Shimano R9100-P Powermeter

Discuss light weight issues concerning road bikes & parts.
User avatar
Conza
Posts: 96
Joined: Sat Oct 07, 2017 3:28 am
Location: Melbourne, Victoria, Australia
Contact:

by Conza

mds wrote:
pantelones wrote:Have you filed a report to SRM?

The following forum is related to SRM PC8 and associated firmware development.
http://www.srmservice.com/forum/viewforum.php?f=2

Yes. SRM says

"If the PC8 would cause this issue, we would have Power Spikes with all systems transmitting read calculated Power. ... Its more likely that something is rolling over in the DA PowerMeter at a continuous interval, but I could only tell for sure if I could examine the ANT+ Data that is transmitted by the PowerMeter. The PC8 does not do any filtering of the received data, because its ready calculated and should be correct."

Shimano asked me to update R9100-P firmware, which I did, to no improvement.

I am still investigating and soon will be trying different equipment parings (eg PC8/power2Max, Wahoo Bolt/R9100-P, E-TUBE app/R9100-P).


Thanks. Keep up the good work :beerchug:
It's all about the adventure :o .

MoreRideTime
Posts: 118
Joined: Sun Sep 17, 2006 9:09 pm
Location: Canada

by MoreRideTime

Another thing to try is the Quark Qollector (https://www.quarq.com/product/quarq-qollector/) if you can get your hands on one.

by Weenie


bas
Posts: 35
Joined: Thu Jun 30, 2016 3:58 am

by bas

mds wrote:
bas wrote:I find power data does not upload to Strava from device agent/PC8 correctly. If i have done intervals the lap power on strava is always way different to actual recorded. Don't use Strava for anything other than the odd bunch ride upload so doesn't matter but interesting to note

Regarding Strava, PC8 GPS sometimes has trouble with accuracy, causing Strava to miss segments occasionally.

Most of the time it is OK.

But here is an example where PC8 GPS is not good. First map is iPhone 7 Plus, second PC8. With PC8, Strava missed segments along this ~5km canyon climb.


Its not the power data of Strava segments that is off, but the interval power data recorded. For example if i did a 5min effort at 300W for example, the lap data on Strava would show 180W or some other number way lower than actual.
As i said i don't particularly care its just unusual that the data doesn't feed through correctly

mds
Posts: 192
Joined: Sat Dec 13, 2008 6:28 pm
Location: California

by mds

kulivontot wrote:Get an ant+ usb stick off Amazon (i bought the knockoff docooler one, its the same chipset and drivsr), plug into a pc and run golden cheetah, trainer road or similar utility to capture the raw power data. There should be no filtering applied there.

Here is a chart comparing ANT+ USB/TrainerRoad and SRM PC8 with my R9100-P. The spikes do not appear in the former, but I don't think I can conclude anything because the data does in fact appear to be filtered. SRM tells me PC8 does not filter, but I am skeptical. I asked TrainerRoad on what filtering they apply, no info yet.

Charts for heart rate and cadence are similar but not identical. Not sure what to make of this.

Update: TrainerRoad replies that they record the exact data sent by the power meter, they don't modify it, filter it, or affect it at all. This is basically what SRM told me regarding the PC8. Assuming both are true, maybe my PC8 unit is defective and is not working as a normal PC8 should. My next test is to pair my PC8 with a power2max Type S.
Attachments
Test.png
HeartRate.png
Cadence.png
2009 Look 595

User avatar
Conza
Posts: 96
Joined: Sat Oct 07, 2017 3:28 am
Location: Melbourne, Victoria, Australia
Contact:

by Conza

mds wrote:
kulivontot wrote:Get an ant+ usb stick off Amazon (i bought the knockoff docooler one, its the same chipset and drivsr), plug into a pc and run golden cheetah, trainer road or similar utility to capture the raw power data. There should be no filtering applied there.

