Accuracy & repeatability

[Viktor Chebrikov]

Been shaken by people telling me Concept2 is not as accurate as claimed.

Someone quoted a study, I can't find any others.


https://ojs.ub.uni-konstanz.de/cpa/article/view/2893

I need reassurance please as I really thought we had accuracy and repeatability across machines etc.

Thanks.

[sharp_rower]

The study is from 1992. My hope and guess is that Concept2 ironed out these kinks since then...Also, the discrepancy was observed only for one athlete, according to the abstract.

I must admit that rowing on my CRASH-B erg in 2014 felt different than the one I have at home. The chain at CRASH-Bs came out really really smoothly compared to my machine; pulling it out felt way easier than on my home machine (my home machine is a gray model D 1993-US with PM4 made in 2012), which I lubricate every 50 hours of use...But in the end my score was consistent with my training and past 2k scores. I PR'd and it felt easier than I thought it would, but I chalk that up to having followed a formal and intense training program, and also to the fact that I was really amped up because it was CRASH-Bs.

[Edward4492]

I row on a fairly new model D at home (less than a year old with about 2.5 million meters on it), I also row at a public gym that has fairly new Model D's. With in the past year I rowed at a gym that had an old model C and at another gym that had two model C's that were in horrid condition. They sounded like a coffee grinder( I quit that gym when the owner declined my offer to service the machines and would not get them fixed). Even though the machines felt horrible, the work output and monitor readings for my distance work felt about right (I know, pretty subjective). At CRASH B's the machine felt perfect, probably because they are brand new (as was my home machine last year). I set a PR by 2 seconds which seemed about right given my preparation, nutrition, etc. was dead on. I also attended four other events and my performance always seemed consistent with what I expected. And some of those high school machines were pretty beat up.
In addition, I don't think I've ever read on any of these forums any comments or concerns regarding accuracy and repeatability. I would not concern myself with a 22 year old study citing possible inconsistencies with one athlete on an old PM2.

[gregsmith01748]

I have rowed on a very large number of machines. I have raced at least 5 times, each time with a different erg and in different locations. I have a model D at home and use a model c at work. When I travel, I use model D and model E machines in gyms and crossfit boxes.

What I have found is that all the machines are frighteningly alike. With the bounds of normal maintenance, I get the results I expect. If the machine makes grinding noises, or the chain slips or the seat is loose, or the fan housing is choked with dust, then all bets are off. But if the machines are well maintained, then they match very closely.

There is a reason for this. The repeatability of the machine really comes down to having the PM measure and compensate for most of the sources of errors.

There is a good write up of this here:
http://www.atm.ox.ac.uk/rowing/physics/ergometer.htm

Here's an excerpt:
By measuring the damping, the ergometer will automatically compensate for any of the following:
- Opening/Closing the vents to increase/reduce resistance
- Changes in friction on the flywheel bearings with time
- Changes in air pressure, density, viscosity etc.
- Environmental factors such as proximity to walls or other ergs
Things that are not compensated are:
- Changes in the chain friction
- Changes in the tension of the return mechanism
- Manufacturing variations flywheel moments of inertia (probably negligible)
- Changes in flywheel moments of inertia (unlikely with the solid flywheels)

[lindsayh]

I have rowed on a very large number of machines. I have raced at least 5 times, each time with a different erg and in different locations. I have a model D at home and use a model c at work. When I travel, I use model D and model E machines in gyms and crossfit boxes.
What I have found is that all the machines are frighteningly alike. With the bounds of normal maintenance, I get the results I expect. If the machine makes grinding noises, or the chain slips or the seat is loose, or the fan housing is choked with dust, then all bets are off. But if the machines are well maintained, then they match very closely.

Me too - it is a bit amazing IMO. I have raced at least 20 times here in Aus as well as in NZealand and the UK (EIRC) and trained in gyms all over the place (as well as E at home). Numbers are repeatable every time.

[PaulG]

It's hard to evaluate this study from the abstract only, but the correlation statistic they cite (r=0.98) between total mechanical work and distance covered is about as high a correlation you will ever get in biological data. r=1.0 is perfect correlation so 0.98 is pretty darn good. Furthermore they appeared to test individuals on different occasions and we all know how things can vary form day to day in human performance. It would be interesting to see the entire article.