Piano World Home Page Forums Our Most Popular Forums Piano Tuner-Technicians Forum Stanwood Touch Design question

My own work, which I referenced above in my response to Ed Foote, showed that at middle C in the particular Steinway B that I analyzed, that hammer inertia was 83% of the total, the key and its lead weights was 14% of the total, and the wippen was 1.5% of the total. In general, my measurements and calculations correlate very well with those of Fandrich and Rhodes.

Interesting, however, I have to question the completeness of information. Is strike weight the overwhelming source of inertia at all levels of play? I think as the speed of movement increases the mass of the key becomes more influential, so at what speed does Fandrich and Rhodes' research make the claim? Any speed, or something specifically slower that full blast?
Regards,

As the Five Lectures info describes it, the key is moving well before the hammer, so inertial levels of the hammer may as well be infinite in relation to accelerating the key.

Actually, the value is closer to 89%
Fandrich/Rhodes is onto something but not quite there . . .

There are a couple of things at work here. Roy is quite right in his analysis of the effect of the different components of the piano action; hammers are, by far, the biggest contributor to the overall inertia of the action. As Roy also points out, key leads are an indicator of an inertia problem; they are not all that much of the problem in and of themselves.

The article by Anders Askenfelt & Erik Jansson, ”From Touch to String Vibration,” tells only a small part of the story when it comes to explaining how the piano action works. In all piano actions there is a time delay between the initial key strike and the motion of the hammer. How much of a delay between the key strike and the motion of the hammer—and then the actual acceleration curve of the hammer—depends on many factors including the length, shape and stiffness of the key, the compliance of the many felt and leather components in the action, the stiffness of the various wood action components and, most significantly, the mass of the hammer itself. None of these are constants.

When a key is struck very lightly the lag between key motion and hammer motion is relatively small. That is, there is little key bending, little compression—beyond the static compression—of the felt and leather parts and little bending in the wippen levers and the hammershank. The motion of the hammer tracks the motion of the key fairly accurately.

When the key is struck with a very hard blow the lag between the motion of the key and the resultant hammer motion is significant. Indeed, in all piano actions there will be a point at which a harder key strike will no longer result in any additional hammer velocity. This is the point at which the key actually strikes bottom before the hammer starts to move. This is called the point of ”action saturation.”

The point of action saturation is different in every action and, indeed, from key to key within a given action. Action saturation will be reached sooner with keys having a significant dogleg than in keys that are relatively straight. It will be reached sooner in an action using thick, soft balance rail punchings. Sooner in an action—like the concert grand action I looked at last week—using relatively soft felt under the knuckle leather. Sooner in an action using heavy hammers. Etc.

All other things being equal, action saturation will occur sooner in an action using heavier hammers than in an action using lighter hammers. This, because the inertia of the hammer is so great that it takes more force to accelerate it and the action components are only capable of transferring so much energy from the keys to the hammers. Beyond this they simply twist, bend and compress.

Generally in these discussions we tend to think of the energy transfer from the key front to the hammer as 100% efficient and instantaneous; it is not. How energy is stored and released within the various action components is greatly affected by hammer mass. A relatively compliant action that works very nicely with light hammers will be quickly saturated with heavy hammers. In this case the quality of touch and/or feel has nothing to do with how well the action is going to function; it is going to be overwhelmed fairly early on no matter how many leads there are or are not in the keys.

ddf

the key have a chance to rebound from the punching under heavy staccato playing. probably the hammer have those 89% value.

as the key react to an impact, only its initial inertia may be of some interest (?)

ABsolutely very well stated Del .

Del, Your getting into explaining why harder is not louder. When the enthusiastic teenager starts breaking strings, I tell them to view the piano like a drum head. Any knucklehead can put a stick right through the drum head. It takes a skilled musician to make it sound like a gun shot.

I don't see any mention of the rebound portion of the stroke.

Mostly, so far the discussion is, at least as I read it, concerned with the "effort" part of the stroke, which I interpret to mean the part of the stroke which ends at the key bottoming on the punching.

From the the pianist's perspective, especially in rapid play,the kickback imparted by the parts already in motion, which have a mind of their own at this point, is a significant factor in what an action feels like in play...at least that's my take.

Although the SW has been mentioned as the determining source of inertia, a concept that is kind of "duh" science, in play, the returning key, with shank in free motion and then the reversing of that weighted key's direction impart significant information that determines how a pianist reads the action in play.

The SW/ratio/friction determines how much lead is going into the key, but once the lead is in the key, the key's momentum and/or inertia (if these are not the same things...physicists?)becomes a significant factor in determining the dynamic touch....No?

Jim Ialeggio

Not so much, probably the "free key" behavior have its importance on touch, but even somewhat large difference in leading can remain discrete.
Yesterday I worked on a piano (Steinway O) from 1908

in the high basses a white key have zero lead, while the neighbors (white too) have 3 leads. [edit : I am sorry I only looked on one side, possible half leads are used...]

