Details on #2 methodology, and the real world
Lyn J. Mangiameli
>" But *if* Lyn's testing and results were based upon several planes and blades of each brand/type, and *if* these tests were run multiple times on each and every subject in the samples, results plotted and averaged, and *if* the methodology was distributed, then seperate samples were obtained by independent, multiple other testors and using the same methodologies tested, plotted and averaged those results and the whole shebang then published, it would approach scientific."
There have been two formal studies. The first, though formal in how it was conducted, and IMO a vast improvement over what preceeded it, was very lacking in it's methodology. Failures that I went to some trouble to point out in the report of it.
The second study, which due to my only failings still has yet to be formally reported, sought to improve on the original in several respects.
Here are the planes that went into the second study:
Cast Iron BD:
#2 Lee Valley #4.5
#1 Lie-Nielsen #4.5 (both frogs)
Cast Iron BU:
#2 Lee Valley LA Smoother
#1 Lie-Nielsen #164
Infill Planes:
#2 S&S (45 & 50)
#6 Shepherd (#2 @45, #1 @ 47.5, 50, 55, 60)
#1 Ray Iles
#1 SMT (later confirmed with second SMT)
#1 Knight Japanese bladed infill
WoodBodied Planes:
#5 Knight Finish Smoother (45, 50, 55, 60, 65)
#2 Knight Coffin Smoother (50, 63)
#1 ECE/Primus (50)
#1 Mujingfang (63)
#1 HNT Gordon (60)
Not formally in the study but later compared to the others, have been the LV LA Jack (#1) and new LA BU SuperSmoother (#2).
There were multiple samples of blades, sometimes as many as a half dozen for each plane, in the case of some planes, I was able to compare Japanese to HSS to Carbon style blades.
Woods:
At least two sample of each, usually three.
BlackWood Acacia
Bloodwood
Bocote
Bubinga
Cocobolo
Casuarin
Dogwood
Ebony
Goncolo Alves
Ipe
Kauri
Lacewood
Lignum Vitae
Manguis
Fiddleback Maple
Curly Oak
Padauk
Figured Purpleheart
Brazilian Rosewood
Teak
Zebrawood
Testing:
Two testers, at least one trial for each wood sample for each plane configuration.
Three conditions:
Planer #1, Consensus evaluation by both testers
Planer #1, Blind evaluation by Tester/Planer #2
Planer #2, Blind evaluation by Tester/Planer #1
Quality of surface finish was the criteria for ratings.
The tactually smoother finish was found superior.
The finish with the lease visible chipping, tearout, stringing, etc was found superior.
The finish with the brightest, most reflective surface was found superior.
The surface with the most evenly bright finish was found superior.
The surface that offered the greater clarity of detal was found superior.
All the data was entered into the SPSS statistical database for analysis. Primary analysis found two main effects and one interaction. Angle within plane, Angle within wood are main effects. Angle within plane with respect to the type of wood is interaction. Note absence of effect of plane type.
So, not the level of scientific rigor that might be desirable, but I would assert the highest level related to plane type and blade angle that has been made available for discussion to date.
Perhaps you can see why the study and write up has taken a while.
I would end with this. Studies such as my more recent study outlined above, and my earlier one, form an objective, comparative foundation for my statements, but it doesn't end there. Though some would like to imply otherwise, I think the regular reader of this forum has the information to recognize my assertions are also developed from actual experience using hand tools to make things. What I share with you all is developed from BOTH the practical experience of using hand tools (see for example my comments on the ergonomics of plane handles) and the addition of objective, comparative, data.
I can't control what you make of it, but I can be content in knowing that I've offered you some of the best and least biased information available to date.