Tilt Cal R.I.P.

[daidnik]

Some explanation of the algorithmic basis of eliminating the Tilt-Cal might be appropriate. When I saw this in the firmware revision notes I decided to wait until finding out more about the basis for this change.

I can see how the Newton could establish this from an Out & Back ride on the same course. Is this the basis of establishing this? If so, then it would seem valid so long as nothing changed on the mount.

Personally, I've never found the Tilt Cal very burdensome and I rather like seeing what the Newton reads a 'level' incline to be. It gives me some reassurance on repeatability on measuring the same thing day after day.

Removing it from the Setup menu does not seem to be any 'performance enhancement' at all unless there is some algorithmic change to the calculation of the work done per unit of time which would seem unlikely to me, for the user is not obligated to do the Tilt Cal if he/she doesn't want to.

The Newton MUST have a means of correlating the inclinometer with gravity. The user could update/re-establish that correlation via Tilt Cal, but did not have to do so unless they wanted to; "fine & dandy". Now, you remove the user's ability to do except by re-doing a Cal Ride and this is an 'enhancement'?

I just can't hear the truth behind it without some explanation; JMO.

[dtrousdale]

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.

I have a few questions about this, and just so you know, I am using the latest firmware:
(1.) I'm not clear on the steps here. My ride today has three segments (please reference the 3 attached files dated 5/9): (a.) slightly less than a mile to the beginning of my out-and-back, (b.) the out-and-back ride feature selected on the Newton+, and (c.) slightly less than a mile on my way back home after finishing the out-and-back. What do I do next? Select either the first or the last segment and then use Check Calibration? Or do I select the out-and-back ride itself and then use Check Calibration? I ask because I did this twice today: same route, two different bikes each with their own DFPM. From the first ride, I opened the first segment (before starting the out-and-back), and then did Check Calibration. I accepted the recommended changes. From the second ride, I tried both the first and third segments, and each time I got a message saying the ride was too short. Did I do this correctly except for not having a segment long enough to analyze? If so, what would be the proper length of the segment to be analyzed? Or was I supposed to select the Calibration Ride itself and use Check Calibration on it?
(2.) I did open the Calibration Rides (the second segment in each of the two rides I did today) and was startled to see HUGE power spikes (1500W to 2000W) and STEEP slopes (18 to 26%) for the first mile and a half on both Calibration Ride Segments. This does not seem right to me, or am I missing something?
(3.) The reason I'm doing this is to make sure I have my Newton+ dialed in correctly. I was pretty certain I did because I used Check Calibration on rides I did last January, and I accepted the recommendations at that time. However, now I'm doing serious training and have noticed that my TSS numbers as calculated by the Newton+ with DFPM turned off are more than 3 times larger than the TSS numbers with View DFPM turned on (see ride file dated 4/26 in my next post; 3 is the max # of files per post). There are also some pretty impressive power spikes and slopes that I am certain were nothing I encountered on the ride. Looking at the Calibration Ride segments I recorded today now is making me wonder if there is something wrong with my Newton+.

[dtrousdale]

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.

(3.) The reason I'm doing this is to make sure I have my Newton+ dialed in correctly. I was pretty certain I did because I used Check Calibration on rides I did last January, and I accepted the recommendations at that time. However, now I'm doing serious training and have noticed that my TSS numbers as calculated by the Newton+ with DFPM turned off are more than 3 times larger than the TSS numbers with View DFPM turned on (see ride file dated 4/26 in my next post; 3 is the max # of files per post). There are also some pretty impressive power spikes and slopes that I am certain were nothing I encountered on the ride. Looking at the Calibration Ride segments I recorded today now is making me wonder if there is something wrong with my Newton+.

Please find the file referenced above attached.

[Davis6022]

Can't open the csv files in Isaac. Need the ibr files.

I think the power numbers and slopes in the calibration files are meaningless since when you ride the first leg of the out and back, the Newton doesn't know the tilt yet. It computes the tilt on the return leg. I have questions about calibrating to a DFPM too, so I'm hoping we'll get some good guidance here.

