Official Luthiers Forum!

What I think we need is an in-house set of standardized testing proceedures that we can use to communicate our experimental data or compiled testing data with one another in a meaningful way.

For example: I am thinking about the different ways that many of us seem to have to test deflection of a top or of top material, or of a brace, under a load. Frequently we have adopted a load that turns out to be a jug of water or an old plane or a hunk of lead centered in the middle of a spruce top that has some thickness that we like to use between two supports that we made in the shop that span a specified distance. While all these measurements are consistent "in shop" they do not translate well between luthiers.

What I think might be useful is a set of testing procedures that are standardized and well defined within the olf community that so that we can refer to a deflection of, say, .2 inches when describing a top and it means something to everyone because we are all using the same weight and the same top testing thickness and the same distance between centers, in other words the same testing procedures. That way, we could communicate and compile a lot of meaningful data that amplify our experiences. A lot of this data is going to be meaningful statistically but only if we have collected a lot of it.

It may be more trouble than its worth, and it may mean that a lot of us would have to change our testing procedures, but it does seem to me that if a little data is good, a lot of data is better. Of course we would have to rate our results objectively before the shared numbers make any sense and that could be a problem too.

Just thinking outloud.

John

Keep thinking John! I'm game! Thats a fantastic idea --

_________________
Support the OLF! Bookmark our STEWMAC link Today!
Lance@LuthiersForum.com

Sure, that seems reasonable. Couldn't we all just pick up some used (or cheap) scientific weights and measures. A kg or 500g weight or something. I would think that those would be the way to go.

Here is a place that sells them for $10 for a
500g Unit.

_________________
Brock Poling
Columbus, Ohio
http://www.polingguitars.com

I'm in ! Seems like a great thing to me. That even helps to tighten our sense of "community" here.
I know that Bob Steidl uses some methods he got from David Hurd, I believe. Perhaps if we adopt some specific standard already in place, we can be one step ahead.
I just was given a small laptop computer for my shop, and this would be a great way to compile the data for myself. I also plan to start doing some sonic alalysis ala John O. as well.

_________________
"I want to know what kind of pickups Vince Gill uses in his Tele, because if I had those, as good of a player as I am, I'm sure I could make it sound like that.
Only badly."

I wouldn't hurt to adopt some tried and true methods if any are out there.

Those scientific masses are not too expensive and would certainly make our methods sound technical and universal.

John,

Before you start reinventing the wheel, you might want to check into Brian Burns' quite comprehensive work on this topic.

I had the opportunity to use his setup for evaluating tonewood when I was up there at his place last year. The way he does it is quick, easy, and accurate.

Here's a link to a video that was made of his lectures on the subject, available from LMI:

http://www.lmii.com/CartTwo/thirdproducts.asp?NameProdHeader =Material+Testing%2C+with+Brian+Burns


Best,

Michael

_________________
Live to Play, Play to Live

John
I think thats a great idea and have thought the same thing for a few years now. just never really new to propose it to. Agreeing on test methods may be a bit tricky but should be doable, static Vs. dynamic
My basic setup are the gallon of water you mentioned. I like them because they are cheap, readily available, and very repeatable as far as weights go. A gallaon of water + bottle weighs the same here in New Mexico as it does California or Florida.
2 gallons of water typically deflect a .700 x .375 x 12.0(span between supports) about .04 inches. You want to get a reasonable deflection so the percent of error goes down.
I would also propose a gram scale, which is probaly the most expensive part needed and can be had for $100 or less, or more if you really want.

_________________
Jim Watts
http://jameswattsguitars.com

Ervin Somogyi recommended to me an Ohaus triple beam scale. Anyone know of a good place to purchase such a scale?

_________________
From Nacogdoches...the oldest town in Texas.

http://www.stephenkinnaird.com

Steve,

I have one. It's on indefinite loan from a chemist friend of mine. Very hard to beat in terms of accuracy, ease of use, etc.

I would suggest maybe eBay, but even there, don't count on being able to steal one -- unless the Auction's closing like at 3am on a Monday morning or something. :)

Best,

Michael

_________________
Live to Play, Play to Live

If we do this as an "organization" we would have to adopt a standard. If anyone has a standard that they recommend why not describe it. I am heading out of town for a few days but will think on this and do some research myself and make some suggestions.

I think it would be really cool to have an OLF procedure that perhaps we could post in some place on the forum.

