Tim, I cannot disagree with your own results and/nor your reasoning. However, perhaps i did not make clear that in my own experience over the years, the "theoretical" (your term) results recorded from the rolltester seem to broadly track observable real-time rollability under the car on the layout. This is not always, of course, but it is enough for me to use the recorded data as a reliable guide without the routine of additionally testing the results under a car.
toggle quoted messageShow quoted text
The one oz. test load is not ideal, but most of the time it seems to be enough; and of course, it is simply additive to whatever the weight of the truck cum wheels might be- relatively little fora Tahoe or Accurail truck, relatively more for Kadee with their metal-bearing plastic frames. I don't routinely weight my cars to exact NMRA standards, but when challenged, the cars actually weigh out pretty close- mostl very slightly on the light side. No problems.
As for the Lindberg trucks, placing weight on them did, and does make them roll by themselves even on a flat surface. However, they also will -by themselves- roll uphill after replacement by Reboxx wheels- with or without weight.
--- In STMFC@..., Tim O'Connor <timboconnor@...> wrote:
Denny, a 1 oz load is about 1/2 the typical load of a freight
car. There are also rolling dynamics involved. And under load,
the shape of the bearing and journal, and the materials, makes
a big difference. Kato ASF A-3 trucks for example roll incredibly
well without a load, but as the load increases, the performance
is affected -- it's almost a straight-line correlation.
I roll test every car that I put together, or make ready for
operation. It takes a little time, maybe 20 minutes, to identify
the proper truck design for that model and then find a model truck
and wheelset combination that gives good results. It's makes no
difference to me how well a truck rolls "in theory".
For example, remember those old Lindbergh trucks? Those things
rolled like crazy, we thought they were miraculous in the 1960's.
Oh wait, until you put a car on them. Then your results may vary.
Tim O'C inquires about the effect of weight on truck rolling test data.
I do not routinely test truck rollability with/without weights, although I do so on occasion. My reason for not doing so is that I use the test data to conveniently guide my choices of wheel sets going forward (not excluding leaving OEM alone!) without having to parse out with testing anew with each install the usual small inherent differences resulting from both variable weighting, but also rollability differences caused by the common variability between nominally-identical trucks. Sometimes the weight improves rollability, occasionally in dramatic fashion; while at other times, the same weight can all but stop a truck dead in its tracks. Most of the time, in real time, it makes no difference.
As to the new Kadee trucks, I did indeed test the Barber S3 with and without weight. The weight (a 1-oz. lead block balanced on the truck bolster) caused the rollability to improve, but ever so slightly- remaining still in the midst of the Acceptable/Good field. This was why it was not mentioned.
Denny S. Anspach MD