Piano World Home Page Forums Our Most Popular Forums Piano Tuner-Technicians Forum What are the pros for strip muting?

Hi,

Normally I check that unisons stay put all along the way, while I am tuning. And I need to check that the bridge-soundboard is behaving properly, according to my anticipations/calculation.

My conjecture is that our routine has been and is changing in relation to our target.

When I started, I was taught to expand the temperament with 'copied' octaves, and try to reduce cumulative errors by ear, were my ear offended. I think that, for that type of practice, tuning with a single wedge could be ok. Personally, I stopped 'copying' octaves long ago.

Open unisons as you go? Hmm..., that, for me, is a problem. I have tried in the last months, since rxd and Isaac where talking about that. I said... why not, let's give it a go. It did not help me at all, albeit I could tune a perhaps 'reasonable' interval and a reasonable unison. Nothing to compare with the accuracy I am now.. devoted/addicted?
Many more moves, from one pin to the other and back again, down to a complete unison that perhaps sounds ok, as 'unison' as possible, but still hypertrophic.. perhaps you would say rich, sonorous, but for me it is still an ok-dirty, with no much control onto the final color and duration, nor on 'combined pitch'.

That, IMO, might be the reason why some of you have added problems with pitch-shifting and ET. The unison must be built coherently, together with pins that must be 'active' coherently; the leader string and the other two strings must be 'coupled' in the same way, note after note, and the NSL dealt with, making sure that... hope you already know.

I cannot say scientifically what the difference is, when loading the bridge-soundboard with new tensions, with one string at the time (the middle one) or with three strings at once (in your way). I guess that experience and 'feeling' can be spent either way, and still anticipate correctly how the bridge-soundboard system will react/behave. Time&Energy-wise? Difficult to say, easier if you see how tired you feel at the end, and the result you got.

But one thing, I believe, would not be possible for me, if tuning with only one wedge: to see how I am steadily shaping the 'form', how I am tempering the whole piano, check that all intervals are interrelated correctly, that the tuning curve is the one I want, which is what I do with many available intervals, tuning both the trebles and the bass.

Hi rxd,

You wrote:..."We usually never listen to a unison as closely as when we try to tune another note from it. Certainly not after removing the strip and assuming we infallibly tuned unisons that will stay there."...

I do not see myself in your description.

..."I am not a control freak but I do like to have as much control over my tunings as possible,..."...

That is what I may well be, a freak :-)

Regards, a.c.

IMO, the drift due to the Weinreich effect is much smaller than the drift provoqued by a pitch raise. If we are able to compensate/correct the drift in a pitch raise then the Weinreich effect should be no problem neither.

I'am not saying this drift does not exist I am saying it is useless to take it into account in a pitch raise as it's negligible compared to the drift due to bridge and soundboard moving.

You are absolutely right. This effect comes into play as a noticeable element in fine tuning at pitch on larger pianos. It is possible to "guess" the drift and come close, but that's not 100%.

The effect is very obvious even on my 120 cm (48") upright. And it is most evident in the 5th and 6th octave, so piano size shouldn't matter much.

Also, observing this with an ETD is perfectly fine. Why wouldn't it be? It shows the frequency detected of the partial you choose and shows how much it is off in cents etc. It is very obvious that the individual string can be perfectly spot on and still and then when two strings couple, the detected frequency is often flat by a cent or two. Very easy to demonstrate.

So, given that, does an ETD tuner tweak each string in a unison till the pitch is perfect when all three strings are played together? Virgil showed me to tune the first string slightly sharp to allow for Weinreich. The resulting unison is then dead on. Within limits, it's better to err to slightly sharp, than flat.

For me, I change up tuning with strips, and not. Sometimes ETD, sometimes aural. Sometimes I tune up and down by 5ths, not octaves. Sometimes I tune the left string first, sometimes the middle. I get bored tuning the same way all the time. The most important aspect of tuning is the second pass and final checks, where all the issues from the first pass are found and corrected.

Only if the ETD is reading only one string and reporting only 1 string's frequency. If you then try and read the frequency of 2 strings, its averaging or doing something else, and in my opinion, giving you a false composite reading.

Robert Scott....are you out there? What is the software doing when we ask it to report the frequency of 2 unison strings at once?

Jim Ialeggio

Doesn't Stopper's software help with tuning unisons? And allow for coupling effects?

