Jump to content

Gil

Recommended Posts

  • Replies 220
  • Created
  • Last Reply

So now i am using the 9B Metrics App too. Can anyone tell me the save limits of the Ninebot One E+? So far i dont know if i was riding with too much aggression until i know where the limits of my 9Bot are.... 

Link to comment
Share on other sites

Just now, Paco Gorina said:

The Ninebot already tilts when speed is too much (near 24 km/h) or when battery is too low (under 20%, forces reduction of speed increasing tilt).

I have not tested a tempreature warning.

Oh i think you dont understood whatvi ment. Thats all clear, i wanted to know the limits of the Ninebot compared to the data read out from NB Metrics.... Is it ok if i have over 600 watt in peaks (like jumping from curbs) or over 500 watt continious at accelerating.... How much Ampere are to much? How much voltage is to much? When will i be in an critical range? 

Link to comment
Share on other sites

Oh, I think the community must find the standards.

The hardest I have run is a over 20 km/h climb of 200m (Avg 21 km/h)9B_20160213_Bena.txt9B_20160213_Bena.txt. Average power was 560W with a maximum of 1232w.

No problem till battery went under 20%. Then if forced a titlback so speed was reduced to about 12.5 km/h.

Battery recovered when stopped and descending.

Temperature was fresh so I have no idea about temperature limits.

https://www.dropbox.com/s/kd2i2ncuzf2lfej/9B_20160305_130242.txt?dl=0

The pitch is very interesting, most spikes are tiltback related to speed. The more sustained spikes between 13' and 15' are battery related

Link to comment
Share on other sites

Sorry for the mess. First and second link are more or less same climb (descent a bit different). The Bena climb is slower and the other is the one at more than 20km/h.

As you may see average climb rate shoud be about 5% but first part is near 7%

Link to comment
Share on other sites

1 hour ago, Ecko said:

Oh i think you dont understood whatvi ment. Thats all clear, i wanted to know the limits of the Ninebot compared to the data read out from NB Metrics.... Is it ok if i have over 600 watt in peaks (like jumping from curbs) or over 500 watt continious at accelerating.... How much Ampere are to much? How much voltage is to much? When will i be in an critical range? 

9B E+ batteries and motor are absolutely in normal range with around 500W continously. Batteries should be able to deliver 1000W continously (have to look at the cell specification in craniums thread again, but i am quite sure). 2000W should be ok for 2-3 secs (for the batteries).

Max peak for the motor was about 1500W - i read sonewhere imho.

At my last ride i was always under 1500W - the highest peak was about 23 A. "Normal" peaks about 15A.

The problematic limits are imho temperature - i had once a 15m incline with 10-15 degrees temp rise. Unfortionately 9BMetrics did not log the current then...

The other incline test i managed to somehow not save the log ? This was a couple of kms going up some 100ths of meters - i will repeat this once and show currents/power and temp.

The second problematic situation could be with not so full/?stressed? batteries - with voltage drop and/or batteries not able to deliver the current anymore...

But i had never any probs till now... And thanks to @Paco Gorinalogging times begun so we will soon know more!

Link to comment
Share on other sites

4 hours ago, Chriull said:

9B E+ batteries and motor are absolutely in normal range with around 500W continously. Batteries should be able to deliver 1000W continously (have to look at the cell specification in craniums thread again, but i am quite sure). 2000W should be ok for 2-3 secs (for the batteries).

Max peak for the motor was about 1500W - i read sonewhere imho.

At my last ride i was always under 1500W - the highest peak was about 23 A. "Normal" peaks about 15A.

The problematic limits are imho temperature - i had once a 15m incline with 10-15 degrees temp rise. Unfortionately 9BMetrics did not log the current then...

The other incline test i managed to somehow not save the log ? This was a couple of kms going up some 100ths of meters - i will repeat this once and show currents/power and temp.

The second problematic situation could be with not so full/?stressed? batteries - with voltage drop and/or batteries not able to deliver the current anymore...

But i had never any probs till now... And thanks to @Paco Gorinalogging times begun so we will soon know more!

If 9BMetrics has recorded Energy it is possible to recover Current *aprox

1. Derive Energy. Take (E(i+1)-E(i))/(t(i+1)-t(i)) = Power at (t(i+1)+t(i))/2

2. Compute V at that time by intepolating nearer values

3. I = Power/V

In newest versions I Store Energy. At that moment I Don't remember in the 1.0. Have to look at the sources.

Think it is not perfect but reasonable.

Link to comment
Share on other sites

Hi  Paco

 

Request, is  here any chance of automatic reconnection after the bot is turned off and on.

It asks which NB I would like to connect to after an, on off, since I only have one, this isn't so necessary

I have to do a lot of portages.

Nonetheless works splendidly. 

