Jump to content

Breaking crash analysis


Paco Gorina

Recommended Posts

 

8 hours ago, Frode said:

To me it looks like you will have it for only about half a second, and it will only be able to take away between 10% and 15% of the velocity energy before something brakes down.

Unlikely that producers would even bother to implement active braking if this would be the case.

Quote

It is very strange that the pitch increases that fast when (if) the braking force was increased 2-3 times.

If you overlean the wheel it is perfectly understandable that the pitch increases. So it is entirely consistent that we see maximal force and pitch increase, when the driver applies too much leaning.

8 hours ago, Frode said:

He must have pushed hard, much harder than in the other no-crash case.

The difference between overlean with crash and everything is fine can be arbitrary small. Either you go over the cliff or you don't and any small difference can bring you over the cliff. In the crash case the wheel was faster, probably the battery was hotter (after all it was the last of several tests), and looking carefully at the figure reveals that the wheel was already in more serious tilt-back while the speed was still increasing the initial push to brake was maybe a little harder, as it lead to a notable acceleration, which was not the case in the OK trial. 

It seems pretty clear that the current flowing between point 4 and 5 is not used for acceleration, because the wheel is still decelerating and we can safely assume that the driver is pushing the wheel with increasing force from behind forward before he goes buttplanting.

Link to comment
Share on other sites

  • Replies 53
  • Created
  • Last Reply

@Mono I agree with your analysis. I think there is no active braking but there is tiltback for too much speed mixed with the sequence of events. 

I am trying to compare with other cases with tiltback. Tiltback is a mixed blessing. It tells you that you must slow but uses some moment from the motor.

Also agree that difference between no problem and crash is like a razor blade , just a very small difference.

 

Link to comment
Share on other sites

6 hours ago, Paco Gorina said:

I think there is no active braking

Sorry if I have missed it, but how do you then explain the graph starting from t=0.8, where we see a positive current growing from 10A to 50A, associated with a deceleration from 22km/h to 18km/h within a little less than 0.2 seconds? Would you agree that the deceleration cannot be (re-)generative due to the sign of the current? Then where does the deceleration come from?

Link to comment
Share on other sites

4 hours ago, Paco Gorina said:

Tiltback is a mixed blessing. It tells you that you must slow but uses some moment from the motor.

I was in the belief that this must be a minor amount of energy. Am I wrong in that keeping a given angle costs always the same amount of energy irrespectively of what this angle is?

My main problem with tiltback is that it makes it more difficult to brake without (getting the feeling) to fall/slide over behind the wheel. For any given tiltback angle set-point one can eventually only slow down by further exceeding this angle, which I find at times scary beyond the comfort level.

Link to comment
Share on other sites

9 hours ago, Mono said:

My main problem with tiltback is that it makes it more difficult to brake without (getting the feeling) to fall/slide over behind the wheel.

Really! My main problem with tilt back is 50% of the time it throws me violently off.  The other 50% I have an "oh shit" moment, and almost get thrown off.  I know why it's there, but on my wheel it's quite violent.  I guess with a bigger battery and more conservative 3rd speed warning setting, and not accelerating too fast, or braking too hard, or respecting, low battery limitations, I'll be good to go. ?

Link to comment
Share on other sites

12 hours ago, Mono said:

My main problem with tiltback is that it makes it more difficult to brake

I have never hit tiltback at the top speed, (too scared to fall off at 15 mph)  but if I turn on the speed limiter (it only lets me go a little over 6 mph before tiltback) I get strong tiltback and the wheel slows down by itself, I guess it's in the firmware.  The fastest I have gone was 14.6 mph according to the app but I immediately slowed down some.  I only wear a helmet and wristguard gloves, to go that fast I would wear knee and elbow protectors but I don't have those, I usually go about 10mph max.

Do you mean you only get tiltback and to slow down you have to lean back even further?

3 hours ago, Smoother said:

My main problem with tilt back is 50% of the time it throws me violently off.  The other 50% I have an "oh shit" moment, and almost get thrown off.  I know why it's there, but on my wheel it's quite violent.  