Here is a chart comparing ANT+ USB/TrainerRoad and SRM PC8 with my R9100-P. The spikes do not appear in the former, but I don't think I can conclude anything because the data does in fact appear to be filtered. SRM tells me PC8 does not filter, but I am skeptical. I asked TrainerRoad on what filtering they apply, no info yet.

Charts for heart rate and cadence are similar but not identical. Not sure what to make of this.

Update: TrainerRoad replies that they record the exact data sent by the power meter, they don't modify it, filter it, or affect it at all. This is basically what SRM told me regarding the PC8. Assuming both are true, maybe my PC8 unit is defective and is not working as a normal PC8 should. My next test is to pair my PC8 with a power2max Type S.


So from everything you've learned so far; if I was to get the DA power metre - using a garmin 520, and uploading to Strava - I shouldn't have any issues?

None with the spikes? :shock:
It's all about the adventure :o .

mds
Posts: 192
Joined: Sat Dec 13, 2008 6:28 pm
Location: California

by mds

Conza wrote:So from everything you've learned so far; if I was to get the DA power metre - using a garmin 520, and uploading to Strava - I shouldn't have any issues?

None with the spikes? :shock:

Likely OK, Garmin 500 shows no spikes. So far only pairing that shows spikes is SRM/Shimano.

I just tried SRM/power2max. No spikes.

:noidea:

Update: Just tried 510 - no spikes.

Tests so far:

Garmin 500/Shimano - no spikes
Garmin 510/Shimano - no spikes
TrainerRoad/Shimano - no spikes
E-Tube/Shimano - no spikes
SRM/Shimano - spikes
SRM/power2max - no spikes
2009 Look 595

User avatar
charmette1261
Posts: 23
Joined: Tue Jan 02, 2007 11:36 am
Location: France

by charmette1261

This powermeter 9100 founction with my Suunto spartan Bluetooth capability?

mds
Posts: 192
Joined: Sat Dec 13, 2008 6:28 pm
Location: California

by mds

charmette1261 wrote:This powermeter 9100 founction with my Suunto spartan Bluetooth capability?

No.

Bluetooth LE only with E-Tube Project for smartphones/tablets. Currently use for firmware update and testing.
2009 Look 595

User avatar
charmette1261
Posts: 23
Joined: Tue Jan 02, 2007 11:36 am
Location: France

by charmette1261

mds wrote:
charmette1261 wrote:This powermeter 9100 founction with my Suunto spartan Bluetooth capability?

No.

Bluetooth LE only with E-Tube Project for smartphones/tablets. Currently use for firmware update and testing.


Thanks ! Smartphone App compatible can replace an bike computer ?

mds
Posts: 192
Joined: Sat Dec 13, 2008 6:28 pm
Location: California

by mds

charmette1261 wrote:Thanks ! Smartphone App compatible can replace an bike computer ?

The app can display power and cadence for short tests, see photo. The app/BLE may or may not work for longer rides, I am not sure. I see no option for logging data.
Attachments
E-Tube.jpg
2009 Look 595

User avatar
Conza
Posts: 96
Joined: Sat Oct 07, 2017 3:28 am
Location: Melbourne, Victoria, Australia
Contact:

by Conza

mds wrote:
Conza wrote:So from everything you've learned so far; if I was to get the DA power metre - using a garmin 520, and uploading to Strava - I shouldn't have any issues?

None with the spikes? :shock:

Likely OK, Garmin 500 shows no spikes. So far only pairing that shows spikes is SRM/Shimano.

I just tried SRM/power2max. No spikes.

:noidea:

Update: Just tried 510 - no spikes.

