Comparing efficiency at different stroke rates

[Annisotropic]

As described in my other thread, I've been consciously working on technique.

Is there a simple formula for predicting pace at different stroke rates? eg. If the pace is 2.xx @ 18spm, what pace should I expect at 20spm (over the same distance)?

[dabatey]

Google 'pace to watts' and the c2 calulator should appear.

You can then work out your watts per stroke for your chosen baseline pace/power.

Then multiply for new stroke rates to get power and using the same calculator convert back to pace.

It should give reasonable ballpark figures for everything but faster/shorter stuff where your watts per stroke will probably be higher.

[Annisotropic]

Perfect, that's exactly what I hoped for!

[gvcormac]

There's no formula. I can row at the same pace (e.g. 2:00) whether I pull 15 spm or 35 spm. You need to pull harder at slower stroke rates, to achieve the same pace (or power or calories).

[Annisotropic]

Yes. At a higher stroke rate, one doesn't need to pull so hard, to achieve the same rate. I was asking what rate could reasonably be predicted if one pulled equally hard, but more often.

[Ombrax]

Everything else being constant, the equation for HP says that it is linearly proportional to RPM.

If you apply that to the erg, stroke rate corresponds to motor rpm, so the increase in watts is proportional to the increase in SPM.

[jamesg]

For equal stroke length and force, a 10% higher Rating (so 10% higher Power too) gives about 3% faster pace, due to the cube law: 1.03³ = 1.093

This is due to aero- and hydrodynamics which say that to go twice as fast we need eight times as much Power. This cube law applies to both fans and boats, which is why C2 machines have a fan.

The exact formula is not so simple: Watts = 2.8 V³ on a C2, and Pace is 500/V, with V in m/s.

Ergdata has a page where we can choose to see both Pace and Watts.

[JaapvanE]

For equal stroke length and force,

By definition this isn't true when you change SPM. Increasing SPM will reduce time per stroke, so either drive or recovery has to become faster (or both). Reducing recovery time, reduces the time the flywheel is allowed to spin freely, thus reducing the distance travelled per stroke. Increasing SPM by shortening the drive, requires a higher acceleration during the drive to become faster, thus increasing the force on the handle. Thus, either way these assumptions will not hold.

And this is besides the fact that changing stroke rate changes the feel of a stroke. So rowing 2.xx at 40 SPM will feel light as a feather (but will wear you down by going up and down the slide more often) where 18 SPM typically results in a slow heavy stroke. From there on, it is just a matter of personal preference what you like most.

[jamesg]

It was a hypothetical question by the OP, not an assumption. However that is what oarsmen do.

Try MCs L4 tables, they're very effective. The roughly constant factor is Work: force x length.

[iain]

For equal stroke length and force,

By definition this isn't true when you change SPM. Increasing SPM will reduce time per stroke, so either drive or recovery has to become faster (or both). Reducing recovery time, reduces the time the flywheel is allowed to spin freely, thus reducing the distance travelled per stroke. Increasing SPM by shortening the drive, requires a higher acceleration during the drive to become faster, thus increasing the force on the handle. Thus, either way these assumptions will not hold.

Yes, but it also increases the number of strokes performed in the given time which will compensate for that.

Some of us find that our work per stroke is fairly consistent and so the ratio holds (ie drive stays the same and the recovery is slowed), while others actually increase the work per stroke with the rating (ie shortening the time of the drive and accelerating the fan faster) and so the power increases faster than rating. What you want to avoid (except for sprinting) is the work per stroke reducing as rating increases (ie stroke becomes weaker and rushing the slide). As for efficiency, each stroke takes more energy at higher rates as we need to accelerate our bodies quicker and so the efficiency actually drops.

[JaapvanE]

For equal stroke length and force,

By definition this isn't true when you change SPM. Increasing SPM will reduce time per stroke, so either drive or recovery has to become faster (or both). Reducing recovery time, reduces the time the flywheel is allowed to spin freely, thus reducing the distance travelled per stroke. Increasing SPM by shortening the drive, requires a higher acceleration during the drive to become faster, thus increasing the force on the handle. Thus, either way these assumptions will not hold.

Yes, but it also increases the number of strokes performed in the given time which will compensate for that.

Regardless, when keeping force during the stroke the same and increasing SPM, stoke length will shorten up by sheer definition alone, as the time for the boat/flywheel to travel will decrease.

Increasing SPM can compensate for stroke length. I do that almost daily when fatigue sets in on SS sessions and those long heavy strokes become too much so I switch to a lighter faster stroke. But there is no easy magic formula there. There is only one way to find out: that is just row the SPM and see what happens.