[Velocomp]

Some explanation of the algorithmic basis of eliminating the Tilt-Cal might be appropriate. When I saw this in the firmware revision notes I decided to wait until finding out more about the basis for this change.

I can see how the Newton could establish this from an Out & Back ride on the same course. Is this the basis of establishing this? If so, then it would seem valid so long as nothing changed on the mount.

Personally, I've never found the Tilt Cal very burdensome and I rather like seeing what the Newton reads a 'level' incline to be. It gives me some reassurance on repeatability on measuring the same thing day after day.

Removing it from the Setup menu does not seem to be any 'performance enhancement' at all unless there is some algorithmic change to the calculation of the work done per unit of time which would seem unlikely to me, for the user is not obligated to do the Tilt Cal if he/she doesn't want to.

The Newton MUST have a means of correlating the inclinometer with gravity. The user could update/re-establish that correlation via Tilt Cal, but did not have to do so unless they wanted to; "fine & dandy". Now, you remove the user's ability to do except by re-doing a Cal Ride and this is an 'enhancement'?

I just can't hear the truth behind it without some explanation; JMO.

I don't know the details of how my electric toothbrush works, but that hasn't stopped me from using it.

More seriously, if we went into the all of the details about how the iBike works, it likely would be somewhere between, boring, fascinating, and stupid--for we would be giving away all of our intellectual property.

Remember, the iBike has a rich variety of sensors in it, and among other things we can use them together to cross-check each other. In this case, we use accelerometer and barometric pressure sensor data to automate the tilt calibration. These algorithms have been in use now for about 5 years, so it is very battle-tested.

[daidnik]

John,

I appreciate the work in design of the Newton.

I raised a few points of doubt and concern about this change in the firmware.

I am discouraged that you could not make any effort at all to offer a meaningful response to even one of those points.

Is a meaningful answer to the question of How/Why a Tilt Cal is no longer needed going to burden you that much or reveal a trade secret?

The question is more specifically posed as "Why was it necessary to remove the ability of the user to perform a Tilt Cal?"

I honestly can't see what you hoped to accomplish with your last response.

[Velocomp]

We're not trying to be evasive or uncooperative, even though the last post might have had a bit of that tone. Sorry!

Yes, the O&B ride is the key to getting things right. It always has been, so in that sense the latest FW is nothing new.

We completely agree that Tilt Cal is not burdensome. But that's ONLY because we know how to do this simple procedure.

REMEMBER, most folks have no clue about bike computers, power meters, etc.. They want to ride their bikes, and for them, ANY setup procedure is complicated.

We've had a lot of internal debate about whether or not to keep the Tilt Cal in the setup screen. Once a Cal Ride is done, the initial Tilt Cal is done as well. Furthermore, on any subsequent ride, any minor tilt calibration corrections are performed automatically as well. This has always been the case.

And critically, if a Tilt Cal is truly needed, say because the user has moved the position of the mount on the handlebars or stem, or gotten a replacement Newton, then what's REALLY needed is a new Cal Ride. And, of course, when a new Cal Ride is done, the Tilt Cal is done too.

Our view is that anything that makes setup easier, and less prone to mistakes, is a meaningful enhancement.

Leaving in the Tilt Cal may be a comfort for experienced users, but for everyone else it requires explanation, clarification, and qualification, which makes it a potential source of confusion, complication, and errors.

But we also know that it's fun to "open up the hood" and tweak things.

We're open to alternative opinions; now, at least, you know the logic that lead us to this change!

[daidnik]

John,

Thanks for the response.

Based on what you stated about the Out/Back ride establishing an accurate or more accurate 'Tilt Cal', it raises the question of "What happened when the user performed a subsequent 'Tilt Cal'? In the previous firmware, were these subsequent Tilt Cal's tweaks on the stored value or simply overwriting the stored value? I would have thought that the latter was true, but your posting makes me less sure of this.

I also infer from you post that some dynamic re-calc/eval/updating of this Tilt Cal may be done "automatically". Is this the case?

