How can determine the accuracy of my PT? (or, it must be the taper talking)

Dave Tallo

A few weird wattage experiences with recent rides, a PT that has been ridden in 8h of pouring rain over the weekend, a recent replacement of hub batteries, and a few rides where the numbers seem to be anywhere from 10 – 30 watts off from the normal make me think my PT might be a little wonky.  Not “it shows 500 watts as my NP” wonky, but “off” wonky.   

 Torque has been zeroed many times through rides. 

 Is there a solid protocol to determine if the numbers that are appearing on the LYC today represent the same numbers I have been seeing all season?   I would guess that a steady cadence, same gear on the trainer over a few minutes would give a consistent number, but unfortunately, I don’t have a pre-rain, pre-battery baseline to compare this to.   

 

Supplemental:

Anyone replaced a PT wheel the week of a race?   I guess the same guidance from Rich’s “Read this before renting a Racedaywheels PT wheel” would apply, yes? 

charris

The only truly reliable method would be if you had done a stomp test before. Other methods (like you describe) might get you in the ballpark but are going to be subject to some error.

It's funny, sometimes when I'm reading an Andy Coggan post I imagine that he must show up for a ride with a set of weights, a scale, an anemometer, weather station, etc. and I chuckle to myself. But I'm starting to see his point.

I'd suggest retesting to establish a new baseline but obviously that's not ideal. And even if you did, I'd worry that the PT was on its way out.

FWIW, I have a PT Elite+ on an EA90SLX with wheelbuilder cover sitting in my basement that is, ahem, not getting much use currently. I could drop ship it to you so you had it before the weekend if you want to do some A/B testing.

Chris G

I would trust the thing.  When  PT does not work they tend to really not work as in crazy numbers or no numbers at all.  I think that you are making yourself crazy.  You kinda have to trust the tools and yourself.  If you think, wow at around 300 watts I think this thing is off by 10 or 20 watts sure seems that way.  How about running down the road with a 6:45 pace showing on your garmin and thinking that it seems 20 seconds too fast or slow.  Likely just you right?  Maybe not...just cause you are paranoid does not mean they are not after you...

charris

Fair point Chris - my thinking was if it was just 10 watts and/or a newb rider I'd chalk it up to taper tantrums. But the upper end of that range - 30 watts - is probably somewhere around 10-15% at the numbers he's riding which, for someone who has been staring at a dial for a long time, seems pretty significant.

You could do a stomp test, but all that would tell you is whether or not it's accurate *now* - not whether it's been accurate in the past. Although that would be some comfort I suppose.

 

This isn't helping is it Dave?

Coach Rich

Dave,

I know that WB could turn this around in a day if you wanted to get it looked at. I could tell Rich and crew to jam it out quickly. Shipping would suck but you probably know that.

Dave Tallo

Thanks a lot, all.



Craig, that is incredibly generous of you, and my plan is to do this weekend's RR, and if things are still crazy - like 10 - 30 watts crazy - I'll reassess and might take you up (I've done something like 25 RRs on this route, so I should have a pretty good baseline of data there). And hopefully shipping the wheel isn't going to interrupt your current 200k/week of run work!



Chris, you're probably right. Only real outlier was a millisecond max power of 985 on Saturday, and there is no way, no how, that I make that number come from my legs.



Rich, your post sounds suspiciously like an offer to hand-deliver a WB PT to CDA. Which I might entertain, after I make a final assessment this weekend! (seriously, I might pursue buying a new one, in which case, it'll absolutely positively come from WB, and I'd just opt in their for expedited turnaround for the surcharge, and have it shipped to my hotel in CDA. )

mgraffeo

Craig, any links to this stomp test for those who are curious?

charris

@Mike. Simple:

1. Sell your PT

2. Buy a Quarq

3. Buy a Wahoo Dongle, an iPhone/Pad/Touch, then download Qalvin from the App Store.

 

But since I still have a PT, I'll try do up a Wiki entry. Last year I just hobbled together something from several ST and Wattage posts - I'll dig those up (or I'll go looking for one - I imagine someone must have consolidated something somewhere by now).

mancona

I'm almost positive I have seen instructions on how to test a PT before, however IIRC it requires you to have a LYC instead of the Joule. I think in the setup/test menu of the LYC but it has been 2 years since I used one. The process was the exact same as the Quarq (i.e. take a bunch of measurements with crank arm at 3:00 with no weight, then add a know weight, etc). What I don't recall is what you actaully do with the measurements and how/if you can actually change the calibration. You may have only been able to check it and not set it.

I tried it once and then gave up as I just wasn't that interested and my PT was working just fine.

charris

@Matt - ah, but how can you be sure your PT is working just fine? Answer: do a stomp test.

You pretty much described the process. You can do it with the LYC and I'm pretty sure you can do it with the Joule as well now. You've also been able to do it with the Garmin 500 and 800 with recent firmware for the last while and the 310XT just added torque display.

 

@Mike - did a quick search and found something that didn't exist when I was looking last year. This lays it out pretty clearly:

http://www.cyclepowermeters.com/pow...k-76-c.asp

 

I will probably still do a Wiki article though with a few other notes.

 

With PT, if you discover during the test that it's out of spec, you send it back to Saris and they fix it. With Quarq and SRM, you can actually adjust the slope yourself.

 

If you want to see how slick the process is with Quarq's app, check out:

http://www.quarq.com/i/quarq-qalvin...110413.pdf

JustMike

I've done the test described here: http://www.cyclepowermeters.com/powertap-garmin-calibration-check-76-c.asp to figure out that my race PT was about -15% from my training PT. It's currently being repaired.