Is the tiltback stronger the faster you are accelerating up to the tiltback point?  If you are just slowly getting faster and faster up to the top speed, does the tiltback become more of a soft tiltback?  With the speed limiter on, the tiltback seems to vary according to how fast I accelerate through the speed limiter on my ninebot one E+

Link to comment
Share on other sites

4 hours ago, Smoother said:

Really! My main problem with tilt back is 50% of the time it throws me violently off.  The other 50% I have an "oh shit" moment, and almost get thrown off.  I know why it's there, but on my wheel it's quite violent.  I guess with a bigger battery and more conservative 3rd speed warning setting, and not accelerating too fast, or braking too hard, or respecting, low battery limitations, I'll be good to go. ?

Sorry to hear, that doesn't look like a good implementation of tilt back at all then.

Link to comment
Share on other sites

37 minutes ago, steve454 said:

Do you mean you only get tiltback and to slow down you have to lean back even further?

Yes, this is how it works: tilt back changes the setpoint at which the wheel keeps the speed constant. If the driver does not react (keeps the pedals more or less level), the wheel starts to accelerate moving in front of the driver thereby creating a lean back situation. Then the driver has to react (quickly) to slow down the wheel beyond the current setpoint (i.e. possibly quite strongly), thereby initiating the slow down and preventing a buttplant. If the driver acts instantaneously on the setpoint change, nothing happens accept that the pedal tilt changes. This is what usually happens to me now, after I got used to cruise within the tilt back speed. 

AFAICS the wheel cannot limit the speed by itself if the driver does not comply, unless it gives up balancing the driver. 

44 minutes ago, steve454 said:

Is the tiltback stronger the faster you are accelerating up to the tiltback point?  If you are just slowly getting faster and faster up to the top speed, does the tiltback become more of a soft tiltback?

Possibly, but I am not sure about any of this.

Link to comment
Share on other sites

3 hours ago, steve454 said:

 

Is the tiltback stronger the faster you are accelerating up to the tiltback point?  If you are just slowly getting faster and faster up to the top speed, does the tiltback become more of a soft tiltback?  With the speed limiter on, the tiltback seems to vary according to how fast I accelerate through the speed limiter on my ninebot one E+

That's a very good question.  I know I got a violent one, accelerating away hard from rest with a low battery.  It literally shot out in front of me and dumped me on my ass and wrists and it flew 20 feet down the road.  That was over 3 weeks ago and I still can't put full pressure on my left wrist.  My interpretation of the events: accelerating hard put a current draw spike on my low battery that it couldn't supply (340wh). BMS or cobo sent low battery and tilt command at almost the same time.  Wheel has to accelerate even harder than me to initiate tilt back, why?  the reason it is severe is because it has to get the contact patch of the tire in front of my centre of gravity, which is already way ahead of the contact patch because of the extra leaning needed for my rapid acceleration.  It really felt like someone pulled the rug out from under my feet. I felt like Charlie Brown trying to kick that damn football. Butt plant

i think I had a mostly violent, high speed tilt back the other day.  Battery fullish, speed gps at around 26/27kmh gps I think I was going at a reasonably steady speed. I certainly WASN'T accelerating.  Suddenly out of nowhere, beeps, and an immediate wild tilt back that almost threw me down the road, and I do mean public road with other vehicles. I have it on film, with gps speed in corner of screen.  I don't know why it had to be so violent, except that my max was set to 30 app ( which is not a real 30, more like 27 I guess) which is the wheels max, so if it was going to get ahead of me before it ran out of head room it was going to have to do it quickly. Maybe if my max was set at 28 app it would still have room (30-28) to get ahead of me by a more relaxed tilt.  I'm quite reluctant to test this theory, as you can imagine.  Despite ATGATT this 54 year old body doesn't like being thrown Down the road like a bowling ball.  This should have been a gentle tilt back because I wasn't accelerating.  Maybe it WAS more gentle, because I didn't actually fall off, but it was close, but if that's gentle, then I'm the King of Siam. Etc,etc.

Link to comment
Share on other sites

45 minutes ago, Smoother said:

Wheel has to accelerate even harder than me to initiate tilt back, why?

