I have a 1904 Weber 7’4” Grand piano in the process of restoration in my shop. Below is a look at the string and rib scale relationship.
This piano had a cracked soundboard that was flat, a short sustain, however, the tone had a nice fullness to it, which showed lots of potential.
The string scale was basically 10 notes for one gauge and the tension curve showed that in between gauges would benefit for smoother transitions. I went with a 6 /4 pattern which worked well. Altering the scale only added another 40lbs to the downbearing.
This piano had been rebuilt before, and the bass string scale was aweful and uneven. The lowest bass string had a 4.3 tension to length ratio 286lbs/66in. Since the lowest tenor string was 160lbs, I scaled the bass section like this- 8 Unichords 286 – 322,8 bichords 244-276, 6 trichords 203- 230. This way each was in the 39- 50 breaking percent range.
So next, I examined the rib scale to see if it could support the downbearing.
The rib scale was beautifully engineered and precise. Only a little improvement could be found. Ribs 8 and 13 were the hard working ribs and needed to be a little larger to bring down the stress level to match the others. The rib profile was at 76% originally, and with a little adjustment, I was able to get it down to 73%. This will give the piano tone a little more depth. I accomplished this by raising the average stress from 1,977 to 2,007 lbf (still within acceptable range)and changing the rib profile to a little wider and not as tall.
Here’s the specs:
Resistance….. .10.72 Sq.In.
Board Sq.Ft ....15
Rib Count ……15
Rib Stress…….2,007 lbf
Now with the assurance that the string scale is supported by the rib scale, the rebuild is off on a solid foundation.
I particularly like this Weber design as it is well engineered, inexpensive compared to known brands, but on equal footing in my opinion
It looks to be a fine project for sure. The numbers you include seem to come from a very specific protocol. Would you mind explaining them to those of us who don't know how they were arrived at ? Thanks in advance.
Thank you for the inquiry.
First the bass string scaling method is heavily inspired by the 3-part articles “Some Thoughts on the Design of Bass Strings” by Richard M. Brown August 1988
The rest is a practical system I came up with to analyze a soundboard that will pinpoint the problems it has (if any), improve its resonance, and make sure it can properly support the downbearing. I came up with six elements- Force, Resistance, Board Square Feet, Rib Count, Rib Profile, Rib Stress. After comparing several boards to one another a clear picture begins to form. Most boards are not engineered properly! Because of that the tone suffers. For example boards that “sound” stiff have ribs that are too large with a high Rib Profile. I’m finding boards in the 93-98% range. Boards with a low tone and plenty of resonance have a rib profile in the 70% range. A Steinway O that came in had a rib profile of 83%, with leaves much room for improvement. I have a piano in my shop with a 68% profile and it has a fantastic deep boom sound. It is a large upright with a 12 sq. ft. board that really sounded much “bigger”. And I found that it really just came down to good engineering.
The six elements are not independent from each other, but interrelate. So if one element is changed, it has an effect on the others. As an example, a common ailment is having one rib short. This creates a situation in which all the ribs are larger than they should be, and at the same time, they are over stressed and under-engineered. Let me tell you that it is an amazing feat of engineering for that designer (Jacob Gross) to get such a low rib profile of 68%, all while maintaining the other elements in the proper proportion.
My system evolved by asking myself two questions that kept coming up.
Why copy soundboards with tonal problems and obvious design flaws and bad engineering?
How did some boards achieve such a low tone resonance, and most do not?
Now with the ability to answer those questions and make the necessary adjustments, allows me to rebuild a clients piano to exceed their expectations.
I apologize. When I just hit reply, it seems to insert all those pesky Question marks. I don't know whay, but being an ex-teacher, it bothered me enough to come on the website and reply that way. Sorry. Clark
By sound I am referring to the the Tap test that is used when the strings and plate are removed. The Rib Profile element is the "sum of the Height to Width ratio" of the ribs. As a crude example, if all the ribs were 1/2" tall and 1" wide that would be 50% Rib Profile.
I guess that's learned through experience. Comparing one to another. Most common are the high pitched boards with their engineering problems. I'll be posting examples of soundboards with problems to correspond to the chapter presentations.
I think you should teach this at national next year.
Sheffey Gregory, RPT
Speaking as an executive of Westpac Regional Conference it is understood that any person teaching classes must have some credentials of professional peer review. I would think this would apply to the national conferences.
Hello Sheffer I am submitting the paperwork.
Roger, I find your statement offensive, insulting, and unbecoming of a professional. Not something that inspires other people to become members.
