Zeroing a Powertap - how often?

doyler78

Manual says to do it often but then says to do it if watts are not zero while coasting. I've never zeroed my powertap in the 4 months that I've had it as it always goes to zero when I coast. Is there any reason why I should do it more often and if so how often?

Just Alex's blog on Sins of Sins #2 got me wondering.

http://alex-cycle.blogspot.com/2009/07/ ... ftp-2.html

My data seems consistent so that hasn't given me any rise for concern but that's perhaps because any problems aren't dramatic enough to be noticeable but important enough to be significant.

deal

you will likely find that the computer is using the autozero function by default (zeros the powertap when you are coasting)

Alex_Simmons/RST

A Powertap has a torque autozero function, so for the most part people get away with not zeroing their meter (providing the auto zero function has been enabled). But it doesn't always do the trick, so it is a very good habit to get into to at least do a manual zero before every ride and sometimes doing a check during a ride as well.

It only takes a few seconds and having confidence about collecting good data is worth that minor investment of time.

There are a number of reasons for doing it:

- if the torque zero is out beyond a certain value (I forget exactly but > 8ft-lbs I think but it might be another value) , the auto torque zero function doesn't work

- negative torque zero values aren't displayed, only shown as zero, so seeing zero doesn't necessarily mean zero, just not positive

- if you have your meter in rapidly changing temperatures, it can impact the zero offset. e.g taking off from roof of car, from warm house to cold outside etc

- if you use the same CPU for different hubs/meters, then the torque zero can be significantly different for each

My understanding is the zeroing method has changed in newer models of Powertap (for the worse IMO).

doyler78

Ok. Just checked I have auto-zero on yes so it is auto-zeroing.

Thanks Alex for the reason's why not zeroing may allow compromised data. I'm in the habit at the minute of clearing my powertap data at the start of my first ride of the day so I will simply introduce the zeroing procedure to that. It only takes a couple of seconds to do as you say.

Zeroing on my model is done off the bike however I believe in previous versions it was done on the bike whilst coasting. Not sure why that would be important however I do remember seeing a thread on wattage about it long before I even got the powertap so it went over me but I will go have a look for that too just out of curiosity.

nmcgann

I zero every ride (even with auto-zero enabled).

I didn't do it as often until I got a second hub for my road bike as well as the turbo/TT bike. That hub has a very different zero torque reading (It normally shows 20 or so before manual zeroing) and I didn't know how big an offset the auto function can remove, so I don't risk it. It takes seconds to clear so it's just a habit now.

Neil

doyler78

Managed to drop the headunit face first onto a slate floor this morning whilst doing the zero. Display now only showing the bottom screen options so that now needs to be fixed/replaced :evil: I'm such an idiot. If I'd just stuck to working with in the mount instead of in my hands it wouldn't have happened

Alex_Simmons/RST

doyler78 wrote:

Managed to drop the headunit face first onto a slate floor this morning whilst doing the zero. Display now only showing the bottom screen options so that now needs to be fixed/replaced :evil: I'm such an idiot. If I'd just stuck to working with in the mount instead of in my hands it wouldn't have happened

Bugger.

Alex_Simmons/RST

doyler78 wrote:

Zeroing on my model is done off the bike however I believe in previous versions it was done on the bike whilst coasting. Not sure why that would be important however I do remember seeing a thread on wattage about it long before I even got the powertap so it went over me but I will go have a look for that too just out of curiosity.

Well in the older models, torque zero would kick in when wheel speed is non zero but cadence was zero. That always guarantees that torque at the hub is actually zero.

Of course you could also do a manual zero when wheels weren't turning, but the auto zero function wouldn't operate in that situation.

However now it requires the wheel speed to be zero. That's dumb because you can still apply torque even though the wheel ain't turning. What happens when you track stand, or simply have your foot resting on the pedal when stopped at the lights?

As I said, IMO it is a big step backwards and I really don't understand why they would have changed it and significantly increase the chances of invalid data occuring.

