Auto tilt adjustment

Post Reply
GWPOS
Posts: 178
Joined: Mon Aug 24, 2009 2:47 pm
Location: Yucaipa, Ca

Auto tilt adjustment

Post by GWPOS »

I have posted on this before but wanted to make sure before asking the question again. Ok now that I have my unit calibrated I have noticed that after climbing a long 4-9% grade then turning around my units watts are off for 5 minutes until it recalibrates the tilt...again. While coasting without pedaling or softpedaling I get watts between 50-100 at around 25-31 mph. After that first five minutes of going downhill the unit will then recal. the tilt and all seems fine until I get to the bottom and go to climb again. My tires are properly inflated and the unit is mounted rock solid. Is anyone else encountering this?

Is there anyway to turn off the auto recal. for the tilt after it does it the first time 5 min into the ride? I know it keeps adjusting as we ride every 5 min, just wandering if it has to or am I just going to have to deal with these numbers.

When doing intervals that require staying at a certain power number it is a little bit of a problem when swithching from climbing to descending. Thanks for any in-put.
Clint
rruff
Posts: 445
Joined: Wed Apr 23, 2008 10:48 am

Re: Auto tilt adjustment

Post by rruff »

IME a 50-100W error on downhills isn't bad. Gravity is contributing the equivalent of 500W or so at 30mph, and the change in weight distribution is going to cause a tilt error. With the Gen2 I noticed ~+.1% on uphills and -.1% on the way down.

The problem I encountered with hill intervals was that the altitude readings at the bottom and top of the hill varied quite a bit... enough to result in a variance of ~5% in the power readings on the climbs. On another forum someone speculated that the altitude sensor takes some time to adjust to changes, and if so the iBike might have been missing the altitude of the top and bottom... ie it thought I was climbing less than I was. I wish there was a way to turn it off also.
User avatar
racerfern
Posts: 1356
Joined: Fri Jan 11, 2008 12:47 am
Location: Baldwin, NY
Contact:

Re: Auto tilt adjustment

Post by racerfern »

GWPOS wrote:I have posted on this before but wanted to make sure before asking the question again. Ok now that I have my unit calibrated I have noticed that after climbing a long 4-9% grade then turning around my units watts are off for 5 minutes until it recalibrates the tilt...again. While coasting without pedaling or softpedaling I get watts between 50-100 at around 25-31 mph. After that first five minutes of going downhill the unit will then recal. the tilt and all seems fine until I get to the bottom and go to climb again. My tires are properly inflated and the unit is mounted rock solid. Is anyone else encountering this?

Is there anyway to turn off the auto recal. for the tilt after it does it the first time 5 min into the ride? I know it keeps adjusting as we ride every 5 min, just wandering if it has to or am I just going to have to deal with these numbers.
This has been discussed before but we never heard an official answer. Does the iBike recalculate every 5 minutes? I don't think so. I think the iBike constantly calculates tilt based on the accelerometer readings after the initial 5 minutes of a ride. I don't think it takes 5 minutes after a big change like turning around on a hill interval to re-calculate as the unit would constantly be way behind. The key is for the unit to be current, not 5 minutes behind.

If you're getting 50-100w while soft pedaling going downhill and assuming your wind calibration is correct, then make a slight adjustment in your aero number by lowering it slightly. That will make you more aero (producing fewer watts). It's very sensitive for adjustments like this. Try soft pedaling all the way down the hill then download the ride. Of course you'll see watts but try making a copy of your profile and making the slight adjustment there. You'll get your answer pretty quickly by switching profiles.
Fernando
rruff
Posts: 445
Joined: Wed Apr 23, 2008 10:48 am

Re: Auto tilt adjustment

Post by rruff »

The iBike is always calculating tilt based on accelerometer and speedometer readings... but the issue we are discussing is something else. At 5min intervals it calculates the change in elevation based on the tilt readings and compares it to the altimeter. Then it makes an adjustment to the "zero" tilt so they match. If the altimeter readings are off, then it will make a bad adjustment. Also, since it is always backward-looking, it assumes that the future 5 minutes will be like the last 5 minutes. On a series of uphills and downhills, this won't be the case.
GWPOS
Posts: 178
Joined: Mon Aug 24, 2009 2:47 pm
Location: Yucaipa, Ca

Re: Auto tilt adjustment

Post by GWPOS »

