Official Luthiers Forum!

I'd like to see how you are setup to do these test. What equipment have you built or use to do this with.

Well I guess I'll just answer my own question with a little help from Edzard.
http://www.ukuleles.com/Technology/statmeas1.html
Anyone else interested in this stuff?

Chris, have you seen this old topic?

http://www.luthiersforum.com/forum/viewtopic.php?f=10101&t=17956

I've pretty much copied Brock's setup...

Christian

Sure haven't, Thanks.
I like the flexability of the adjustable span. I think I'd also would like to be able to put the dia indicator on the top or on the bottom. I also would like to have the weight span the plate.

Chris,

The method by David Hurd on his ukulele site that you quoted is definitely the way to go.
it has the advantages of

- true uniform loading across the width due to the long bar used as the load.
- true centre of span loading due to having the dial indicator underneath.
- deriving a stiffness result "E" rather than just a deflection number which is dependent on span, width, and thickness of the sample as well as material stiffness.

Knowing whether the E value of the sample is High, low or medium before thicknessing will allow you to make sensible decisions about the thickness you go for.

I am and I even took a deflection measurement on my last bandura top but I lost the paper where I wrote it down!

I use a similar setup to the ones shown but I use triangular supports for the top. I haven't looked at the E equation lately but when deflecting on a curved surface like a dowel, the length of the span will decrease depending on the amount of deflection which will affect the calculation. Using a triangular setup will minimize this. I made those supports shaped like 90o triangles to make measuring the span easier.

The thing that this method misses though is the cross grain stiffness and according to many, that's where the real magic is, or at least the ratio of lengthwise to crosswise can better help determine what type of guit a particular top will work well on. I recall reading that Alan C. uses a sonic method of determining both lengthwise and crosswise E to within 10%. Anyone know if that's in the DVD?

_________________
http://www.birkonium.com CNC Products for Luthiers
http://banduramaker.blogspot.com

Wouldn't turning the plate 90 deg. and measure the deflection do that? Although I've found that material I haven't read through it yet as I'm trying to get a tone generator system amoung other things together.
Also when or if you measure your brace deflection do you just put the ruff brace across the span and measure the deflection of each brace? That's where I see the adjustable span also being usefull.
Would not a square bar do the same as a triangle and do you really think that it makes that much of a difference?

For a bare top I use 2 2X4's flat under the edges of the top. I see how much weight it takes to make the top touch the bench. I do this both directions. For a braced top or back I use 2 strips of 3/4" plywood and do the same thing. I'm sure there's alot more elaborate set ups but I like stuff simple.

_________________
I didn't mean to say it, but I meant what I said.
http://www.brackettinstruments.com/

Yes, turning the plate 90o should do that. I think we'd just have to be careful not to overload the plate. I'm not yet applying any of these tests mostly due to a lack of time, but I thought I should at least take the measurements so that I could sift through the data later.

As to braces, it should probably be close enough to just measure one rough brace while it's still rectangular in shape, assuming they're all from the same billet. I realize that even within the same billet there can be variations but c'mon... Final tuning of the bracing will be done either by ear or by pattern anyway.

From David's website:



We can see that the length term is cubed which means that any error in measurement will have a cube effect on the end result. That's pretty big. I think you're right though that using a square support should be about the same. I just think the dowel is a bad idea because as you load the plate, the free length will shrink and the error will be cubed.

_________________
http://www.birkonium.com CNC Products for Luthiers
http://banduramaker.blogspot.com

Hi Chris, did you top your guitar with the top from your video? Was it too loose? What happened?

_________________
Here is what a Parlor Guitar is for!
http://www.youtube.com/watch?v=oEa8PkjO6_I

I'm not sure what top from what video you are talking about? I'm trying to get setup for the top that is in the lastest video where I'm doing the rosette.

Andy how about this? This I beleive is from the Left Brain Luthier.

I think that looks pretty good. The only thing I'd add is I'd cut two lengths of wood to the exact same length to use as spacers for the sample holders. You could use different spacers for different spans.

BTW, I did a quick calculation using the spreadsheet on the link and a .05% error in length measurement led to around a 1% error in calculated E. I did this using a 18" span compared to a 17.99" span (assuming a loss of .005" per side using a dowel instead of a point edge). This sounds like a big deal but probably it's not really that big a deal. I think the bigger deal would be making sure you use good measurement on the length.

