tuning exam

[Billbrpt@AOL.COM]

---------------------- multipart/alternative attachment
In a message dated 9/1/01 3:40:00 PM Central Daylight Time, gmcc@pipeline.com 
(Gary McCormick) writes:


> Is it appropriate to ask how the scoring is done on the tuning exam?
> I'm trying to practice the temperament test using RCT; pulling up the record
> of my own piano, after tuning a temperament, I manually record what I've
> done, and compare it to  the tuning record.
> On a couple of notes, I may be off 1.10 cents, and the rest anywhere from
> .12 to .50 cents.  Is this acceptable?  At the moment I feel like I'm trying
> 

Yes, it's appropriate to ask and Kent S. answered you well.  I presume you 
are trying to tune by ear then judge your results against electronic values.  
You must designate a Temperament Octave.  That is usually F3-F4 but can be 
something else if you prefer.  You must be reading the 3rd, 4th and 5th 
octaves all on octave 5.  If you are reading the wrong partial, you will get 
far different numerical values.

Try first by using the RCT to tune a 2 octave span, C3-C5 using the usual 
techniques.  Now, record the values for each note as they appear when read on 
octave 5.  Example, record the value for C3 by focusing on C5.  Record the 
value for C4 by focusing on C5 and finally, record the value for C5 by 
focusing on C5.  Do like wise for all of the notes in that 2 octave span, 
making sure you record the values for each note by reading in Octave 5.

Now, detune your 2 octave span by offsetting your RTC by an arbitrary amount, 
say 3-6 cents sharp.  Starting on C3, tune in whole steps up to C5.  Now, 
offset the RCT by the same amount you chose to tune sharp but set it by that 
amount flat this time and tune in whole steps starting on C#3 and ending on 
B4.  This will produce a thoroughly out of tune span but will not upset the 
basic tension on the piano.

Now, test your fork against the RTC and calibrate it if necessary.  It should 
be as exact as possible, not more than .5 cents variant from the RTC.   To 
score yourself, first test you note A4 by reading it on A4.  Any amount more 
than 1 cent either sharp or flat will give you a point off for each 1/10 
cent.  More than 20 points fails.  Use the RCT to correct A4 electronically 
to as exactly 440 as possible.  Now using A4 as your starting reference, tune 
your two octave C3-C5 span aurally, the best you can.

Now read your aural tuning the same way as you did the electronic tuning, all 
on Octave 5.  Compare the difference on each note.  The tolerance for error 
is 1 cent.  Any more than 1 cent is a point, any more than 2 cents is 
another, etc.  In your temperament octave (usually F3-F4) you must not have 
more than 8 points total.  In your entire 2 octave span, you must not have 
more than 13 points.

Now, the Exam itself does make one other adjustment that can give you a 
little break if your pitch is passing but not perfect.  It adjusts the 
average pitch of your aural tuning to the electronic, so-called "Master" 
tuning.  But grading yourself in this way will give you an indication of 
whether your aural tuning is within passable tolerances.

Before judging your work, try to correct any errors yourself.  Play up and 
down 3rds, 6ths, 4ths and 5ths.  Listen for any unevenness.  Any "busy" 4th 
should have a corresponding 5th which sounds too pure.  Any "wobbly" 5th 
should have a corresponding 4th which sounds too pure.  If you don't know the 
4:5 ratio of contiguous 3rds test, you need to learn it.  Any two contiguous 
thirds will beat at a ratio of 4:5.  This is a little difficult to discern 
but you know when you don't have it right more plainly than when you do.  Is 
the bottom 3rd slow and the top one very fast?  Are they too much the same?  
Is the top one slower than the lower one?

If the answer to any of these is "yes", then you have an error somewhere.  
You should be able to find that "busy" 4th or 5th ar that 4th or 5th which is 
too pure if your contiguous 3rds test indicates an error.  The point here is 
to try to *prove* which note is in error or at least which note has the 
greatest error to start correcting the temperament.  When you find the note 
which is too sharp or flat by proving it this way, you can make your fine 
correction.  Such a correction will be on the order of less than 2 cents, 
maybe even less than one cent which is very fine work.  Don't make the 
mistake of changing things without proving the need to do so because you may 
increase your errors rather than eliminate them. 

When your two octave span sounds perfectly smooth and even, it is time to 
score it.  If it really is smooth with no rough spots, no wobbly 5ths, no too 
pure 4ths and 3rds & 6ths which ascend and descend very smoothly and the 
contiguous 3rds tests also seem to check out, you'll have a temperament which 
should score above 90.  A temperament that only scores an 80 will have easily 
detectable errors but will still usually not be so bad that it would produce 
unmusical dissonances.  That's why the tolerance is set where it is.  It 
takes a very highly skilled tuner to produce a score of 100, so don't be 
disappointed if you get passing scores but still have a few errors.  Indeed,  
a temperament which only scores about an 80 can still produce a good sounding 
piano if the unisons and octaves are also within Exam tolerances.

If you have any more questions about this, feel free to post them.

Bill Bremmer RPT
Madison, Wisconsin

---------------------- multipart/alternative attachment
An HTML attachment was scrubbed...
URL: https://www.moypiano.com/ptg/pianotech.php/attachments/21/1f/6a/a4/attachment.htm

---------------------- multipart/alternative attachment--