PowerTap vs Calibrated CompuTrainer - 2 Hour Ride



Porkyboy

New Member
Apr 28, 2006
234
0
0
Hi

If anyone is interested I have today done a comparative study of the 2 units above over a 2 hour indoor ride and have posted the results to my blog. The link is in my signature, the entry is for Saturday 17th May 2008.

Regards to all.

PBUK
 

RChung

New Member
Sep 12, 2006
306
2
0
Did you do a static calibration check on the PT?

Re: your point #3: depending on how big gthe difference was, it could be due to high torque tire slip.
 

Porkyboy

New Member
Apr 28, 2006
234
0
0
Hi

RChung said:
Did you do a static calibration check on the PT?

Re: your point #3: depending on how big gthe difference was, it could be due to high torque tire slip.
Have to confess that not done static calibration on the PT, did the zero torque check of course but have no reason the suspect the unit at all, it's almost straight out of the box, and of course very recently manufacturer calibrated, gotta trust someone! I think if there was a problem it would be out of whack with the C/T, it would be a heck of a coincidence for both the PT and the CT to be out by the same margin.

Re the tyre slip suggestion, under the circumstances I made the observation in #3 the torque is actually very low, biggish gear yes but high speed so torque low, far lower than on the climbs (7-10N-m compared with 20 N-m). Press on force in recommended range and freshly cleaned tyre, pretty confident tyre slip wasn't an issue, has never been so far.

Thanks for taking the time to read through it all.

Cheers.

PBUK
 

Porkyboy

New Member
Apr 28, 2006
234
0
0
Hi Alex

Alex Simmons said:
How does the CT calculate power?
I'm just waiting for a "proper" answer from RacerMate and I'll post any reply as soon as I can. I'm pretty sure I know how it works but I'll get the facts!

Cheers.

PBUK
 

tonyzackery

Well-Known Member
Dec 23, 2006
3,517
80
0
56
Something struck me as being quite odd: "only" 1700+kj expended in a 2 hour ride????? I work with calories, so I understand 1700kj to be in the neighborhood of 400+ calories. An average of ~200cal/hr at 223w/hr seems extremely low to me.

Appreciate the info pertaining to the comparison, but something's amiss...:confused:
 

RChung

New Member
Sep 12, 2006
306
2
0
tonyzackery said:
Something struck me as being quite odd: "only" 1700+kj expended in a 2 hour ride????? I work with calories, so I understand 1700kj to be in the neighborhood of 400+ calories. An average of ~200cal/hr at 223w/hr seems extremely low to me.

Appreciate the info pertaining to the comparison, but something's amiss...:confused:
Hmmm. Seems right. 223 watts * 7200 seconds / 1000 = 1600 kJ.

I'm guessing you've forgotten to take gross efficiency into account.
 

Porkyboy

New Member
Apr 28, 2006
234
0
0
Hi RC

RChung said:
Hmmm. Seems right. 223 watts * 7200 seconds / 1000 = 1600 kJ.

I'm guessing you've forgotten to take gross efficiency into account.
Thanks for the input, appreciated. I'm taking from this that you refer to the 75% of the energy we produce that heats the room up so effectively and turns the wheel, sadly, not at all?

Cheers.

PBUK
 

Porkyboy

New Member
Apr 28, 2006
234
0
0
Hi Alex

Alex Simmons said:
How does the CT calculate power?
I got a reply to the question on the RacerMate forum from one of their staff. The answer given was:

"The wattage is calculated based on speed. We have a chart that says that based on the speed you're going on any grade how many watts you have to be putting out."

I'm sure the electronics that go into it are very complicated but it seems to work really well. I think the system, from memory from the manuals, adjusts itself (the applied load) x30 per second.

Hope this helps!

Thanks.

PBUK
 

rmur17

New Member
Oct 5, 2004
1,066
0
0
Alex Simmons said:
How does the CT calculate power?
simply {applied + RRC_zero_factor} load times wheel speed.

Alex the true 'feedback' portion is wheel speed or should I say more precisely: roller speed as there can be some slip from time to time.

I've done about five million cross-checks on mine vs. my old trusty PT Pro (and lesser millions on the SRM) since 2004. It does a darned good job of 'estimating' power in free-form, ergo and 3D modes.

Racermate are often pretty sketchy about this sort of thing. Some of the answers on their forum are downright laughable to an "old hand" :D .

In any case, the very simplest way to check for 'important' errors is to run multiple step checks in ergo mode and average the two across steps of 3-5min. Peak is pretty meaningless for any PM really ...

Anyhow, the procedure is out there :p ..
 

Porkyboy

New Member
Apr 28, 2006
234
0
0
Hiya!

rmur17 said:
In any case, the very simplest way to check for 'important' errors is to run multiple step checks in ergo mode and average the two across steps of 3-5min. Peak is pretty meaningless for any PM really ...
Thank you for the input. The step/ramp type test in ergo mode is precisely a test I plan to run, probably later this week, I've already prepared the .erg file to use, I'll post the graphs on the blog in case anyone is interested. I wish I could say I was looking forward to my FTP test on Thursday as I'll be collecting data on the P/T and the C/T - but I'm not, too uncomfortable!:eek:

Cheers.

PBUK
 

rmur17

New Member
Oct 5, 2004
1,066
0
0
Porkyboy said:
Hiya!

Thank you for the input. The step/ramp type test in ergo mode is precisely a test I plan to run, probably later this week, I've already prepared the .erg file to use, I'll post the graphs on the blog in case anyone is interested. I wish I could say I was looking forward to my FTP test on Thursday as I'll be collecting data on the P/T and the C/T - but I'm not, too uncomfortable!:eek:

Cheers.