This would seem to me to be a more difficult matter that would be handled at the level of downloading and analyzing the data on the PC in the case where the start & end of the ride were the same; like tweaking the slope parameter until the integrated sum is zero net elevation gain/loss then adjusting the stored value for Tilt Cal to be consistent with the "zero sum" 'back-modeled' value?

Is this the sort of "automatic" 'error-correction' for 'Tilt' that you allude to?

Not trying to steal/reveal any trade secrets here, but just want to be clear about these matters of "automatic" correction vs what was happening when the user previously performed a Tilt Cal?

[hiroad]

We completely agree that Tilt Cal is not burdensome. But that's ONLY because we know how to do this simple procedure.

REMEMBER, most folks have no clue about bike computers, power meters, etc.. They want to ride their bikes, and for them, ANY setup procedure is complicated.

Yes, it truly is! I therefore praise your concept to "go as simple as possible". However, for we old-timers (I am an iBike user since your Gen 2) it is easier to stay in the same pond sometimes with few infos. I'd rather asking you to give us a few explanation to encourage the leap out of the pond.

Also I think we old-timers can act as evangelists of iBike for newcomers (we 'd happy to do!), so it can be beneficial for you to make us clear with some of your words.

We've had a lot of internal debate about whether or not to keep the Tilt Cal in the setup screen. Once a Cal Ride is done, the initial Tilt Cal is done as well. Furthermore, on any subsequent ride, any minor tilt calibration corrections are performed automatically as well. This has always been the case.

And critically, if a Tilt Cal is truly needed, say because the user has moved the position of the mount on the handlebars or stem, or gotten a replacement Newton, then what's REALLY needed is a new Cal Ride. And, of course, when a new Cal Ride is done, the Tilt Cal is done too.

This arise me some questions instantly,

1) Is my understanding correct, about the "initial Tilt Cal" value and the "additional Til Cal achieved through Cal Rides" value are separate and both are needful, even on OS400?

2)Are the values of "initial Tilt Cal" and /or "additional Til Cal achieved through Cal Rides" tweaked through ANY of the rides afterward, even the ride is one-way hill-up or down?

3)How above #2 is possible? If possible with or without PC download? (A trade secret?)

[dgf]

Mine hasn't worked for chit after the elimination the tilt cal. I have to Cal Ride before every single ride but IMO, the data then varies from ride to ride making actually training with power with this unit impossible. It's become a deal breaker for me.

[Velocomp]

Mine hasn't worked for chit after the elimination the tilt cal. I have to Cal Ride before every single ride but IMO, the data then varies from ride to ride making actually training with power with this unit impossible. It's become a deal breaker for me.

Something is not right. From these postings we can't tell what it is, but we will put Tilt Cal back. Look for it in the next release.

[Velocomp]

John,

Thanks for the response.

Based on what you stated about the Out/Back ride establishing an accurate or more accurate 'Tilt Cal', it raises the question of "What happened when the user performed a subsequent 'Tilt Cal'? In the previous firmware, were these subsequent Tilt Cal's tweaks on the stored value or simply overwriting the stored value? I would have thought that the latter was true, but your posting makes me less sure of this.

I also infer from you post that some dynamic re-calc/eval/updating of this Tilt Cal may be done "automatically". Is this the case?

This would seem to me to be a more difficult matter that would be handled at the level of downloading and analyzing the data on the PC in the case where the start & end of the ride were the same; like tweaking the slope parameter until the integrated sum is zero net elevation gain/loss then adjusting the stored value for Tilt Cal to be consistent with the "zero sum" 'back-modeled' value?

Is this the sort of "automatic" 'error-correction' for 'Tilt' that you allude to?

Not trying to steal/reveal any trade secrets here, but just want to be clear about these matters of "automatic" correction vs what was happening when the user previously performed a Tilt Cal?

Lots of questions!

Yes, the O&B establishes the basic Tilt number. And yes, subsequent manual Tilt Cals are nothing more than tweaks of the basic calibration.

And importantly, even if you do a manual Tilt Cal prior to your ride, a few minutes after you begin your new ride, the "manual" tilt cal you just made is, itself, is checked and tweaked automatically by the Newton, as needed! This has always been the case.