I don't understand exactly how all of this works as far as the tilt recalculating during the ride. I understand my weight does shift slightly but not much depending if I am climbing or on the flats or even descending. I just wish this tilt auto recal. would do its thing the first time and then stop doing it for the rest of the ride. I know its not an aero thing that needs adjustment because watts read zero when coasting exactly 5 min after I turn around and go back downhill. It is only the first 5 min after turning around that I get these "ghost watts".
User avatar
racerfern
Posts: 1356
Joined: Fri Jan 11, 2008 12:47 am
Location: Baldwin, NY
Contact:

Re: Auto tilt adjustment

Post by racerfern »

Um, that's a new one on me. I guess you learn something everyday so now I can take the rest of the day off.

Seriously, having never experienced this (or paid attention to it) I have a couple of questions.

Would this be a problem only when the altimeter is thrown off by barometric pressure changes? I do 5 minute hill repeats but the elevation graph happens to stay pretty accurate so there may not be a problem. However, sometimes the trip down is only 4 minutes, what happens then? Occasionally the barometric changes give me a series of ever increasing hills and valleys or ever decreasing hills and valleys if there are barometric pressure changes at that time of day. I wonder how critical this is?

Wouldn't this also affect readings on a flat when there are significant barometric changes? Why would it be only on hills? If I'm riding along the coast and the altimeter drops from my start of +30' to -100' because of weather changes the same problem would manifest itself. This is very common for me and I at times will adjust the altimeter during a ride only because I realize the weather changes are affecting the readings.

So does it make sense to change your altimeter to the current altitude if known, just like a pilot would?
Fernando
User avatar
racerfern
Posts: 1356
Joined: Fri Jan 11, 2008 12:47 am
Location: Baldwin, NY
Contact:

Re: Auto tilt adjustment

Post by racerfern »

The problem I encountered with hill intervals was that the altitude readings at the bottom and top of the hill varied quite a bit... enough to result in a variance of ~5% in the power readings on the climbs. On another forum someone speculated that the altitude sensor takes some time to adjust to changes, and if so the iBike might have been missing the altitude of the top and bottom... ie it thought I was climbing less than I was. I wish there was a way to turn it off also.
One of my regular rides is an area with 4 sets of rollers from about sealevel to about 125 and back down. If I switch to the elevation screen I can clearly see the elevation updates lag behind the actual elevation. More than likely this happens since I'm going a good 30+mph and the changes happen quicker than the unit can update or calculate. On the uphill, the updates happen quickly enough because I'm generally going under 10mph. This can be evidenced by clicking a lap at the low spot then when downloading the data it appears you kept on going down a little further as the unit kept updating. I don't see this delay in the tilt measurements while looking at both on the enviro screen.

By coincidence when I do my hill intervals I usually arrive at the bottom of the hill about 15-30 seconds before its time to launch again. That's more than enough time for the unit elevation to settle in so I may not be seeing the elevation issue others are seeing.
Fernando
travispape
Site Admin
Posts: 392
Joined: Fri Jan 11, 2008 8:52 am
Location: Durham, NC

Re: Auto tilt adjustment

Post by travispape »

Hey guys,

John asked me to check in on this thread.

First of all, let me mention what this on-the-road tilt correction algorithm is for. (I'm going to call it the slope_offset algorithm since that is what I'm used to calling it.) What we found with the initial generations of iBike is that things happen while riding to cause the iBike to tilt nose-up or nose-down during the course of a ride. Since gravity is such an important component of the power balance sheet, these events could lead to errors and in the initial generation of the product there was no way to know that it was happening.

The thing to remember is that depending on many factors, the actual nose-up or -down tilt of the iBike is subject to changing mid-ride. To name some of the candidates that come to mind: frame flex, fork flex, head tube flex, play in the headset bearings, stem/headset interface, stem flex, stem/handlebar interface, handlebar flex, interface of the iBike mount, unequal tire loading (downhill vs. up), rider position & weight distribution, rider applying torque to handlebars or aerobars (can be a large effect depending on mount position). Has anyone else seen the picture showing how much the back wheel can flex when a rider is climbing out of the saddle? I couldn't find it just now.

The slope_offset algorithm takes advantage of the fact that we have an independent measurement of slope in the altimeter data. In essence, the tilt measured by the accelerometer system is "tied to reality" based on the altimeter data. As the running average of the accelerometer-based measurement deviates from the running average of the altimeter-based measurement, a bias is applied to the accelerometer-based measurement to bring it back into line with what it has to be according to the independent data.