Curious as to why you find it so. Your name is one of many in this wonderful invisible world that we call the Internet. Like my name is to you, I'm sure. :)
If I'm going to a national piano technician convention, I would expect to find teachers who are qualified and have been vetted. There are people who do not really know their subject matter. And there are people who know but cannot teach well. I am not saying either applies to you, BTW.
Teaching at chapter levels, then at regional, then at national seems to make perfect sense to me. Hopefully by then anyone not qualified has been weeded out by then.
Hi John, You are right and I am currently going through the steps that I have been encouraged to do by some very nice people here.
Roger and I have a history, and that was nothing but an example of a bully pulpit.
I am getting wonderful emails from techs around the country so I will continue contributing.
With all due respect to you, i have been very precise, and not dancing around the topic. This is basic engineering. Engineering 101. In fact, I would not dare go onto an engineering forum with this simple stuff. But since soundboards are engineered, and my specialty, in my opiniion worthy of discussion with other piano experts willing to learn.
I realize you may be having trouble with the engineering principles. I assure you, i have not invented anything new here. i simply have put in the hard work (research) and have eliminated the unnessecary terms from one trade and made it presentable to another trade.
For example; I have already posted twice, the formula for stress. It is Moment divided by Section. Bending Moment divided by the Section Modulus. Google those. By the way, you use those two principles every time you tune a piano. Or to be a little more precise Moment of Torgue and Section Modulus. As a rough guess, you are probably using 5-10 lbs of force and exerting 200lbf of torgue.
Hope that's a little more clear. I will go into a little more about Moment, Section, and Stress on my next post.
P.S. I have not discussed compression or downbearing because of non relevance. I've been discussing principles, not constants and methods.
Here is an exercise/challenge if someone out there wants to try it.
Given the Height, Width,Stress load, Force, and soundboard thickness of the rib below, solve algebraically for the length.
I'll confirm it if you are right.
Stress- 1.854 lbf
Force in Center- 60lbs
soundboard thickness- .30
------------------------------ChrisChernobieffChernobieff Piano and Harpsichord Mfg.Lenoir City TN865email@example.com/ChernobieffPianoandHarpsichordMFG
------------------------------ChrisChernobieffChernobieff Piano and Harpsichord Mfg.Lenoir City TN865firstname.lastname@example.org/ChernobieffPianoandHarpsichordMFG
Mr Chernobieff: Great post! Could not find formula quick on Google (Eng.101) but am inspired. 2 ?s for now please. 15sq ft seems too small for board size on a 7 ft piano. How is this figured? Can panels routinely be reused replacing only ribs?
------------------------------ChrisChernobieffChernobieff Piano and Harpsichord Mfg.Lenoir City TN865email@example.com/ChernobieffPianoandHarpsichordMFGOriginal Message:Sent: 04-15-2016 14:14From: Chris ChernobieffSubject: A Weber Semi-Concert Grand Piano Analysis
Thank You Paul.
Is that small? Thats 2,160 sq, inches and it looks big when I stand next to it. LOL
Regarding reusing a panel. Good question, I have never done it. The problem is getting it out of the piano without half destroying it. With some of that beautiful Adirondack Spruce that used, it sure would be nice.
Maybe turn piano upside down and put wet cloth around rim overnight. Have not tried.Thank you.Thought sq.ft. measurement only included part of the panel.
Calculating soundboard area can be complex using math formulas trying to figure out odd shapes. So i ended up coming up with a crude way that works for me. After popping out the board, i get 1/4 masking tape and physically map out each square foot. It makes it so easy that you can figure out the square foot from a photo.
The length of what is 27"? There are 15 ribs. This conversation is non productive.
testy remarks on this forum
George; I count 1 possible testy remark by one and at least 6 elsewhere. Refers where? Rhetorical question better suited for PTG-L.or ignored as usual.
Most people I know are calculating rib scales based on beam formulas which require length,width height and load in addition to the modulus. I would also find it helpful to know if you're using fixed ends or simple ends. I use fixed ends but some don't. Center loading is the typical format even though the upper end of the piano deviates quite a bit from that.
There will be variations based on the fact the board designers use somewhat different criteria in their decision making and sometimes include things like how much crowning of the ribs is done and to some extent the EMC of the panel at glue up, though it should be noted that the range other than NY Steinway is pretty small, probably 5-6% EMC. All panels under normal conditions have some compression. Some, as in the case of NY Steinway traditional methods, have more. The actual amount of compression, of course, will vary with the ambient humidity and the panel's ability to withstand compression loads. Load distribution between the ribs does vary from designer to designer. The panel does add some stiffness but since it the panel contribution is cross grain it doesn't deliver the same stiffness as you would see if you were to increase the rib height by the same amount along the grain. But it does add some stiffness. At the same time the rib scalloping will reduce the effective strength of the beam at the edge since most designers use beam formulas with a constant cross sections. Some will treat those two things (scalloping and the added stiffness of the panel) as offsetting and therefore use the constant beam cross section as the basis for their design..
