Quest for "best calibrations"

Post Reply
User avatar
dtrousdale
Posts: 94
Joined: Fri Jan 11, 2008 2:11 pm
Location: Mountain View, CA

Quest for "best calibrations"

Post by dtrousdale »

I'm starting this as a new topic though I started it under "Tilt Cal R.I.P."--http://www.ibikeforum.com/viewtopic.php ... =30#p15553--since my concerns goes beyond Tilt Cal.

Today I decided to create a new profile by starting from scratch, that is, re-entering basic information via Device/Setup Device. I did this very precisely, even to the point of measuring the actual circumference of my rear tire with a tape. In addition I zeroed my DFPM (Quarq) yesterday and calibrated its slope using the Quarq utility, Qalvin, and following precisely the Quarq recommended protocol for this. BTW I do this regularly using Qalvin, and the factory calibrations have changed little since I purchased the unit more than two years ago.

As I reported in my previous post, the first problem I ran into was that the Newton+ would not exit from Calibration after my 10 minute out/back ride. This was in spite of turning it off and on several time and trying to exit from different screens. The only thing that worked was a hard reset.

I then checked the new profile in Isaac. All the details of the new profile were within the guidelines for profile ranges as posted by Velocomp in this Forum, so I figured I was good to go. I then set out to do a longish ride that was also an out-and-back (12+ miles) and then to use that with Check Calibration against my DFPM, as recommended by John.

Here are some results:
1. The needle on the Change-o-Meter was all the way to the right in the red zone.
2. Isaac informed me that my tilt calibration was invalid and that I should do another one before trying this again. Of course, I can no longer do this with the new firmware. Prior to Check Calibration, the tilt was -0.7, and afterwards it was -1.2.
3. Two other parameters recommended for change as a result of Check Calibration were outside the recommended guidelines for profile ranges: (a.) Wind Scaling--recommendation is 1.3 to 4.5, prior to Check Cal it was 2.2, and after Check Cal it was 1.259; and (b.) Friction--recommendation is 3.0 to 17, prior to Check Cal it was 12.570, and after Check Cal it was 17.742.
4. Finally I compared some statistics that would seem to indicate to me that my devices are giving me usable data:
  • a. Isaac before Check Cal with View DFPM turned off--Average Power = 282.2, IF = 1.6, TSS = 194
    b. Isaac after Check Cal with View DFPM turned off--Average Power = 167.8, IF = .83, TSS = 52
    c. Isaac after Check Cal with View DFPM turned on--Average Power = 182.6, IF = .83, TSS = 53
    d. DFPM (Quarq) recorded by Garmin 800 and viewed in WKO+--Average Power = 169, IF = .783, TSS = 63.8
Frankly these statistics seem all over the place to me. The statistics using the Quick Start procedure (a.) seem way, way, way out of line. The statistics after Check Cal look better, but a 15W difference in Average Power on a 45 minute ride seems very suspicious to me. The difference between the DFPM data as recorded and interpreted on Garmin/WKO+ versus Newton+/Isaac also seem very suspicious to me, especially for a 45 minute ride.

Can someone at Velocomp look at this and give me their interpretation? At this point, I'm highly inclined to trust my DFPM more than Newton+, especially because I have a utlity (Qalvin) that independently checks precisely to make sure the components are working properly.
Attachments
kuota_05_10_2013_0041_12_Miles_cal ride.csv
(258.18 KiB) Downloaded 322 times
User avatar
racerfern
Posts: 1356
Joined: Fri Jan 11, 2008 12:47 am
Location: Baldwin, NY
Contact:

Re: Quest for "best calibrations"

Post by racerfern »

If you put a cloth measuring tape around the tire then all bets are off. If you did a proper rollout, http://ibikeforum.com/viewtopic.php?f=2 ... 471#p15443 then you've got it right.

Please post the ibr file as the csv file doesn't have enough info for proper analysis.
Fernando
User avatar
dtrousdale
Posts: 94
Joined: Fri Jan 11, 2008 2:11 pm
Location: Mountain View, CA

Re: Quest for "best calibrations"

Post by dtrousdale »

Here you go:
kuota_05_10_2013_0041_12_Miles_cal ride.ibr
(222.51 KiB) Downloaded 391 times
User avatar
racerfern
Posts: 1356
Joined: Fri Jan 11, 2008 12:47 am
Location: Baldwin, NY
Contact:

Re: Quest for "best calibrations"

Post by racerfern »

I agree there are issues here and likely their basis is in the original cal ride. Since you did a hard reset after experiencing a lock up I suggest you do another controlled cal ride.

As far as the ride you uploaded, I think 12 miles is way too long to think about checking calibration on an out/back. There are just too many variables. Using the ride you uploaded perhaps the middle 2-6 miles would be more appropriate as a sanity check after you complete your cal ride.

I would expect you to get WS about 1.2 assuming your unit is on the stem and tilt around -.6 since we're similar in size and my CdA is just slightly lower than yours.