Alfredo mentioned the behaviour of the bridge and soundboard. Giordano explains the results of his research into this subject in his book, the Physics of the Piano. He showed that the movement of the bridge affects the vibration of an individual string. When you have two or three strings vibrating together there are bridge movement effects and coupling effects.

I'd guess that the random tuning variations Weinrich mentions at the end of his paper are to compensate for variations in string/bridge/soundboard behaviour as well as the hammer imperfections he cites.

Stopper recommends tuning individual strings of the unison to the display. I use Stoppers program, and composite readings seem to display an average. I tune unisons aurally as I go, and set octave placements by "committee", setting aurally and comparing with the display...then questioning either me or the display as needed. Bass, especially small high IH instruments, aurally.

In any case the Weinrich effect, if we assume it is real, is still only a very minor portion of the accuracy advantages of tuning to whole unisons, in my opinion.

The major benefit in tuning whole unisons, is quite simple. That is, that using previously tuned whole unisons as references for subsequent intervals, gives you multiple efficient references to check and recheck what you have done previously. For me, the major source of reference note movement will be tuner induced drift...uhhh...me. That is me, rather than a micro, hard-to-quantify coupling effect. The drift can be minor, with the unison slipping as a unit or the individual string(s) may slip out of unison...sometimes pushing a cent. In some cases, the unison may sound apparently reasonably clean, but usually not. This means previously tuned unison drop or rise is easy and quick to diagnose and correct.

So when proceeding, I initially do a mini-pitch raise on each unison, getting the unison very slightly high with just a few movements, paying atterntion to the NSL condition but not so much that it would be a final setting. Then, after tuning the other 2 strings to the center (as doublets,2 mutes), check them again and tweak minimally as needed. When correcting for my own tuner induced drift, the slight corrections end up leaving the system more stable, and the tuning really clean and coupled. Interesting thing is, that proceeding with these min-pitch raises per unison, shooting for close, but not final pin or pitch setting in the initial movements of a string, I have sped up considerably. The accuracy has improved very nicely to boot.

I think Mark is correct in making the point that reading of the front segment well, in minute placements and re-placement of notes, plays a huge role in "stability" during the tuning. And that many of the effects we have been taught to finger as global tension resolution, are really related to inaccurate front segment placement...ie, tuner technique related.

Jim Ialeggio

This has been my experience. I'm not a piano tech, I'll say that right off, but I have a strong physics background. On the two pianos I tune regularly, I have only ever had the coupling effect pull the pitch slightly flat, never sharp. I realise that some people have reported the pitch going sharp, and to that I can only assume that the different construction of the belly/bridge assembly itself causes this to happen. Btw, the change is both audible and measurable on Tunelab. Tunelab is able to "see" the resultant drop in pitch when the Heinrich effect occurs and the drop (in cents) is representative of what I hear if I were to detune a single string by the same amount.

So, I think that setting the temperament with a mute strip, recording the temperament with Tunelab, then tuning the piano with Tunelab (although setting the stretch by ear) is the most efficient method for a novice like me. Sometimes I still tune the whole piano by ear - just for the fun and challenge. But Tunelab gets me to the same result faster. I still prefer to set my own temperament with the strip though. I don't know how you guys can set a temperament without muting. All that jangling drives me nuts!

For clarification, ETDs, in general, have two separate functions.

One is to measure iH, using a transform algorithm. This has limited, but useful accuracy, and can detect and display two or more nearly coincident peaks in a mistuned unison.

The second, much more important, and very accurate function, is to compare the sensed pitch with a reference pitch - that is a strobe type effect used to tune.

An ETD, when, and only when listening to a string being tuned, can determine the instantaneous pitch of that string within a few thousandths of a hertz. If it is listening to two or more strings simultaneously, it will indicate the coupled interaction of the strings by the activity of the strobe. If the the strings were perfectly in tune and perfectly in phase, the strobe would be still. If any one of the strings is not in tune and/or not in phase, the strobe will dance about wildly.

Stoppers program, however it is calculating the display, does not dance about with 2 strings sounding. Rather is displays a steady reading that seems to me to be some rough sort of an average.

ji

Interesting. Thanks for that info.

Edit: Presumably, if it uses a strong smoothing function, you should be able to tune any unison trichord without ever using any mutes. Try it and let me know how it works.

This is a popular delusion. Many people think ETDs are scientific measuring devices providing some direct report of reality. They are not. They do have an input device from the real world, but that input goes into a "black box" (virtual or real) where it is manipulated, massaged and interpreted with that final result being displayed on the readout. What happens to the objective data is at least obscure and often proprietary. An ETD is not a frequency counter.