_________________
http://www.birkonium.com CNC Products for Luthiers
http://banduramaker.blogspot.com

Of course getting an E value that is as accurate as possible is a good thing but what concerns me more is that my test setup generate repeatable and consistent results with as little fussing about as possible. I am more concerned about the relative numbers that I collect for myself. The key to repeatable measurements is to eliminate as many variables as possible. So, when I build my test setup, which I'll do before I do another top, I'll want a way to index the crossbars in one of several positions and also build in brackets or whatever so the dial indicator always ends up in the same place. Wood dowels are probably fine but I will probably go with 1" square aluminum tubing set so that the plate rides on an edge just to eliminate the possibility that the wood may sag over time and change the results.

_________________
Steve Smith
"Music is what feelings sound like"

I like that one. I measure from the bottom also. I feel that measuring from the top is not as accurate and will cause repeatability issues as the weight and the measuring device should be in the same place. Putting the dial indicator on the bottom solves that problem. It does make it harder to read the indicator.
I have not figured out how to put the weight on the bottom yet.

_________________
"An adventure is only an inconvenience rightly considered. An inconvenience is an adventure wrongly considered." G. K. Chesterton.

Steve Smith,
I was think that one could use 1" X 1/4" or 1/8" alu. bar stock (or if you want to get fancy brass) and then slot grooves into the side rails to position the span rails on edge don't you think? Or slots for the square tubing.

Steve,
I was also giving some thought about maybe making a seesaw type of linkage from the middle to out front so the dial indicator could be mounted upright. It might be tricky to balance it though.

Chris, the bar stock would probably be easier since you could just slot the rails on the table saw.

_________________
Steve Smith
"Music is what feelings sound like"

I agree that precision is more important than absolute accuracy in something like this. Accuracy would be great too as it would allow different luthiers to compare values of E.

To your point however, I think that simple wooden spacers (or Al) would be the way to go rather than index slots or something like that. Those sound quite sloppy to me and actually wood is more stable in length than Al (though that's not really an issue in a temp/humidity controlled shop). Just put both spacers on the chop saw at the same time. I think about 2 to 4 thou is going to be the limit of this length measurement anyway.

I'm glad this came up as my last setup (which was done in kind of a hurry) was pretty sloppy and I think this would be much nicer.

_________________
http://www.birkonium.com CNC Products for Luthiers
http://banduramaker.blogspot.com

Andy, that's a good point. Whatever method is used needs to eliminate as much slop as possible. It's likely I'm overthinking this, as usual, and a relatively simple setup, such as the photo from LBL that Chris posted, with some sticks to set the length may be more than sufficient.

I'm sure those of us who haven't tried this yet will learn lots after we start doing. I want to make sure I get good data from the start since I am such a slow builder. I'd hate to have to throw away the data for the first 10 tops because I didn't do it right.

_________________
Steve Smith
"Music is what feelings sound like"

You could always set your rails spacing and then drill and pin the rail. The pin will always locate the rail in the same position. What if the termites eat you spacer sticks? I really don't like a lot of parts to keep track of. After all you just want to get stuff in a range and then use it and tune the top.

Don't mention termites - I had to restructure 37' of exterior wall and replace a vanity (believe it or not) because of the little buggers.

_________________
Steve Smith
"Music is what feelings sound like"

Only if it's a perfect fit.

What if zombies with acid blood invade your basement hmm...yea, what would you do then?...(I forgot to mention that wood is immune to zombie acid blood). That's what I thought.

p.s. what if you lose your pin? Sticks are definitely the most fool proof and easiest to implement x infinity.

edit:

_________________
http://www.birkonium.com CNC Products for Luthiers
http://banduramaker.blogspot.com

You glue in the pins. They have been using pins to locate vise plates for CNC and tooling for years my friend. It's a pin used to locate your fret slotting template isn't it? You could use a piece of 1/4" peg board as a drill guide if you want. Lots of way to skin that cat.

True, but I did end my post with x Infinity.

_________________
http://www.birkonium.com CNC Products for Luthiers
http://banduramaker.blogspot.com