To all I know the tilt back only changes the inclination angle setpoint where the speed remains constant. If this change happens and you keep the pedals untilted, the wheel will shoot in front of you, exactly as if you would have strongly pushed the front foot down (without leaning forward) at setpoint 0º inclination (without tilt back). On your wheel, this setpoint change apparently happens too quickly, ideally it should be smooth and slow enough to allow the driver react to it.

Link to comment
Share on other sites

1 hour ago, Smoother said:

 Maybe if my max was set at 28 app it would still have room (30-28) to get ahead of me by a more relaxed tilt.

That might be interesting to try a slightly slower max for awhile just to see if with, like you said, more reserve might lessen the severity of the tiltback.  Maybe, like @Mono said the setpoint change happens too quickly at the highest speeds.  

Maybe even set it to 24 for awhile.

What wrist protectors are you using, they must have worked pretty well.

Link to comment
Share on other sites

On 10.12.2016 at 6:31 PM, Mono said:

Unlikely that producers would even bother to implement active braking if this would be the case.

Why do you hold so strong belief that there is "active braking" implemented in the wheel?

 

Link to comment
Share on other sites

5 minutes ago, Frode said:

Why do you hold so strong belief that there is "active braking" implemented in the wheel?

All the data I remember to have seen suggest that some wheels do implement active breaking. Out of my head:

- braking on some wheels gets quite weak after a few hard braking attempts (suspected cause: heat dissipation problem)

- braking on some wheels gets quite weak under low battery (battery load problem)

- graphs from brake sequences show currents and voltages that are (only?) consistent with active braking

- a Solowheel representative was quoted in this forum that Solowheels do both, brake regeneratively and actively

- I believe I can feel the switch between generative and active braking on a Gotway

- IIRC I checked myself battery statuses after emergency braking exercises which were consistent with heavy loads

- I suspect that discharge currents tolerated by the batteries are larger than the tolerated charge currents

- a "first generation" IPS 132 has very weak brakes which would be consistent with only generative braking (I don't have any specific information as to whether this is the case or not)

- I suspect that decelerating firmly from 3km/h to zero (like for idling) isn't even possible purely generatively, which presents a smooth (developmental) path to active braking at any speed. 

I probably forgot some other points. 

Now, I am less convinced for what the Ninebot does, as I have much less information, but

- you would need to explain away the very high positive current under deceleration from 22 to 18km/h in a different way

- the Ninebot is what I would consider a "second generation" wheel from a comparatively big manufacturer, so I assume some level of sophistication (which doesn't mean perfection by any stretch)

 

Link to comment
Share on other sites

6 hours ago, Mono said:

Is the tiltback stronger the faster you are accelerating up to the tiltback point?  If you are just slowly getting faster and faster up to the top speed, does the tiltback become more of a soft tiltback?

Yes and I am quite sure it increases if you don't reduce speed (perhaps intgrates or low filter speed). 

Link to comment
Share on other sites

On 12/10/2016 at 0:56 AM, Frode said:

Energy is not cooked when you accelerate. It is much like charging a battery, but instead of converting it to chemical energy in a battery you convert it to speed (E = 1/2 m v^2).

Are you saying that the efficiency of an electric motor when accelerating is invariably 100%? To all I know, the efficiency (at max torque) goes to zero when speed goes to zero which seems to imply that most of the energy is cooked if we accelerate strongly from low or zero speed. This is consistent with the (thought) experiment where we generate a force against a wall at standstill without changing speed at all. In this case all used energy must be cooked.

Link to comment
Share on other sites

12 hours ago, Mono said:

- I suspect that decelerating firmly from 3km/h to zero (like for idling) isn't even possible purely generatively, which presents a smooth (developmental) path to active braking at any speed. 

I don't know in detail how they control these EUC motors, but think this is possible by manipulating the phase between current and voltage in the motor or by switching the mosfets. Remember motor windings are coils. If you load them with current and then swhich the current off abruptly it can generate a high enough voltage to charge the battery.

I beleive my KS-16 has regenerative braking only, Braking with full battery may give alarm, tiltback and loss of braking ability altogether, for example going down a steep slope. I have experienced that, and cannot find any other reason for the wheel behaving like it did (repeatedly braking for a short time and loosing brakes, resulting in mean speed increasing down the slope). I wish it had plugging braking or dynamic braking. It would have prevented problems like that, at least for long enough time to stop the wheel without having to jump of at speed.