A side benefit of the slope_offset algorithm is that if there is any error in the tilt calibration (performed in the unit's setup menu) or in the profile's riding_tilt, the slope_offset algorithm will "save the day" and repair the data after the first 5 minutes.
GWPOS wrote:While coasting without pedaling or softpedaling I get watts between 50-100 at around 25-31 mph.
and
rruff wrote:IME a 50-100W error on downhills isn't bad. Gravity is contributing the equivalent of 500W or so at 30mph…
Ron is making an important point. The power you see on such a descent is a difference between very large numbers. Coasting down hill you might have 1000 W of gravity opposing 800 W of air loss and 200 W of rolling loss. Don't worry about the numbers, the point is that if you soft-pedal in order to see the power number, what you see will be heavily influenced by your ride position at such large speeds. If you tuck even a little more than you did when you did your calibration, it will be magnified by the high speeds that you hit while descending. Or it could be that your CdA is off slightly. Any such error is magnified at high speeds.
GWPOS wrote:Is there anyway to turn off the auto recal. for the tilt after it does it the first time 5 min into the ride?
No, but notice that in the post-processing in iBike4, you can improve it by doing Analyze Route. The post-processing has the benefit of being able to look into the future in addition the past for the correction.
rruff wrote:The problem I encountered with hill intervals was that the altitude readings at the bottom and top of the hill varied quite a bit... enough to result in a variance of ~5% in the power readings on the climbs.
But that is not the same thing as saying that instantaneous errors in the altimeter measurement could contribute such large errors to the slope_offset algorithm. Since the slope_offset algorithm is based on altimeter readings that are five minutes apart, instantaneous errors on the altimeter don't matter that much.
racerfern wrote:This has been discussed before but we never heard an official answer. Does the iBike recalculate every 5 minutes?
Sorry I didn't catch the question before. The answer is that it is calculated continuously. It starts getting applied to the ride at the 5 minute mark.
racerfern wrote:If you're getting 50-100w while soft pedaling going downhill and assuming your wind calibration is correct, then make a slight adjustment in your aero number by lowering it slightly. That will make you more aero (producing fewer watts). It's very sensitive for adjustments like this. Try soft pedaling all the way down the hill then download the ride.
Sounds like a good idea, but don't go crazy with slope of the descent that you choose. You have to be true to your ride position that you use when you do work on the bike, and in order to do that you might want to pick a descent that puts you in the same ballpark speed that you can achieve on flats.
rruff wrote:If the altimeter readings are off, then it will make a bad adjustment. Also, since it is always backward-looking, it assumes that the future 5 minutes will be like the last 5 minutes. On a series of uphills and downhills, this won't be the case.


The slope_offset algorithm isn't that sensitive to instantaneous altimeter errors. After 5 minutes of riding, you are typically 1.6 miles down the road, so a few feet of error in the altimeter reading doesn't translate into a huge error on the altimeter-based slope measurement. I'm not saying there isn't some error, but it is not enough to undermine the benefit of the slope_offset algorithm.
racerfern wrote:Would this be a problem only when the altimeter is thrown off by barometric pressure changes?
Actual weather changes can contribute some error, but again, the altimeter errors can only be so big due to weather changes on average.
racerfern wrote:Wouldn't this also affect readings on a flat when there are significant barometric changes? Why would it be only on hills? If I'm riding along the coast and the altimeter drops from my start of +30' to -100' because of weather changes the same problem would manifest itself. This is very common for me and I at times will adjust the altimeter during a ride only because I realize the weather changes are affecting the readings.
You can use the Analyze Route feature to get a handle on how big the errors are.

Travis
User avatar
lorduintah
Posts: 642
Joined: Fri Jan 11, 2008 9:37 am
Location: Plymouth, MN

Re: Auto tilt adjustment

Post by lorduintah »

The 5 mimute adjustment period is probably quite reasonable - one has to find enough data to average out random noise from all of the road vibrations and rocking of the bike along with the possible rolling incline or decline across some distance to make corrections. Yes hills that have a short period for "rollers" make that a less than optimal situation and here there could be some deviations.

The whole point of any power measurement training is to gauge against a previous value. Riding the same terrain for training day in and day out, one can judge effort (measured effort not perceived) and improvement regardless of the absolute correctness of the power expended. You base your training on a relative scale.