The initial tilt is throwing this ride way off and as I said this appears to be an issue with the original cal ride. I know you've been using the unit for quite some time but double check that something like the headset isn't loose just to make sure.
Fernando
Davis6022
Posts: 38
Joined: Mon Mar 11, 2013 10:03 am

Re: Quest for "best calibrations"

Post by Davis6022 »

If you use the Power Meter Comparison in Isaac on your ride, it looks like Newton and your DFPM are fairly close from about 8 minutes to the end of your ride. The first 4 minutes of your ride are off by a factor of about 4.2 (423 Watts average for Newton vs. 100 Watts average for the DFPM). It appears your temperature setting is either way off, or the temperature of the Newton is much different than the ambient temperature when you start your ride. I would suggest riding around for ten minutes first before using the Newton to calibrate to a DFPM.

Edit: I just looked at your CSV file in Excel. It reads 77 degrees at the start of the ride (Time Stamp 16:00:32) and it cools off to 61 degrees at Time Stamp 16:11:07 which is about 10.5 minutes into your ride.
User avatar
dtrousdale
Posts: 94
Joined: Fri Jan 11, 2008 2:11 pm
Location: Mountain View, CA

Re: Quest for "best calibrations"

Post by dtrousdale »

racerfern wrote:I agree there are issues here and likely their basis is in the original cal ride. Since you did a hard reset after experiencing a lock up I suggest you do another controlled cal ride.
As far as the ride you uploaded, I think 12 miles is way too long to think about checking calibration on an out/back. There are just too many variables. Using the ride you uploaded perhaps the middle 2-6 miles would be more appropriate as a sanity check after you complete your cal ride.
I would expect you to get WS about 1.2 assuming your unit is on the stem and tilt around -.6 since we're similar in size and my CdA is just slightly lower than yours.
The initial tilt is throwing this ride way off and as I said this appears to be an issue with the original cal ride. I know you've been using the unit for quite some time but double check that something like the headset isn't loose just to make sure.
Thanks Fernando. I always appreciate your insights and feedback.
First a response to some of your earlier comments: (1.) my Newton is on my handlebars, (2.) I measured tire circumference as you reommended and found I was off by one centimeter, and (3.) headset is not loose.
I've completed two more Quick Starts (one yesterday and one today) from scratch with each one followed by the 10-minute Cal Ride. I am happy to report that the Newton did not lock up in Calibration mode either time, and I did not get a message that my tilt was invalid. However, I repeated the Quick Start today because yesterday I noticed that the 10-minute Cal Ride made no difference in the profile in the Newton before the 10-minute Cal Ride. I repeated the Quick Start again today very, very carefully just to make sure I did not miss something, and I got the same result: the 10-minute Cal Ride made no difference in the profile. In fact, this was exactly the same profile I had after the hard re-set the other day.

I followed each of the Quick Starts by out/backs (5+ mile rides on a paved bike path away from traffic; no stops), completed a Check Calibration against my DFPM, and got similar results to what I originally reported a few days ago: the Change-o-Meter all the way to the right in the red and Wind Scaling and Friction outside the guidelines for profile ranges and Tilt right up against the maximum guideline. Here are the profiles from today's tests at the different stages of the process:
  • a. Quick Start profile before and after 10-minute Cal Ride: CdA .362, Crr .0054, Wind Scaling 2.2, Tilt -.7, Aero .796, Friction 12.570.
    b. Quick Start profile after 5+ mile out and back and after Check Calibration against my DFPM: CdA .415, Crr .0054, Wind Scaling 1.184, Tilt -1.2, Aero .491, Friction 17.642.
And here are some of the ride metrics from the 5+ mile out/back that would indicate to me whether or not the Newton is giving me usable data:
  • a. Isaac before Check Cal with View DFPM turned off using the initial Quick Start profile--Average Power = 195.1, IF = .76, TSS = 20
    b. Isaac after Check Cal with View DFPM turned off--Average Power = 137, IF = .54, TSS = 10
    c. Isaac after Check Cal with View DFPM turned on--Average Power = 138.7, IF = .54, TSS = 10
    d. DFPM (Quarq) recorded by Garmin 800 and viewed in WKO+--Average Power = 141, IF = .549, TSS = 10.2
So this seems a little better to me, but I still have concerns:
1. Why was there no change in the Quick Start profile after the 10-minute Cal Ride? Given the radical changes as a result of my subsequent out/back and the Check Calibration, this would make me very leery about trusting a Quick Start profile with one of my non-DFPM bikes. Also the metrics of Average Power, IF, and TSS seem wildly inaccurate when using just the Quick Start profile.
2. Are the facts that the revised Wind Scaling and Friction values are outside the guidelines, and Tilt up against the maximum, for profile ranges an issue?
3. Why are the DFPM metrics different when recorded by Newton and viewed in Isaac as compared to those recorded by my Garmin and viewed in WKO+? The DFPM transmits exactly the same data, doesn't it? I am most concerned about IF and TSS, the calculations of which I would expect to be almost exactly the same. I have noticed differences like these in DFPM recording in the past between Newton/Isaac and Garmin/WKO+, and sometimes they are quite significant. The out/back ride here was short and easy (5.47 miles, 21 minutes, no hills, little wind, etc.) but I'm still expecting what seem like small differences here to sometimes be very big when I do a long, hard ride.
Attachments
kuota_05_12_2013_0754_5_Miles_cal ride.ibr
(88.85 KiB) Downloaded 301 times
User avatar
racerfern
Posts: 1356
Joined: Fri Jan 11, 2008 12:47 am
Location: Baldwin, NY
Contact:

Re: Quest for "best calibrations"

Post by racerfern »

Hmm, the out/back ride you sent was not the original as recorded and it isn't a "real" cal ride. The recorded ride had an original WS of 2.2. After check calibration it drops to 1.186 and correctly so. I'm not sure why the numbers skew like they do.