There are a lot of scientific supply houses, Flinn, Carolina, Wards, Sargent-Welch and they all have these O-hause tripple beam balances for about 99 dollars. Price fixing to be sure.

John,

I will reiterate -- check out what Brian Burns has been doing for years. He's the authority on this topic right now.

Best,

Michael.

_________________
Live to Play, Play to Live

I'd be happy to adopt a standard but as a relative beginner I haven't been doing deflection tests. I'll just follow the thread and wait for a consesus on equipment to develop. The triple beam scales are great but you need to take care that you don't buy a used one that has been abused.

_________________
Don Atwood
Arlington, VA

I'll give the gist of what Brian Burns does. I have detailed notes around here somewhere, and when I find them I can supply you with a comprehensive explanation.

First Brian determines the density of the wood being evaluated. To do this, he must accurately as possible measure its length, width, and thickness, multiply these three values together to get the wood's volume, then weigh it on an accurate scale. He uses an Ohaus scale. Keeping the units consistent (ie, don't mix metric with imperial systems), he then divides the weight by the volume. This gives the wood's density.

Next, he has a fixture that is a rack of sorts with several parallel bars running across it. The bars are spaced so that he can measure cross-grain stiffness one way and longitudinal stiffness the other. At the bottom of this fixture, he has mounted a dial indicator, which can be zeroed out against the bottom surface of the wood.

To make the deflection measurements, he has a set of weights of known measurement that he sets atop the wood. He uses one weight for cross grain and another, heavier one for longitudinal stiffness. He zeroes the dial indicator against the bottom of the wood, then places the weight atop it, then measures the amount of deflection, doing this both cross grain and longitudinally.

After that, he inputs these numbers, including the density, into a formula and comes up with a Factor that is a ratio of density to flexibility. I'll have to dig out my notes to provide you with the exact formula, but this is the gist of it.

Brian also has an electro-acoustic setup that he's cobbled together from various surplus items which uses these values in addition to finding out the responsiveness of the tonewood at various fundamentals and overtones. All this comes together with an overall evaluation number he uses to determine the tonewood's overall quality.

Brian really makes no claims that his method is THE way to do it. Instead he cites subjective impressions from people who have evaluated the guitars built from these woods, and by all accounts, the results are quite positive.

I'm a believer. I used his setup to evaluate a couple of guitars worth of wood while I was there. The first one I built came out sounding very nice. The second one is still in the works.

I plan to duplicate his setup as soon as I have the room for it. Not that it takes up that much space, it's just that I have a very small shop.

Best,

Michael

_________________
Live to Play, Play to Live

And don't forget that some of us don't live in the US, but we'd also like to participate. We haven't used gallons as a unit of measure here for over thirty years. I'd never find a gallon sized bottle.

I'd suggest that whatever methods are developed, they use standard SI units to measure everything as does most of the scientific community, and there's a reason we do. Then anybody on the planet can perform the same test, do the same math, and get the same result.

I don't think that it's necessary for everyone to buy the same brand of weight, scale, balance, whatever. Just so long as they are accurate to within acceptable limits, and use the same unit of measure. There's always going to be various degrees of error and therefore variations in results from experimenter to experimenter, but that's ok as long as the sample size is large. As John said in his original post "A lot of this data is going to be meaningful statistically but only if we have collected a lot of it. "

Dangit, my microwave oven interferes with my wireless network, hence I ended up duplicating the above post. Please disregard this one.
Michael McBroom38555.8436574074

_________________
Live to Play, Play to Live

What Paul B said: get the equations, set them up on your compter in a spreadsheet or something, and then report things like the denstiy, Young's modulus and Q value, specifying the units you're using: centimeter-gram-second (c-g-s) or meter-kilogram-second (m-k-s) would probably be best, but anything can be converted simply (please avoid microfurlong-carat-femtosecond: the units get too hairy). Anything else really would be reinventing the wheel.

Deflection testing is just one form of testing, Chaldni patterns are another, and I am sure that there are more...I think that a more consistent thing is to adopt a vocabulary that everyone agrees on...Young's modulus is well understood, Brian's Q, those kinds of things...it would be useful to have an article posted to the site or at least a FAQ (frequently asked questions) so that when the topic comes up we can refer to those terms...I have not checked if there is a similar FAQ or list of terms on any other forum such as 13th fret or Acoustic Guitar, etc...if there is then we should use that as a starting point, if one doesnt exist then it would be good to collect it from the various posts we have had on the subject...

