SRM vs Computrainer power readings



colobikelaw3

New Member
Jan 2, 2011
5
0
0
I've used SRM for a while now and know the power readings are accurate. However, using Computrainer Ergo I'll set power to 400, but SRM reads only 250. Why so different? I've warmed up CT and calibrated accordingly, but always off. Any thoughts?
 
IME, the accuracy of individual Computrainers varies quite a bit and they also tend to drift a fair amount at even moderate workloads unless you actively cool the braking unit.

My CT was way off relative to a torque tested PT hub a few years ago until I voided the CT warranty and manually adjusted the CT slope via the hidden adjustment potentiometer accessible through the small access hole on the inside of the CT load unit. I swapped to SRM cranks about a year ago and other than the slight power offset due to drive train losses my power readings are very close at least they start that way after 10 minutes of warmup and the coast down calibration for the CT. Unless I dedicate a fan to cool the CT braking unit the two power readings start to drift after about half an hour to forty minutes of sustaining 250-280 watts.

I trained last winter at an indoor cycling center equipped with Computrainers. I learned quickly to try to get the same trainer for each session as the individual units were all over the map in terms of some reading high relative to my torque tested and zeroed PM and some reading quite low even after a solid 20 minute warmup and CT coast down calibration. Even when I used the same numbered CT unit week after week I'd see variation relative to my PM. I learned to take the CT power readings with a large grain of salt and to trust my PM for workout intensity.

Anyway, they do seem to vary a lot but if you don't mind voiding the warranty on your CT you can adjust the slope to get a better match to your SRM readings. Basically you just get a really good warmup for your CT, do the roll down calibration and make sure theh press on force comes in around 2 to 2.5 or so, readjust the tire press on force knob if your much outside of that range and make sure you get consistent roll down values (which BTW are in units of pounds of press on force). Then do a few three to five minute efforts at varying power levels across your primary range of interest for instance maybe 200 watts, 250 watts and 300 watts or something like that. Then compare your average SRM recorded average power for each of those efforts to the CT power for each of those efforts, tweak the adjustment potentiometer in tiny adjustments and retest till you get good agreement. It took me about an hour to do the adjustment but other than the drift issue with heating it hasn't changed in the past few years.


Good luck,
-Dave
 
How do you know SRM is correct?
That's far too big a difference to be accounted for by a simple out of cal unit.
Something else is wrong
 
This all very interesting. I appreciate your comments. I'm new to CT, but have had SRM for a year. My coach uses CT to determine various thresholds. My SRM corresponds closely with his CT. However, my CT is way off. I notice there is a drag factor setting, but haven't fully understood it's significance. Looking forward to the new CT/software next month. Fortunately, I'll get that for free since it's so new. I assume the SRM more accurately reflects my power since it's close with others. I would like to think I can produce 400+ watts for an hour, but I have my doubts there. I would agree temperature does affect CT wattage.
 
Originally Posted by Alex Simmons .

How do you know SRM is correct?
That's far too big a difference to be accounted for by a simple out of cal unit.
Something else is wrong

The first CT that I attempted to calibrate using an accurate powermeter (PowerTap, so no drivetrain losses to worry about) was off by nearly as much. Based on internet chatter, they do seem to have improved the consistency of the factory calibration over the years, but I'm still not sure that Racermate "gets it".
 
Originally Posted by acoggan .



Quote: Originally Posted by Alex Simmons .

How do you know SRM is correct?
That's far too big a difference to be accounted for by a simple out of cal unit.
Something else is wrong

The first CT that I attempted to calibrate using an accurate powermeter (PowerTap, so no drivetrain losses to worry about) was off by nearly as much. Based on internet chatter, they do seem to have improved the consistency of the factory calibration over the years, but I'm still not sure that Racermate "gets it".


I'm hearing ya. What I mean is, if the unit is out by that much, then something is wrong besides a calibration issue. It's a fault, probably a bad load controller, and not really a calibration issue.

As well just making sure the SRM is right as well, although that's probably far less likely, just want to make sure - I sometimes see SRM power files that are way wrong because either slope or ZO is accidentally way off mark. One day a masters aged rider sent me a 90-min race file at 6.x W/kg and they got excited about how well they were going. I had to let them down gently. LOL. Have seen a few others.
 
I have a similar issue with an ebay-purchased Computrainer. It's reading ~100W higher than my Powertap.

Where on the CT unit is the adjustable pot?

I'm going to run my other bikes over the CT to confirm that it is the CT that's incorrect before I adjust anything but it's well out of warranty so I'm not too worried.

EDIT:

"OK, I'll reply now that I can again....
I'll reiterate -- on our dyno, these units read correctly at the calibration point. Lab units are verified over a broader range -- but none of this will account for slippage or calibration routine errors, which almost always occur. We are displaying power from a load curve that operates the Load Generator and is based upon speed. The same errors people have indicated in speed variations from other devices to CompuTrainer can be equated to this too. The Load Generator is producing the correct load for the speed it sees and displaying the correct wattage for that speed. Period. If there is slippage, we have no way to factor that back into the equation except to use a constant [or assumed curve] of what that would be, which we know is neither. That said I'm going to let everyone in on a little secret -- you can adjust the CompuTrainer if you are so inclined. This adjustment will only raise and lower the curve, but not change the curve shape. DISCLAIMER: If you make this adjustment and then subsequently need me to put it on the dyno to recalibrate it -- it'll cost you $75. Proceed at your own risk. To adjust the CompuTrainer remove the sticker on the inside edge of the Load Generator [on the side closest to the tire]. This is the one indicating the power supply to use, etc. Behind this sticker is a series of rectangular holes. In the uppermost left hole is a potentiometer. Turning this pot to the left lowers the load at a given speed. Turning this pot to the right increases the load. If the CompuTrainer reads low on the watts reading, then chances are the load at the Load Generator is too high, which is limiting your speed, which results in a lower wattage display. Roger
RacerMate Inc."

http://www.postfinder.com/post/24419438.html

and more discussion here:
http://www.cyclingforums.com/t/319578/srm-vs-computrainer-wattage-discrepancy/15