Help with settings

[danelmanperry]

Hello John,

Can you please look at a ride file or two and let me know if there are any changes I can make to improve the accuracy?
I am trying to resolve two trends I notice:

1) First 8 minutes of each ride show huge power numbers leading me to think the PP is recalibrating. This would make sense if the position had changed, however I notice it on back to back days when I rode my bike from my front door. No car travel, stored flat on two wheels leaned against a wall, not washed, PP not removed or hand movable in its mount. I don't even charge it after every ride (run a cord to a wall plug, not into PC).
2) Huge power spikes on a downhill. This seems to happen when I resume pedaling after coasting downhill, such as the transition in gradient at the base of a hill. My most recent ride, which I will attach, shows a 2015w spike! I wish

Thanks John!

-Daniel

[Velocomp]

1) If your PP is angled at all differently from ride to ride, then first 8 minutes the PP will correct its internal calibration. Follow the directions carefully to make sure you install your PP at the same angle.

2) The easiest thing to do is ignore the power spikes--as you say, they are artifacts. If you do a lot of hill climbing I would add a cadence sensor to your bike; this will reduce spurious spikes.

[danelmanperry]

Thanks for the quick reply John!

1) What I'm getting at is that the PP is not moved, at all. I cannot move it by hand in the supplied mount after following the directions. The bike is parked in the same position it was ridden and leaves the house for the next ride in the same way, however I get an 8 minute calibration each time. This doesn't seem correct so I thought maybe I have an inaccurate setting somehow which the PP is trying to compensate for on each ride?

2) I will search for the instructions to connect my cadence sensor to the PP; thanks for the suggestion. I don't notice the spikes on-screen so they are easy to ignore in that way; it's more of an issue that my training software picks up on them from the uploaded ride file. This affects TSS tracking and it's hard to get an accurate graph of 10, 30, 60 second power over the last 6 months when there are superman spikes every other ride

-Daniel

[Velocomp]

Are you removing PP to download rides?

[danelmanperry]

A couple of times per month, yes.

[Velocomp]

I suspect your problem happens after you download your rides.

When you reattach PP to your mount, unless it is in exactly the same position, you'll get potential problems with watts for the first 8 minutes, while PP is recalibrating itself.

If you leave PP attached between rides, then you should not see this problem, as long as PP is attached so that it cannot rotate in its mount.

[danelmanperry]

That makes perfect sense.

That is not what I am experiencing though. I'll attach back to back ride files to this thread next time I have them (switching bikes today so of course I will have 8 min calibration).
I'll also read the directions again; and post back even if it works properly and I'm crazy

Thanks for your time John,
-Daniel

[Velocomp]

OK, let us know what you find.

[danelmanperry]

Looking good!
I've completed two rides with no 8-minute recalibration.

In the process of switching bikes, I played with settings in Isaac. After doing that the PP would not connect to my Garmin. Then while trying to get it to connect I must have held the button too long and I think it tried to do a calibration ride because the light would flash green/red constantly. I tried this on three rides and gave up; reset to factory defaults.
After resetting I was able to connect to my Garmin again. I also made sure the cadence sensor paired at the same time as the speed sensor.

With a new out and back calibration ride including cadence sensor, I have not seen the 8-minute recalibration at the beginning of rides, I have not seen huge power spikes going downhill, and the power readout is WAY lower; like 50-100w lower. This is more in line with reality and in the ballpark of the people I can keep up with.

So I think the answer to my original question is yes, something was wrong to cause the 8-min recal every ride, but we'll never know what, and it is solved. The device was attached properly; rotated all the way forward, couldn't move it without forcing it by hand, and the allen bolt so tight I was worried I'd strip the plastic.

Thanks again for your help John! The cadence sensor seems to make a huge difference!

-Daniel

[Velocomp]

[BBQcyclist]

Just a random thought ... Perhaps sometimes when riding, following a HARD hit into pothole, the PP may not move, but the handlebars may move/rotate ... depending on where the hands/weight of the rider were/was at the time..

Essentially, the PP will then still be "rotated" in a different angle, hence potentially causing temporary power sampling glitzes.

Does this make sense John?

[Velocomp]

Yes.