rock steady unisons in the treble for a year as long as the Damppchaser system is watered.

If it is the second, I have tried different things and am returning to one that has been pooh-poohed, but works pretty darn well. Start well below pitch and start applying only a flagpoling force until the pitch just starts to drop. Then while keeping this force constant, also turn the pin. When the pitch gets to where it belongs release both forces, then reverse both forces just a bit, bang the note a few times and evaluate how it worked. You can expect to need more flagpoling force on the strings with a shorter nonspeaking length. If you are uncomfortable with the amount of flagpoling force that is necessary, add some overshoot to the mix. What I see this doing is making the apparent MZ smaller and more manageable. Another way to deal with the two forces is to image a plane at an angle to the pinblock that the hammer force is applied at.

Doel:

By any chance did the piano have one of these stickers inside of it?

I think so.
That's what I did up to octave 6, where I would have had to pull over pitch by something like 30-40c and I was afraid to break a string so I did a mini multistage pitch raise for each string. Maybe I should practice my knots some more so I won't be afraid of breaking strings.

It did really feel like the pin was stuck deep in the pinblock, as I'm now convinced by the helpful experts here it actually is. I used a heavy Schaff extension hammer and wished it was lighter to get a better feel for the MZ.

This was an older Yamaha. What was different from piano's I have tuned before (which is not a whole lot) is that the tuning pins were much thinner than usual. This would be consistent with less pin stiffness and hence more of the
pin twisting and getting stuck at the far side of the pinblock.

I'm getting paid for this tuning by critical user feedback and permission to drop by again to evaluate the tuning myself. I measured and recorded my final tuning. How long should I wait before remeasuring to check how stable my tuning was in your opinion? Would 2 weeks be enough?

An another interesting datum is that this piano had not been tuned in over 5 years but was quite playable, though overall flat. I guess the previous tuner knew his stuff.

Kees

I think the pin twists far less than what is perceived, and that flagpoling accounts for a significant part of pitch change before the embedded part of the pin moves. Standing in front of a grand, tuning with the right hand with the lever at 2:00 (pulling toward you to raise pitch), flagpoling is unavoidable and significant, and is a major part of the "righting" of the pin to a neutral position that becomes necessary after the energy pass. This is why I believe that any technique that causes the embedded portion of the pin to move sooner in the stroke greatly reduces any perceived marshmallow zone. Really, the pin twists extremely little.

If the pin twists extremely little, then why use a technique that causes the embedded portion of the pin to move sooner in the stroke? The two ideas seem contradictory to me.

Because though it twists very little, it still twists. It would only be contradictory if I took the position that there is no twist at all.

*edit* In other words, the less twist, the better. I think there can be a tendency to "untwist" too much; that is, overcompensating, which introduces yet more twist.

Loren:

Do you want some syrup for your "waffle"? Please speak clearer.

"I think the pin twists far less than what is perceived ... ... the pin twists extremely little." In the context that this is written the reader is led to believe that pin twist is insignificant.

And then as a clarification you post: "... In other words, the less twist, the better. I think there can be a tendency to "untwist" too much; that is, overcompensating, which introduces yet more twist." So now, let's not talk about twist, let's talk about untwist!

Can you see how I think you are "blowing smoke"?

If we can admit that there is pin twist and it is significant (as has been proven with tests including those by Reyburn) then we can talk about how an impact hammer can reduce the twist that is left after rotating the pin. That would be a really good thing. As it is now we are only chasing our own tails.

Insignificant? Where did I say that? I simply said LESS THAN WHAT IS PERCEIVED.



No Jeff, I can't. I can see how you are simply making things more difficult than they need to be. I'll try again.

1. The pin twists, but very little
2. The lest twist introduced, the better.

BECAUSE...

3. That means less untwisting is required.

That better?

Geez, Jeff. Setting the pin by definition means untwisting whatever twist is introduced, be it a small amount or a large one. So it follows that any discussion on this will include both.






Insignificant is your word Jeff, not mine! Reread my thread.

Thank you! Now I have something concrete to respond to.

Who says pin twist is a bad thing? There is more to the whole picture than just the twist left in the pin.

Twisting the pin can be used to render the string. Just because two strings have the same pin characteristics does not mean that they have the same rendering characteristics, especially when comparing two different pianos. But even on the same piano, for the most stable tuning, the left string should be treated differently than the right at least from the treble break upward. This can often best be accomplished with greater, not less, pin twisting.

I don’t think measuring the tuning is what is needed. What is needed is listening to the unisons. At this stage in your learning curve I would check the very next day, then two days after that, then like four after that, and so on. Look for any patterns between the left, middle and right strings. Like if you strip muted, and the middle string tends to be out with the other two, then work on setting the pin when expanding the temperament.