It would be simple enough if you (Chris) were to provide complete rib dimensions and the species modulus to determine under a narrow range of parameters how your design compares with others'. I assume you posted this as a basis for discussion. Rather than shroud the data in mystery why don't you just post it. For me that means L, W, H and species at least. I'm not really interested in being quizzed. If you are crowning the ribs more than, say 16-18 M radius that might matter. Also if you are crowning the ribs are you accounting for the reduced volume as the rib tapers away from the center height? If so, how? For me it doesn't really matter whether you are relying on pure compression crowning or building some in in terms of my ability to analyze the rib scale on its own merits. If it's under engineered then it will be easy to see and conclude that you may need more supporting compression. Likewise If it's over engineered.
That being said I do think that for these types of construction there is a "sweet spot" that gives a balance between power and sustain. But as we've seen that can vary some and still produce a satisfactory result. At some point, however, I believe you can personalize the designs to where they begin to sound strange. Downbearing settings are a separate, though related, discussion.
Let's just post the relevant data and see how our ideas compare. There's room for differences.
Thank you for the good questions.
First I have avoided talking about methods of construction because that is not the point. I believe in high compression boards, but not as a 'make up for' any bad engineering. What I am doing and have done is come up with a simple way to compare soundboards from one make to another. Discover their mistakes if any and fix them with small improvements. What conditions or methods those boards were made under doesn't matter. Just comparing the boards physical attributes.
The Weber is the first I posted and i pointed out possible mistakes and how they can be corrected. As more boards get analysed I will have more useful data. I have observed so far that boards that are under engineered are flat a 100 years later. The two that are over engineered still have full crown and sound great a 100 years later. So I think the engineering matters.
I am soon to be popping a board out of a M&H BB. I will post the rib data side by side with the rib data of the Weber for your study.
Center loading beam formula is the least useable. More useable and accurate is the One off center formula (I already posted an example), and the two force beam formula.
On the open vs fixed beam test to determine the proper MOE. Both are incorrect for soundboards, and will give you an incorrect deflection amount. Iv'e done both and rejected both. MOE for Spruce is roughly 1.9 Million on an open end. For fixed it is roughly 3.6 million. I actually applaud you for noticing and checking into that. As for soundboards the MOE is roughly 3 times that, and vary to size.
You asked me a lot of questions so I dont know if i got them all or not. Its much easier to convey the info in a class setting with actual soundboards to compare and look at, along with their data.
Again I will post the rib data on two rib scales side by side mid may.
Thank you David.
And on the question of rib shape.
From an engineering point of view (and simplicity),it made sense to use an unaltered rib profile as a basis. I assume that anything that is taken away from that, goes to what i call an "uncertain engineering" and has a further weakening effect. So if the rib resistance upon design is "just" strong enough, then it had been severely weakened when planed by hand.
Actually all I've asked is dimensions, materials (modulus), and load, several times now. I'll assume you don't want to give that information. But without it I can't participate in the discussion.
John Formsma wrote:
"If I'm going to a national piano technician convention, I would expect to find teachers who are qualified and have been vetted. There are people who do not really know their subject matter. And there are people who know but cannot teach well. I am not saying either applies to you, BTW.
Teaching at chapter levels, then at regional, then at national seems to make perfect sense to me. Hopefully by then anyone not qualified has been weeded out by then."
John, that sounds like a good way to go. However, I am not sure that what you have outlined is how instructors for the national convention are chosen, either now, or in the past. In fact, I am often frustrated at how so many in our organization do not seem to have a clear notion as to what constitutes a good teacher, particularly when we have so many great instructors among our members. (For example, if you look it up "teacher" in the dictionary--and your dictionary is any good--you should see a picture of LaRoy Edwards!)
Thank you for your questions, I appreciate them. I admit I was taken somewhat aback by your aggressiveness and so I shortened my replies, and because of that, my response may have had a harsh tone. So I apologize for any mis-spoken words on my part.
George I am sure meant well, as any good leader would.
My system of analysis, (as would any system) has built-in assumptions. It would be impossible to calculate every nuance. Nor would that be practical. I have kept it to 6 basic elements (Force, Resistance,Board sq ft, Rib Count, Rib profile, Rib Stress) that I feel are the most important for analysis. It also keeps it simple and makes analysis quick.
