Jump to content

Safety speed with your EUC


volavoile

Recommended Posts

Hello,

French Wheeler, but not as good as some other, I had a bad accident (broken shoulder, now with some metal!!) with my small ninebot one E+... and I spent some time in order to understant why we face cut out or accidents.

So, on the forums, I could recognize 4 kind of accidents :

 

1- Pilot did ignore alarm.... nothing to say.... too optimistic,

2- driving accident (bumps....),

3- Sudden cut off.... (looks like my accident, but I'm not sure it wasn't a technical failure),

4- Technical failure. Rare since construcors did improve the safety...or correct bugs (BMS for instance).

 

For the point 3, I did some calculation, that shows that specification of constructors seems to be far too optimistic, and the speed alarm does not consider the wheight of the pilot... which has a huge impact.

What I did consider is a common case of accident : you run on flat road, no alarm, straight... and you have a small slope (not big, couple of meters long!).

So immedialtely, since you do not consider you have to slow dow, your Wheel has to give all power in order to maintain the speed during the slope.

In this condition, considering a 10% slope (not big at all, it is a 6° angle road), what is the maximum speed you can run?

I did consider air drag, which has a big influence over 30 km/h), the solpe that I consider as a safety margin, and the wheight of the pilot.

The power was considered as constant whaterver the speed is (optimistic), and, since nobody know the duration of the peak's power, 0,5 second for this peak (which is only use for bumps in fact!).

Yield of the Wheel in order to transform electrical energy to mechanical energy is estimated at 75% (tests on some gotways).

What do we calculate show that with a 500 W, at 25 km/h, you're not save at all... except if you are a kid.

On a gotway ACM our MS3, 45 km/h is not a realistic speed since air drag consume most of the power...

What do you think?

Conclusions :

- Power, and slow speed is the key for safety...

- Adapt speed limit cosidering your wheight,

- If you want to keep the hight speed limit, go quiet, and in all case, use protections!!!

Of course, this is only my caclulation, an you can disagree!

EUC safety speed.xlsx

safety speed EUC.png

Link to comment
Share on other sites

In the spread sheet, the power(500W, 800W,,,,2000W) are not referenced except for legend.
When I alter the power, or even delete the power, the curve is not changed at all.

There are many recursive references. For instance, how cell B6 is calculated? 
View from the formula, it says B6 is from B88, plus or minus a very small number, but B88 is equal to B6!

Link to comment
Share on other sites

29 minutes ago, zlymex said:

In the spread sheet, the power(500W, 800W,,,,2000W) are not referenced except for legend.
When I alter the power, or even delete the power, the curve is not changed at all.

There are many recursive references. For instance, how cell B6 is calculated? 
View from the formula, it says B6 is from B88, plus or minus a very small number, but B88 is equal to B6!

It is an iterative calculation where circular references are on purpose.

Link to comment
Share on other sites

16 hours ago, volavoile said:

Hello,

French Wheeler, but not as good as some other, I had a bad accident (broken shoulder, now with some metal!!) with my small ninebot one E+... and I spent some time in order to understant why we face cut out or accidents.

So, on the forums, I could recognize 4 kind of accidents :

 

1- Pilot did ignore alarm.... nothing to say.... too optimistic,

2- driving accident (bumps....),

3- Sudden cut off.... (looks like my accident, but I'm not sure it wasn't a technical failure),

4- Technical failure. Rare since construcors did improve the safety...or correct bugs (BMS for instance).

 

For the point 3, I did some calculation, that shows that specification of constructors seems to be far too optimistic, and the speed alarm does not consider the wheight of the pilot... which has a huge impact.

What I did consider is a common case of accident : you run on flat road, no alarm, straight... and you have a small slope (not big, couple of meters long!).

So immedialtely, since you do not consider you have to slow dow, your Wheel has to give all power in order to maintain the speed during the slope.

In this condition, considering a 10% slope (not big at all, it is a 6° angle road), what is the maximum speed you can run?

I did consider air drag, which has a big influence over 30 km/h), the solpe that I consider as a safety margin, and the wheight of the pilot.