So I would not expect to find the ideal results from riding a new route every day. That is not to say that using the iBike in racing is not beneficial - it should still help you from bonking because you threw too much effort into an early portion of the race or you did not conserve energy in places where you should (or could) be backing off.

I like very much the fact that the iBike encompasses many factors to produce power results as accurately as it does and on the same route should yield relative effort values that are indicative of improvements or over training.

Tom
Zoltan
Posts: 213
Joined: Mon Jan 23, 2012 9:47 am
Location: HUNGARY

Re: Auto tilt adjustment

Post by Zoltan »

travispape wrote:The slope_offset algorithm takes advantage of the fact that we have an independent measurement of slope in the altimeter data. In essence, the tilt measured by the accelerometer system is "tied to reality" based on the altimeter data. As the running average of the accelerometer-based measurement deviates from the running average of the altimeter-based measurement, a bias is applied to the accelerometer-based measurement to bring it back into line with what it has to be according to the independent data.
It is a wise approach to stick to the altimeter data. In one of my posts I had supposed that iBike cleaned the tilt data using the speed data from the sensor, but it is much easier and maybe more precise to use the altimeter data to do so. But what if there are strong wind bursts and thus false altimeter data? In this case even the running average of the altimeter-based measurement can differ from the real alimeter data, so slope_offset will be false, wont it? Does iBike also use the wind data for compensating the altimeter data?
rruff
Posts: 445
Joined: Wed Apr 23, 2008 10:48 am

Re: Auto tilt adjustment

Post by rruff »

I don't think wind is going to effect it... at least not enough to matter. But there are things that do. See my recent topic about elevation change errors.

http://www.ibikeforum.com/viewtopic.php?f=6&t=2456
Zoltan
Posts: 213
Joined: Mon Jan 23, 2012 9:47 am
Location: HUNGARY

Re: Auto tilt adjustment

Post by Zoltan »

rruff wrote:I don't think wind is going to effect it... at least not enough to matter. But there are things that do. See my recent topic about elevation change errors.

http://www.ibikeforum.com/viewtopic.php?f=6&t=2456

I still have no experience with iBike, just waiting for a Newton working with polarized sunglasses, but I cna confirm that all the gadgets like Garmin 60 series or Cycleops Joule give you false vertical velocity and grade data, respectively, if wind is blowing. Of course Velocomp could opt for different averaging which can help say that wind conditions do not affect on grade/slope/tilt calculation.

Actually I just wondered if Newton has a well placed thermo sensor, a wind sensor and a barometer, it should use all the three to improve the accuracy of tilt calculation.

To be honest I do not focus on power calculation of Newton because I am plannig to use Newton with a PT hub.
rruff
Posts: 445
Joined: Wed Apr 23, 2008 10:48 am

Re: Auto tilt adjustment

Post by rruff »

I'm guessing that you mean the Garmin and Joule give fluctuating grade readings if it is windy. The iBike would give less accurate grade data also, if it only relied on the baro pressure sensor. It uses the accelerometer for grade and the baro sensor only as a 5min correction to grade.
Zoltan
Posts: 213
Joined: Mon Jan 23, 2012 9:47 am
Location: HUNGARY

Re: Auto tilt adjustment

Post by Zoltan »

Yep, i meant what and how you phrased. Eg I am sure that the grade is around 0% or +1%, I am riding on a log flat, while I see eg. the following slopes each after each on Joule +1%, +4%, +5%, +1%, 0%, -4%, -3%, -2%, -1% ,0% It is clear that it is because there is an "acceleration" of the headwind, so pressure is lower and lower, then the wind speed suddenly becomes constant, so the slope is 0 or +1% again, but since there s an artificially high elevation (=low pressure) measured by Joule which has to decrease to the correct one when wind calms, it happens to get -4% at that time. The other phenomena come fro mjust the smoothing.

And yes, i did figure out from some post that grade is mainly calculated from an accelerometer and using baro as a secondary tool for correction.
A) My inquiry is that why there is no correction for the very short term errors of tilt calculation due to the acceleration of the bike? Normally I would opt for speed data from the speed sensor as a secondary tool to calculate the acceleration of the bike and to correct the measured data of the accelerometer?

