Duplex

[Overs Pianos]

Hi Ron N,

>  >They tend to go out of tune to about the same degree as the speaking
>>lengths in the extreme treble,
>
>Where the hitch to rear bridge pin lengths are most nearly the same as the
>front bridge pin to tuning pin lengths. About half, actually, or a 
>little less.

Yep, the same length exactly for the top six notes.

>  > to a bit more towards the lower end of
>>the second top section.
>
>Where the hitch to bridge pin lengths are more like 30% the length of the
>front bridge pin to the tuning pin.

25% of the speaking length actually. You would expect these to be the 
big problem on account of their relative shortness, but they seem to 
be quite OK in practice.

>  >The initial positioning will always have errors of measurement, and
>>will require adjustment. The act of tuning will, in the initial
>>months, require a combination of moving the block and adjusting the
>>duplex tension. As the block positioning improves (by being adjusted
>>and refined at the time of tuning) the duplex tuning will come to be
>>accomplished more by tension adjustment and less by block
>>repositioning.
>
>I understand. As you said, it takes a while to find the ideal position.

In fact, once the blocks are properly sorted, it becomes absolutely 
unnecessary to move the blocks at all. When tuning such a piano, the 
duplexes are tuned first before tuning the speaking length - by 
rubbing the speaking length for tuning up, or rubbing the duplex 
segment for tuning down.

>. . . . . . My understanding is that to most
>of the folks moving aliquots to tune duplexes, tuning the duplex consists
>of merely moving the aliquot to a "better" position than it was in before
>they moved it. The test being to pluck segment lengths and listen to get
>them close.

If this is their approach it would be a surefire way to wreck tuning 
stability. When we tune duplexes we are not just interested in 
getting them close, we are tuning them just as accurately as the 
speaking lengths. When the piano is next tuned, you can tell if the 
block is in the correct position by the string creep (if any) across 
the bridge. If the block is in the correct position there will be no 
string creep.

>After the aliquots are positioned, by this single procedure,
>the duplex isn't touched again. Ever, as far as I know, because it is now a
>"tuned" duplex. That's the way I understood the idea that Mr Franklin
>supports, and unless my memory has holes in it (which is quite likely),
>that's the procedure demonstrated in his classes. Dan, If I've screwed this
>up and have it wrong, please straighten me out here.

Tell us Dan. How be it?

>As far as I know, Ron,
>you are the only one actually specifically tuning the duplex tensions with
>each and every tuning.

If this is so then it is disappointing. Such an approach as you 
mentioned above would surely result in 'screaming' duplexes and most 
everything else as well probably before the kit gets back into the 
vehicle. Duplex segments can't be left nearly in tune.

>(RN)>>How can this possibly be? Haven't we already determined that 
>strings will
>  >>render across bridges?
>  >
>(RO)>No. The wire only renders across the bridge when the speaking length
>  >or duplex length is 'rubbed' down, or when the differential becomes
>  >large enough for the string to pull through the friction created by
>  >the bridge pin offset angle.
>
>(RN)And isn't that strings rendering across bridges?

Yes, but it won't happen if the tension is equalised across the bridge.

>And why do individual
>strings sometimes creep high after you are through with them when lowering
>to pitch, if it's not from strings rendering across bridges?

Sure, it's rendering again, but this is because the tension becomes 
relatively higher out the back when the speaking lengths are let 
down. In the event of pitch lowering, it would be an advantage to rub 
the back scale to encourage any pullover to the back scale to occur 
before it is tuned. Just the opposite of the pitch raise situation, 
when you would rub the speaking length to pull the backscale tension 
up.

>. . So the tendency of your tuned duplex system
>(technique) to be more stable is the result of taking the time and effort
>to individually test and adjust the tension in every duplex section in the
>instrument . . .

Yes.

>As you said, it
>isn't the tuned duplex at all that accounts for this, but the specific
>effort to adjust segment tensions that makes the difference. Terry Farrell
>brought this up a while back too, without tying it to tuned duplexes. Yes,
>it would almost certainly produce more stable tunings.

Exactly - rock solid, wonderful isn't it? Mr Farrell seems to be a 
very fast learner. You haven't been doing this stuff long have you 
Terry?

>So how do you equalize segment tensions in the tenor and bass where 
>the duplexes aren't "tuned" and are braided off?

Similarly, the back scale or speaking lengths should be rubbed to get 
the back scale as close as possible to the speaking length tension (I 
use a hammer shank). Sure, you don't really know where the tension 
is, but you know that if you just pulled the piano up 25 cents, then 
the back scale will need to be raised. As I also mentioned a week or 
so back, I rub all the speaking lengths mildly every now and then 
before a tuning, particularly when I really want the piano to stand 
up to a thrashing (for the covered strings I use the back (squared) 
end of a felt tuning wedge - after lifting the dampers with the 
sustain pedal to avoid damper felt elongation).

>This is the point I was trying to make. Positioning the aliquots to produce
>a "tuned" duplex won't produce a duplex that will stay in tune with it's
>corresponding speaking length.

Certainly won't. Now I understand what you were referring to. 
Adjusting the duplex position alone to bring the duplex segment into 
tune with the harmonic of the speaking length, without considering 
where the tension is would be an absolutely and utterly inadvisable 
approach. This would be one sure way to ensure that the piano 
wouldn't stay in tune for five minutes.

This must be the problem with tuned rear duplexes, folks don't seem 
to understand how to deal with them. There is a procedure one must 
follow to get them in tune and tensioned very close to the speaking 
length. Maybe this would be a good convention class topic at some 
time - complete with piano, tuning lever and other necessary stuff.

Regards,

Ron O
-- 
______________________________

Website:  http://www.overspianos.com.au
Email:        mailto:ron@overspianos.com.au
______________________________