Ygduf

are we in the correct lunar phase for that?

Racer Ex 

My sundial says yes.

Creakyknees

so, how much does the readout wattage change after you do that? e.g. will an uncalibrated srm read at 200 instead of 300? etc.

Creakyknees

Also: I know a guy who's employed by the USDA to calibrate the scales at the cotton gins near Lubbock. He's a former cat 1, now old guy who takes juniors to Belgium every year.

tetonrider

the powertap portion is relevant. i'm pretty sure you can check but not update the slope on a power tap. this makes it useful for identifying a change in slope but not fixing it.

^^ this tells you the process. with an SRM or quarq one can take the info and update the slope so that future readings are correct. the PT and LYC -- to my knowledge -- do not allow one to make this update.

getting the weight right is critical to the calculation of slope. don't forget to weigh anything else that you use to attach the known weight to the pedal.

a few grams can make a difference.

my recommendation to those looking for a weight is to either find a scale that is accurate to +/-2g for a weight that is in the 20kg range. if you're going lighter, the weight needs to be more accurate. (ROT is error should be 1/1000th of the thing being measured.)

use the heaviest weight you can manage as it makes the process easier and more tolerant of minor variation. i believe 44# is a functional minimum.

tetonrider

^^ yes. in the earlier post you wrote +/-2%...i think perhaps you actually meant +/-0.2%. your example here agrees with the ROT i've always used.

tetonrider

i very much agree that there is not all that much to be done about past measurements. it's in the past, and unless it was fixing the workout that just happened there's not much point in fretting over it. checking slope gives users some power; learning how to check (and set) slope is actionable.

i'm also in the camp that we are measuring a moving target. stated accuracy for these devices is in an optimal scenario (and some manufacturers have just made a claim because it is what the industry leaders have proven) -- simple actions like checking slope periodically gives us the best chance to attain that optimal number, but there is still error no matter what.

tetonrider

creaky, it depends on the magnitude of the slope change.

a healthy meter may take a few months to stabilize but then should be rock solid. i've tested 5-6 yo SRMs that are dead-on. ex may have some experience here.

as a concrete example, if i switch from 53/39 rings to 56/44 on a particular unit i have, i will see a 0.1 change on a slope of ~23. this is about 0.5% error -- meaning 300w would be reported as 301.5. definitely in the noise, but that is a bomber unit. SRMs tend to be less influenced by ring choice (manufacturer) for a given size (in fact, not at all influenced by it in my experience), and minimally influenced by changes in ring size.

i have tested other brands and have seen much more variability (6% and up) -- what i would classify as material.

if it's a broken meter, it could be anything. if you were seeing 200 when you are putting in a 300w effort you would notice something is immediately wrong based on PE alone.

more subtle, but material, efforts such as reporting 290 for a 300w effort are harder to detect based on PE but can be quite important for quality data.

Creakyknees

cool, good answer, thanks man

shovelhd

What was really frustrating for me (and certain other important people) was trying to recover from a major injury with a power meter that was failing. In my case, it was reading low and kept drifting downward. So I would struggle with the hard workouts, not being able to finish them, and no matter how much rest I was getting, it was like swimming upstream after a dam release. It kept getting harder and harder even though I was getting stronger.

Hopefully I am not jinxing myself but so far, the Riken replacement has been rock solid. I did not calibrate it myself because I do not have the tools, but the offset has been solid and predictable since day one. Temperature compensation seems to work well. The Riken also has the Omnical feature which allows changing chainrings without a need for recalibration. I don't know how well that works as I don't change chainring sizes, but whatever they improved on the power meter side seems to be working out well.

thechemist

Good catch TR!
that sucks shovel. I am crossing my fingers for a solid unit myself.

rbart4506

I'm with Shovelhd...S975 was flaky from the get go! Riken has been solid. Been using it on the rollers primarily and power numbers are consistent when compared to PE/gearing/cadence...

Of course that doesn't mean I shouldn't check it

danielrbaer

So, my wired powertap was consistently reading ~50 Watts below what it should have on my ride today (based on RPE). I tried zeroing out the torque, but that didn't help. I just replaced the hub batteries, too. Any thoughts on how I could fix this?

shovelhd

Take a rest week. Seriously.

danielrbaer

Ha, this week actually turned out to be really light regarding cycling; it's not as if i'm overtrained to the point where producing 250 W actually feels more like 300 W. It's just annoying not having the PM read correctly relative to the output i've been able to associate with previous efforts.

shovelhd

RPE is the least dependable metric. How many races have you done? You're better off using Strava.

Saris will calibrate your hub for a fee, but it's probably in the same ballpark as buying a calibrated weight.

thechemist

Went out today for the first time with the my new quarq. It was nice seeing the unit variation from initial ride to end only be 6 units. I expect a bit of a shift as things settle on the bolts but so far so good. Did some manual and auto zeros and will be checking the slope at some point soon.

danielrbaer

I agree. I'm certain that there's something wrong with the hub. I'm having trouble running the offset torque value test on the LYC, though, and was asking to find out if there were any solutions other than sending it to Saris or using a calibrated weight.

JandersUF

I got a "new" Red quarq about 18mo ago. It has been very dependable since the get go, only with some dropout issues while riding indoors (I always suspected interference from multiple nearby wifis, etc.)

Last week I was out for a ride, about 10mi from home looked down and I had a 5minute average of 500w... something fishy. A few auto-zeros and coasting later, and I was only putting out 20w at most. I had changed the battery 2 days prior. When I got home, I checked the offset, and it was varying wildly +/- 20k.

I emailed quarq, and they got back to me within a couple hours. Free fedex the next AM, and they replaced the entire spider component, shipped it back, and I had it about 5.5 days after it failed. Cross country.

New unit seems to be very stable, and actually no long seems to suffer from in door drop out.

So minus one point for failing, but plus 3 points for excellent customer service!

mike868y

typical sram approach

Racer Ex 

It's probably more of a commentary on how bad most customer service is that people value good customer service over a product that doesn't fail.

I'd rather find out 10 years later that a company has suck ass CS then 1.5 that it has good CS.

Ygduf

In guest service work, a customer that has a problem in a hotel and has it satisfactorily resolved on 1 contact leaves with a better impression of the hotel than if they had no problem at all.

It's not unique. It's like reverse negativity bias. You attribute the problem to an unavoidable circumstance, but find the service to be exceptional. It's a + for the company, for sure.

mike868y

my bad experiences with sram products (despite good experiences with their cs) turned me off to the brand in a big way. no experience with quarqs, but it sounds like the same story.

jsutkeepspining

My first quarq was perfect! ZO moved about 1 unit most of the time, and it was always consistent until one day it just stopped working. Sent it off, and got a new Riken. THings a piece of ****. ZO always moves 30+ units.

happybday29475

I thought I would post my performance manager chart for my CX season for educational / feedback purposes. Feel free to dissect/comment/criticise/snark. I am surprised by the steadiness/consistency - I've never done that before.

FYI I'm an old cat 4; this reflects the CX season from start to finish; I was consistently in the top 25% but never on the podium; this was a fairly consistent 9 hours per week; the typical week was Sat & Sun race; Monday off; Tue longer easier ride / technique drills; Wed hard workouts with short intervals; Thurs off; Friday openers. Probably 90% of these rides are on dirt on a CX bike, and only a few are on a road bike.