The power was considered as constant whaterver the speed is (optimistic), and, since nobody know the duration of the peak's power, 0,5 second for this peak (which is only use for bumps in fact!).

Yield of the Wheel in order to transform electrical energy to mechanical energy is estimated at 75% (tests on some gotways).

What do we calculate show that with a 500 W, at 25 km/h, you're not save at all... except if you are a kid.

On a gotway ACM our MS3, 45 km/h is not a realistic speed since air drag consume most of the power...

What do you think?

Conclusions :

- Power, and slow speed is the key for safety...

- Adapt speed limit cosidering your wheight,

- If you want to keep the hight speed limit, go quiet, and in all case, use protections!!!

Of course, this is only my caclulation, an you can disagree!

EUC safety speed.xlsx

Dear @volavoile,
thank you very much for that analysis of uphill sustainable speeds as a function of power and weight.

I have a few questions (of course, wouldn't I :rolleyes:!!).

1) Did you really implement the 75% yield? I could not find it anywhere in your excel sheet 'Speed limit'.
2) The terms for air resistance are of the form =0,5*1,7*0,5*(C7/3,6)^3*1,2. I assume

  • 0,5 (1/2 factor in formula)
  • 1,7 (I don't understand this. Is it the area of a person in m*m ? If so, I think it's too big a value: 1,7m tall and 1m wide?!.)  
  • 0,5 (this could be the cW value, although I found 0,78 for upright person listed in Wikipedia),
  • (C7/3,6)^3 (This should have an exponent of "2" instead of "3". Or am I wrong?)
  • 1,2 (This must be air density, I found the value of 1,25, but that's close enough.)

I started another calculation of the same thing but initially based on force not on power. I also used an analytic instead of iterative approach and some other table tools in Excel. Here are my results. They come actually quite close to your numbers. Perhaps you could have a look those?

slope1.pngslope2.png

uphill.xlsx

Link to comment
Share on other sites

6 minutes ago, RenaissanceMan said:
7 minutes ago, RenaissanceMan said:

Did you really implement the 75% yield? I could not find it anywhere in your excel sheet 'Speed limit'.

It is in the second datasheet "Emergency braking calculation"

Link to comment
Share on other sites

23 minutes ago, RenaissanceMan said:

I started another calculation of the same thing but initially based on force not on power. I also used an analytic instead of iterative approach and some other table tools in Excel. Here are my results.

I think you are very good at physics and Excel. In order for further analysis and assumptions, I attach one of my real test chart/data of battery voltage, battery output current and speed, when I rode my Gotway V3 climbing up and down a hill of about 10% slope. I'll come back tomorrow as It's getting very late here.
Chart3.gif

2016_10_23_22_08_13.xlsx

Link to comment
Share on other sites

 

4 hours ago, volavoile said:

I'll work on your file...

Just to let you know!

0.5 = 1/2

0.5 = S=area (m²)

1.7 = Cx

 

F = 1/2 * S*Cx*V²

Ok, so the large difference is in the cW alias Cx value. That should be the form specific resistance factor. You say it is 1.7, my value is closer to 0.8, about half of it. Where did you get Cx from?

I made another attempt on velocity as a function of power and weight using Maxima. Here is the plot, below are the source files.

 

Clipboard14.png

uph4.pdf

uph4.tex

uph4.wxmx

Link to comment
Share on other sites

:)

4 hours ago, zlymex said:

I think you are very good at physics and Excel. In order for further analysis and assumptions, I attach one of my real test chart/data of battery voltage, battery output current and speed, when I rode my Gotway V3 climbing up and down a hill of about 10% slope. I'll come back tomorrow as It's getting very late here.
 

Thanks for the file and your kind words <smiley should go here>!

I will also try to get back to this tomorrow evening.

Link to comment
Share on other sites

13 hours ago, RenaissanceMan said:

 

(Please ignore the empty ref above, which I sometimes cannot reply without it)

Attempting to calculate the safety speed is a very good and important thing, I appreciate very much for anyone who is trying so.