Between Alan Carruth, Brian Burns, and others there is a rich knowledgebase to build a common set of principles from...

For deflection board testing of tops I use a simple 5lb dumbell weight purchased at Dunhams sporting goods for < $10.00. Seems as though a gallon jug (approximately 7lbs) could vary in weight depending on how full you fill it. Steel remains pretty constant.

Digital scales are fairly cheap on eBay and will measure in ounces and grams. Accuracy to 2 decimal places which is close enough for my work.

_________________
tim...
http://www.mcknightguitars.com

Actually, I think there is quite a bit of variation in weight equipment.

I think that if you can pick up a kg or 2 or 3 for under $10 - $20 that seems like the way to go. I think that if we could standardize on 1) a testing protocol 2) A data protocol and 3) accurate tracking and reporting systems that this could be a cool resouce that lots of people could tap.

_________________
Brock Poling
Columbus, Ohio
http://www.polingguitars.com

[QUOTE=Brock Poling]
Actually, I think there is quite a bit of variation in weight equipment.

I think that if you can pick up a kg or 2 or 3 for under $10 - $20 that seems like the way to go. I think that if we could standardize on 1) a testing protocol 2) A data protocol and 3) accurate tracking and reporting systems that this could be a cool resouce that lots of people could tap.


[/QUOTE]

I bought one off eBay in the price range. It will weigh up to 2000g but has a graduation of 1g and 2-5g tolerance which is fine to weigh hide glue. On the other hand the budget triple beam scales are accurate to 0.1g and run about $67. It depends on how accurate we want to be (do we need better than 1g accuracy). If there is enough interest, I'll check with the triple beam dealer and see if we can get a better price for a bulk buy.    

_________________
Don Atwood
Arlington, VA

Go, Don! I'm interested.

_________________
From Nacogdoches...the oldest town in Texas.

http://www.stephenkinnaird.com

Does anyone else have David Hurd's book on the subject? He goes into quite a bit of detail in "Left Brain Lutherie."

Left Brain Lutherie

I have an ohaus triple beam balance I been using for a few years also. The best price I found was from Rio Grande jewlery supply for $89. But I'm sure most any triple beam would work.

I find it hard to imagine finding a heavy weight more repeatable than a gallon of water (if you purchase the water from the store) for a reasonable price. Believe it or not the USDA inspectors make that if your selling a gallon you actually get a gallon. A gallon weighs 3780 grams, if I'm off 10cc that's only 10 grams out 3780. I include 50 grams for the container in my spreadsheet.

One last thought, I'm an engineer and have a hard time understanding what dynamic testing really brings to the table here. I'm sure someone could convince me but as of yet I just don't get it.

_________________
Jim Watts
http://jameswattsguitars.com

[QUOTE=Jim_W] A gallon weighs 3780 grams, if I'm off 10cc that's only 10 grams out 3780. I include 50 grams for the container in my spreadsheet.[/QUOTE]

But it weighs 3780g at what temperature? Water, like everything else contracts or expands with changes in temp. At what temp are you going to measure your gallon, so that it weighs 3780g?


[QUOTE=Jim_W]One last thought, I'm an engineer and have a hard time understanding what dynamic testing really brings to the table here. I'm sure someone could convince me but as of yet I just don't get it.
[/QUOTE]

What I think we could get out of it is a formula that will tell us the optimum thickness for a top based on the results obtained by testing the materials we're working with. What we need to be finding out is how lightly we can build so that the the guitar still withstands the stresses on it, but is as light and responsive as we can get away with without the whole thing colapsing under the tension of the strings.

We can all experiment independantly and in a few decades of building we each might have a better idea of how to build guitars for optimal results, or we could colaborate and get better data sooner. We'd all win.

paul, even if you could find a 30+ year old gallon container it would do you no good as the old imperial gallon is(was) bigger than the us version.

agreed, if this is to be done it should be in standard scientific units, but finding inexpensive metric dial indicators may not be easy in the us, and for some of the non-technical us participants the metric measures may be a mystery.

perhaps our huge industry could utilize its great economic and politcal influence get congress to finally adopt the metric system as mandatory instead of the ridiculaous optional situation which has applied for over a hundred years.

oh, well uh, that was a nice little dream now, wasn't it!!

it can work though; when gough declared the imperial system kaput in '73 aus managed the conversion without too great a dislocation. and even the good old uk has been in the process for i don't know how long.

crazymanmichael38556.4878587963