Erg Fan Moment of Inertia

[Bob]

I know this is a longshot but does anyone know the moment of inertia
of the original RowPerfect flywheel/fan? Surely, it can't be proprietary information after all this time.

I was hoping to compare the numbers from my old DOS version of the RowPerfect Care program with calculated values derived from papers on
the subject. It is interesting because erg fan speed is so nonlinear i.e. a lot of effort goes into small increases in fan speed when the fan is spinning quickly. This makes me wonder how accurate ergs really are since relatively small
changes in measurements convert to large changes.in readings.

Best,

Bob

[carl]

All ergs are partial simulators, with many assumptions built in.

The original RP flywheel is a disc of steel, 430mm diameter & 3mm thick,
with 8 radial blades, 50mm wide & with 13mm right-angled flanges, each
blade 183mm long & 2mm thick, attached to one face by their 13mm
flanges. That should be enable you to derive the wheel's MI?

The flywheel accumulates and stores kinetic energy with a V^2
relationship, as do the moving masses of your body & boat.

As the fan element of the flywheel is anything but an ideal piece of
turbomachinery, it is not as peaky in its efficiency vs angular velocity
relationship as more refined fan systems & is thus more likely to
reasonably represent the square-law relationship of drag vs angular
velocity over the operating range.

To properly measure the power input requires a dynamometer - constantly
measuring force in the chain or belt & its speed of movement.
Calculations based on flywheel inertia & rate of run-down make
assumptions of a relationship between velocity & drag which are not
necessarily accurate, given the inevitable peakiness of the
speed/power-absorbed relationship of the various fan types. This has
been demonstrated by applying dynamometers to ergs & comparing the
results. And, please note, due to the time lags in surrounding air
flows resulting from the constantly varying flywheel speed & air
throughput, erg computer results can be further at variance from actual
work being done & are not necessarily representative of rower power
performance.

Next there's the increasing fixation on erg results for comparison and,
worse still, for crew selection. Rowing technique is so very different
from static-erg technique, since those ergs' dynamics involve rapid
changes in body mass velocity which are entirely absent in rowing. Thus
you have to decelerate the body as you approach the catch on a fixed erg
& much less so on a RowPerfect, while in a boat (despite what coaches
like to tell you to do) the notion of decelerating your movement down
the slide is actually meaningless since you are not moving down the
slide but are actually pulling the foot-stretcher, with the boat,
towards you. (Distant cries of "anathema" & disbelief?)

So perhaps we should not be too fixated on what the erg computer tells
us, but see the erg as another kind of weight-lifting machine with a
rather closer relationship in the necessary movements to a real boat,
while the boat itself is a delicate, responsive instrument demanding a
range of techniques, skill & finesse which are entirely absent from the
erg room. And doesn't some of the rowing we see highlight that crucial
difference!

Cheers -
Carl

--
Carl Douglas Racing Shells -
Fine Small-Boats/AeRoWing Low-drag Riggers/Advanced Accessories
Write: Harris Boatyard, Laleham Reach, Chertsey KT16 8RP, UK
Find: tinyurl.com/2tqujf
Email: ca...@carldouglasrowing.com Tel: +44(0)1932-570946 Fax: -563682
URLs: carldouglasrowing.com & now on Facebook @ CarlDouglasRacingShells

---
This email has been checked for viruses by AVG.
https://www.avg.com

[Bob]

Carl,

Thank you for that explanation. I bet you did what I should have done which was measure the
flywheel on my old RowPerfect. I did think of that but, as an Electrical Engineer, I thought
the complex shape would be to hard to calculate. As soon as I saw your measurements, I realized
the calculation could be broken down into a disk and an eight bladed propeller whose mass vs
radius is constant and just look up the answer, So much for my being a practical engineer!

More to the point is your statement that a flywheel moving air is more complex than the relatively simple
relationships I'd use to calculate the power curve from flywheel speed. I don't see how ergs can be
accurate enough for comparison purposes given all the variables like stoke rate, temperature, humidity,
air pressure etc. that influence flywheel speed. I know they measure the rate of slowdown of the flywheel
order to compensate for the conditions but how accurate is that? Of course some coaches only want
a number, some rowers need that number for their place in the hierarchy of the boathouse and, in the US,
to get that coveted University scholarship.

Comparisons are a fact of life in rowing. Erg scores are the most used form of comparison and
much to the delight of Concept 2, their erg is the gold standard. From a practical standpoint,
if I had to come up with an erg score could I do one piece on the CII, repeat it on my RowPerfect and
apply a correction to the RowPerfect result so the numbers match? If so, I could track my improvements
on the much preferable RP while scaling them to match the CII.

I have my little project along to the point of picking up the magnets on the RowPerfect flywheel with a Hall
Effect sensor and sending the time deltas between flywheel magnets to my computer. I plotted the data and can
see the fan spin up and slow down. I got started on the nonlinear aspect when I observed how long
it takes the fan to stop and how much harder I have to pull to reduce the deltas between magnet pulses
once the flywheel is up to speed. It seemed to me that getting accurate splits from such small changes
in flywheel speed was problematical. That got me started on putting my faith in any erg result.

Bob

[A. Dumas]

Bob <head...@gmail.com> wrote:
> More to the point is your statement that a flywheel moving air is more
> complex than the relatively simple
> relationships I'd use to calculate the power curve from flywheel speed.
> I don't see how ergs can be
> accurate enough for comparison purposes given all the variables like
> stoke rate, temperature, humidity,
> air pressure etc. that influence flywheel speed. I know they measure the
> rate of slowdown of the flywheel
> order to compensate for the conditions but how accurate is that?