Tilt Cal was required as part of the original iBike, introduced in 2005. The automatic correction tilt algorithm was implemented in early 2008, over 5 years ago. It works so well that we realized that the original Tilt Cal was no longer necessary. That's why we eliminated Tilt Cal from the startup process.

[Velocomp]

We completely agree that Tilt Cal is not burdensome. But that's ONLY because we know how to do this simple procedure.

REMEMBER, most folks have no clue about bike computers, power meters, etc.. They want to ride their bikes, and for them, ANY setup procedure is complicated.

Yes, it truly is! I therefore praise your concept to "go as simple as possible". However, for we old-timers (I am an iBike user since your Gen 2) it is easier to stay in the same pond sometimes with few infos. I'd rather asking you to give us a few explanation to encourage the leap out of the pond.

Also I think we old-timers can act as evangelists of iBike for newcomers (we 'd happy to do!), so it can be beneficial for you to make us clear with some of your words.

We've had a lot of internal debate about whether or not to keep the Tilt Cal in the setup screen. Once a Cal Ride is done, the initial Tilt Cal is done as well. Furthermore, on any subsequent ride, any minor tilt calibration corrections are performed automatically as well. This has always been the case.

And critically, if a Tilt Cal is truly needed, say because the user has moved the position of the mount on the handlebars or stem, or gotten a replacement Newton, then what's REALLY needed is a new Cal Ride. And, of course, when a new Cal Ride is done, the Tilt Cal is done too.

This arise me some questions instantly,

1) Is my understanding correct, about the "initial Tilt Cal" value and the "additional Til Cal achieved through Cal Rides" value are separate and both are needful, even on OS400?

2)Are the values of "initial Tilt Cal" and /or "additional Til Cal achieved through Cal Rides" tweaked through ANY of the rides afterward, even the ride is one-way hill-up or down?

3)How above #2 is possible? If possible with or without PC download? (A trade secret?)

Answers:

The purpose of Tilt Cal is to establish, with high accuracy, the angle at which the Newton sits on the bike. If, by chance, the Newton were to be mounted perfectly parallel to the road beneath it and, if by chance, the accelerometer on the circuit board were to be perfectly parallel to the road beneath it, then a Tilt Cal would not be required.

But in reality neither will ever be the case, and there will always be an offset angle between the Newton and the road. The Tilt Cal measures the offset angle, making it possible to digitally align the accelerometer to be parallel to the road beneath it.

In the original iBike, the Tilt Cal procedure determined the offset angle by having the user turn the bike 180 degrees, twice (the offset angle is determined in the first 180 turn; the second 180 turn confirms the accuracy of the calibration).

In 2008 we figured out a Tilt Correction algorithm, that tweaks the tilt calibration offset value during the ride, using data gathered from the baro and speed sensors. How does that algorithm work? It works very well, but for reasons we choose not to share. Suffice it to say this: since 2008, any ride you've done with an iBike has benefited from the tilt correction algorithm.

The Newton introduced latest-generation sensors, far better in precision than Gen I/II/III sensors. Newton sensors work so well that we realized that we could eliminate the initial Tilt Cal altogether.

Answers to other questions:

1) Yes, "Initial Tilt Cal" is achieved through the manual calibration. "Additional Tilt Cal achieved through Cal Ride" is a "tweak" of the Initial Tilt Cal. If, for any reason, the user made a mistake in the Initial Tilt Cal, tilt would be tweaked as needed in the Cal Ride. But here's the magic: since the Additional Tilt Cal precisely corrects any errors of the Initial Tilt Cal, then the Initial Tilt Cal is not needed!

2) Tilt offset is monitored continuously throughout a bike ride. Any variation in calibration, caused by changes in rider position, tire pressure, or other factors is detected and corrected dynamically. This has been true since 2008...

3) The correction algorithm runs in real time, in the firmware. It is extremely robust and proven. In fact, Isaac takes the raw sensor data, and reprocesses it in software, to check the calculations. The firmware and software always produce the same results.