B) on the other hand what is the real benefit for the tilt coming from the correction used some 5 min baro data? I suppose it has nothing to do with any error coming from the acceleration of the bike, certainly has nothing to do with the wind. My only guess is that it has something to do with some empirical approach that if one sums up the ascents calculated from tilts estimated by the accelerometer and it is far from the total ascent calculated from the measured elevation data by the baro, then something must be done without knowing the real cause of the difference. Do you have a precise answer in this matter?

Double thanks!

EDITED: I use the words grade, slope and tilt with a meaning 'incline', but I suppose you may use tilt with a different iBike-related meaning, too.
Last edited by Zoltan on Wed Jul 04, 2012 7:34 am, edited 3 times in total.
Zoltan
Posts: 213
Joined: Mon Jan 23, 2012 9:47 am
Location: HUNGARY

Re: Auto tilt adjustment

Post by Zoltan »

rruff wrote:The iBike is always calculating tilt based on accelerometer and speedometer readings... but the issue we are discussing is something else. At 5min intervals it calculates the change in elevation based on the tilt readings and compares it to the altimeter. Then it makes an adjustment to the "zero" tilt so they match. If the altimeter readings are off, then it will make a bad adjustment. Also, since it is always backward-looking, it assumes that the future 5 minutes will be like the last 5 minutes. On a series of uphills and downhills, this won't be the case.
It seems that the answers to my just posted inquiries can be more or less found in this post of you.
rruff
Posts: 445
Joined: Wed Apr 23, 2008 10:48 am

Re: Auto tilt adjustment

Post by rruff »

I haven't tested this, but I think it continuously upgrades the tilt based on the last 5 minutes of data... not at 5 minute intervals.
User avatar
racerfern
Posts: 1356
Joined: Fri Jan 11, 2008 12:47 am
Location: Baldwin, NY
Contact:

Re: Auto tilt adjustment

Post by racerfern »

rruff wrote:I haven't tested this, but I think it continuously upgrades the tilt based on the last 5 minutes of data... not at 5 minute intervals.
Agreed.
Fernando
Zoltan
Posts: 213
Joined: Mon Jan 23, 2012 9:47 am
Location: HUNGARY

Re: Auto tilt adjustment

Post by Zoltan »

racerfern wrote:
rruff wrote:I haven't tested this, but I think it continuously upgrades the tilt based on the last 5 minutes of data... not at 5 minute intervals.
Agreed.
Yes, it is how me too understood from the "official" posts from Travis.

What about http://ibikeforum.com/viewtopic.php?f=26&t=2167 , Fernando? Can you please have a look at it?

Sidenotes: The closer I am to getting a Newton, the more unrelaxed I am. I checked many posts not just about slope measurement, but also about temperature measurement like the detailed report from Travis having 3 iBikes on his bike.
rruff
Posts: 445
Joined: Wed Apr 23, 2008 10:48 am

Re: Auto tilt adjustment

Post by rruff »

Since you aren't concerned about power, what are you using the slope, wind, temperature, etc data for?
Zoltan
Posts: 213
Joined: Mon Jan 23, 2012 9:47 am
Location: HUNGARY

Re: Auto tilt adjustment

Post by Zoltan »

I know it sounds crazy, but just for fun.

It gives me a great hobby to achieve as accurate environment data as I can WHEN CYCLING (EDITED). I started with a handy anemometer, a thermometer with a hidden sensor under the down
tube and a Garmin for vertical speed.

a) Slope: I could change the vertical speed to real slope with Cycleops Joule working well when accelerating, but working poorly in windy weather, and also added a USD15 worth inclinometer (Monti or something) working well in windy weather, but working poorly when accelerating.

b) Thermometer. I gave up because even the white frame of my KTM branded trekking bike got hot after 30 minutes in direct sunshine, so hiding the sensor was not enough. It was close to ambient air, but for some reasons it measured too high data.

c) Wind speed. There is no way to hold an anemometer in your hand when cycling. :lol: :lol: :lol:

With iBike I guess that slope and wind speed can be quite perfect, but I would like to be sure before burning 500 bucks, temperature can be also very accurate considering the placement of the thermo sensor in Newton. Even if I buy a Newton, I will still miss the wind direction as something I'd like to see constantly.