10 hours ago, Mono said:

Are you saying that the efficiency of an electric motor when accelerating is invariably 100%?

No, but you compared this to cooking energy when pushing against a wall. You will have less than 100% efficency when accelerating, but not by much (unless you have a very bad controller). Electric motors are quite efficient decices, so talking about cooking away energy when you accelerate strongly as heat like happens when pushing the wheel against a wall is not a very good description. They are more like the opposite to each other. How could you else "accelerate strongly" if you cooked away a large part of the energy as heat?

What you said that i commented:

On 9.12.2016 at 1:53 PM, Mono said:

I guess the answer is similar to where are they cooked when accelerating strongly from low speed or when pushing backwards against a wall without accelerating at all?

 

Link to comment
Share on other sites

1 hour ago, Frode said:

I don't know in detail how they control these EUC motors, but think this is possible by manipulating the phase between current and voltage in the motor or by switching the mosfets. Remember motor windings are coils. If you load them with current and then swhich the current off abruptly it can generate a high enough voltage to charge the battery.

That's how i understood regenerative braking works from some links i read. The lower side mosfets are switched and by this the motor coils generate a high enough voltage (controlled by the switching frequency) so the battery can be charged. But i did not find anything about in which range this works, like for the 3 km/h mentioned by @Mono - but i also doubt this.

Normaly the motor is accelerated by generating a magnetic field which is a bit "before" the magnetic fields of the permanent magnets. So i always had the idea, that by generating a magnetic field which is a bit behind the permanent magnets the motor can be decelerated "actively" - but from my readings i reckon imho that this is what is called "plugging" breaking...

@Mono's points in http://forum.electricunicycle.org/topic/6022-breaking-crash-analysis/?do=findComment&comment=73356 are more or less why i did not exclude the possibility of plugging breaking - especially with the braking telemetry data posted here from  @Paco Gorina. ( 1) )
What speaks against plugging breaking is the vast energy that needs to be "burnt" (in the motor coils, battery and Mosfets) - normally a resistor is "used to burn the energy" with plugging breaking, which is not existent in EUC's...

The second point is, that the Mosfets would have to withstand up to more or less the double voltage (battery + back emf from the motor) - which they don't. Imho the mosfets used in the Ninebot One E+ are rated for somthing like max 80V.

So for "low" speed with back emf below (80V-battery voltage-safety margin) it could at least work for some short time?
All this would maybe also explain the many reports about burned motherboards of the Ninebot One P with strong braking?

Quote

I beleive my KS-16 has regenerative braking only, Braking with full battery may give alarm, tiltback and loss of braking ability altogether, for example going down a steep slope. I have experienced that, and cannot find any other reason for the wheel behaving like it did (repeatedly braking for a short time and loosing brakes, resulting in mean speed increasing down the slope). 

Could be! I never had breaking probs with my KS16 - but i also never started down a slope when it was fully loaded. Maybe also my charger is adjusted to a little bit lower voltage, so the cells never get charged 100% and there is always enough reserve for regenerative braking with the 828Wh pack... 

Unfortionately the KS16 does not report the direction of the current - so it is much harder to distinguish (imho only works by looking at the voltage - if there is a sag or the voltage gets higher. But that's quite imprecise)

Maybe @Zymlex could once record same data with his self made logger from some KS/Gotway braking and idling? Or maybe also from @Jason McNeil, if he has his logger still in use in some "modern" wheel?

Quote

You will have less than 100% efficency when accelerating, but not by much (unless you have a very bad controller). Electric motors are quite efficient decices, so talking about cooking away energy when you accelerate strongly as heat like happens when pushing the wheel against a wall is not a very good description. They are more like the opposite to each other.

BLDC's have max efficiency of somewhere between 60%-97% - don't know in which range the BLDC's for the EUCs are.

But at stall torque (zero speed) and no load speed (0 torque) the efficincy is 0%.

Here a graph of efficiency over speed:

http://www.pedelecforum.de/forum/attachment.php?attachmentid=9&stc=1&d=1216669492

and over torque:

BLDC-performance-graph.jpg

So at very low speed the efficiency is very low and only a small amount from the power delivered from the battery is converted to acceleration. The rest is "burnt" in the battery, the motor coils and mosfets...