Pin twisting can indeed be used to help render strings on some pianos with high friction in the front duplex area. I often use my hammer at 3 O'clock position to maximize twist, and I even try to flex the pin back. When the speaking length finally reacts, this means you will not have turned the pin nearly as far.

The pin twists a lot in a tight pinblock. Have you ever seen the demo where a tuning pin is driven into a piece of pinblock that is clamped to a table? A wire is extended off the bottom of the pin under the block and from the tuning pin hole at the top, The wires are aligned so that are even with each other. Then when you turn the pin you can actually see how much more the top wire move then the bottom. It can easily be 5 degrees, which in tuning terms is a LOT. Then it is interesting to watch how the bottom almost immediately catches up once the pin actually moves in the block.

Ryan:

Do you mean that you put the hammer at 3 O'clock so that flagpoling helps render the string in addition to pin twisting? That is how I look at it.

Exactly. With low friction pianos it is the opposite: hammer at closer to 12 and flex as little as possible.

In addition to pin twisting, there is a flex of the tuning hammer shaft. The more rigid the hammer the less MZ.

That's why I love my Fujan, what I feel is the elasticity of the tuning pin, not the flexing of my tuning hammer.

Gadzar:

I might agree with you if I percieved the MZ as an arc. But I percieve the MZ as a window of pitch, not a window of hammer movement.

And in addition to the flexibility of the tuning hammer shaft, Owen Jorgensen says that "the sensation of pin bending and twisting actually arises from events in the wood fibers of the hole, not from actual bending and twisting of the pin".

So the elasticity of the pinblock's wood comes also into play. Should we talk about setting the pinblock's hole instead of tuning pin setting?

Gadzar:

I agree that the wood fibers has something to do with it, but not because that is what someone wrote. It is something that I can feel. In fact, I think the wood fibers are why tuning from flat to sharp is different and more stable than from sharp to flat. But just what is going on, I am not sure.

For me what Owen Jorgensen says has more value than what I, in my short career, or you in yours can feel.

I think the experiments referred to earlier in this thread prove him wrong.

Kees

Well! I see your Owen Jorgensen bet and raise you an Arthur Reblitz!

this thread is alive and kicking for sure

What experiments?

Wood has more elasticity than iron.

I have never seen a wooden coil spring. I think steel is much more elastic.

Here you go again.

I am tired to argue with you.

The pinblock, as anything else in this material world, has some elasticity in it and it flexes under the forces execed with the tuning hammer. What I am saying is that this contributes to the MZ.

Please give me a break!

Extracted from "Different Strokes"

Appendix D: The Marshmallow Zone.

As written by Ken Burton:

Causes for the "Marshmallow Effect"

... After tuning one of those Wurlitzer pianos with the tuning pins set in a cast iron pin block (no kidding, no wood, all iron!), and after some "messing around" (I really can't call it experimenting) with tuning pins in my shop bench vise, I have come to believe that the pins do actually bend and twist a bit, but not as much as it feels like in some pianos. The wood fibers gripping the pin are probably also dynamically involved, somehow adding to the sensation of flexing and twisting...

That´s what I am talking about when I said wood has more elasticity than iron.

So the elasticity of the pin block (as the flexibility of the shaft of the tuning hammer) has its own place in the Marshmallow Effect.

And that's exactly what I'm saying: the pins do twist but not as much as what it feels like.

This has gotten beyond silly. Eventually, you have to stop marshmallow zoning and start tuning.

The experiments referred to earlier in this thread. A wire is attached to the back and front of the pin. It's just a few postings back.

Hard to believe the pin twists 5 degrees, but I assume Uri Geller was not involved in the experiments.

Now everybody here agrees the wood is also elastic and the perceived MZ is the sum of both.

What one does not agree on is how much each of these 2 factors actually contributes.

Kees

I made the claim of 5 degrees on a tight pin, but that is from my memory which is like a steel sieve.

I totally agree.

Thanks, Bill! TUNE THE PIANO FOR GOD's SAKE! 45 minutes oughta do it.

Gadzar:

I will admit that some pianos have mushy feeling pinblocks and the pinblock component of the MZ is greater than the tuning pin component. These are the minority, though.

How do you know? (Question, not trying to be argumentative.)

Kees

I checked unisons after 4 days and they were as I left them. Now an arctic front is moving in. I'll be very curious to try after that.

Kees

Good question, I am trying to run it through my mind. I think it was most noticeable in the lack of "springing back" of the pin. If I think of a better explanation I'll post it.

Can anyone else describe it?

On the particular mushy piano I had under my hands, to bring the pitch up I would go "jerk-jerk-jerk-jerk". (No insult intended.) At each jerk I imagined I felt a "steel spring" and something moving in quicksand. The "steel spring" part immediately bounces back, the quicksand stuff just creeps back under test blows. Now I suspect the quicksand stuff is a mix of wood fibers deforming reluctantly and string rendering issues.

Does that sound right?

Kees