Stroke Rate and Perceived effort

[gregsmith01748]

Nav,

Thanks for your insightful comments. To your point about "solving the tradeoffs". One thing that I have noticed is that I get into "grooves", where the stroke cadence, split time, and breathing rate required to sustain the pace balance. If you think that oxygen supply is limited by lung capacity x breaths per unit of time, then maybe you have different sweet spots for different power outputs.

[Tinus]

I reread the physics of ergometers and that suggests that higher strokes rates are more work on the erg because of the ratio of energy pushing yourself up and down the slide versus the energy going into the flywheel decreases. It says that it takes a 75kg rower 37 watts to slide back and forth at 30 SPM. At 20 SPM, this will only take 11 watts. So, say that you are rowing at 1:55 pace (230 watts). At 30 SPM, you are putting 267 watts into moving yourself plus spinning the flywheel. If you pull the same split at 20 SPM, you are putting 241 watts into the effort, or about 10% less.

This seems totally backwards to what I experience sitting on the machine. I can do an hour at 1:56 if I don't cap the rate, but if I limit myself to say 20SPM, I'm gassed trying to pull faster than 2:00. Is this really just a matter of personal physiology? Are there really people out there that can do an hour faster at 20SPM than 25SPM?

I am now officially puzzled.

Without making any assumptions about optimal modes of muscle contraction one can already make calculations about optimal stroke frequency and this is not the lowest stroke rate possible (as put forward in the quote).

1) Indeed, as a rower moves with higher speed and higher frequency up and down the slide it may be assumed that more energy is lost merely due to losses from the movement alone (absorption/loss of the energy of the movement each turn point).

2) But, there is another opposite effect. In order to measure the distance/speed the rowing monitor does not make a direct measurement of power. Instead it measures flywheel rotation. This means that the same speed may possibly be acquired by different power input for different stroke rates.
At a higher stroke rate less energy/power is needed for the same number of flywheel rotations/speed. This is because higher speed fluctuations require more power for the same average speed. This same principle is also true for rowing on water.

Putting 1 and 2 together an optimum can be found around 30 strokes per minute and there is a large range in which stroke frequency only has small effect (e.g around the optimum only the order of 1 second difference on a 2km for a stroke rate 5 per minute higher).

Doing this on the back of an envelope: The first effect scales like frequency to the power 3 (The energy per stroke scales like frequency squared and the numbers of stroke per second scales linear with frequency. Together these result into the effect of power 3). The second effect scales like frequency to the power -2 (approximately, using the integral of a+(1/f)*x^3).
We could fill in some values. Using 10 watts for 20 spm as the first effect we have F^3/800. Using a speed fluctuation of 2 m/s at 5 m/s average, 300 Watts and 20 spm we have 4800/F^2. Totalling F^3/800+4800/F^2 which has a minimum at 19 spm.
The difference with the 30 spm above is probably in assumptions. 10 watts energy loss for moving at 20spm is the case for bad technique and loosing all kinetic energy each turn. When using higher speed and power as assumption the optimum will be higher. Also the scaling is not completely correct as it assumes certain speed profiles which are naturally not found. Numerical models improve this.

The consequence of the optimum around the stroke rates often used and the small effect of deviations is that one does not really have to focus on the mechanical/kinetic aspects of stroke frequency optimisation and one can focus on optimal muscle movement (which is actually much harder to determine).

[NavigationHazard]

More on the business of stroke rate vs. effort, perceived or otherwise.

Here are some graphs based on a session I did last night. My hope is that they'll illustrate some of what I've been talking about....

The rowing in question was 30" on/30" rest, the work intervals at 20/ 22/ 24 ... 40 strokes/minute. To the extent possible I attempted to keep my grunt level (rower effort) and stroke profile constant. The idea was to show what happens when input on the drive stays the same and stroke rating is used to manipulate pace.

First, here's the pace results:



Interval 1 was 1:40 r20, 2 was 1:37.4 r22, 3 was 1:35.5 r24, 4 was 1:34.3 r26, 5 was 1:32.6 r28, 6 was 1:30.9 r30, 7 was 1:29.8 r32, 8 was 1:28.8 r34, 9 was 1:28.8 r36 (oops; forgot where I was on the ladder until halfway through), 10 was 1:27.2 r38, and 11 was 1:26.2 r40.