Well, I have always said, you have to find what works for you. I actually tuned for 10 years without strip mutes and I had a lot of problems. When I saw the lecture and demonstration by the late George Defebaugh RPT and Jim Coleman Sr. RPT at the PTG Annual Institute in Minneapolis in 1979 that showed strip mutes made of action cloth (rather than the red felt sold by supply houses for this purpose), it changed my life and tuning technique forever. I never turned away from that approach.

What I see missing in all of the arguments pro and con is all of the details. The people who tune without strips make it seem (as I told my apprentice, Lucas Brookins) that all you have to do is tune a string one time to pitch and you are done with it! It will stay! (If you don't have a strip in there. If you have a strip in there, it won't!).

It simply does not work that way. Fine accuracy and stability are achieved one way or the other and neither is easy and neither is achieved by tuning each string only one time.

So, this afternoon, after spending considerable time after several years, re-aligning the Genuine Steinway Parts that I had replaced in a Teflon Steinway several years ago at the local technical college theater, and after again using voicing techniques that I had learned at the Steinway factory, I roughed in the tuning which was already rock solid at A-440 on the note A3 (but both flat and sharp elsewhere).

So now, the apprentice had a piano that had already been at pitch and had already been tuned once. Here is where I personally see the divisiveness in this argument come to life. The non-strip muters would have already had the job done! At least, according to what they claim (and which I do not believe for a minute and never will).

What I will always believe about that kind of technique is the unwillingness to concede the amount of drift that takes place. Sure, the unisons may sound good. But the whole enchilada is basically not there and a grand excuse for tuning the 7th octave flatter than any pianist ever wants to hear it. (Probably the 6th octave too). As they say, the higher you go, the flatter you get. And they want to make you accept it! If you hear it that way long enough, you will! (George Orwell: 1984)

So, today, I had my apprentice, Lucas Brookins tune the Steinway Concert Grand at the local technical college theater after it had been precisely aligned and regulated and tuned roughly to pitch. Lucas participated in the fine capstan adjustment and back checking.

After one pass, he played the piano, newly voiced and the theater technician and I too took notice of the splendid sound. It was not finished, however as I left Lucas to make his final pass, concentrating on unison quality (pure and not any other ridiculous claim). What I heard as I left was solid quality tone and nothing less.

He had the strip mutes there but also the single felt mutes for when they were necessary. He also had the custom tuning program on Tunelab.

Here is a photo that I took of Lucas Brookins tuning a Steinway Concert grand for a piano concerto event that will be held on Saturday, January 31, 2015. He is a novice but with the right kind of guidance, he can perform at concert and artist level expectations. When he turns age 18 next month, he will pass the PTG Tuning Exam with high scores. You can count on that.

I would never claim to only tune each string one time and be done. There is still plenty of tweaking here and there depending on the piano.

I tuned a fairly new Essex today and the stability was giving me all kinds of trouble in the midrange/temperament area. I'd set what felt like a solid unison, but as I moved up into the treble I had to keep going back and correcting unisons. I usually don't have to do that so much - it was partly because the piano was green.

But there are other times, when once the midrange is set it stays fairly solid while tuning the rest of the piano. However, it is still not uncommon for me to change a note hear or there in the midrange during bass or treble tuning.

On the other hand, all your assumptions above about ETDs could be called delusions as well, no offense meant.

Virtually all ETDs use some form of fast fourier transform to detect and analyze audio. In this case it is focusing on a particular partial or the frequency band around this. In the case I describe, there is no averaging or smoothing involved. We are talking about two strings in a unison that when played individually are exactly the same (+- the resolution of the ETD, which is much smaller than what we are talking about here). If you smooth or average these, the result will be the same, not 1-2 cent lower.

Also, I am not just looking at a frequency number. I am looking at the phase display and every other information available in TuneLab when doing this. If you tried what I described, you would see very quickly that there is a big difference in the way TuneLab reacts if you play two unison strings that slightly out of tune with each other: phase band jumping chaotically; and the way the phase display reacts when playing two unisons in tune with each other or just a single string: phase display still if the resulting frequency is what the ETD wants, or a slow roll to the left if it is flat, to the right if it is sharp.

Also, you can test it aurally with beats by playing some lower note with a coincident partial at the first partial/fundamental of the unison in question. When muting all but one string, you will get one beat. When playing the full unison, you will get another beat. Again, assuming we're talking about one of the unisons that actually does this on one of the pianos that actually does this. I can't speak for other pianos but it is easy to demonstrate both with TuneLab and aurally on my piano, mostly in the 5th and 6th octave.

