Paco Gorina Posted March 15, 2016 Share Posted March 15, 2016 The other day I suffered an inconsequent Overlean. There was a step apron 25/30cm height not absolutely step (90º) but quite . I was approaching it slow and result is I overleaned but was capable or running it (as speed was very slow) Anyway it is interesting to analyze the wheel magnitudes : Pitch begins to increase and current and power jump. As there is no way to maintain it and I jump out current and pose go down and as the wheel falls speed goes up till the controller stops it. If you look at the values of the Excel file Pitch continues to increase after maximum current and power (so overlean). Unfortunately sampling period is only 0.2s aprox. but we can see it is not a shutdown because all data is running till the wheel is flat. Probably if we have more of this logs for accidents it would be possible to gain a lot of insight. Unfortunately 9BMetrics is stopped for the moment in Apple Store Review as they need a Ninebot to test it. Will see if a simulator is enough for them but you may download the sources and compile yourself from : https://github.com/fgorina/9BMetrics Overlean.xlsx Link to comment Share on other sites More sharing options...
esaj Posted March 15, 2016 Share Posted March 15, 2016 2 hours ago, Paco Gorina said: The other day I suffered an inconsequent Overlean. There was a step apron 25/30cm height not absolutely step (90º) but quite . I was approaching it slow and result is I overleaned but was capable or running it (as speed was very slow) Anyway it is interesting to analyze the wheel magnitudes : Pitch begins to increase and current and power jump. As there is no way to maintain it and I jump out current and pose go down and as the wheel falls speed goes up till the controller stops it. If you look at the values of the Excel file Pitch continues to increase after maximum current and power (so overlean). Unfortunately sampling period is only 0.2s aprox. but we can see it is not a shutdown because all data is running till the wheel is flat. Probably if we have more of this logs for accidents it would be possible to gain a lot of insight. Unfortunately 9BMetrics is stopped for the moment in Apple Store Review as they need a Ninebot to test it. Will see if a simulator is enough for them but you may download the sources and compile yourself from : https://github.com/fgorina/9BMetrics You could probably ask @Kevin what he did to get the custom Xima Lhotz app through, maybe he sent a video of it or something. I recall he was having the same problem. 2 hours ago, Paco Gorina said: Overlean.xlsx Is the voltage curve missing from the graph or am I blind? Link to comment Share on other sites More sharing options...
jer Posted March 15, 2016 Share Posted March 15, 2016 3 hours ago, Paco Gorina said: Impressive stuff, I would love to have this. Thanks Jer Link to comment Share on other sites More sharing options...
Paco Gorina Posted March 15, 2016 Author Share Posted March 15, 2016 3 hours ago, esaj said: You could probably ask @Kevin what he did to get the custom Xima Lhotz app through, maybe he sent a video of it or something. I recall he was having the same problem. Is the voltage curve missing from the graph or am I blind? Thanks, I will ask Kevin. The voltage is out of scale (over 50-60 V) but precision of voltage seems quite low. Link to comment Share on other sites More sharing options...
Chriull Posted March 16, 2016 Share Posted March 16, 2016 Just out of curiosity: The negative current peak was from (regenerative) braking before going up the step? 0,2 secs is the maximum sample rate of your application? Did you ever had (constant) readings of more then 25 A? Or could this just be some "soft shutoff" limit of the motherboard? What do you mean with "precision of voltage" - is has too low resolution or the absolute values seem off? I can't wait until your app gets available in the store - looks really great! To compile it myself and use it on my own iphone (before the approval process) i need to enroll to the "apple developmer programm" and pay the yearly subscription amount of USD 99? Link to comment Share on other sites More sharing options...
Paco Gorina Posted March 17, 2016 Author Share Posted March 17, 2016 >>The negative current peak was from (regenerative) braking before going up the step? Yes, I reduced speed before the step. >>0,2 secs is the maximum sample rate of your application? 0,2s is the answer from trhe NB. I am asking for 0.01s but unfortunately is what I get. I will see if it can be better but really had no time for the moment. To be fair may be also a problem of the iOS Loop. Timers go in the event loop and I really don't know how fast it is. >>What do you mean with "precision of voltage" - is has too low resolution or the absolute values seem off? It means the precision. That is the number or decimals it has. Absolute values are OK (if you look at the Excel they are OK, I just scaled the graphic diferently), and for the moment I don't have more time resolution. >>To compile it myself and use it on my own iphone (before the approval process) i need to enroll to the "apple developmer programm" and pay the yearly subscription amount of USD 99? Probably not, just download XCode from the AppStore, clone the fit repository and give it a try. It uses iCloud to access iCloud Drive but it is posible to comment it so you should not have any problem. Also in the git theere are 2 branches, AppStore_1.0 is the one that is for aproval at AppStore, master is the continuing developement and has been much changed in the graphics and sending riding level and max speed to the Ninebot Link to comment Share on other sites More sharing options...
Paco Gorina Posted March 18, 2016 Author Share Posted March 18, 2016 Made some tests and got sampling periods of aprox 0.05 s (varying) by changing to another thread and just don't waiting. Just made preliminary tests so I don't know if it has any impact in the wheel performance Link to comment Share on other sites More sharing options...
Paco Gorina Posted March 21, 2016 Author Share Posted March 21, 2016 On 16 de març de 2016 at 7:47 AM, Chriull said: Just out of curiosity: The negative current peak was from (regenerative) braking before going up the step? 0,2 secs is the maximum sample rate of your application? Did you ever had (constant) readings of more then 25 A? Or could this just be some "soft shutoff" limit of the motherboard? What do you mean with "precision of voltage" - is has too low resolution or the absolute values seem off? I can't wait until your app gets available in the store - looks really great! To compile it myself and use it on my own iphone (before the approval process) i need to enroll to the "apple developmer programm" and pay the yearly subscription amount of USD 99? V 1,0 ready IN App Store Link to comment Share on other sites More sharing options...
Recommended Posts
Archived
This topic is now archived and is closed to further replies.