PBUK
the only tip I can think of now is don't make the steps too short as when you make a step in ergo mode it take the CT controls a little while to settle into the new load. As well, there's a natural tendency to bog down just a little at the new load, cadence drops then rises. Long way of saying that the rising edge of each step in PT true power terms will be a little wobbly.

I tried for 5-min most of the steps except when hitting 4xx Watts which back in 2004 I couldn't hold for very long (very often).

FTP test ah - bring on the cooling, hide the clock and it'll pass :)
 

tonyzackery

Well-Known Member
Dec 23, 2006
3,517
80
0
56
RChung said:
Hmmm. Seems right. 223 watts * 7200 seconds / 1000 = 1600 kJ.

I'm guessing you've forgotten to take gross efficiency into account.
Gross efficiency??:confused: The formula you reference fails to account for bodyweight. Doesn't the CT take that variable into account when calculating work performed?

I've seen numerous formulas on the net that attempt to estimate caloric expenditure for a cyclist with a certain bodyweight cycling at a certain speed and none of them would return a figure as low as the above.

I'm not attempting to hijack this interesting thread so this will be last post on this matter...Thanks in advance for your reply.
 

RChung

New Member
Sep 12, 2006
306
2
0
tonyzackery said:
Gross efficiency??:confused: The formula you reference fails to account for bodyweight. Doesn't the CT take that variable into account when calculating work performed?
Doesn't need to. The power the CT is measuring is the power needed to spin its roller. No matter how long you pedal that damn thing you're still going to end up 10 feet from the wall in front of you. You're not moving so your bodyweight is irrelevant.
 

tonyzackery

Well-Known Member
Dec 23, 2006
3,517
80
0
56
RChung said:
Doesn't need to. The power the CT is measuring is the power needed to spin its roller. No matter how long you pedal that damn thing you're still going to end up 10 feet from the wall in front of you. You're not moving so your bodyweight is irrelevant.
LOL! Last time - promise...
Okay, you say BW is irrelevant - but what about going uphill while riding on a Computrainer??? BW is most definitely relevant, and you're still going to be 10 feet from the wall in front of you...correct???? So back to my initial comment...it appears that something is amiss (to me anyway) with the OP only burning ~200+/hr while riding at 220+ watts for 2 hours. The formula you provided for kJ calculation leaves out the cyclists' weight, which does impact how many kJ will be burned...no?
 

Alex Simmons

Active Member
Mar 12, 2006
2,471
40
38
tonyzackery said:
LOL! Last time - promise...
Okay, you say BW is irrelevant - but what about going uphill while riding on a Computrainer??? BW is most definitely relevant, and you're still going to be 10 feet from the wall in front of you...correct???? So back to my initial comment...it appears that something is amiss (to me anyway) with the OP only burning ~200+/hr while riding at 220+ watts for 2 hours. The formula you provided for kJ calculation leaves out the cyclists' weight, which does impact how many kJ will be burned...no?
No.

Energy (J) = Power (watts) x time (seconds)

200W x 7,200 secs (2hrs) = 1,440kJ

If you are measuring power already, weight has nothing to do with it.

Now to produce that 1,440kJ of "mechanical" work at the rear wheel, our bodies need to burn about 4 times that since we are about 23-24% efficient at converting our total energy output (heat etc) to mechanical energy at the rear wheel. Since that's about the conversion rate of kJ to Cal, then you can basically say we burn same number of Cal as kJ of "mechanical" work done at the rear wheel.
 

Alex Simmons

Active Member
Mar 12, 2006
2,471
40
38
rmur17 said:
simply {applied + RRC_zero_factor} load times wheel speed.

Alex the true 'feedback' portion is wheel speed or should I say more precisely: roller speed as there can be some slip from time to time.

I've done about five million cross-checks on mine vs. my old trusty PT Pro (and lesser millions on the SRM) since 2004. It does a darned good job of 'estimating' power in free-form, ergo and 3D modes.

Racermate are often pretty sketchy about this sort of thing. Some of the answers on their forum are downright laughable to an "old hand" :D .

In any case, the very simplest way to check for 'important' errors is to run multiple step checks in ergo mode and average the two across steps of 3-5min. Peak is pretty meaningless for any PM really ...

Anyhow, the procedure is out there :p ..
I'm not concerned about how good it is (I have a pretty good idea of that), was just interested in how the CPU determines and then records power. e.g wWhat interval does it use? A turn of the crank, the roller, or a time based interval?
 

rmur17

New Member
Oct 5, 2004
1,066
0
0
tonyzackery said:
LOL! Last time - promise...
Okay, you say BW is irrelevant - but what about going uphill while riding on a Computrainer??? BW is most definitely relevant, and you're still going to be 10 feet from the wall in front of you...correct???? So back to my initial comment...it appears that something is amiss (to me anyway) with the OP only burning ~200+/hr while riding at 220+ watts for 2 hours. The formula you provided for kJ calculation leaves out the cyclists' weight, which does impact how many kJ will be burned...no?
uhm -- 1700kJ is approximately equal to 1700 kcal (@ 24% GE) not 400!! As Robert said some posts back you seem to be neglecting efficiency.

this link should help:

http://www.midweekclub.ca/powerFAQ.htm#Q21

Now this concerns the additional energy expenditure during exercise - over and above basal requirements. So yes if you're heavier you WILL likely burn an extra few kcal whilst exercising but that's simply basal ..

Anyhow, I hope that helps.