But perhaps I've got it wrong and someone can correct me.

doyler78

Alex_Simmons/RST wrote:

doyler78 wrote:

Managed to drop the headunit face first onto a slate floor this morning whilst doing the zero. Display now only showing the bottom screen options so that now needs to be fixed/replaced :evil: I'm such an idiot. If I'd just stuck to working with in the mount instead of in my hands it wouldn't have happened

Bugger.

That's the mild version. My French has never been so fluent as it was this morning :shock:

And I never got as far as actually getting it zeroed though I'm pretty certain it is still recording so hopefully only a screen replacement will be required as a full replacement will be very costly.

Jeff Jones

I zero mine each time I ride, when coasting. But when I hold down the select(?) button when Watts is flashing, it doesn't stop flashing. I have to press it normally to get it to stop flashing and go back to watts mode.

I still think it's reading right although my power is lower than last year yet I'm going faster. I'm hoping there's nothing wrong with it (would low hub batteries lower your power?) because that means I've increased speed via aero improvements + good conditions, so there's potentially more to come

To cross check, my best time this year for my 39km hilly TT training loop is 1:04, whereas last year in July-August I was knocking out 1:02 at similar HR. So that would point to a 20-30W decrease in power that I hope to regain. Soon!

One of these days I'll actually start training with the power meter.

Alex_Simmons/RST

Jeff Jones wrote:

I still think it's reading right although my power is lower than last year yet I'm going faster. I'm hoping there's nothing wrong with it (would low hub batteries lower your power?) because that means I've increased speed via aero improvements + good conditions, so there's potentially more to come

Low batteries tend to show up as weird and wonderful symptoms but I'm not sure if reporting a drop in power is one of them.

How about doing a "stomp test" to check the PT's calibration?
http://www.midweekclub.ca/powerFAQ.htm#Q23

Ric & I have reported some PT Zipp Disks of late that are well under spec and reporting under power by ~ 10%.

Jeff Jones wrote:

One of these days I'll actually start training with the power meter.

How about a training podcast? Might be easier than writing articles LOL. Since you're now the podcast king.

Jeff Jones

Alex_Simmons/RST wrote:

How about doing a "stomp test" to check the PT's calibration?
http://www.midweekclub.ca/powerFAQ.htm#Q23

Ric & I have reported some PT Zipp Disks of late that are well under spec and reporting under power by ~ 10%.

I'll try that, ta. 10% is huge!

How about a training podcast? Might be easier than writing articles LOL. Since you're now the podcast king.

Good idea

doyler78

Is there any reason why the weight for the stomp test has to be above 50lb because the biggest weight I could find was only 29.875lb and with 172.5mm cranks and the gear 34x14 that meant that the measured torque would be 83.5in-lbs whereas the powertap measured 80 (hovered in and out of 79/80 and then settled on 80). That means, if correct that my powertap is 4.2% under recording my watts if my calcs are right.

Is this significant other than to know that my watts are actually 4.2% higher or does it indicate a bigger problem with the powertap and can I rely on the result given the low weight used?

Alex_Simmons/RST

You calcs are right provided the inputs are correct.

If your unit is really out by that much, something is wrong with it as that is significantly out of spec (+/- 1.5%).

At 300W the Average Effective Pedal Force is ~ 20kg (depending on cadence) and the peak forces are about double that. Hence why many use about that mass or more for doing a calibration test. Where possible, using different (known) masses is also a good idea.

Make sure you repeat the test on a few gear ratios to see if the variance is consistent, and also make sure the unit is zero'd beforehand.


I am thinking about seeing if I can get some special pedal spindles made for PM calibration. Imagine a speedplay zero pedal without the pedal - just a horizontal axle that you can screw into the crank liek a normal pedal on which you can easily place some typical gym weights, which are easy and cheap to obtain and a quick trip to the post office to have them weighed will give you the mass to sufficient accuracy.

doyler78

Alex thanks for the tips. Will go and a do a bit of checking.

scherrit

These are all road/freewheeling PT's right? The Track adapted old PT's need re-zeroing as often as possible, I tend to re-zero after each segment of training at the track. It is a bit of a marginal application, I keep breaking them....
Cheers,
Scherrit.