Quote

How could you else "accelerate strongly" if you cooked away a large part of the energy as heat?

By supplying even more energy than the energy that is burnt... ;(

 

1) In @Paco Gorina's braking telemetry data there is quite some braking while the current is positive (discharging the battery). And this after a quick raise in pitch (from ~8 to 12°). My interpretation(s) of the graphs would be the following:

- This quick pitch increase is the more or less Paco loosing balance (starting to fall back). The next to be seen is the current changing from charging to discharging, the speed decreases and the pitch stays quite constant - Imho this is because the wheels tries everything to keep Paco upright but fails, since the imbalance was already to strong. So Paco falls back further (increasing pitch) and this goes on until BMS cut-off and the wheel does not resist Pacos fall anymore and is so accelerated by him.

- The current direction change and the pitch increase starts at the same time (3m41.85s) - so could also be that the firmware decided to stop regenerative breaking and so the pitch increased (Paco falling back by the wheel stopped braking). once the current got positive (~3m41.87s) the pitch increase stopped and deceleration started.
As Paco already mentioned, while this incident the data was delivered with a lower sampling rate than normal, maybe also the timely order was not precise and no real interpretation is possible?

Link to comment
Share on other sites

On 5.12.2016 at 11:25 AM, Paco Gorina said:

...
So we have a lower voltage (may have affected) but curiosly a big increase in sampling period.

I don't know if it is an artifact of the sampling and transmitting to 9BMetrics but linked to the crazy current reversal of the controller, one may think that the controller really choose a wrong way, was working with excesive delay and got messed.

...

 

Just a thought: I have an old log where i accalecerated my ninebot a bit to fast into the tilt-back. There seems to exist afirst stage of the tilt-back at about 7-8° pitch:

sjcSruZ.png

 

This is about the pitch angle at which at your incident the regenerative breaking stopped. And then there is the second point ~14-15° pitch after which everything got out of controll. This could be the second tilt-back stage? At my graph max pitch is "only" at around 12° - but maybe because i tried everything to keep balance and brake a little bit, so maybe i managed to stay just a bit below this threshold...

So because you drove faster than the 20 km/h were normaly tilt-back is about to start somehow the tilt-back handling routine kicked in while you were braking and so the normal breaking behaviour got totally messed up?!

PS.: Could also be pure BS and nothing like tilt-back stages exist - could be that this plateau at ~7° just happened as some kind of equilibrium of the tilt-back kicking in and me desperately trying to keep balance...

 

 

Link to comment
Share on other sites

6 hours ago, Frode said:

I beleive my KS-16 has regenerative braking only, Braking with full battery may give alarm, tiltback and loss of braking ability altogether, for example going down a steep slope. I have experienced that, and cannot find any other reason for the wheel behaving like it did (repeatedly braking for a short time and loosing brakes, resulting in mean speed increasing down the slope). I wish it had plugging braking or dynamic braking. It would have prevented problems like that, at least for long enough time to stop the wheel without having to jump of at speed.

Agreed that observing downhill brake failure under full battery is an indication that regenerative braking takes place. It is not an indication that the wheel does never brake actively though. Unfortunately, active braking is also problematic downhill, as you run very quickly into heat dissipation problems, I assume the time frame is rather seconds than minutes. 

6 hours ago, Frode said:

Electric motors are quite efficient decices, so talking about cooking away energy when you accelerate strongly as heat like happens when pushing the wheel against a wall is not a very good description.

Pushing the wheel against a wall is a very reasonable approximation for the low speed scenario: efficiency is roughly speed divided by max speed. That is, if max (rated) speed is 40km/h, efficiency is 10% at 4km/h, that is, 90% of the used energy under max acceleration at 4km/h is converted to heat.

motorcurve.gif

In reality, Amps might be upper bounded which would change the picture at the left end. I am not exactly sure how this affects efficiency precisely, but I assume it will remain to be monotonously increasing.

6 hours ago, Frode said:

No, but you compared this to cooking energy when pushing against a wall. You will have less than 100% efficency when accelerating, but not by much (unless you have a very bad controller).