Merci beaucoup

Jer

Link to comment
Share on other sites

FYI: Battery update from my last post. Thanks to @Craniumthe datasheets from the 9B E+ batterie cells ( LGDDBMG11865 which translates to LG 18650 2900mAH batteries (2850mAH nominal)):

 

So the datasheet "only" says standard Discharge 0.576A and max discharge current 10A. So since there are 2 cells in parallel, 15 in serie we have ~55,5V nominal Voltage and 20A max current -> 1110W maximal "safe" power from the batteries. Unfortionately i did not find anything from the 20A for 2 secs per cell - was maybe from the 9B P which uses(used) the LG HG2 cells...

So the 1110W / 20 A should be considered as max "safe" rating. But it's no problem to get a higher loads with the E+: http://imgur.com/lduqAeI

45 minutes ago, jer said:

...Request, is  here any chance of automatic reconnection after the bot is turned off and on.

It asks which NB I would like to connect to after an, on off, since I only have one, this isn't so necessary...

Should be included in the next version:

Unfortionately we have by now 2 9BMetrics threats.... ;(

Link to comment
Share on other sites

... and my ideas/wishlist :D for the next version(s) of 9BMetrics:

- the new high speed sampling! Maybe with zipped export (or some compression the iphone sdk hopefully supports) or binary export (gnuplot supports binary datafiles - double float (64 bits) e.g. for the timestamp (do not know if it could interpret an uint64 containing fractions of seconds) and values could be saved as raw data...)

-- for this also the possibility of "oversampling" would make sense: To save for the values just the average, min and max for a given time period.

-- An "save detailed data" option: So 9BMetrics just holds in memory the last x minutes of detail data and if this "button" is pushed the detail data gets saved (in case of an shutoff or special situation one wants to know everything in detail), if this "button" is not pushed only the low detail average,min,max data gets saved.

- Your derived Energy idea sounds very nice - should/could be a way to a more accuraty "battery capacity" value?

Link to comment
Share on other sites

Hard work and next version near finished.

The gnuplot experts would say if the new format for files is more useful for  making the graphics.

Changes include :

  • Bug correction. Explicitly the "no current" one
  • Faster sampling
  • Possibility of changing the rinding level directly from the application
  • Much better dashboard for when recording
  • Reconstruction of connection process to run independent of User Interface and have better reconnection
  • GPS data is recorded with the workout each 5 m. It is good for long runs, not so good for slaloms etc. (anyway think GPS precision may be about 6 m, sometimes 3m but sometimes 10m). Does not load too much a modern iPhone as uses deferred recording. Tested in a running and skiing application and loads about 6% battery in 1hour for an iPhone 6s.
  • Much better exporting of data
    • Full session data is exported as a .zip 
    • There is a file for each variable. I removed the center column with the variable code and changed time to shorter ones as seconds from beginning of recording. First line shows file version, variable and first time date. That way size is much smaller.
    • There is a .gpx file with the track that may be uploaded to different sites or presented with .gpx applications
    • Better graphics and better selection computes statistics of selection between bars.
    • Export option in graphics which allows to export just the selection between the vertical cursors to a .csv file easy to import in excel

I enclose a link to a directory with :

  • a .zip file exported with GPS data
  • a longer .zip file exported without GPS data
  • a part of the second file selected from the graphics around a tilt back from excess speed
  • the selection imported to Excel and presented as a Graphic

 

https://www.dropbox.com/sh/t7wgzwf6vt5j19m/AACwQRWyD3oZMoWp8RnKiyY7a?dl=0

Link to comment
Share on other sites

Impressive! We did not expect so much improvements! Really looking forward testing this new version. 

One question : are you sure that increasing the sampling rate is safe ? Couldn't it saturate the processor which is - marginally ;) - used to keep us alive ?

Link to comment
Share on other sites

I have not found any problem, must be tested.

I believe problems with applications are more related as having the phone in the hand than the application. The Ninebot applications flushes the wheel with requests without waiting for answers :(

 

 

Link to comment
Share on other sites

I had forgotten that you had analysed the official ninebot app. I share your wise risk analysis ! For my part, I never use my phone while riding. I have a single core, mono task brain ;-)

Link to comment
Share on other sites

15 hours ago, Paco Gorina said:

Hard work and next version near finished.

Great and thanks again!

15 hours ago, Paco Gorina said:

The gnuplot experts would say if the new format for files is more useful for  making the graphics.

I had the chance to revive my old Mac Mini (upgraded to  El Capitain and installed the new xCode - it took forever.. ;( ) and tried your last github source. Looks already very fine!

The csv export is great for gnuplot! Graphs can be made now without awk, just "gnuplot -c 9b-new.dem 'Datafilename.csv'" is enough and in the gnuplot script 9b-new.dem the values can now be identified by their names from the title row.

15 hours ago, Paco Gorina said:

Changes include : ...

Wow - really impressive list! I can't wait to try this new version!

9b-new.dem

Link to comment
Share on other sites

chriull compiled your 1.1 version from github for iOS9.0 for me. I repacked the .app file to a .ipa and installed it manually on my jailbroken iphone 6 on 9.0.2. it works perfectly fine. love your new dashboard btw :) 

Link to comment
Share on other sites

  • esaj pinned this topic
  • Chriull unpinned this topic

Archived

This topic is now archived and is closed to further replies.

×
×
  • Create New...