They re-calibrate (calculate the drag factor) every stroke. See
http://eodg.atm.ox.ac.uk/user/dudhia/rowing/physics/ergometer.html#section7

[carl]

Correct. But still inaccurate as the flywheel's energy-dissipation vs
time on slow-down & vs angular velocity are not log-linear, & fan energy
consumption is affected by flow transients, so that power consumption
during acceleration differs from that during run-down. And when the fan
moves with the feet of the rower, e.g. RP, then the intake vortex formed
in the vicinity of the fan is being dragged to and fro - which is also
suboptimal.

And the intake geometry of other machines is scarcely ideal. Fans are
turbomachinery (of a kind) & intake design is a key part of the overall
system - just as hull design really matters for racing boats

I guess (blindly) that it would not be too hard to program a form of
magnetic induction resistance to replace the fan and flywheel with a
resistance that better matches real-life boat dynamics. But we can use
a sculling boat to settle those arguments, & it's a lot more fun ...

[Bob]

A.

The link you provided led to the article I planned to use to calculate my own curves hence the need for the
flywheel inertia.

Carl,

I believe the Hydrow uses magnetic resistance. The thing weighs and costs a lot but at least you can turn the TV down when you erg!

https://www.fitrated.com/hydrow-vs-concept2-model-d/

The non-steady state fan effects you mention have me concerned. It might be easier to put a strain gage on the handle chain and measure
force directly. The distance moved by the chain (old RowPerfect) might be hard to measure. However, I LEARNED MY LESSON AND LOOKED
at my RowPerfect and saw that the nut on the flywheel spins with it. This would lend itself to an optical encoder. The force and distance
over time sounds like work to me.

Thanks,

Bob

[A. Dumas]

Bob <head...@gmail.com> wrote:
> The non-steady state fan effects you mention have me concerned. It might
> be easier to put a strain gage on the handle chain and measure
> force directly.

The drag calculation by the erg is imperfect, yet the power at the handle
is mostly just off by a constant amount:
https://www.researchgate.net/publication/7166945_Power_Responses_of_a_Rowing_Ergometer_Mechanical_Sensors_vs_Concept2R_Measurement_System

[Andy McKenzie]

I use a WaterRower for my winter sessions of self torture when light and weather preclude sculling. The water rower's split calculation is well known to be more generous than that of a Concept 2, by about 10% at 'normal' paces. You can now equip a WaterRower with a 'Smartrow', which is a strain gauge inserted between handle and flywheel (actually a replacement of one of the pulleys). These measure force and stroke length, and have been programmed to match Concept 2 results more directly.

However I think there is a bit of over pessimism about the 'accuracy' of the Concept calculations. The fact that rowers in gyms don't tend to care which erg they sit at and don't get all superstitious about only producing their 'best' times on a particular machine seems to be empirical evidence of consistency.

Andy

[Bob]

A. Dumas

That is a very interesting link! They discuss how the errors increase as the effort and SR change on the CII erg which supports Carl's statement that transients affect flywheel energy dissipation. There is one thing about their graphs that I don't understand. I think they scatterplot the instrumented vs CII Power values and create
an average line. That is fine but the plotted values are all over the map so to speak with significant outliers. This seems surprising to me. Can anyone enlighten/correct me? I'm also curious about the authors surmise that there is an intentional error in the power reading - "It can be supposed that the manufacturer has modified the C2D with the main objective to reproduce the speed of the boat rather than to assess the actual power developed by the rower".

Your comment about gym ergs and consistency makes sense but different damper settings can result in differing variations in flywheel speeds which induces errors
according to the paper and Carl. That cumulative error could depend on stroke rate so I wonder how it all works out or if there is some innate faith in the CII Model D
as a tribute to CII marketing? In some ways, it doesn't really matter what the readout says as long as the CII is consistent from unit to unit and under all conditions and settings which, I think, is a tall order. I know I can't help taking a childish pleasure in beating someone's split by even a second! As an engineer, I'm not astute enough to know what that says about me and human nature but it probably isn't good.

Andy,

Can the Smartrow be programmed to yield power directly and, if so, how does it compare to the original WaterRower display on a stroke by stroke basis? The paper mentions that CII underestimates power so is the fact that the Smartrow needs to be scaled down to match CII, the reality of a gold standard that is not pure gold?

Thanks,

Bob

[Kit Davies]

FWIW, the C2 MoI is 0.1001 kg m^2 according to an old post on a C2 forum
I found. I don't know how close RP made their wheel to the C2 one. AIUI
measured watts are roughly comparable between the 2 machines.

Kit

[Andy McKenzie]

I haven't got a Smartrow myself. The original WaterRower display gives distance/speed, rather than any attempt to display power, so direct comparisons would be hard, but users have posted a fairly consistent 10% difference between pieces, even at differing intensity. From my perspective it doesn't really matter as I'm racing myself and no one else.

Andy

[carl]

And, of course, no ergs precisely simulate a boat, & (though RP comes
pretty close on the dynamics) none of them teaches you about blade-work,
or those other trying aspects of technique. So there's an element of
sales hype in promotions that tell you otherwise, or kidology in telling
ourselves that erg performance can represent on-water performance. The
erg is a land-training device which can help you to develop strength &
fitness, as can weights & body-weight circuits, running etc. But it
will not make you into a rower.

[Bob]

I want to thank everyone who contributed to informing me on the nuances of my little project to measure erg power.
I've abandoned the flywheel measurement approach and ordered an optical encoder from ebay and a pull pressure load cell from Amazon?!.
Despite the fancy names they were surprisingly inexpensive. I still plan to pursue making a precision measurement of power in
comparison with the numbers from my original RowPerfect.

The real takeaway of this discussion is just as Carl said. Ergs can make you stronger but if you want to become better, you need to get on the water.

Many Thanks,

Bob