For your concerns on compression, how would you quantify that? And input that strength value into a computer? How would that be helpful?
I do measure the thicknesses of each board, and that becomes part of the Height of the ribs.
Downbearing for analysis purposes, I feel is a constant and not an element. In my software, I keep an eye on 3 downbearing types. The actual downbearing, the Max downbearing, and the % between downbearing.
I also wanted to avoid methods of construction as there are most likely many “recipes” and systems to construct soundboards. And people get protective of their methods. I believe that I have created a valuable tool that is all inclusive not exclusive. Believe it or not, I respect all those methods. I just love soundboards all together.
------------------------------Alan Eder, RPTHerb Alpert School of MusicCalifornia Institute of the ArtsValencia, CA661.904.6483
If you really believe that a CC board is totally panel supportive, then don't put any ribs on the next one. I'd like to see that.
Chris: Sorry. I do not get you on this (CC?) either. Thanks Ron for explaining difference with 6% and board support as that was another ? for me. Chris: Still mostly on taxes but will try your formula eventually. I also wonder how species of wood figures in calculations. Any old wood would do. Ha,ha Balsa/hickory ok?
CC was Rons abreviation for Compression Crown. He and i believe differently on compression theory and the strength of it to support downbearing. From an engineering perspective mass is mass. An arch an arch. Sure there are fluctuations. For example, could a 2x4 of the same species and size be stronger than another?
I think of compression like that as far as added strength is concerned. I think one should make a soundboard in the way that they are compfortable with the results. Engineering has built in safety factors to compensate.
My system will simply show if a board is under engineered. I suppose you have to assume that the invisible compression force is performing when the mass is under performing. But gravity and the downward force will eventually overcome an under engineered board. Will that take 5 years? 50 years? Do the advantages outweigh the disadvantages?
Different species would only matter in the deflection formula. Each species has its own MOE so make sure that you are using the correct one in the deflection formula.
Good luck with the balsa wood soundboard! LOL
Chris: I know CC stands for compression crowned but one needs the ribs to form the crown by restraining one side of the panel as it re-hydrates from being dried to 4%.
Sorry Paul the answer was 9. But not a bad try.
David here is the data you wanted. I didn't include any MOE as that is program specific and would depend on what parameters you want to use. So my MOE would do you no good. The weber soundboard thickness is.31 throughout. And not sure the totals will do you any good as you have expressed the disapproval of my dispersion method. Enjoy the rest of the weekend fellas.
OK. What species of wood for the ribs: sugar pine, white spruce, Sitka? And what is the total load? I thought I saw ~900lbs earlier in the discussion.
I'm not certain of the exact species of the Weber board. Could be Red Spruce. Load is 19.53 Tons = 39,058lbs. I came up with 901lbs downbearing.
So I've attached my analysis using just a couple of metrics that I employ. I've analyzed this based on sugar pine ribs (which they probably are), Red Spruce would add some stiffness though the trend would be the same. I've posted your rib scale (I've reversed the rib numbers since Rib #1 is usually the lowest bass rib, I think you had them reversed based on the lengths), followed by how I would approach this design were I making a set of ribs from scratch using sugar pine, which I usually don't use. I've spread the design load as I typically would in a scale design. Others may have a slightly different approach. I've purposefully arranged it so some ribs are taller and some are wider to demonstrate that the MOI doesn't much pay attention how the rib achieves this particular output. The ratio itself doesn't really matter as a determinate value. A wider rib will need to be less tall and a taller one less wide. The height of the rib contributes more in terms of stiffness by a factor of 3, as you know. I use fixed ends formulas not simple ends as I explained earlier. Rib scale performance is based on a constant cross section. The scalloping of the rib removes some strength and the addition of the panel, even with some panel thinning, adds some. I do thin the panels. The amount depends on the panel species. If the ribs are more or less crowned than the 15M radius with or without compression one might choose to alter the design slightly. I use a 18M radius with some compression (EMC = 5 - 5.5%, pretty typical) which results in something near a 15M radius with some compression.
I prefer my rib scale over yours for this one. The deflection characteristics and MOI are smoother and transition better. Yours is weak in the bass end in particular. I do use other metrics as well which I have not included but this gives a very basic view of how the board will perform.
Thank you for sharing, that was very educational to see how you would rescale the weber. Below are a side by side comparison of the original, my modification and your modification. One thing I can say is that your changes have no comparison to anything historical I have studied so far. I have tried to stay close to the original, with the intention of finding small improvements to their design if any. You totally changed it to something different with the rib profile at 102%. Your braver than me!
Thank you again, hope to meet one day!