Before asking me, no-no-no, i am not interested in humidity, dew point and ambient pressure realtime, although my Garmin can show the latter. Also miss a bit precipitation data and calculated wind chill. Especially during cold winter riding ;)
Last edited by Zoltan on Thu Jul 05, 2012 1:43 am, edited 2 times in total.
User avatar
lorduintah
Posts: 642
Joined: Fri Jan 11, 2008 9:37 am
Location: Plymouth, MN

Re: Auto tilt adjustment

Post by lorduintah »

I suppose that were we to have all the formulae for all the interactions of the sensors, you could have a field day - except you would be worrying about an R^2 < 0.999 :lol:

Tom
User avatar
racerfern
Posts: 1356
Joined: Fri Jan 11, 2008 12:47 am
Location: Baldwin, NY
Contact:

Re: Auto tilt adjustment

Post by racerfern »

I can't think of any device that even comes close to a Newton for environment recording. Temperature might jump a bit in bright sun but it seems with the movement of the temp sensor toward the inlet stream of air it is not affected as much by extreme sun. One thing for sure the barometric sensor and algorithms for determining altitude is vastly superior to the G3. I say this because there is a local short hill I do repeats on. I used to get variations in the climbing and altitude numbers before analysis. The screenshot is typical of my 4 or 5 efforts with the Gen 3. The picture from the Newton shows consistency that I have not seen before. Perfect? of course not, but really really close.

Truly flat rides don't accumulate as much elevation from "scatter" and climbing rides show slightly higher totals which are more in line with other websites or devices. Of course you don't get wind direction but as you know Zoltan, SportTracks has plugins that grab weather info depending on your location.

The slope info from the Newton just like previous versions is, IMO spot on. Where there used to be issues on repeats, I believe those are gone. I have only done a couple of rides with longer repeats and those were spot on. The first hill repeat shows 154' minimum, 686' maximum, the second effort shows 155' minimum and 681 maximum. How much better does it have to be?

BTW, I always record at 1s resolution. Questions/comments, post away.
Attachments
longer effort
longer effort
longer_effort.jpg (143.63 KiB) Viewed 19419 times
Gen 3 Elevation Raw File
Gen 3 Elevation Raw File
gen3_elevation.jpg (154.07 KiB) Viewed 19423 times
Newton Elevation Raw File
Newton Elevation Raw File
newton_elevation.jpg (139.53 KiB) Viewed 19425 times
Fernando
rruff
Posts: 445
Joined: Wed Apr 23, 2008 10:48 am

Re: Auto tilt adjustment

Post by rruff »

The gradual downward drift of the Gen3 elevation is most likely caused by barometric pressure changing. The Newton would do this too, if it was happening.

I generally look at the average slope on the "up" portions, since this is a number that directly effects power.
User avatar
racerfern
Posts: 1356
Joined: Fri Jan 11, 2008 12:47 am
Location: Baldwin, NY
Contact:

Re: Auto tilt adjustment

Post by racerfern »

Agreed that the Newton would also be affected by the barometric pressure. However, my starting and ending numbers are closer to reality than with the Gen3.

AFA the longer climb with two efforts, the first time has 6.09% and the second has 5.98%. I would have to do more tests and/or more repeats to establish a pattern. My point is this is closer to what was actually climbed than with a Gen3.
Fernando
Zoltan
Posts: 213
Joined: Mon Jan 23, 2012 9:47 am
Location: HUNGARY

Re: Auto tilt adjustment

Post by Zoltan »

racerfern wrote:I can't think of any device that even comes close to a Newton for environment recording. Temperature might jump a bit in bright sun but it seems with the movement of the temp sensor toward the inlet stream of air it is not affected as much by extreme sun. One thing for sure the barometric sensor and algorithms for determining altitude is vastly superior to the G3. I say this because there is a local short hill I do repeats on. I used to get variations in the climbing and altitude numbers before analysis. The screenshot is typical of my 4 or 5 efforts with the Gen 3. The picture from the Newton shows consistency that I have not seen before. Perfect? of course not, but really really close.

Truly flat rides don't accumulate as much elevation from "scatter" and climbing rides show slightly higher totals which are more in line with other websites or devices. Of course you don't get wind direction but as you know Zoltan, SportTracks has plugins that grab weather info depending on your location.

The slope info from the Newton just like previous versions is, IMO spot on. Where there used to be issues on repeats, I believe those are gone. I have only done a couple of rides with longer repeats and those were spot on. The first hill repeat shows 154' minimum, 686' maximum, the second effort shows 155' minimum and 681 maximum. How much better does it have to be?

