Are watts per stroke just finished or an average of last strokes?

[greg]

Hi All,

I am wondering if the wattage per stroke showed by PM5 is really per stroke just finished or is it just an average from the last x strokes?

I observed the following sequence:
1. a stronger pull, just 56W, but the force curve is quite high
2. a light pull, 86W, but the force curve is small and almost flat

By this simple test I saw that the force curve reflects the last stroke which has been just finished. Whereas Watts are somewhat still the average. Please note that I don't mean Ave Watts, but Watts.

Do you have the same observation? Is it correct behaviour of the machine?

Thanks

[Citroen]

It's the power averaged since the previous split.

Use Ergdata and you get stroke by stroke data (in the log).

[JaapvanE]

I am wondering if the wattage per stroke showed by PM5 is really per stroke just finished or is it just an average from the last x strokes?

I use OpenRowingMonitor next to a PM5, where ORM has no smoothing across strokes (thus really displaying the last finished stroke data). And ORM calculates pace/watts twice per stroke, where the PM5 updates this once per stroke.

The formula we all use for power calculation have the side effect that during an acceleration, the power calculation is a bit slow to respond. For ORM, and I suspect the PM5, that is a conscious decission as it improves robustness of the data produced. As the force curve is created independent of the power calculation, it isn't bothered by this. So there is some discrepancy between power and force curve to be expected regardless of monitor.

When comparing the PM5 to ORM, I notice it will display the same data eventually, but it is indeed a bit sluggish to respond. I suspect it displays the running average of the last two or three strokes.

[jamesg]

wattage per stroke showed by PM5

In the split screen, the large figures in the top half are last stroke. Those in the lower half are averages and forecasts.

[Ombrax]

Not to be too pedantic, but if you're doing these calculations yourself, be sure you use the correct units:

Most folks talk about watts divided by the stroke rate (not just watts / stroke). Since 1 Watt = 1 Joule / sec divide that by the stroke / minute / 60 to get strokes / sec, so you then get

Watts / sps = Joule / sec x ( 1 / (strokes / sec) ) = (Joules / sec) x (sec / stroke) = Joules / stroke which is energy or work done per stroke, which could be a useful measure if you're interested in that sort of thing. (If you want to tweak it further, you could divide by your mass, to get J / stroke / kg.

Alternatively, since 1 W ~= 0.74 ft lbf / sec you could calculate ft lbf / stroke if you prefer that set of units.

Bottom line, check your units...

[jamesg]

A handy amateur Work level is about 200W at 20, ie 10 Watt-minutes = 600J.

[JaapvanE]

Not to be too pedantic, but if you're doing these calculations yourself, be sure you use the correct units:

That, in fact, can't be done because of limitations by C2.

In theory, the area underneath the force curve is the work done. You can get the average force of a stroke and the drive length from ErgData, so that can be calculated.

However, in rowing it is customary to calculate the power across the entire stroke, thus one needs the time of the total stroke. There you run into trouble, as the indicated stroke rate on a PM5 is way too imprecise to be used for any calculation. An indicated stroke rate of 20 can be a stroke rate of 19.5, or a stroke rate of 20.4, which makes a significant difference in calculations. And our side-by-side comparison to ORM suggests that stroke rate also seems to be smoothened to make it more stable.

Even if you use tools like PainSled to record all data, you can only get single digit stroke times, which is also lacks the required precission for this calculation to produce meaningfull results.

And that is beside the fact that the OP has already posed a very valid observation: a strong stroke (as shown by the force curve) doesn't align with a high pace, nor does a weak stroke (as indicated by the force curve) align with a low pace.

A valid point you could have made, and I forgot to mention, is that the recovery is part of this equation and thus matters too. So a strong drive, followed by a too long recovery might reduce the power produced. And it might display it after the recovery is completed.

This, in fact, is one of the questions still open decades after prof. van Holst's initial analysis of the PM: when does the PM's stroke end? Is it drive followed by a recovery, or a recovert followed by a drive?

[Ombrax]

If one is happy with an average value (as James showed in his calculation) you can get a very good approximation for power and a very good approximation for stroke rate over a given period of time. The net result (which would be fine for me) is that I can calculate the work / stroke. I don't need an instantaneous number, simply something to let me know where I am now, and as my training continues and I hopefully improve over time, where I'm going. Sure, the recovery time affects how much time is spent applying the force in a stroke, but if all I want is the work I've done in a stroke, it doesn't matter if the recovery is 70% of the stroke time or 30% of the stroke. If you want to fine-tune that part of your technique then you don't use this basic calculation, you do something else. Over, say, any generally steady-state condition lasting a few minutes (or more) I'm happy to use the PM's stroke rate.

[JaapvanE]

If one is happy with an average value (as James showed in his calculation) you can get a very good approximation for power and a very good approximation for stroke rate over a given period of time. The net result (which would be fine for me) is that I can calculate the work / stroke. I don't need an instantaneous number, simply something to let me know where I am now, and as my training continues and I hopefully improve over time, where I'm going.

That is the OP's key question: is it really instantanuous, or some running average?

[jamesg]

Depends how you set up the screens. PM and Ergdata let us see whatever we want, including units, averages and last stroke. In ergdata there's a page with twelve numbers, and more to choose among.

However Watt/Rate is not shown. I do that ex post in an excel line, using stroke count, meters and time, with one decimal. Clearly it's easy to follow on the fly too: 120/20 = 6, which is not enough to make me go blind.

[JaapvanE]

Depends how you set up the screens. PM and Ergdata let us see whatever we want, including units, averages and last stroke. In ergdata there's a page with twelve numbers, and more to choose among.

Please read OP's question carefully instead of blindly reciting the manual.

OP wants to know if the data on his display (he already chose the data he likes, and his bottom display shows the force curve, so no split averages to be displayed!) is true instantanuous data of the last stroke (and thus should match with the force curve displayed) or is a smoothened running average (that can deviate from the force curve).

[jamesg]

The split display with the force curve shows both the average and the last stroke Watts; and says so.