[pianotech] GH-1s

[David Love]

I'm not clear about what and why you are asking for what you are.  For your
"apparent information" the bridge distributes the load between ribs.  If the
way you calculate the load between two adjacent ribs is based on how the
actual unisons sit over them the and that, you believe, represents the
actual load that those ribs carry I think you are mistaken.  The bridge, yet
another beam, distributes the load across those adjacent ribs.  If you
measure the load by virtue of unison placement of two adjacent ribs at 50
and 25 lbs respectively are you telling me that you would set up those ribs
with dimensions appropriate to those specific loads?  I certainly wouldn't.


I've given you my answer to your questions comparing your theoretical rib
with mine using the same criteria under a specific load.  My rib is much
lighter as you can see.  Of course, a rib that is 1090 mm will be larger in
cross section.  What's your point?  You should be able to glean the
difference in our approaches from what I provided.  I have no intention of
chasing every little test you feel like throwing out there.  You have enough
information to outline our differences.  

I did not, btw, say that I don't use cut-off bars.  I have and do use them
sometimes.  But not always depending on the customer's comfort level with
redesign issues and how I view the project.  My own preference, if the truth
be known, is for a modest cut-off bar, very modest.  The long ribs I prefer
to see just under 1000 mm.  But I don't feel that unless I do that the board
will fail.  I can tell you from having done several without cut-off bars
that they don't.  Compared to the original rib scales that I find on many
pianos there is no question but that I make changes.  The targets I hit are
very specific.  In compression crowned boards I am almost always increasing
the load bearing capacity of many of the ribs and changing the overall way
in which they transition.  But I don't believe that reducing the long rib by
some 400 mm is necessary or, for that matter, even desirable.  With respect
to tighter radii, no, I don't go to 9M anymore, though I have and have gone
tighter in some designs.  Having done pianos both ways I am unconvinced
about variable radii in ribs.  I can't and am not saying that it's
necessarily a detriment (unless it produces more crown than is desirable for
fit under the plate) I can say that I'm not convinced that it's necessary or
contributes in any meaningful way any more than a more modest radius.  I do
crown ribs and I do press them into cauls slightly tighter than I cut them.


With respect to loading, I distribute the load relatively evenly through the
piano in my calculation phase.  If there are 14 ribs and the load is 900 lbs
then the load per rib would be 64 lbs.  That's where I start.  There are
other considerations and several other calculations that I use beyond spring
rate or percent deflection that impact my final decision.  I also consider
how I feather the ribs in any given section and how much and where I thin
the panel to address different impedance requirements in different sections.
But in my view, your rib, which you've calculated out to support 25 lbs, is
significantly  overbuilt compared to how I would have done it.  I think that
is clear.   

David Love
www.davidlovepianos.com


-----Original Message-----
From: pianotech-bounces at ptg.org [mailto:pianotech-bounces at ptg.org] On Behalf
Of Ron Nossaman
Sent: Thursday, December 20, 2012 3:26 PM
To: pianotech at ptg.org
Subject: Re: [pianotech] GH-1s

On 12/20/2012 4:34 PM, David Love wrote:
> Yes, I understand that the cutoff makes for a smaller rib cross 
> section but it's not without other consequences.  Why not reduce the 
> rib to 550 mm.  You could reduce the cross section even further.  The 
> tonal implications can't be ignored.  We can talk about physical 
> structure and ignore subjective evaluation of tonal consequences but 
> we are building a musical instrument, not a bridge to drive toy cars over.

We're talking load and rib sizing, or we're not talking.


> Your question for comparison is a bit in isolation and so it's hard to 
> be very precise because I don't approach it this way but I will try 
> just for comparison sake.  Just so we are on the same page I'll assume 
> you did these beam calculations with MOE for sitka spruce.

Isolation means nothing. It's a simple straightforward no bullshit, no
diversion rib loading calculation. Yes, Sitka at 1570000 psi.


> I actually use 1.5 degrees, not 1 degree, but since you calculated the rib
> dimensions based on a load of 25 lbs I'll use that.  Twenty-five pounds
load
> on a rib is very low to me.  If I extrapolate that out and use the total
> load on a board as I would load it then that would mean that the board
would
> have some 36 ribs!  Approximately 900 lbs divided by 25 lbs gives 36 ribs.
> Even at 1 degree it would be about 600 lbs and use 24 ribs (I'm using
35000
> lbs total tension, I don't remember what a Steinway B actually is off the
> top of my head).  First of all, I wouldn't build something like that.  Nor
> do I target a specific certain percentage for deflection which is a
somewhat
> arbitrary number depending on the amount of crown you have as you pointed
> out. My approach is a bit different.

That 25lbs is from bearing calculations of unisons actually being 
carried by the rib (one of 18), not an arbitrary equal dispersion. Ribs 
in actual pianos aren't equally loaded, for your apparent information.


> But if I had a rib as you describe and wanted a 50% deflection then my 770
> mm rib with 9M radius that needed to support 25 lb of calculated load
would
> measure more like 15 mm wide by 16 mm high, roughly.  That's 24 ribs at 1
> degree or 36 ribs at 1.5 degrees for the whole piano.  Of course, I'm
using
> fixed end formulas, not simple ends formulas.

That's a start, but you said you don't use tight radii, nor cutoff bars. 
That's what I would do. I asked for your claimed real approach, full 
length with a more conventional bigger radius. What would your sectional 
dimensions be for a 1090mm rib at a 16M radius, for instance?
Ron N