The bottom line is this: for users who have carefully done Tilt Cals all these years, you get great results. For users who do Tilt Cals, but who occasionally make a mistake, you'll STILL get great results, because the Tilt Correction algorithm will automatically tweak any mistakes. And since the introduction of the Newton, even if you NEVER do a Tilt Cal, your watts will automatically come into line after 5 minutes.

What's new with OS 400? It memorizes the corrected value of Tilt Cal, so that, at the beginning of each new ride your corrected value of Tilt Cal from the previous ride is the starting value of your new ride.

In our view Tilt Cal has become like the body's appendix; it was useful at one time, but now is an artifact of the past.

[jayt]

John, thanks for the great explanation. I can certainly say the whole tilt business seems to be taken care of well with Newton. But, back to a scenario I posed earlier that I will face this Sat. I shipped my bike to North Carolina. I will reassemble it on Sat., so the handlebar position will not be "exactly" the same. I can get it close, but it won't be exact. If I understand what you said above, all I need to do is put the Newton on the bike and go for a 5 to 10 min. ride (not necessarily a "cal ride", just a ride) and then all will be fine (this is ok since I will do a "shake down ride" before my event on Monday anyway)?

Thanks,
Jay

[Velocomp]

Yes, if you ride your bike for 5-10 minutes, then do a Trip Reset, your Newton will be dialed-in again. Or, if you do a long ride, only the first 5 minutes will be slightly off; after that, everything will be fine.

All of this said, for all you "Power Users" (pun intended) we're going to put Tilt Cal back into the next FW release. You'll find Tilt Cal in the Racr sequence. This is going to be an undocumented feature, since we really do want Tilt Cal to go away over time...

If you have already done a Cal Ride, then you will use Tilt Cal if you want to tweak your Newton manually. For the first ride AFTER a Tilt Cal, it's possible your watts will be slightly off, but then they will settle in and be just fine.

But, DO NOT do repeated, back-to-back manual Tilt Cal and Cal Rides!!!! This scenario will keep your Newton in a state of perpetual confusion.

[hiroad]

John,
Thank you for the kind explanation. Now I understand that,

1) NEWTONs have "Tilt auto-correction" functionality and continuously observe/correct their tilt calibration during everyday rides. So there is no more real reason to resort to "Tilt Cal". After I do "Cal Ride" once per a profile, It will get right by itself within 5 to 10 minutes of ride, every time. So smart!

2)When I noticed a big change of the mounting position or angles, it is better to do a Cal Ride, instead of staying with performing Tilt Cal only. I guess this is because those great changes may affect to the air-flow to the sensor and change the wind sensor variations.

3)In spite of #1, you understand our worrying about not having Tilt Cal in hand, you kindly back it in the User Sequence in coming OS402. Thank you for making me feel that I treated right and kind. However, with all of those new infos, maybe I can be happy without using this hard earned Tilt Cal functionality any more. Ironic, I'm afraid. Anyway I'll see.

***
I have one more to ask to you,

But, DO NOT do repeated, back-to-back manual Tilt Cal and Cal Rides!!!! This scenario will keep your Newton in a state of perpetual confusion.

Are you saying that right-before or right-after the Cal Rides, we SHOULD NOT perform Tilt Cal manually, right? Just for confirmation, please.

[Velocomp]

John,
Thank you for the kind explanation. Now I understand that,

1) NEWTONs have "Tilt auto-correction" functionality and continuously observe/correct their tilt calibration during everyday rides. So there is no more real reason to resort to "Tilt Cal". After I do "Cal Ride" once per a profile, It will get right by itself within 5 to 10 minutes of ride, every time. So smart!

2)When I noticed a big change of the mounting position or angles, it is better to do a Cal Ride, instead of staying with performing Tilt Cal only. I guess this is because those great changes may affect to the air-flow to the sensor and change the wind sensor variations.

3)In spite of #1, you understand our worrying about not having Tilt Cal in hand, you kindly back it in the User Sequence in coming OS402. Thank you for making me feel that I treated right and kind. However, with all of those new infos, maybe I can be happy without using this hard earned Tilt Cal functionality any more. Ironic, I'm afraid. Anyway I'll see.