I agree with @RenaissanceMan that start the calculation based force not on power. Power is more complicated, I can identify five:
#1. Motor rated power. 
This is a fixed value. In the case of Gotway MS3, it is 1500W. This is the max. continuous output power at a given set of parameters(at rated voltage, rated speed and temperature etc) above which the motor will be over heated.

#2. EUC peak power
May be called Motor peak power, but often unknown, vary according to speed, and is very difficult to test. Gotway only specify as greater than 3000W, but how much greater?

#3. Battery output power(and is the board input power, ignoring the wire loss)
It is an instant value(varies all the time) and is the product of battery current(blue curve of my chart) and battery voltage(red curve). There is a column(T) in my spreadsheet for that.

#4. Board output power(and is the motor input power, if wire loss ignored)
It is an electric power, can be calculated as the product of the armature current(green curve below) and the motor voltage. It is also difficult to obtain because the motor voltage is an PWM signal and three phased.

#5. Motor output power
It is a mechanical power at the shaft(not at wheel brim). It is an instant value too. The effeciency of the motor(at a given condition) is #5 divided by #4.


On the other hand, there are many forces as well
#6. hill drag(or hill push)
equals to weight * g * sin(alpha), where g is the gravity constant, alpha is the slope angle(negative when down-hill), weight is the total weight including EUC.

#7 air drag, although proportional to the square of the relative speed(not necessarily EUC speed to the ground), proportional to the air density, proportional to the section area, but also proportional to the drag coefficient. The later is probably difficult to calculate, but can be determined by experiment.

#8 Friction forces
Such as tire friction. According to my test, these forces are small compare to #6 and #7, but may not be fixed with speed. There is a portion of which is proportional to the speed.

#9 Accelerate and deceleration force
can be calculated by the total mass times the acceleration.
The acceleration force may be very large and evidenced by my tests for even mild acceleration. There are many faceplant that happened because of acceleration.

#10 ?


All these forces, algebraically sum up, will applied through the rider's feet to the pedals in order to get the driving(or braking) force/torque necessary to overcome the force sum, and the EUC will counter-react by producing equal but opposite torque through the application of ever changing armature current. If the armature current needed beyond the limit of the EUC, a "cut off" will happen. The limit can be determined by the speed-torque curve which varies according to EUC, battery level, temperature etc.

One thing make the calculation of safety speed very difficult is the riding skill. For instance, whether the rider choose a good path when an uneven road ahead, whether the rider relaxes(or what ever) when come across a bumper or a ditch, whether the rider brake or lean back a little when there is a short but steep slope or a gust of wind. One example is @EUC Extreme, he can ride very fast but still maintain balanced on the EUC, so the safety speed for him is at the high end.

My method is to look at the torque spike(equivalent to the armature current spikes) and determine the torque reserve, but this only apply to my skill. I hope more people upload their data/chart recorded by Wheellog or similar apps in various riding situations so that a better understanding can be obtained for different EUC and riders.

The chart below is almost the same as the previous one with added blue curve for armature current. Data is from Wheellog(already in my previously uploaded spreadsheet) which in turn measured by current sensor ICs directly in series with the armature.

Chart4.gif

Link to comment
Share on other sites

Here is another graph. It shows the theoretically admissable maximum safe speed when entering a climb of 10% for a given weight (wheel plus rider) with a yield of  70% (i.e. assuming of the 100% energy that comes out of the battery, 70% are converted into mechanical and kinetic energy. (Is this yield realistic? Can somebody pls confirm/deny?)

z -axis and color code denote max speed in km/h.

Clipboard12.png

Maxima code:

plot3d(velocity(P*0.7,m),[P, 500, 2000], [m, 50, 130], [grid, 15, 15], [legend, false], [elevation, 0],
    color_bar, [xtics, 200], [ytics, 5], [ztics, 5], [color_bar_tics, 2], [mesh_lines_color, true],
    [title, "Safe speed for running into a 10% climb at given weight and power available (70% yield)"],
    [xlabel, "Power [Watt]                                                   "], [ylabel, "Weight [kg]   "],
    [palette, get_plot_option(palette, 3)]);

Link to comment
Share on other sites

We agree!

what a pity seeing people commiting suicide by trying to reach possible speed on their wheel. Finding the limit, why not. Reducing the safety margin... ok.... but why ca they remove  hard limits?

