Plate and Rim (was Re: Tension Resonator)

[Delwin D Fandrich]

----- Original Message -----
From: "Phillip Ford" <fordpiano@earthlink.net>
To: <pianotech@ptg.org>
Sent: May 08, 2002 11:53 AM
Subject: Re: Plate and Rim (was Re: Tension Resonator)


> Del,
>
> Actually I'm not assuming that there is a substantial amount of load
> transfer from plate to rim.  In my initial post I said that I thought that
> there was essentially none.  After discussing this with Robin I conceded
> that I thought there could be some because there is capability to transfer
> load.

Sorry, I came into this late and didn't catch that.


>
> >I once joked that if all we had to worry about was the string load a few
#10
> >wood screws would be sufficient to hold the plate in the rim. Of course
> >there's a bit more to it than that. The greatest strain is placed on the
rim
> >bolts/screws during moving or shipping. Even a lightly built plate is
rather
> >heavy and dumping a wheel in that first pothole would probably rip those
#10
> >wood screws right out of that Select Hardwood rim and the whole thing
would
> >end up on the floor. While it is a formidable task holding a 200 to 500
> >pound (90 to 230 kgf) iron casting in place and it would seem that a good
> >friction fit between the plate and the rim would be essential, even the
> >Baldwin system seems to do an adequate job.
>
>
> I hadn't thought about this and I can see that it would be an issue.  I
> don't agree about the friction fit being essential.  What is essential, in
> my view, is a load path from the plate to the point where the load would
be
> reacted, in this case the flat side of the case that is sitting on the
> skid.

Good point. There are several ways to accomplish this.


>
> I suppose the most direct load path would be to the straight side of
> the rim as in a full perimeter plate.  But since most pianos don't have
> that then the load has to go through the rim (or through the rim braces
> over to the rim) and bend around to the straight side.  The load from the
> plate to the rim for this particular loading will be a shear load (as
would
> the load from the strings if the plate and rim were sharing the load).  So
> there needs to be a shear path.  Friction would be one way, but I think
not
> a very reliable way in a structure made of wood or partly of wood, since
> the friction is only developed if the plate screws are tight, which I
> believe is often not the case because of wood movement.  The Baldwin
system
> provides a very good and reliable shear path from plate to rim if the
bolts
> are big enough.

As I recall they are hardened 3/8" x 5" fully-threaded machine bolts.


>
> >Among the primary functions of the plate are (first) to support string
> >tension and (second) to establish and maintain a desired relationship
> >between the string plane and the soundboard/bridge assembly. If nothing
was
> >changing or shifting my #10 wood screws might well be adequate. But even
the
> >most well-designed rim assembly is going to shift and twist with changes
in
> >relative humidity--at least they will as long as they are made of wood.
If
> >the rim is able to do this independently of the plate there will be
changes
> >in the string plane to bridge/soundboard relationship that will result in
> >tuning instability. If the plate is fastened to the rim such that the two
> >essentially form an integral whole these changes in relationship will be
> >minimized and tuning stability will be improved.
>
>
> I think the point of my posts was that if the plate is resisting movement
> of the rim, then in turn the rim has to be resisting movement of the
> plate.  How well the two are coupled will depend on the effectiveness and
> stiffness of the fastening system.  If the plate wants to move under
string
> load (or maybe more correctly, if the points on the plate where the plate
> screws are, want to move under string load) then the rim has to move with
> it if the connections are effective enough for the plate to also resist
> movement of the rim.  I was, however, fixating on string load being shared
> by the rim.  I hadn't been thinking about the plate resisting rim movement
> or twisting.  If the plate is resisting rim movement but is not putting
> much string load on the rim then my conclusion from this statement would
be
> that the plate is considerably stiffer than the rim.  I think this is
> probably true.

Hmm. I suspect that would depend on where and how the stiffness was
measured. With the plate screwed to the pinblock of any model Steinway
grand, for example, but free from the rim it will be much easier to twist
the tail of the plate than the tail of the rim. At least in the short term.
Over a long period of time the rim might twist somewhat more, I suppose.

My thinking was (is?) that if there is going to be twisting (or whatever)
you would like both to twist (or whatever) together such that the
relationship of the strings to the soundboard bridges remained the same.


>
> However, if part of the function of coupling the plate to the rim is to
> resist rim movement then I think things may not have been set up very well
> on the typical piano.  The plate is only attached to the rim along the
> curved side, not along the straight side or the belly rail.  Perhaps a
full
> perimeter plate with some lugs down to the belly rail would be more
> effective.  Perhaps better yet is my idea of a metal spider or frame to
> replace the wooden rim braces and not depending on the plate to perform
> this function.

Or, at least a bracing mechanism that was designed to prevent the
deformation of the rim over time.


>
> >Taken from this perspective there are any number of fastening systems
that
> >are adequate to the task.
>
>
> Here again I don't agree.  I think the standard system of a plate sitting
> on wood dowels with a screw holding it down to a wood rim is a bad
> design for an application like this.  It may be adequate, but depends
> on friction between the plate, the
> screws and the dowels, and this friction is too easily lost in normal
> service.  If you view tightening plate screws as normal service which will
> be done as a matter of course then maybe this system is acceptable.

Unless the screws go into Select Hardwood in which repeated tightening will
surely strip out the wood. Especially with the relatively small screws found
in some of these pianos.

Still, pianos such as the Steinway (of all ages) get shipped all over the
country without mishap and without any screw tightening going on first. What
holds those plates in place? I suspect the pinblock screws and whatever fit
exists between the plate pinblock flange and the pinblock.


>
> But if
> you look on this as service which is not likely to get done then this
> system can be improved upon.  The Baldwin system doesn't have this
> shortcoming.  As I see it, the Baldwin system is essentially maintenance
> free.  I'm sure there are any number of ways to design an adequate system
> to perform this function, the Baldwin system is just one that I'm familiar
> with and which serves as a good example.

OK. I hadn't though about it from that perspective but I see your point.
I've not investigated the coupling between the plate and rim overly much.
I've used the threaded plate system on a number of pianos that I've
remanufactured and on several that I've designed (including the Walter
plate) and I like it a lot. I'll also be using it on the new grand. Once I
got hooked on that system I really stopped considering the others.

Del