You are saying the above graph is wrong. Could you explain further why this is so and point to the source where you have this information from?

Quote

Electric motors are quite efficient decices, so talking about cooking away energy when you accelerate strongly as heat like happens when pushing the wheel against a wall is not a very good description. They are more like the opposite to each other. How could you else "accelerate strongly" if you cooked away a large part of the energy as heat?

Because enough energy is available. One intuitive "explanation" is that going from 0 to 4km/h adds less kinetic energy than going from 10 to 14km/h (a factor of six times less if I am not mistaken) hence there is quite some energy available to cook at low speed.

EDIT: sorry for the double post, @Chriull gave pretty much the same answer already above which I recognized only after my post. 

Link to comment
Share on other sites

5 hours ago, Chriull said:

Unfortionately the KS16 does not report the direction of the current - so it is much harder to distinguish (imho only works by looking at the voltage - if there is a sag or the voltage gets higher. But that's quite imprecise)

From the graphs I remember to have seen, voltage is quite telling, if it is displayed on a scale where we actually see the changes. Why do you think it is imprecise?

5 hours ago, Chriull said:

Maybe @Zymlex could once record same data with his self made logger from some KS/Gotway braking and idling? Or maybe also from @Jason McNeil, if he has his logger still in use in some "modern" wheel?

I think there had been some other recordings posted in this forum already, we just would need to re-find them :ph34r:

displays two brake sequences where, I believe, we (also) see the switch from regenerative braking (where voltage increases) to "active" braking (where voltage drops).

graphs for idling. Voltage drops and current peaks correlate to the most part with deceleration, kind of funny, but understandable. Small negative currents are observed only very close to (ie. slightly after?) the speed peak. 

Link to comment
Share on other sites

3 hours ago, Mono said:

efficiency is roughly speed divided by max speed. That is, if max (rated) speed is 40km/h, efficiency is 10% at 4km/h, that is, 90% of the used energy under max acceleration at 4km/h is converted to heat.

motorcurve.gif

I doudt this efficiency, because:
1. it is not a real test data, it may come from a text book of wrong type of motor or wrong type of controller that different than current EUC.
2. It is not in accordace with my test result(I'll put more if I got the time)
3. There are some product cats and articals contradict this.
This I believe is real test: https://www.researchgate.net/figure/260626434_fig4_Fig-8-BLDC-Motor-Torque-with-and-without-dynamometer
Fig-8-BLDC-Motor-Torque-with-and-without
As can be seen that the efficiency is not very much dependant on the speed.

This is a high efficiency sevomotor: http://powerelectronics.com/motion-systems/more-efficient-servomotor
servomotor-graph03.jpg
The efficiency at 10% of the max speed is even higher..

There are some discussion here about the efficiency: http://www.diyelectriccar.com/forums/showthread.php/ac-vs-series-dc-efficiency-56409p5.html
bc64d573-66bf-4ab0-be3e-208e102bb058_zps

Efficiency=Torque*Speed/(Current*Voltage)
If efficiency is roughly speed divided by max speed, it means the current will not changed if speed is lower.
But this is not the case. My test(by climb the same slop) shows that low speed requires low battery current(the armature current is about the same).
As can be seen from my test chart that at very low speed, the ratio of (armature current):(battery current) is very large.
Therefore, the test book must be either
- regarded the armature current as the supply current, just like I originally did. Or
- regarded the voltage as the supply voltage, which does not change( much) with the speed.

Link to comment
Share on other sites

3 hours ago, Mono said:

From the graphs I remember to have seen, voltage is quite telling, if it is displayed on a scale where we actually see the changes. Why do you think it is imprecise?

...

graphs for idling. Voltage drops and current peaks correlate to the most part with deceleration, kind of funny, but understandable. Small negative currents are observed only very close to (ie. slightly after?) the speed peak. 

Thanks - imho a very nice/perfect example. Here one sees what i meant with imprecise - its a combination of the low sampling frequency and the missing sign for the current. Without the seperate logging of @zlymexone could never tell if the lower current values are in reality positive or negative...

