Nomath wrote: ↑January 13th, 2022, 4:44 pm
Very strange and worth investigating in depth because it conflicts with the basic principles explained in
The Physics of Rowing Ergometers and the claim of self-calibration in C2 ergs. As described by Tsnor, C2 ergs are quite robust and tolerant for poor maintenance.
Yeah, I know it is is odd and it goes against anything that a decent product (like C2) would stand for. Across rowers of the same brand and type (i.e. same physical construction of the flywheel and cage) you'd expect the behaviour to be quite similar. That is why it struck me as odd as well. Bad maintenance, especially the air intakes/outlets, can have a significant effect, but will only result in a lower dragfactor for a given specific damper setting. But the dragfactor very effectively describes the physical behaviour of the flywheel and if the person rowing keeps it constant, so should be his/her results (more or less, dependent on fatigue, etc.).
There are some things outside the dragfactor that do affect rower performance. For example, the chain friction and tension of the shock cord. But that isn't that determening for machine performance, IMHO.
I must say, thinking about it further last night (sorry, I see rowers as interesting physics puzzels). It could be that these machines have seen some abuse. Due to Corona measures, many gyms moved their trainings outside for the last two years (as you couldn't have any trainings inside with larger groups), these machines and monitor could have seen rain and similar moisture. In all fairness, that would be a more likely explanation than physical differences.
Nomath wrote: ↑January 13th, 2022, 4:44 pm
It reminds me of a remark that I read somewhere on OpenRowingMonitor, written by you if I remember well, that the drag factor changes with the power of the stroke. This would also be against the Physics of Ergometers.
During the recovery the flywheel loses speed. For an air-braked ergometer, the rotation period Tr increases linearly with time at a slope equal to 2π * C/J (C is the drag coefficient in kg.m² ; J is the moment of inertia of the flywheel in kg.m²). The slope should not depend on the speed or the power : the drag factor is nearly a constant during a rowing session.
For a pure air rower, you are correct: you'd expect the rower to coast down in a linear fashion (i.e. D = k * w^2). And you see that the drag factor changes slightly from stroke to stroke, but typically is within one or two points on a concept2.
However, I probably mentioned this in the context of hybrid or magnetic rowers. Magnetic rowers have a more constant friction (i.e. D = k * w, see Note 1 on Physics of Ergometers). When you have hybrid rower, you have both types of friction, and the balance between the two shifts when the speed changes (as the air friction increases linear with speed, and the magnetic friction remains constant). It is a hard to issue to callibrate as it requires a lot of rowingdata to be able to isolate both types of drag. But for hybrid rowers it still works quite well, when you accept the changes in drag factor across a training. However, last week a user applied OpenRowingMonitor (with currently only a pure air rower model) on a magnetic rower, which thus uses the wrong calculations making the drag factor unstable for that reason. Given that the PM5 is a dedicated and very well-calibrated air-rower monitor, I guess we can safely exclude that source of error in these cases.
Nomath wrote: ↑January 13th, 2022, 4:44 pm
We are both from The Netherlands. If you need some help, I am happy to assist.
Cool, I'll keep you to that promise
. I have some electronics on the way to start reading the data from the Concept2 into OpenRowingMonitor (optocoupler/isolation board). You have some quite intersting ideas (like stroke length and powercurves) I'd like to look more closer at when I get both running on the same flywheel. First, I like to validate OpenRowingMonitor's physics model, and then we hopefully can build these functions.