So I'm wondering where the WS of 2.2 came from, not from the cal ride just prior to this ride?

Did you actually do a cal ride or just the out/back and called it a cal ride? I'm not pointing fingers, just asking questions and trying to understand.
Fernando
User avatar
dtrousdale
Posts: 94
Joined: Fri Jan 11, 2008 2:11 pm
Location: Mountain View, CA

Re: Quest for "best calibrations"

Post by dtrousdale »

racerfern wrote:Hmm, the out/back ride you sent was not the original as recorded and it isn't a "real" cal ride. The recorded ride had an original WS of 2.2. After check calibration it drops to 1.186 and correctly so. I'm not sure why the numbers skew like they do.

So I'm wondering where the WS of 2.2 came from, not from the cal ride just prior to this ride?
The WS of 2.2 came from the original profile created by Device/Setup Device and before the 10-minute Cal Ride. I think I figured out what I'm doing wrong. See my latest post here: http://www.ibikeforum.com/viewtopic.php?f=55&t=2801
racerfern wrote:Did you actually do a cal ride or just the out/back and called it a cal ride? I'm not pointing fingers, just asking questions and trying to understand.
Sorry if I'm being confusing. Thank you for your patience. I misnamed the file. This ride was just an out/back that I could run Check Configuration against. This is what John advised me to do to get the best calibrations:
If you have a properly calibrated DFPM (recently serviced, and zero offset done properly) then the easiest way is to fine-tune your Newton is with your DFPM. I would do a 5 mile long, out-and-back ride. Then, I would use the Check Calibration feature.
Davis6022
Posts: 38
Joined: Mon Mar 11, 2013 10:03 am

Re: Quest for "best calibrations"

Post by Davis6022 »

Why are the DFPM metrics different when recorded by Newton and viewed in Isaac as compared to those recorded by my Garmin and viewed in WKO+? The DFPM transmits exactly the same data, doesn't it? I am most concerned about IF and TSS, the calculations of which I would expect to be almost exactly the same. I have noticed differences like these in DFPM recording in the past between Newton/Isaac and Garmin/WKO+, and sometimes they are quite significant. The out/back ride here was short and easy (5.47 miles, 21 minutes, no hills, little wind, etc.) but I'm still expecting what seem like small differences here to sometimes be very big when I do a long, hard ride.
Is your FTP set the to same value in both your Newton and the Garmin?
User avatar
dtrousdale
Posts: 94
Joined: Fri Jan 11, 2008 2:11 pm
Location: Mountain View, CA

Re: Quest for "best calibrations"

Post by dtrousdale »

OK. I went through the whole process once again, and I believe I did everything right this time and under close to ideal conditions. The only thing that looks off to me is the Wind Scaling number. After entering my personal data using Device/Setup Device and completing the Cal Ride, WS was 1.139. After my 5.45 mile out-and-back and after Check Calibration against my DFPM, WS was changed to 1.136. Both of these numbers are below the guidelines for profile ranges. Is this a problem?
kuota_05_13_2013_0737_5_Miles_out&back.ibr
(98.16 KiB) Downloaded 310 times
User avatar
racerfern
Posts: 1356
Joined: Fri Jan 11, 2008 12:47 am
Location: Baldwin, NY
Contact:

Re: Quest for "best calibrations"

Post by racerfern »

Your profile looks spot on with only the slightest of differences between the before and after.

I think I remember you said the Newton was on the handlebars so 1.1xx is low for a handlebar position but certainly believable . For a stem position it's spot on.

You need to be careful with your settings as you have the tire size set to 26 x 2.35. If so then your Crr is quite a bit higher. The devil is in the details. :)
Fernando
User avatar
dtrousdale
Posts: 94
Joined: Fri Jan 11, 2008 2:11 pm
Location: Mountain View, CA

Re: Quest for "best calibrations"

Post by dtrousdale »

racerfern wrote:You need to be careful with your settings as you have the tire size set to 26 x 2.35. If so then your Crr is quite a bit higher. The devil is in the details. :)
Thanks. I measured tire size precisely as you recommended in an earlier post. And it was off by one centimeter from the measurement I got using a cloth tape measure.
User avatar
racerfern
Posts: 1356
Joined: Fri Jan 11, 2008 12:47 am
Location: Baldwin, NY
Contact:

Re: Quest for "best calibrations"

Post by racerfern »

I understand that but then you should just enter the result. I don't understand where you got 26 x 2.35 because that will affect the speed and distance.
Fernando
Post Reply