One would expect the difference in touch to be fairly noticeable, it is not, at last in most basics mode of play.

I will be there today, and I will check against fast staccato play, and slow forceful, so to say, other modes, and I let you know the differences I can perceive.

There are no assist springs on that Steinway (from the start visibly) the action is light but with a good balanced feel.

When I talk of the rebound it is more or less anecdotal, but a key without lead on an action with assist spring will have little inertia, possibly not enough to be at ease in some modes of play (but it can be also appreciated by some pianists, as I have been told yet, to my surprise, just feeling a "transparent key") .If the assist springs where not prejudicial to the upward motion of key/action I would have used them more often.

I understand now that even if they lower/suppress the whippen inertia, this last have not a really noticeable inertia.

What count more in action is probably the quality of acceleration progression between levers. indeed assuming the hammer SW is well related to the action ratios.

(changing from 4.5:1 to 5.5:1 between rest and let off moment, for instance...)

(dont know if is a good ratio "evolving" but this is what can be find in a short grand action, for instance.

This is a very well produced document, I like the analysis (did not go thru the numbers of course but the reasoning is easy to follow)

Thanks for providing the result of your work, it is so interesting.

To me what make action feel is the good relations between the leverage progression , we expect the hammer gravity center to move in a predictable way at each note same, that is what gives control on the acceleration, with a very tiny moment at the end of the stroke where the breaking induced by the jack friction during let off, allow to correct or manipulate a little more the acceleration in some modes of playing.

the way the hand take control on the acceleration is all but a "yes or no" thing.

the pianists hand mass is managed with its own inertia so to match the action inertia, but then muscles manage acceleration.

I'm not so sure that kickback or rebound, as you called it, is much of a factor in the action feel as the key and other action components return to their rest positions. The hammer does bounce off the string with some energy--energy which, of course, depends on how hard the hammer strikes the strings. However, most of that energy is lost during check; the hammer tail scrubs against the backcheck, turning the hammer's rebound energy into heat. What's left is only the energy stored in the balancier's spring. I suppose the motion of the hammer as it leaves check is somewhat felt in the key, but I'm not so sure that it is significant.

Once again, however, we find that hammer mass and action ratio are significant, as they greatly affect the time it times for the action parts to reset. Forgetting any effects of hammer and felt rebound for the moment, the only force trying to reset the action is the up weight. This force has to accelerate the action components. Think back to Newton's famous formula f = ma, or in the world of rotating components, T = Ia, where T is torque, I is moment of inertia, and a is angular acceleration. We see that the acceleration of the action as it resets is inversely proportional to moment of inertia. We can go further and solve for the time it takes to reset, and find that
t = sqrt(2*alpha*I/T), where alpha is the angular travel of the key.

BTW, your description of the hammer mass as dominating the moment of the inertia as duh science seems a bit gratuitous. Like lots of things, it's only after someone makes the effort to do the numbers that it seems obvious. It's also obvious that many, if not most, people in the piano community are not aware of this fact.

the pianist can throw a key that have enough mass, so the jack follows the roller (or does not loose the contact too long) and the end of the move is just obtained by the key mass.

on the other side (DW) sure nothing remains of the initial move unless some energy can stay stored within the action stack and key frame. (something may happen at the back of the keyframe, too)

Um, the spring, compressed and held at check is a major player in the reset speed of the action. It accelerates the key, and since upweight is usually measured with the spring out of the equation, I don't think we can ascribe exclusive force of resetting to upweight, alone.

Well, I'm not so sure about that. As long as the hammer is held in check, the spring exerts no force. As the hammer starts to leave check, some of the spring's energy is dissipated as friction between the hammer tail and the back check. Finally, any spring energy that's left can supply force only as long as the hammer is accelerated upward or there is contact between the end of the balancier and and the drop screw. But, we must take into account the fact that friction at the balancier action center, which is usually pinned quite tightly, absorbs more of the spring's energy.

I thinks it's obvious that there's enough complexity here to require measurement and analysis before a firm conclusion can be reached.

I have to disagree. As long as the hammer is held in check, the spring is exerting its maximum force. It is potential, not kinetic. It is this force, acting against the checked hammer,via the rep/knuckle contact that is mainly responsible for the key's upward speed. When the key is released, in fast repetition, the hammer remains motionless, acting as the inertial resistance to the spring, which dumps all of its potential,(minus friction), into moving the key upwards. It is still motionless when the jack resets, (at least, according to the high-speed films I have seen of fast repetition).

