Free Internet Rowing Model (FIRM)

[NoEyeDeer]

Haven't done any more modelling, but have a question for Leo.

At the start of the manual it lists good and daft ways to try and use this app. What I'm interested in is how the differences in stroke length and boat speed will influence optimal catch and release angles, as well as torso angles, for a fixed seat boat compared to a scull.

My question is, would it be reasonable to attempt to use FIRM to get quantitative comparisons for these details?

 

[Leo Lazauskas]

Haven't done any more modelling, but have a question for Leo.

At the start of the manual it lists good and daft ways to try and use this app. What I'm interested in is how the differences in stroke length and boat speed will influence optimal catch and release angles, as well as torso angles, for a fixed seat boat compared to a scull.

My question is, would it be reasonable to attempt to use FIRM to get quantitative comparisons for these details?

If you exert the same force over a longer stroke then, yes, you
will cover more distance for each stroke, but you will do more work.
A better comparison would be to look at the mean hull speed per stroke
for the same "impulse". If you increase stroke length in the model,
decrease the maximum force so that the area under the curve is the
same.
Kleshnev has grappled with the same issue in his August 2014
newsletter:
http://www.biorow.com/RBN_en_2014_files/2014RowBiomNews07.pdf

Changing the torso angle will change the boat acceleration, but it
will have little or no effect on mean hull speed.

Version 2.30 will be released in a couple of days.
Old anthro.csv files will need to be changed to include radii of
gyration of limb segments. Standard values for these will be included
in the example files.

 

[NoEyeDeer]

With the torso angles and catch/release angles I was mainly wondering if FIRM could give insights on optimums, or if that's not an appropriate use of the app.

What I'm thinking is that past certain torso angles you're likely to become biomechanically inefficient, so there would be a useful limit on torso angles. This is probably more the case towards the catch where (if rowing fixed seat) you're basically trying to pick up a load from touching your toes, if you see what I mean.

The other thing is that any lift force at the catch, which seems to be trendy among scullers, is likely to largely dependent on the square of the boat speed. That means slower fixed seat boats may be more efficient with shorter catch angles. I don't know if this is actually the case, but it seems plausible and is something I would be interested doing comparative modelling for, if this is an appropriate usage of FIRM.

Shortening the stroke is probably going to mean less variation in hull speed, but OTOH more energy chewed up throwing your torso, arms and oars around, so there will be trade-offs there too.

Old anthro.csv files will need to be changed to include radii of
gyration of limb segments. Standard values for these will be included
in the example files.

I saw your grumbles about the source of data being retired Air Force personnel. TBH this is pretty much ideal from my perspective, since data for aging white blokes is what I need for myself.

 

[Leo Lazauskas]

With the torso angles and catch/release angles I was mainly wondering if FIRM could give insights on optimums, or if that's not an appropriate use of the app.

What I'm thinking is that past certain torso angles you're likely to become biomechanically inefficient, so there would be a useful limit on torso angles. This is probably more the case towards the catch where (if rowing fixed seat) you're basically trying to pick up a load from touching your toes, if you see what I mean.

FIRM might give you some insights into how the mean hull speed and
instantaneous propulsive acceleration varies with catch angle and torso
angles. I suspect you would be better off looking at how to improve the
mid-drive, where most power is generated.

The other thing is that any lift force at the catch, which seems to be trendy among scullers, is likely to largely dependent on the square of the boat speed. That means slower fixed seat boats may be more efficient with shorter catch angles. I don't know if this is actually the case, but it seems plausible and is something I would be interested doing comparative modelling for, if this is an appropriate usage of FIRM.

I think there is too much emphasis on the lift force generated by the blades.
Firstly, it occurs for a very short time near the catch, when blades are close
to the surface. Secondly, there is also an attendant drag that goes with that
lift. Many people think that lift comes with no penalty.

Shortening the stroke is probably going to mean less variation in hull speed, but OTOH more energy chewed up throwing your torso, arms and oars

That's right. It is a very complex system. I will try to explicitly calculate "internal" power losses at some stage.
In one reference I saw a few years ago there were 17 definitions of rowing "power"!

I saw your grumbles about the source of data being retired Air Force personnel. TBH this is pretty much ideal from my perspective, since data for aging white blokes is what I need for myself.

They are probably Ok for oldish, smaller men. Fighter pilots tend to be short
wiry types. I doubt that there are many 2m tall, 100 kg fighter pilots.

I read that the differences in segment masses and CM location can be as
much as 15%-20%, but fortunately it doesn't make quite as much difference
to hull speed and acceleration predictions.

There are many new techniques for estimating limb masses, CM, and other
quantities. Pinning a severed limb and swinging it like a pendulum to estimate
its radius of gyration is not for the faint-hearted I imagine.

 

[NoEyeDeer]

If you exert the same force over a longer stroke then, yes, you
will cover more distance for each stroke, but you will do more work.
A better comparison would be to look at the mean hull speed per stroke
for the same "impulse". If you increase stroke length in the model,
decrease the maximum force so that the area under the curve is the
same.
Kleshnev has grappled with the same issue in his August 2014
newsletter:
http://www.biorow.com/RBN_en_2014_files/2014RowBiomNews07.pdf

Just read that. Makes sense, as far as it goes, but doesn't mention the fact that if you shorten the stroke, you'll probably also be increasing the stroke rate. So in that case you wouldn't need to hold WpS constant. All that would matter would be mean speed for a given power output over the course (including allowances for internal losses at the different stroke rates).