This thread, by implication, is about the final stages of a tuning on a stable piano. How one gets to that stability that will accept a fine tuning is not really the issue.

The self same checks and balances are used by all responsible tuners whichever system is used, strip, double string or single string unisons, (or triple string unisons those who can do it).

Faults in the temperament and intervening noted can be spotted as the tuning progresses in either method by the responsible tuner. The main advantage I find with a single wedge is the additional ability to check unisons along with all the other intervals early in the procedure, correct them and ensure they stay corrected.

Of course, faulty unisons cannot be tuned from. Deliberately impure unisons, if they are of musical value, should be uncomplicated to tune from by the tuner who tuned them that way and should be tuned from, however difficult, to ensure consistency in the octaves, among other intervals, which would be quite random otherwise.

Much of my work is refining an existing tuning that may be only a few hours old. Much of that refinement is often a refining of an already good unison. this is also true of a stable tu

Time limitations also apply. It sometimes happens that a tuning has to be concluded quickly or even abandoned, a class or audience coming in or a family that wants to go out, etc. we've all been there. Who would want to have to hurriedly put in a whole piano full of unisons, (more than half of the strings) with severe time limitations?

My main reasons are purely practical. When I first started to use a strip, it was for the temperament octave only, the original ones were shorter than they are sold today. It was the practice to use them only as an aid for putting in the temperament. The strip was then immediately taken out and the rest of the piano tuned single wedge.

The other side of that coin is that I have often put in a strip in the whole piano and found most of the middle strings in tune and I only need to make a few corrections and put in unisons.

Both techniques are useful. I prefer the final pass to be single wedge.

I think my main issue with the strip for temperament tuning is this: You finish with the strip in and are very happy with the results of your carefully tuned temperament. Then when the unisons are pulled in you find some drift. But it is not as clear which notes drifted as the results of pulling in the unisons, and I found myself suspicious of every note, which is stressful after doing careful work. At that point, when I was still using strips, I would sometimes put the strip in again and have another go at it.

A breakthrough occurred after reading a Journal article where putting the strip in between every other unison was suggested as an efficiency booster. So I changed my approach and after tuning the midrange with the strip in between every unison, I would pull it out and reinsert it between every other unison to bring in a second string. For the final pass I'd pull it out, shift it over and get the rest of the strings. That still left 2 strings per note open. That is when I began exploring the unison shimming technique to refine the results, which was a major breakthrough. Then when I pulled the strip completely out, I would just have to touch up the few unisons that I had tweaked during the previous pass.

I read about and discovered for myself that when two strings of a unison are in tune and fit well with the rest of the temperament they will be solid and unlikely to drift. That's why the "every-other-unison" strip method worked well for me for several years. But it still felt like there was some redundancy. No matter how carefully I'd set the string, or firm I'd pound the note, there would be some shifting around after that first pass with the strip fully inserted. It wasn't until I was "forced" to tune with single mutes (due to leaving them at a previous appointment) that I finally explored tuning unisons as you go along.

The main pain was having to move mutes around more and not having the peaceful clarity of tone that a temperament strip provides. But once I completed the first pitch raise pass, and started fine tuning the piano, I was surprised to discover that the notes seemed to stay put as I worked through the temperament. That doesn't mean I was always happy with where I put the note. Temperament, for me, is still a circle of refinement with a tweak here and a tweak there as I try to get my ducks all lined up in a row. With the temperament strip, my ducks seemed to fiddle around on the fence more. With open unison tuning, the ducks didn't move hardly at all. They behaved more like dead ducks! Or at least like ducks with their feet nailed to the fence. (Sorry duck lovers!)

I still use a strip occasionally under certain circumstances - mostly if I'm in a major time crunch and I'm not concerned about an accurate temperament. But that's probably only a couple of times a year.

When you tune all the center strings, then come back later and tune all the outside strings, it's not as stable as tuning all strings of the unison before tuning all strings of the next unison, Starting at the low bass and ending in the high treble, tuning unisons as you go is more stable yet.

Virgil tuned unisons as he went for that reason - and he would occasionally return to a previous unison because it slipped.

The thing I noticed when I used strip mutes is this: Sometimes things would work out well, other times it did not. It was unpredictable. When I switched to tuning unisons as I went, especially when combined with pre-tensioning the section, things became much more predictable, stable, and less stressful.