***
I have one more to ask to you,

But, DO NOT do repeated, back-to-back manual Tilt Cal and Cal Rides!!!! This scenario will keep your Newton in a state of perpetual confusion.

4) Are you saying that right-before or right-after the Cal Rides, we SHOULD NOT perform Tilt Cal manually, right? Just for confirmation, please.

Answers:

1) Correct!!
2) Correct!!
3) Mostly Correct!! It's in the Racr Sequence
4) Correct!!

[jayt]

John,

Thanks for helping with setting my bike back up this weekend. I think adding the Tilt Cal back to the RACR menu is a good idea and will accommodate all of us. I appreciate your responsiveness to your customers.

Jay

[NorBike]

John

You say the auto-correction algorithm work so well, but I often use my Newton on my MTB finding rare watt readings after riding longer than 5 minutes on gravel roads or small tracks.

I have two questions I hope you could answer me:

1.I can see the point in removing the tilt cal because of the auto-correction algorithm. On flat tarmac it works very well, and since there is no on/off auto-correction algorithm I understand the tilt cal gets "overwritten" by the auto-correction algorithm after 5 minutes. BUT I don’t understand, if tilt cal is removed, why Newton needs the riding tilt number/offset?...if the riding tilt is wrong, shouldn’t the auto-correction algorithm also correct/override this?

2.A few weeks ago I turned my stem upside down, and got a 25% slope error on my Newton. For fun I did a ride before the cal ride. The first 5min the Newton was totally off. 25% slope “hill climb” at 1200watt+ on flat road…. After 5mins it slowly started to get closer and closer to “real life” slope. BUT, when I stopped the Newton showed a -25% slope. When I start riding again the Newton quickly got back to correct slope, and back again to -25% slope when I stop riding. Why?

Just a note. After Isaac 1.0.10 there's a small Wind Scaling error/difference between Isaac and Newton when doing "check calibration". I hope this can be corrected soon.

(The printscreen shows the ride with four stops at the end)

[mttvrtn]

John,

I think adding the Tilt Cal back to the RACR menu is a good idea and will accommodate all of us.

Jay

+1 from me for restoring Tilt Cal (and hiding it under RACR menu would be great).

Also, 4.01 at least seems to be missing the 12h / 24h selection for time (unless it's been moved somewhere). I'd also vote for its return.

Matt

[racerfern]

12/24 hour is in the same place it's been. Go to user > clock and hit center button. If the time is fine hit the center button otherwise set the time to what you want then hit the center button. The next screen is 12/24.

[mttvrtn]

12/24 hour is in the same place it's been. Go to user > clock and hit center button. If the time is fine hit the center button otherwise set the time to what you want then hit the center button. The next screen is 12/24.

Doh! Setting up a new unit for the 1st time in over a year so I obviously blanked. Thank you, Fernando!

[KenS]

Just posting some strange stuff that may be related to problems others were having with tilt.
To start with I had the Newton pretty well dialled in with Check Cals showing "insignificant" variation
I did a ride where I got water in the wind port but which also seems to have affected the Altimeter readings. After the ride I reset the Newton and Isaac confirmed that the data was rubbish as I expected. No problem though.
Next day I did an out and back and the first 5 minutes the Newton was clearly gong through its correction routine. After the ride I reset as usual and Check Cal showed some significant changes needed. But I ignored them as I figured the Newton was dialled in and this was somehow to do with the previous day's ride.
Subsequent rides all come up with "insignificant" changes.

So what I'm not clear on is whether the water problem was persisting to the next day; or was the Newton remembering the corrected tilt from the wet ride? I had thought a trip reset would take it back to its tilt setting from the cal-ride. But if not then I can see how that might cause problems making folks want to do the old style tilt cals.

iBike_2013_05_30_1413.ibr

Rain affected ride

(232.33 KiB) Downloaded 303 times

iBike_2013_05_31_1243.ibr

Out and back with wrong starting tilt

(410 KiB) Downloaded 288 times

[Zoltan]

John

You say the auto-correction algorithm work so well, but I often use my Newton on my MTB finding rare watt readings after riding longer than 5 minutes on gravel roads or small tracks.