 

[Leo Lazauskas]

Just read that. Makes sense, as far as it goes, but doesn't mention the fact that if you shorten the stroke, you'll probably also be increasing the stroke rate. So in that case you wouldn't need to hold WpS constant. All that would matter would be mean speed for a given power output over the course (including allowances for internal losses at the different stroke rates).

Maybe that's why impulse isn't used very often.
For me the shapes of the curve are interesting, and how they vary between
rowers in multiple crews.

If you are really interested in power, then you can get some idea using FIRM.
Oar handle power is handle force X OHCE distance from pin X angular velocity
squared.
You can get power of body segments using their weights and speeds.
I haven't thought much about internal losses yet. I have been looking at a
variety of "bio-energetic" models which could be fun to implement at some
stage, e.g. having small "tanks" to simulate anaerobic and aerobic capacities.

Version 2.30 was released yesterday.
Old anthro.csv input files will need to be changed to include radii of gyration.
Examples use "standard" estimates for those radii.

I've also included two more (real) rowing shells so users can compare
performance, and use them as the basis for their own designs.
(Fixed seat rowing is still very crude because I don't have any on-water data
to work with).

Late edit:
Old main input files will also need to be changed to include wind speed and direction.

 

[NoEyeDeer]

As soon as I can, I'll see about getting you some fixed seat data. It'll take me a while to get anything useful though. However, next year I'm going to try and set up comparisons between my new boat and the Herreshoff/Gardner boat, including test runs with the same person rowing each boat alternately. I want to get a good indication of how the real world performance with the same "engine" stacks up against the Michlet predictions. This should be useful information for future fixed seat designs.

 

[Leo Lazauskas]

As soon as I can, I'll see about getting you some fixed seat data. It'll take me a while to get anything useful though. However, next year I'm going to try and set up comparisons between my new boat and the Herreshoff/Gardner boat, including test runs with the same person rowing each boat alternately. I want to get a good indication of how the real world performance with the same "engine" stacks up against the Michlet predictions. This should be useful information for future fixed seat designs.

That will be an interesting exercise for you!
One day I should try to get some of my acquaintances to fix the seats
on their boats for one or two runs and record (at 100Hz) the gate forces,
oar angles, hull velocity and acceleration. That's all I really need.

Incidentally, the "internal power losses" you mentioned in a previous
post seem to be independent of stroke rate. See:
Hofmijster, Mathijs,
"Mechanics and energetics of rowing",
PhD Thesis, VU University, 2010,
http://hemus.nl/maand/2010/mechanics_rowing.pdf

 

[NoEyeDeer]

Interesting read. I noticed this down on page 37:

At higher stroke rates,
the fluctuations in kinetic energy are larger, suggesting that the internal dissipation
of mechanical energy increases with stroke rate. An indirect way to investigate the
magnitude of the negative muscle power is by considering metabolic energy expenditure.
As both the dissipation of mechanical energy and the subsequent regeneration thereof
involve metabolic energy expenditure, gross mechanical efficiency might be expected
to deteriorate with increasing stroke rate if negative muscle power is substantial.
From this it follows that minimization of negative muscle power could be an important
aspect of intermuscular coordination in rowing. As no data are known to us on the
relation between stroke rate and gross mechanical efficiency or the amount of internal
dissipation of mechanical energy in rowing, this is an area for future research.

It's also clear that athletes can't sustain the higher stroke rates for long, which implies something is being lost somewhere and that these losses increase at higher stroke rates.

 

[Leo Lazauskas]

It's also clear that athletes can't sustain the higher stroke rates for long...

I also wonder how those very high rates would affect their longevity as athletes.
You'd think that very high stroke rates must increase wear and tear on some
joints.

...which implies something is being lost somewhere and that these losses increase at higher stroke rates.

There are a variety of possibilities: increased drag due to surge; losses at
the oarblade; compressing the chest cavity at a rate that is out of phase
with the best regime for breathing and oxygen uptake, etc etc.

 

[NoEyeDeer]

Yeah I was thinking of the last one too, since breathing is very important for rowing. Well, important for continuing to live too.

 

[Leo Lazauskas]

Yeah I was thinking of the last one too, since breathing is very important for rowing. Well, important for continuing to live too.

In my youth I used to think that about drinking too. I also noticed that doubling the drinking rate caused my knee joints to buckle earlier. That's science.

 

[NoEyeDeer]

I had a theory that the reason they try to make sure you're smart before letting you into university is because of the number of brain cells that will be slaughtered by student drinking games. Need to have some left at graduation time.

 

[Atkinsopht]

Leo:
I too have been looking at FIRM.
Do you know how to contact the author?
I can find no name or contact information.
Hope you are well and Happy New Year.
Bill Atkinson

 

[Leo Lazauskas]

A few bugs that have corrected in version 2.31 which will be released
early in 2015.

1. The FIRM window is incorrectly titled FIRM v2.21. This is just a
minor over-sight.

2. Radii of gyration are referenced to the segment CM. The heading of
Table 1 is ambiguous. This has been corrected in version 2.31.

3. Average air and hydro drag power calculations shown on the main
screen and in summary.csv are incorrect. They are actually drag
averages in Newtons. No other calculations are affected. This has
been corrected in Version 2.31.