Imperfect unisons during inharmonicity measurements

Anders

Hi Frank
In the manual it's stated (if i understood correctly) that slightly out of tune unisons is not a problem for the inharmonicity measurements, and that on the contrary it could even be beneficial. Could you elaborate on this? Just want to make sure before I go ahead and do it this way.
Regards, Anders

Frank Illenberger

Hi Anders,
it can be benificial to measure a full unison if the individual strings of a unison have different inharmonicities, as is sometimes the case for bass strings. In this case then pianoscope records an average value which can be better than measuring the individual values. If the unison is unpleasantly out of tune, you should however prefer to measure a single string.

David Kissam

Frank Illenberger This is something I've been thinking about too. After my first pitch raise pass, I delete all of the inharmonicity measurements and re-measure the piano for fine tuning. I have been making the unisons fairly decent during pitch-raising so I can record the inharmonicity without using mutes.

Now I'm thinking that this is a waste of time. I should just make the wound-string unisons decent (which doesn't take long) so I can measure the wound strings without mutes. I'll be super sloppy with the unwound-string unisons and mute all but one string when measuring those. Does that sound sensible?

From reading the manual, and from my own experience using the app, it seems that anything worse than slightly out-of-tune unisons give less accurate inharmonicity measurements. So for the unwound strings it makes sense to measure single strings only. In the bass, where inharmonicity is often not perfectly matched, it's worthwhile to measure both strings. Would you agree with that?

Anders

Ok thank you.

Patrick

I'd like to resurrect this discussion, if I may, to clearly understand the process. Disclaimer: I am a brand new piano tuning apprentice and a new user of Pianoscope Pro.

Before beginning the initial inharmonicity measurement of the instrument, is the recommendation to (a) not mute any strings, so the measurement represents the average values of all strings? (b) mute all trichords and bichords, so the measurement represents only single values? or (c) mute only those trichords and bichords whose unisons are particularly out of tune, so the measurement represents a mix of average and single values?

If the latter, how "unpleasantly out of tune" are we talking? Thank you!

David Kissam

Patrick Hi Patrick, I am also new at this (I have tuned 21 pianos with pianoscope) but I'll give you my thoughts anyway. 🙂

I have experimented with all three of the options you listed. It seems that inharmonicity for strings of the same note varies most often and most widely with wound strings. I've never encountered a mismatch of inharmonicity between unwound strings. So it seems that measuring a single unwound string will give roughly the same result as measuring all three, if the unison is in tune. However, I notice more (seemingly random) variation from the ideal inharmonicity curve when I measure out-of-tune unisons. When I isolate single strings, I tend to get a result that follows the ideal curve more closely, or at least has smoother, more consistent deviations. This makes me think that measuring single strings, or in-tune unisons (if that happens to be available), is more accurate for the unwound strings.

Wound strings are where all the weird stuff happens. It seems that the main value of measuring unmuted unisons is in the wound bichords, especially when you have obviously mismatched inharmonicity. The strangest case I have encountered so far was an old Yamaha G2. There were several bichords in octave 1 that pianoscope gave a negative inharmonicity reading when I measured both strings together. All of the unisons were in tune, and the inharmonicity was well-matched (according to my ears). Oddly, when I measured the strings individually, I got a completely different result for each one. I ended up using whichever string gave me the highest inharmonicity measurement, so that it fit the ideal curve most closely. The lower measurements were so low that pianoscope considered them outliers and ignored them, meaning it did not take those notes into account for generating the inharmonicity curve and therefore the tuning curve.

So my conclusion is this: measure roughly in-tune unisons for the wound bichords, and keep an eye out for any strange deviations from the ideal curve. When you encounter that, try measuring the left and right strings of the bichord individually, and use whichever measurement makes the most sense.

This is how I approach every new piano I tune:

Enter pitch raise mode and measure the entire piano in whole steps (44 notes) without mutes. Only mute notes that are so out-of-tune that pianoscope won't store a measurement for them.
Do a pitch raise or leveling as needed. Leave the wound bichords roughly in-tune (slow beating is okay).
Delete all inharmonicity measurements and measure all 88 notes. Do not mute the wound bichords. (I haven't yet decided whether muting my sloppy trichord unisons is worth the time it takes.)
Correct any sections that are overall sharp or flat.
Fine-tune the piano.
Recheck the pitch and unison of every note; do aural checks.

For a piano that I've already tuned, or one that is already pretty close to in tune, I'll skip the first two steps. I delete my measurements in step 3 because inharmonicity changes with pitch, so after a pitch raise my old measurements might not be accurate anymore.

I welcome any feedback, ideas, or corrections from anyone.

Patrick

David Kissam Hi David. This is incredibly helpful - thank you. May I please ask a few clarifying questions? (my ignorance is showing - sorry! 😉)

David Kissam Do a pitch raise or leveling as needed.

Are you referring to string leveling here? If so, can you please elaborate on that a bit more in the context of the pitch raising process? That has not come up so far in my training.

David Kissam Leave the wound bichords roughly in-tune (slow beating is okay).

Having not yet performed an actual pitch raise with Pianoscope, can you also please elaborate on this? The pitch raising section of the user manual says to tune the piano in fixed order, from A0 to C8, and to tune the associated unisons as you go. So you're saying here that when tuning the wound bichord unisons, it's OK if they are not perfectly in tune?

David Kissam Correct any sections that are overall sharp or flat.

Need some additional clarity here also.

David Kissam Fine-tune the piano.

Would you ever opt to delay the fine tuning to later date - and if so, what criteria might prompt that delay?

Thank you again for this very detailed response, David, and for your patience with my follow-up questions! 🙏

David Kissam

Patrick I'm happy to help if I'm able, Patrick.

A pitch raise is necessary when the piano is significantly flat overall. Sometimes you might encounter a piano that is quite sharp in some places and quite flat in others, and in this case you'd do what my mentor calls a "leveling". It just means getting all of the strings closer to where they should be before doing your fine-tuning pass. You should use pianoscope's pitch raise mode for both situations.

Patrick The pitch raising section of the user manual says to tune the piano in fixed order, from A0 to C8, and to tune the associated unisons as you go. So you're saying here that when tuning the wound bichord unisons, it's OK if they are not perfectly in tune?

Yes. It is a waste of time to fine-tune the unisons when doing a pitch raise. The goal is just to get the tension close to where it needs to be so that the pitch will be stable when you fine-tune it. When pitch-raising, I am very sloppy with the unwound unisons, and a little more careful with the wound unisons because I intend to measure them unmuted after completing the pitch raise.

When I say "correct any sections that are overall sharp or flat", I mean this: Look at the pitch graph after you have finished measuring the pitch of every note. If there are any areas where a bunch of notes are all sharp or all flat, go over that section quickly to get the pitches closer to the target. For example, let's say after my pitch raise, I notice that every note between D5 and B5 is around 5 cents sharp. I will bring all of those notes down closer to the target so that the pressure on that area of the bridge is closer to what it will be when the piano is in tune. For sections where the pitch randomly fluctuates above and below the target, I'll do nothing as long as the median pitch offset that pianoscope reports is less than maybe 2 cents. (Disclaimer: this is just my own method that I made with my own brain. 🙂)

Patrick Would you ever opt to delay the fine tuning to later date - and if so, what criteria might prompt that delay?

No, I will always finish the tuning immediately. When I was just starting out, I was so slow that it took me 3 days to finish a tuning. Now that I can finish a piano in about 4 hours, I always do so.

Patrick

David Kissam Immensely helpful, David - thank you so much for the time you invested in your replies!