I repeat: under even moderately fast repetition, the hammer only moves upward when the jack returns and sends it up. You can observe this by playing a hammer into check, and then sliding you finger off the front of the key, allowing it to rise unimpeded. You will see the hammer drop, not rise. And it may be a quibble, but the drop screw is only in contact with the repetition lever between the beginning of let-off and the hammer's first 1/16" or so rebound from the string and has little effect on repetition, ( unless it allows the hammer to be struck by the vibrating string and impels it below check,causing catastrophic repetition failure.)
Regards,

Consider the hammer held in check. The rep spring is trying to move the hammer upward by pushing on the knuckle. The hammer, while in check, is effectively connected to the back of the key. So, while in check, the rep spring is trying to hold the back of the key up. Therefore, in this condition, the rep spring is at least partially inhibiting the back of the key from falling, or at least not adding any force at all to the reset process. It is only after the hammer leaves check that the rep spring can help in action reset. When one subtracts out friction and considering the fact that the force of the rep spring is only applied for a portion of the key travel, I'm just not sure how significant it's force is. As I said before, I believe some measurements and analysis needs to be done before any firm conclusions can be reached.

I am real sure how significant it is, having dealt with hundreds of actions that were not repeating very well.

I think the difference here is one of perspective:

Consider the hammer in check. The rep spring is trying to move the key back up by pushing against the whippen. And it will begin to do this the instant it is released from check. It is only after the check leaves the hammer that the key rises and the spring moves the key rather than the hammer. It is significant because the spring accelerates things faster than gravity alone, and added acceleration at the very beginning of the return is most valuable. I haven't found that strengthening the spring to excessive levels speeds up the action nearly as much as the checking height. It just makes escapement a ruder interruption than it needs to be at ppp levels of play. This is a waste of sensitivity, since inre repetition speed, is not usually important whether the spring causes as rapid a hammer rise as possible without feeling it or whether it creates a detectable recoil as it throw the knuckle off the jack, repetition will be the same.

I am building fast actions with gentle springs, tight balancier pinning, and 5/8" checking heights. A pianist that requires faster repetition and doesn't need a lot of power in them can usually be satisfied by raising the check height to 3/8", or less if you can get it. There is less power in a stroke that short, but it allows the key to reset without so much upward travel. It can make up for poor technique, I am told.

Regards,

I could test on that Steinway O, in the end there are 2 white keys without any lead, at a fourth interval.


All other keys around them have 3 leads (I made some pics.

The main difference lies in the tone, which is stronger with the lead.

It is also harder to play fast the leaded keys the non leaded ones are easier, the 3 leads can be felt as braking the keys when one want to play rapidly.
seem to slow the rise of the key as the lowering.

Out of that, even knowing the fact, differences are not easy to feel in the fingers at slower speeds.

The tone, yes, the touch, not really...

I did some experiments using an action model. As usual, I found out something that was unexpected.

Test 1
I pressed the key down until just before the start of letoff and before the balancier was displaced by the drop screw. I held the key in this position and then measured the amount weight it took to just start the key in its upward move. This amount of weight was less than what is normally considered the up weight.

Test 2
I slowly lowered the key to the bottom of the stroke. Because of the slowness, the hammer was not in check. I once again measured the amount of weight that it took to just start to raise the key. It was about 5 grams higher than in the first test.

Test 3
I fully depressed the key and manually put the hammer into check. In this case, the weight required to just get the key moving was the same as in case 2. The reason for this is that the hammer was released from check very rapidly.

Result
For the particular key I tested, the compression of the balancier spring provided about 5 extra grams of up force, but only over a very short distance, because the hammer resets quite quickly, using only a small portion of the key stroke.

Unexpected result
I found that the compression of the front-rail felt provided lots more rebound force than the balancier spring. This effect is hard to quantify because the force provided depends on how hard the key is pressed, and how quickly it is released. I would make the tentative conclusion that the felt compression can be a significant factor in how quickly the key resets. If this is true, it would suggest that soft front-rail felts would be more effective than stiff felts, because the soft felts, by compressing more, store more energy. This result is quite interesting as it suggests that front-rail felt could be fabricated and selected based on this criterion as well as its contribution to action feel.

Once again, a quick experiment has potentially revealed information that is poorly understood in the piano community. Experiments beat jaw flapping by a mile any day (and I'm not excepting myself).

I'm the OP. This discussion quickly went beyond my understanding.

Larry Fine, in his Fall 2012 Piano Buyer's Guide (which I read on the web) lists the Stanwood Touch Design under "problem solving".

One of the reasons that my technician was not so keen on this is that (in his words ) it can be used to mask problems with an action. When I mentioned that a piano with Stanwood touch design played very well, his response was - "yes, but for how long?" Is this a valid comment? And if so, would a technician that I bring to inspect a potential purchase be able to opine on whether the Stanwood method applied to the piano I am considering is masking a problem or is in fact an enhancement to an acceptable action?

Thanks
Doug