Tests so far:

Garmin 500/Shimano - no spikes
Garmin 510/Shimano - no spikes
TrainerRoad/Shimano - no spikes
E-Tube/Shimano - no spikes
SRM/Shimano - spikes
SRM/power2max - no spikes


:beerchug:
It's all about the adventure :o .

nemeseri
Posts: 619
Joined: Tue Apr 07, 2015 5:40 pm

by nemeseri

mds wrote:
Conza wrote:So from everything you've learned so far; if I was to get the DA power metre - using a garmin 520, and uploading to Strava - I shouldn't have any issues?

None with the spikes? :shock:

Likely OK, Garmin 500 shows no spikes. So far only pairing that shows spikes is SRM/Shimano.

I just tried SRM/power2max. No spikes.

:noidea:

Update: Just tried 510 - no spikes.

Tests so far:

Garmin 500/Shimano - no spikes
Garmin 510/Shimano - no spikes
TrainerRoad/Shimano - no spikes
E-Tube/Shimano - no spikes
SRM/Shimano - spikes
SRM/power2max - no spikes


Thanks for the details and time you put into this. This is very helpful!
One question: did you set the garmins to 1 second recording instead of the default "smart" setting? That might influence the recorded data.

mds
Posts: 192
Joined: Sat Dec 13, 2008 6:28 pm
Location: California

by mds

nemeseri wrote:Thanks for the details and time you put into this. This is very helpful!
One question: did you set the garmins to 1 second recording instead of the default "smart" setting? That might influence the recorded data.

Thank you nemeseri. Yes, the Garmins were set to the 1 second recording option.

I also see no evidence of spikes when looking at the Garmin displays. On the other hand, on the SRM display, I do see them.

I like to use a 30s power average on my SRM display, and am worried that the spikes will affect the value. I calculate that if the 30s average included one spike (true ~50% of the time), then the average would be inflated by about ~5 watts at my FTP. I can't claim I can actually see this happening, but it worries me anyway.

I am running out of ideas to help diagnosis this.
2009 Look 595

kwakekeham
Posts: 17
Joined: Fri Jul 22, 2016 8:03 pm

by kwakekeham

That breakdown is interesting.

There are 4 power pages in ANT+. Basic (everyone, mandatory), Crank Torque (cranks, arms, pedals), Wheel Torque (Powertap hub only) and Crank Torque Frequency (SRM only). Everyone must have at a minimum basic pages, and almost all use 1 of the 3 others with every 4th page is basic to maintain backwards compatibility with simpler head units -- least that was the idea. Basic pages are also the only place balance is found. It's a violation of ANT+ spec to include multiple option pages and it wouldn't have certified. So 1 + basic.

So based on what has been shown so far I would guess that the Shimano uses Crank Torque pages. Since SRM only has to deal with Crank Torque Frequency in their own products, when connecting another meter it's likely dropping back to Basic pages. I don't have a PC8 so can't check.

The possibility is that either the Shimano PM is not encoding basic pages properly or the SRM head unit is not decoding them correctly.

Also, when testing, what a Garmin records is NOT what it displays. They have different requirements. The specific example the Garmin Edge engineer used at the ANT+ symposium a few years ago when I was there was to imagine going down a hill and you stop pedaling, when you look down you want to see what your cadence was not zero -- so it holds it for a few seconds even though it knows the data is stagnant and not changing. What is displayed in all Garmin head units is filtered, what is recorded on the newer (510/810 and newer) follows good guidelines for proper recording.

These spikes should be easy to track down using the ANT+ development software because it can decode both pages unlike a head unit. My guess is the PM and not the SRM head unit. Likely escaped because nobody saw it in crank torque pages.

by Weenie


PinaF8
Posts: 172
Joined: Mon Jan 23, 2017 8:05 am

by PinaF8

The first is my PC8 paired to Pioneer dual leg power meter.

The second is my PC8 paired to SRM power meter.

Image
Image

Definitely the spikes are being caused by non SRM power meter.
S-WORKS VENGE 1ST GEN
PINARELLO F8
PINARELLO F10
2018 BMC SLR01

Post Reply
  • Similar Topics
    Replies
    Views
    Last post