Also in this graph is imho nicely seen, that active braking is happening! If one looks at the speed curve (Attention: here is also no sign, like with the current. So every second "speed hill" is in reality negative!) one sees that near zero speed (stopping and changing direction) the real current (logged with GL220) has its absolute positive (discharging) maximums. Just shortly after the "speed hill tops" a bit of negative (charging) currents are to be seen.

With the a bit higher speeds (~>3-5 km/h) and gentle deceleration regenerative braking can be seen. Once the deceleration needs to be stronger and/or the speed gets lower there is definitely active braking again!

Link to comment
Share on other sites

27 minutes ago, zlymex said:

I doudt this efficiency, because:
1. it is not a real test data, it may come from a text book of wrong type of motor or wrong type of controller that different than current EUC.

Imho thats real (theoretical) efficiency over speed for BLDC motors. (derived from the max torque over speed diagramm)

If you take a look at the following link (unfortionately in german - but not too much written anyway ;) ):

http://www.pedelecforum.de/forum/index.php?threads/nabenmotor-von-www-at.97/page-6#post-1612

you'll see that the efficiency over speed curves with current limiter ("mit Strombegrenzung") already look much more like your curves...

(P_aufgenommen: "consumed Power, P_abgegeben: "submitted/delivered" power, P_verlust: "lost/dissipated" power, Wirkungsgrad: efficiency)

And these are theoretical (computed) graphs.

Quote

2. It is not in accordace with my test result(I'll put more if I got the time)

Imho its hard to test the max torque over speed curve - you'd need some kind of dynamometer for this?

Quote


3. There are some product cats and articals contradict this.

So including the current limiter and control circuits one comes from the theoretical graphs to the "real measurements". (comparable to your chart of the high efficiency servo motor.)

 

The other two charts you linked, show something different. The first is efficiency over torque for a fixed speed. And the last chart shows... pfff ... hard to say ... "isoefficiency" lines for torque over speed...;) (whatever that means....- ?a 2D representation of an 3D chart, showing the efficiency for each torque/speed couple?)

Link to comment
Share on other sites

4 hours ago, zlymex said:

As can be seen that the efficiency is not very much dependant on the speed.

That seems to be true for a given torque at different speeds, at least to some extend. The theoretical graphs show, to my understanding, efficiency at the maximum torque for each speed, so that is not necessarily a contradiction.

4 hours ago, zlymex said:

This is a high efficiency sevomotor: http://powerelectronics.com/motion-systems/more-efficient-servomotor
servomotor-graph03.jpg
The efficiency at 10% of the max speed is even higher..

Do you know at which current/torque these have been measured? From the given link I wouldn't be able to tell for sure. 

4 hours ago, zlymex said:

 My test(by climb the same slop) shows that low speed requires low battery current(the armature current is about the same). As can be seen from my test chart

sorry, I am missing out, which test chart you are referring to? 

Link to comment
Share on other sites

8 hours ago, Mono said:

That seems to be true for a given torque at different speeds, at least to some extend. The theoretical graphs show, to my understanding, efficiency at the maximum torque for each speed, so that is not necessarily a contradiction.

It might be quite true that the theoretical graphs show efficiency at the maximum/peak torque. In real situation including test of efficiency, the peak torque is seldom reached, unless in a faceplant or 'the test is terminated' situation. Moreover, there seems no fundamental physics behind the motor theory that make the efficiency so different for maximum torque and normal torque. I now clear that that theoretical graphs of efficiency is based on a wrong assumption. I'll explain that later.

8 hours ago, Mono said:

Do you know at which current/torque these have been measured? From the given link I wouldn't be able to tell for sure. 

I don't know. But from the description, it seems the efficiency-speed curve is measured/calculated at the rated torque, and not affected by the torque, at least not very much. 

Note also that the rated power is often the maximum continuous power output(as in Gotway V3, 1500W), but the torque-speed curve and the power-speed curve is based on peak power which may significantly larger than rated power(as in Gotway V3, >3000W).

8 hours ago, Mono said:

sorry, I am missing out, which test chart you are referring to? 


Now, the armature current is added as indicated by the green curve:
V3-slope-test2c.gif

Link to comment
Share on other sites

Archived

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

×
×
  • Create New...