As you can see graphically, the relationship between pace and rating is non-linear. That's because of the cubic relationship involved in the way the monitor calculates output: watts = 2.8*(1/((elapsed time in seconds/distance in meters)^3)), e.g. to go twice as fast you need eight times the watts. You can go faster at higher ratings, given constant input, but the more strokes/minute you take the more the rate at which you gain pace will tend to flatten out. And then there's the problem of sustaining the higher rating physiologically (q.v.).

If you hold rower input and distance/duration constant, spi (watts/rating) will tend to decay as rating increases:



Here it falls quite linearly from 17.5 at 20 spm down to 13.7 at 40 spm. This sort of decay is absolutely natural and inevitable. It follows that trying to hold one spi value constant in training across all ratings and distances/durations, as has been suggested elsewhere on this Forum, is completely misguided.

Here's what would happen to pace if instead of holding effort constant, I tried to hold constant the 17.5 spi I achieved on interval 1:



Could I do it, I'd end up at 1:19.4 pace at 40 strokes/minute. Why can't I do it simply by raising the rate?

Well, here's the wattage requirements at a constant 17.5 spi:



Instead of the 546 watts I factually was able to produce, I'd need somehow to generate 700. This would be impossible without massively increasing my input force. Moreover, the decreased recovery time within the stroke cycle would kill me:



Assuming a constant drive duration of 0.7 seconds, at 20 spm my recovery time is 2.3 seconds per stroke. At 40 spm it's 0.8 seconds/stroke. It simply is impossible physiologically to exert more force on the drive more often for the same distance/duration. Either spi cannot remain constant as rates go up or duration/distance must shrink, or both.

For what it's worth, compare the curves above for factual pace and recovery time. You don't have to be a rocket scientist to realize that there's a strong relationship between the two across the range of ratings. In fact, the correlation between the distance I rowed at a given spm and recovery time at that spm was R^2 = 0.99.

[nharrigan]

As you can see graphically, the relationship between pace and rating is non-linear. That's because of the cubic relationship involved in the way the monitor calculates output: watts = 2.8*(1/((elapsed time in seconds/distance in meters)^3)), e.g. to go twice as fast you need three times the watts. .

Wouldn't you need 8 times the watts to go twice as fast. 1:30 Pace= 480watts 0:45 pace = 3841watts
Maybe I'm missing something.

Good stuff, Nav.

[Bob S.]

As you can see graphically, the relationship between pace and rating is non-linear. That's because of the cubic relationship involved in the way the monitor calculates output: watts = 2.8*(1/((elapsed time in seconds/distance in meters)^3)), e.g. to go twice as fast you need three times the watts.

That should be two raised to the three or, simply, eight.

Bob S.

Edit: Whoops! Sorry. I see that the issue has already been addressed.

[NavigationHazard]

As you can see graphically, the relationship between pace and rating is non-linear. That's because of the cubic relationship involved in the way the monitor calculates output: watts = 2.8*(1/((elapsed time in seconds/distance in meters)^3)), e.g. to go twice as fast you need three times the watts. .

Wouldn't you need 8 times the watts to go twice as fast. 1:30 Pace= 480watts 0:45 pace = 3841watts
Maybe I'm missing something.

Good stuff, Nav.

Yes, I hit the wrong key on my spreadsheet and didn't notice (without my glasses the 3 looks like an 8)....

[Bob S.]

As you can see graphically, the relationship between pace and rating is non-linear. That's because of the cubic relationship involved in the way the monitor calculates output: watts = 2.8*(1/((elapsed time in seconds/distance in meters)^3)), e.g. to go twice as fast you need three times the watts. .

Wouldn't you need 8 times the watts to go twice as fast. 1:30 Pace= 480watts 0:45 pace = 3841watts
Maybe I'm missing something.

Good stuff, Nav.

Yes, I hit the wrong key and didn't notice (without my glasses the 3 looks like an 8)....

Ah come on, Jon. You didn't use the numbers. You typed out the word three in letters. No big deal. We all know that you knew what you really meant and most of us have had similar "senior" moments. In my case, I can palm them off as a pump-head moments. (I also do a lot of editing to catch them.)

Bob S.

[luckylindy]

Nav - awesome charts ... thanks for being this thread's guinea pig!

[jamesg]

Those curves only demonstrate the cubic Watt-Speed relationship, which is given.

