Official Luthiers Forum!

I've got a question.... In all my research about Janka hardness of wood, none of the supposedly scientific findings mentioned how many samples were tested before arriving at a reading. The usual scientific method is to test a specific number of samples, and then average the results to produce a number. I went through at least 12 papers on Janka hardness testing of wood species and found no mention of this. I consider it a salient and cogent point that the authors left out, leaving me to conclude that either (1) they only tested one sample, or (2) they incorrectly assumed that the reader would know what process would be followed.

Any thoughts?

_________________
"Act your age, not your shoe size" - Prince

Any way to contact any publishers of the tests?

Were you looking at actual scientific publications? I did a Google Scholar search for papers that measure Janka hardness and the first three papers I looked at all gave the number of samples of each wood species tested. In those papers, they used 10, 25, and 30 samples in each test.

_________________
Once in a while you get shown the light in the strangest of places if you look at it right - Robert Hunter

You might also ask Eric Meier, the creator of the online resource - "The Wood Database".
https://www.wood-database.com/about/

We all know that hardness can vary quite a bit from board to board, even from one edge of the board to the other. Un fortunately, most sources seem to give an average value, rather than a range.
When I want "numbers" I consult the wood database. It is handy, and I think it gives a rough idea of what to expect.

Are you looking at scientific articles. A proper paper would certainly include the data on number of samples tested as well as data on the variance between samples and how the “average” was calculated.

_________________
Bryan Bear PMoMC

Take care of your feet, and your feet will take care of you.

I'm not sure how useful it is to know the 'average' value of some such measurement, given the range of variation in most properties for any species of wood. Also, knowing hardness strikes me as less useful than some other things, such as Young's modulus and density, except, of course, for a fingerboard wood.

I am asking for a guy over on the Duncan forum.

He'd been trying to do research on the subject and was frustrated with the disturbing tendency to cut and paste among many of the websites and papers he had covered.

And I am sick and tired of being asked to subscribe, pay to download, etc. to get the answer to what I see as a simple question. Surely someone MUST know the basic process without having to give away a state secret.

_________________
"Act your age, not your shoe size" - Prince

Maybe the problem is that he may not be going to the original source of the Janka hardness numbers. If the numbers are in secondary sources, the kind of information you are asking about may very well have been dropped when the numbers were extracted from the primary sources. The details of how averages are calculated and how many data points were used in the calculation are fundamental requirements for scientific papers to get published at all, so they have to be there in actual scientific papers.

A common problem with trying to go directly to the original scientific papers is that many of companies that publish scientific journals require a subscription for access to anything more than the abstract of a paper. However, if you do a Google Scholar search for something like "Janka hardness wood", many of the results will include a link in the right side margin of the results page that takes you to a pdf file of the paper which gives you access to the entire paper.

Beyond that, there wouldn't be any standard specification for the number of samples used to calculate an average Janka hardness number in scientific studies because that would depend on the specific questions being addressed in a given study. If there is a regulatory use for Janka hardness numbers, there could well be standardized specifications about the number of samples required for that purpose.

_________________
Once in a while you get shown the light in the strangest of places if you look at it right - Robert Hunter

I would think that an average would be valuable information at least in a general sense. Even though there may be a lot of variability in say Malagasy Ebony and a lot of variability in something like spruce you would see very different numbers comparing their average. Again for general purposes.

Usually an actual scientific paper has the method clearly outlined.

The average and standard deviation would be very helpful. It would be nice to have an idea of a certain wood is fairly consistent (small standard deviation) or varies wildly. You could also see at a glance that most of your samples would be expected to be at least a certain hardness of -1 Z.

_________________
Bryan Bear PMoMC

Take care of your feet, and your feet will take care of you.