Calibration Error

Since 2001, over 90,000 cyclist's have joined Cycling Forums to discuss topics from general cycling to equipment, training, racing and travel or vacation destinations (especially in europe during the tour de france). We also feature an great deals in our online store, 100's of articles, classifieds and product reviews.

To the SRM users... imagine you have a calibrated SRM, ie it's slope has recently been calibated "properly".

I was wonderining if anyone has done the math to figure out what sort of error is introduced to the power readings as a result of either incorrect zero offset or slope? For example if it's off by 10 points?

Also does anyone have any insights or guesses into the level of error introduced by not recalibrating the zero offset, for example if you ride one day and it's 20 degrees C, then next it's 10?

To the SRM users... imagine you have a calibrated SRM, ie it's slope has recently been calibated "properly".

I was wonderining if anyone has done the math to figure out what sort of error is introduced to the power readings as a result of either incorrect zero offset or slope? For example if it's off by 10 points?

Also does anyone have any insights or guesses into the level of error introduced by not recalibrating the zero offset, for example if you ride one day and it's 20 degrees C, then next it's 10?My understanding of the power calc is as follows (but I could be wrong):

Slope = Frequency transmitted/torque, i.e. Hz/Nm

SRM power = [Frequency(Hz)-Zero_Offset(Hz)]/Slope

Presumably there's a circumferential pedal velocity factor to be added in there as well for the power calculation. Hence you can see the impact an error in ZO or Slope will have.

To the SRM users... imagine you have a calibrated SRM, ie it's slope has recently been calibated "properly".

I was wonderining if anyone has done the math to figure out what sort of error is introduced to the power readings as a result of either incorrect zero offset or slope? For example if it's off by 10 points?

Also does anyone have any insights or guesses into the level of error introduced by not recalibrating the zero offset, for example if you ride one day and it's 20 degrees C, then next it's 10?Alex has most of the answer there. The rest is simply that instantaneous torque is multiplied by instantaneous cadence (well both averaged over a complete crank cycle but that's instantaneous enough!) to yield instantaneous power. Cadence is converted to rad/s BTW.

Here's a tip that I use: always download your SRM files using SRMWin software. Then import the file into WKO+ for normal usage.

now the advantage to doing that is there's a quick and painless method of correcting the ZO (or the slope) using SRMWin. Under the File->Properties menu, there ought to be a dialog box for slope and zero-offset. Enter the corrrect one there and the file will be updated. it's up to you whether to write over the old file or save another with the corrected ZO.

Let's say you did a TT but in the excitement, adrenalin rush etc forgot to zero your meter before the event (not that I've ever done it of course :o :o ). But once you've finished and are desparate to see if you've hit a power PB:) , you remember you didn't zero it.

If you zero it after the TT, and note the zero-offset, you can manually (yet easily) correct the file as I noted above.

The same procedure applies for folks with multiple PC heads and different slopes in each one. DOH!!

Roughly speaking though, for folks who ride at a cadence around 90 rpm, I once worked out that an offset of 10Hz corresponds to around 5W discrepancy in average power. If the stored offset is high compared to actual, the power reported is low (the opposite is also true).

I was wonderining if anyone has done the math to figure out what sort of error is introduced to the power readings as a result of either incorrect zero offset or slope? For example if it's off by 10 points?
You can play around with this in SRMWin and see for yourself. Pick a file and modify the properties (0-offset, slope, etc.) and see how the Pave changes.

Also does anyone have any insights or guesses into the level of error introduced by not recalibrating the zero offset, for example if you ride one day and it's 20 degrees C, then next it's 10?
Depends. I've found that it appears that the 0-offset drops in value as the temp goes down (still confirming this). So if you were to skip the 0-offset on the 10 C day the SRM would under report your wattage (if I got that right) as the slope would have, presumably, dropped.

Dave