I'd always thought the purpose of rating faster was to go faster; which necessarily means more power. Clearly when we get to the top of that hill, the only way forward is down. That's why it's a good idea to stop there, or even a little before.

In numbers, this means holding your stroke constant (though not necessarily to within 0.1%), as seen in the Wolverine L4s where it's around +/- 5% and is anchored to race power. If we use a certain stroke in a race, why should we train using another, once we've found a good one? Good does not mean astronomic, it has to be sustainable.

[hjs]

As you can see graphically, the relationship between pace and rating is non-linear. That's because of the cubic relationship involved in the way the monitor calculates output: watts = 2.8*(1/((elapsed time in seconds/distance in meters)^3)), e.g. to go twice as fast you need three times the watts.

Yes, I hit the wrong key and didn't notice (without my glasses the 3 looks like an 8)....

Ah come on, Jon. You didn't use the numbers. You typed out the word three in letters. No big deal. We all know that you knew what you really meant and most of us have had similar "senior" moments. In my case, I can palm them off as a pump-head moments. (I also do a lot of editing to catch them.)

Bob S.

This is to funny

Watch out Bob, as you know nav is always right and when is not he ............. is always!!!! right.
Making fun of him is not a good idea, he will hound you for the rest of your life lol

After all if you think of it, saying you typed a 3 instead of a 3 (vey close together on the board, not!) but in reality typing 5 times a letter is totaly understandable.

[hjs]

Those curves only demonstrate the cubic Watt-Speed relationship, which is given.

I'd always thought the purpose of rating faster was to go faster; which necessarily means more power. Clearly when we get to the top of that hill, the only way forward is down. That's why it's a good idea to stop there, or even a little before.

In numbers, this means holding your stroke constant (though not necessarily to within 0.1%), as seen in the Wolverine L4s where it's around +/- 5% and is anchored to race power. If we use a certain stroke in a race, why should we train using another, once we've found a good one? Good does not mean astronomic, it has to be sustainable.

Indeed, stating the obvious................

oops I am doing it again

[Tinus]

In numbers, this means holding your stroke constant (though not necessarily to within 0.1%), as seen in the Wolverine L4s where it's around +/- 5% and is anchored to race power. If we use a certain stroke in a race, why should we train using another, once we've found a good one? Good does not mean astronomic, it has to be sustainable.

We train at different stroke rates compared to racing stroke rates for about the same reasons that we train at different speeds compared to racing speeds.

[mikvan52]

Let me introduce another fly in the ointment:

Ever notice the difference in perceived effort based on the physical set-up of your own particular machinery?:

MY specific Example:
Compare the stationary erg to the erg on slides.
IMO: My perceived effort and rates seem noticeably different.
Less effort for slides; more effort for standard (stationary)..

Does anyone have any thoughts on this?
Keep in mind that 2k times have been considered equivalent for both configurations.

My reason:
Anecdotally, I feel I can row at a lower rate on slides and maintain a higher avg. wattage output but I HAVE TO CHANGE MY FORM. Form is not a power, endurance or energy variable... (hmmm)

[luckylindy]

Let me introduce another fly in the ointment:

Ever notice the difference in perceived effort based on the physical set-up of your own particular machinery?:

I notice this same thing with DF. When I started, I rowed at a 170+ DF and quickly hit 19:10 in a 5K without much effort. However, after advice on this forum I dropped the DF to 120-130, and it took me over a month before I could even break 20 minutes, and I just broke 19 ... but it was HARD. Even after a couple extra months of training, my perceived effort for a 19:10 5k is FAR greater than when I had the DF at 170+.

[Tinus]

Let me introduce another fly in the ointment:

Ever notice the difference in perceived effort based on the physical set-up of your own particular machinery?:

MY specific Example:
Compare the stationary erg to the erg on slides.
IMO: My perceived effort and rates seem noticeably different.
Less effort for slides; more effort for standard (stationary)..

Does anyone have any thoughts on this?
Keep in mind that 2k times have been considered equivalent for both configurations.

My reason:
Anecdotally, I feel I can row at a lower rate on slides and maintain a higher avg. wattage output but I HAVE TO CHANGE MY FORM. Form is not a power, endurance or energy variable... (hmmm)

Maybe the times for the different configurations aren't comparable after all.