No one will never go over 48 with a not modified MS3.... why do they try?

Link to comment
Share on other sites

For the benefit of myself and more recent visitors to this forum (and this thread in particular) I collected a few references on previous posts to this subject matter by a forum search for "motor yield". These I found anticipating the discussion at hand and in general substantiating the findings presented here. They all emphasize the inherent safety speed limits based on battery level and other factors. Enjoy the reading AND speed conscious riding :rolleyes:!

 

 

 

Link to comment
Share on other sites

  • 1 year later...

Looking for a little guidance...It's my understanding that everyone has a different riding style, weight etc and therefore it's nearly impossible to predict when you are riding unsafely. The EUC's have built in safety alarms which beep and tilt back but these alarms are user defined (I'm not an expert) so I'm struggling to trust those. I would think a better way to approach the question of "am I riding unsafely?" would be to analyze something like wheel log data (as above). I just started to play around with the data so my question is what is the best way to look at this data and tell if you are pushing the wheel close to it's limit? I'm starting to play around with speeds near and soon to be above 40 kph so I want to do it carefully. I've attached my first stab at it. This is a KS18S and in my mind I'm not even getting close to pushing the capability of the wheel...am I wrong? To me it looks like an approximately 600-800 watts sustained motor output, and less than 20 amp draw on the batteries. KS18S (1680 wh) should be able output 1500 watts and 30 amps...right?

KS18S Log 06JUN18.pdf

Link to comment
Share on other sites

Personally, I ride like the wheel could cut-off at any time, not just because you are near its maximum power output.

Consequently, I minimize the time I go faster than I'm prepared to faceplant.

Probably because I'm an electrical engineer, and just "can't stick my head in the sand" knowing all the single point failures that could cause a faceplant (and not so infrequently, do!)

Link to comment
Share on other sites

14 hours ago, DaveThomasPilot said:

Personally, I ride like the wheel could cut-off at any time, not just because you are near its maximum power output.

So you ride at 10mph max?

Link to comment
Share on other sites

On 6/7/2018 at 5:44 PM, DaveThomasPilot said:

Personally, I ride like the wheel could cut-off at any time, not just because you are near its maximum power output.

Consequently, I minimize the time I go faster than I'm prepared to faceplant.

Probably because I'm an electrical engineer, and just "can't stick my head in the sand" knowing all the single point failures that could cause a faceplant (and not so infrequently, do!)

I understand your point, but I am also an engineer and know that things are designed to a certain design criteria. How many single point failures does an airplane have yet it's incredibly well engineered and safe.  Far better engineered than these wheels are for sure! I am certain that cutout is very much a function of the demand you are putting on the system through load, speed, acceleration demands etc...The motor on a KS18S is supposedly 1500 watt continuous output 3600 peak whatever that means, so you'd think they engineered the other components to more less supply and handle that kind of output. I understand that gravity is about the only thing we'll ever see that NEVER fails so your point is well taken...I try and always gear up in preparation for that time when something does fail. I've had a cutout on an NB1 E+ without any protective gear on and it wasn't much fun! ?

Link to comment
Share on other sites

On 6/7/2018 at 6:07 PM, Evel_Knievel said:

I would think a better way to approach the question of "am I riding unsafely?" would be to analyze something like wheel log data (as above)

If one just considers operator faults (accelerating into the torque speed limit of the EUC) an current over speed plot can show how near one gets to the unsafe area. Should be normalized to the battery voltage, since this limit moves proportionally with the voltage.

One can also compute a "time until overlean" assuming constant acceleration - unfortionately one gets just about 5 samples per second from the wheels, so this value is varying quite strongly and averaging does not improve this with this low sampling rate (imo - did not invest time for this topic lately)

My thoughts about this can be found in 

 

Link to comment
Share on other sites

Archived

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

×
×
  • Create New...