[dabatey]

By definition this isn't true when you change SPM. Increasing SPM will reduce time per stroke, so either drive or recovery has to become faster (or both). Reducing recovery time, reduces the time the flywheel is allowed to spin freely, thus reducing the distance travelled per stroke. Increasing SPM by shortening the drive, requires a higher acceleration during the drive to become faster, thus increasing the force on the handle. Thus, either way these assumptions will not hold.

Yes, but it also increases the number of strokes performed in the given time which will compensate for that.

Regardless, when keeping force during the stroke the same and increasing SPM, stoke length will shorten up by sheer definition alone, as the time for the boat/flywheel to travel will decrease.

Increasing SPM can compensate for stroke length. I do that almost daily when fatigue sets in on SS sessions and those long heavy strokes become too much so I switch to a lighter faster stroke. But there is no easy magic formula there. There is only one way to find out: that is just row the SPM and see what happens.

Does the c2 calculation take into account a freely spinning flywheel? Or is it only the drive phase where flywheel speed is increasing that it works from? I suspect the latter as we don't see decreasing paces on the screen after a pull.

[iain]

Does the c2 calculation take into account a freely spinning flywheel? Or is it only the drive phase where flywheel speed is increasing that it works from? I suspect the latter as we don't see decreasing paces on the screen after a pull

I leave that to those closer to the detail, but the pace is shown as an average for each stroke and so is only shown once per stroke. It certainly includes a decreasing pace during the recovery, but whether that is calculated or based on measurement I do not know. I'm sure we have heard before, but I am not sure which stroke is shown. I suspect that it is from the end of the previous drive to the end of the last drive as you often get low splits when the drive started before the cross over into a new interval, so I suspect that this is then starting with the new interval and ending at the end of the drive so is not reduced by the slowing of any recovery.

[Annisotropic]

Try MCs L4 tables, they're very effective. The roughly constant factor is Work: force x length.

These are surprisingly difficult to track down - none of the links in archived forum threads work for me.
I eventually found them buried in the website of a UK rowing club, but they may have been updated since then.
Mike Caviston - if you see this, are you willing to share them again here, please?


Also, as a tangential question, I see people referring to their stroke length - how are they measuring this, please?

[JaapvanE]

I leave that to those closer to the detail, but the pace is shown as an average for each stroke and so is only shown once per stroke. It certainly includes a decreasing pace during the recovery, but whether that is calculated or based on measurement I do not know.

Power is indeed measured across a stroke. As I have no insight into the inner workings of a PM5, there is no way to know for sure. However, as lead developer of ORM, I have implemented its Open Source counterpart (and am calibrating it against a PM5 to check the correctness of our our physics). If you are interested, the real physics is written down here in detail.

There are two ways of measuring power on the flywheel, the simple one and the complex one. Essentially, in the simple one Power is measured across a stroke by using the average (cubed) angular velocity of the flywheel across the stroke multiplied with the dragfactor. The easiest way to determine the average pace across the stroke is to determine the angular distance traveled throughout the stroke, and divide it by the stroke time. It is a bit inprecise, but robust as hell.

The alternative algorithm is more complex, as it essentially calculates "the volume under the force curve" to calculate the work done, it is more precise and can actually handle inconsistent strokes. Research done by Ulm university seems to exclude the use of latter more complex algorithm, as the PM5 doesn't react well to unstable strokes. But this algorithm also contains the total stroke time as the power is averaged across the stroke. And this algorithm highly depends on the angular velocity at the begin and end of the drive. So in the end, shortening the recovery gets into this formula as well.

I'm sure we have heard before, but I am not sure which stroke is shown. I suspect that it is from the end of the previous drive to the end of the last drive as you often get low splits when the drive started before the cross over into a new interval, so I suspect that this is then starting with the new interval and ending at the end of the drive so is not reduced by the slowing of any recovery.

As prof. van Holst already pointed out, it actually does not matter if the stroke is defined from catch-to-catch (thus a stroke starts at the begin of the drive and ends at the end of the recovery) or from finish-to-finish (thus start at the begin of the recovery). Both are valid perspectives, and as long as a stroke includes a drive and a recovery, it will not matter.

From pure observation of both the display and bluetooth messages, the PM5 seems to use the catch-to-catch definition. Probably because it is much easier to detect reliably (start of a drive is quite an easy event to detect from the flywheel speeds), and everybody thinks this way about a stroke anyway (every session starts with a drive...). Best it can be seen in the PM5 bluetooth specification (which we implemented, see this part of our implementation of the PM5 Bluetooth stack): on a driveStart event, all stroke related metrics are handed over to clients. On a recoveryStart, you only get a simple position report. This is quite consistent with a catch-to-catch implementation.

With ORM we actually implemented both simultanously. So we can update pace, power, stroke rate, distance per stroke, etc. twice per stroke. I find it helpful, as you get a quicker response to changes (especially when your drive becomes more powerful in acceleration).