I have two questions I hope you could answer me:

1.I can see the point in removing the tilt cal because of the auto-correction algorithm. On flat tarmac it works very well, and since there is no on/off auto-correction algorithm I understand the tilt cal gets "overwritten" by the auto-correction algorithm after 5 minutes. BUT I don’t understand, if tilt cal is removed, why Newton needs the riding tilt number/offset?...if the riding tilt is wrong, shouldn’t the auto-correction algorithm also correct/override this?

2.A few weeks ago I turned my stem upside down, and got a 25% slope error on my Newton. For fun I did a ride before the cal ride. The first 5min the Newton was totally off. 25% slope “hill climb” at 1200watt+ on flat road…. After 5mins it slowly started to get closer and closer to “real life” slope. BUT, when I stopped the Newton showed a -25% slope. When I start riding again the Newton quickly got back to correct slope, and back again to -25% slope when I stop riding. Why?

Just a note. After Isaac 1.0.10 there's a small Wind Scaling error/difference between Isaac and Newton when doing "check calibration". I hope this can be corrected soon.

(The printscreen shows the ride with four stops at the end)

Point 2 is an interesting one. Can it be explained by that the result of the auto tilt cal is not something, which is saved by Newton?

[andrep]

I know - I come up with the same question over and over again. But - there is something I do not understand what I like to understand:

EVERY time I come back from a ride, I check the tilt cal WITH THE HILL-FUNCTION on a flat floor in our house. Every time there is a mistake: One direction 1% -180 degree turn - again 1% (not -1% what would be logical). The watts on the ride seem to be ok - but I can´t undrestand, why there is that big "mistake" (or whatever) after the rides?! The ridefils look fine too - only this little thing is irritating me. do you have any explanation for it?!

[Velocomp]

I know - I come up with the same question over and over again. But - there is something I do not understand what I like to understand:

EVERY time I come back from a ride, I check the tilt cal WITH THE HILL-FUNCTION on a flat floor in our house. Every time there is a mistake: One direction 1% -180 degree turn - again 1% (not -1% what would be logical). The watts on the ride seem to be ok - but I can´t undrestand, why there is that big "mistake" (or whatever) after the rides?! The ridefils look fine too - only this little thing is irritating me. do you have any explanation for it?!

The Newton's accelerometer measures total forces caused by acceleration--one of which is caused by hill slope. And the hill slope measurement must also account for the fact that the cyclist is sitting on the bike. So, if you're not sitting on the bike, there may be a slight error in reported hill slope.

The Newton is a power meter; if you want to measure hill slope when at rest, we suggest using an inclinometer.

[andrep]

I like your humor Ok - your are right - I should sit on the bike. And yes - on my tt bike there will be a lot of weight on the front wheel which leads to the wrong measurd hill-slope - sounds logical. And beside all my questions: I am happy with the newton and isaac and keep in mind: I am your brand-amassador in germany

[Velocomp]

I like your humor Ok - your are right - I should sit on the bike. And yes - on my tt bike there will be a lot of weight on the front wheel which leads to the wrong measurd hill-slope - sounds logical. And beside all my questions: I am happy with the newton and isaac and keep in mind: I am your brand-amassador in germany

We love Newton ambassadors, especially those who speak German!

[Russ]

Today I started out and flatted about 10 minutes walk from home. So I walked the (i)bike home : -)

After repair, headed out and noticed zero power unless I really pushed hard. I realized the tilt had gone somewhere South of the border so I stopped and did (fortunately you put it back) a tilt cal then all was really great. Actually a bit too great, man my power numbers were fantastic!

Got home and uploaded the ride, found my CRR had jumped to .019! Wow fixed that and I was no longer superman!

If the firmware guy gets a chance (and you agree at Velocomp) it would be nice to stop the auto tilt when the speed is walking speed and the cadence is zero, might save the hassle.

Thanks,
Russ

[racerfern]

Hey Russ, there's always some unusual situation that people come across and you certainly came across a unique one. If you're walking with the bike why not just shut off the unit until you're ready to ride?