BTW, I always record at 1s resolution. Questions/comments, post away.
First of all thank you both for the answers, comments, sharing your experiences.

Re Sporttracks: You know Fernando, I mentioned that having quasi a full meteo station on the bike is the so-called "hobby" of mine. If I sticked to Sporttracks being pleased with having environment information post ride, I could say why to pay for temperature or wind speed data, both will come with Sporttracks.

It is interesting that although you say "Agreed that the Newton would also be affected by the barometric pressure.",for some reasons you believe that Newton is superior to GenIII in altimeter recording since you state that "The picture from the Newton shows consistency that I have not seen before". I don'tthink that it is just a pure coincidence, there must be some subtle explanation for the improvement. what do you think it is?

Wht does "Truly flat rides don't accumulate as much elevation from "scatter"" mean? My command of English is not strong enough to figure ot what scatter means in this case. Scatter means for me just some disperse. Do you mean the small oscillation/fluctuation?

Finally I promise you that I will check the elevation profiles of my typical calibartion rides that I used for feeding my own "mathematical power measurement model" that I used before buying a Powertap. I will check the raw elevation data collected by Garmin 60csx and the Gauss smoothed elevation data, too
User avatar
racerfern
Posts: 1356
Joined: Fri Jan 11, 2008 12:47 am
Location: Baldwin, NY
Contact:

Re: Auto tilt adjustment

Post by racerfern »

What does "Truly flat rides don't accumulate as much elevation from "scatter"" mean?
http://app.strava.com/rides/11446349 This ride normally shows about 250ft of climbing with the Gen3 and about 280ft on Strava. There are only two overpasses and other that it has two very small rises. I think the actual climbing is 100ft max, maybe a little more if you count every centimeter as you like to do.

The one ride I did with the Newton showed about 150' if I remember correctly. So the annoying "noise/scatter" has been greatly minimized. Of course one ride doesn't set a pattern but I have no reason to suspect future rides would be different.

As you know the weather info on SportTracks is nice to look at, it gives you wind direction but the wind speed is almost useless for analysis purposes. I am like you in that I like seeing all the data, so I use SportTracks. However I import the wind speed data from the iBike into the custom data tracks plugin.
Fernando
Zoltan
Posts: 213
Joined: Mon Jan 23, 2012 9:47 am
Location: HUNGARY

Re: Auto tilt adjustment

Post by Zoltan »

"It is interesting that although you say "Agreed that the Newton would also be affected by the barometric pressure.",for some reasons you believe that Newton is superior to GenIII in altimeter recording since you state that "The picture from the Newton shows consistency that I have not seen before". I don'tthink that it is just a pure coincidence, there must be some subtle explanation for the improvement. what do you think it is?"

You are still kind to answer to me, actually you are also incorporating what you have learned about me, so these are tailored answers. What I still miss is that part above. No idea about it, just an intuition based on short term experiences ?
Velocomp
Velocomp CEO
Posts: 7793
Joined: Fri Jan 11, 2008 8:43 am

Re: Auto tilt adjustment

Post by Velocomp »

racerfern wrote:Agreed that the Newton would also be affected by the barometric pressure. However, my starting and ending numbers are closer to reality than with the Gen3.

AFA the longer climb with two efforts, the first time has 6.09% and the second has 5.98%. I would have to do more tests and/or more repeats to establish a pattern. My point is this is closer to what was actually climbed than with a Gen3.
We are using a latest-generation digital baro sensor in the Newton. The Gen III used an analog baro sensor.

We are seeing much less noisy, much more consistent elevation data out of the Newton.
John Hamann
Zoltan
Posts: 213
Joined: Mon Jan 23, 2012 9:47 am
Location: HUNGARY

Re: Auto tilt adjustment

Post by Zoltan »

And the charts, let me repeat tp clarify not from iBike but from a Garmin.

Raw data
Raw data
Raw on 20111022.jpg (47.79 KiB) Viewed 19309 times
Smoothed data
Smoothed data
Smoothed on 20111022.jpg (44.78 KiB) Viewed 19307 times
Don't regard the minimum levels of elevations, since I did not turn around at the same place on the descents. But you can consider the "zeniths" as really at the same elevation in real life. The y scale is in meters. So there is a difference of appr 3 meters or 9.5 feet . Not perfect at all.
Post Reply