Self built Electric Vehicles, batteries and speed controller

Mitsubishi Outlander PHEV Forum

Help Support Mitsubishi Outlander PHEV Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

MadTechNutter

Well-known member
Joined
Jul 5, 2019
Messages
132
Location
Aussie Land
>>> continuation from a discussion on the Lindqvist thread:

Pomst said:
MadTechNutter said:
As for my bike it does have a 200W 24V motor that originally ran on NiCads. This 8.33A at the rated voltage.
The battery pack consisted of 3 x 3.3Ah 7s batteries = 9.9Ah.
When new and fully charged the voltage would stay above 28.5V on the hill in question. That is 9.89A, less than 1C of the pack.
Assuming I actually had a motor that was capable of 500W, 2C would be 564W and more than enough to deliver and a whole order of magnitude less than what it is rated for, unlike EV batteries.
However the motor has only 200W rated and would have been running at 282W up that hill and maintained the speed at 20km/h. Bike has a speed limiter at 23km/h, it is something legal.
After a year held at full charge (LiPos are charged up to 4.2V it would not go up faster than 9km/h, so relatively speaking it used to 'fly' up those hills like a 'rocket'.

Yes LiPos are not EV batteries we all know that.
They are nevertheless all Li-Ion batteries.
It is common knowledge that all Li-Ion batteries suffer degradation when stored at full charge.
There is a lot of science, research and empirical evidence over the last 25 years that supports the claim that Li-Ion batteries are best stored at 3.7V
There is also some evidence suggesting that keeping batteries stored at temperatures near 0˚C is beneficial.
This MIGHT also have something to do why Australian users of the car are complaining more about battery degradation than from countries with colder climates.


Secondly, you have you calculations way off and it proves my point on experience of batteries.
The voltage of the cell drops under load, depending on the load the cell drops differently when the current increases. My estimates of the load on your cells are based on about 4 packs of hand built bicycle batteries and a dozen of motors with anything from custom controllers to Chinese ones.
A 200w motor is 200w continuously not peak, it could easily peak 600w or more if the controller allows it. Like
I said, it depends on your definition of "rocket", 200w is nothing. Just pedaling gives you 300-400w power.
Assuming you didn't, if you had a proper controller like a Cycle Analyst V3 you would see what I am talking about.
What does 3x 3.3Ah 7S mean? Is it a 7S pack or a 7S3P pack? Because the two are vastly different and effect the C rating dramatically. Irrespective of which, this experiences is not relatable to the cells in an EV pack.


My experience in batteries is not only gathering batteries and looking at them.
I build stuff from them and reverse engineer the BMS systems on the EV packs.
As I stated, EV batteries do not degrade much at all over a 10 year period because of their nature and the strict limits in the BMS.
RC batteries are like a dragster motor, they run hot and fast but die quickly.
The two are not comparable in any way shape or form.
[/quote][/quote][/quote][/quote][/quote]


First of all there is no "peak current" involved with these type of PMW speed regulators.
It is a 200W regulator, not a 500W one.
This is a German made legal vehicle and by law it is not permitted to produce more than 200W at any given time.
It doesn't throw you off the saddle when you apply the throttle, it has a soft start.
If you knew a little about physics, a bicycle or any object for that matter, only requires energy while accelerating which includes dealing with the force of gravity going up a slope and friction.
This friction is quite minimal on a bike, and around 50w would be sufficient to maintain 20km/h on a smooth flat road.
As I could comfortably ride for 2 hours, the average load was no more than 130W from my 260Wh battery, including hills and starts and a bit of head wind.
As for pedalling, I am old and weak, if I were to produce 400W for more than 10 seconds, I would likely faint or get a heart attack!

I was using 3 batteries wired in parallel. The sum is 9.9Ah.
At no time did the motor exceed 280W because that is how the ESC is built by design.
Internally current peaks may be higher but this is for a tiny fraction of a second.
For a PWM brushed controller running at 10kHz this would be 0.1 milliseconds !
These peaks are covered and smoothed internally by low ESR capacitors. When the current sensor detects that the maximum allowed current is exceeded the regulator switches back the duty cycle.
If you lock up the wheel you can actually see the voltage rise a bit again.
The battery never sees a real mean square current that exceeds what the ESC is rated for and in this case it is about 9A (with a but of leeway) and you can cheat out a little extra power by raising the voltage a little.

Brushless ESCs for bikes and scooters are also PWM regulators that additionally vary the frequency, are usually 3 phase but behave similar.

I am currently designing/building a 4WD electric ride-on/mini tractor, essentially an EV, fully drive-by-wire including steering.
There are 4 brushless hub motors rated for 400W. Two of them I ordered with the dedicated speed controllers and they didn't serve the purpose for the exact reason you are questioning.
They would limit the current to only 6.5A, this is for a 36V system, so only 230W.
That is all I got even if I locked up the wheels but then it shuts down.
This is evidence that even cheap Chinese speed controllers fully limit the amount of energy you can take from the battery.

So I ordered 4 new 500W speed controllers(actually 6 because I need more for the slasher motors).
Now I have the power I need to pull tree logs arounds and shift a bit of earth.
When I tried to lock up one wheel it threw one side of tractor up in the air, some serious torque in these geared down hub motors, so I needed to bolt things down for more tests.
However my current sensor never showed more than 16A, no matter how I loaded it.
Oh look, 16A x 36V = 536W, just what the doctor ordered :)

Yes it is built out of hard wood, imagine that :mrgreen:
 

Attachments

  • E-Tractor-1.jpg
    568.6 KB · Views: 64
The law is not followed at all, we have the same laws here and these laws are toothless.
All the motors I have measured used much more then 200w under load and some even peaked at 1000W while all being specified at 200W. The specs are just for show.

Good luck with the project.
Please, for your sake make sure of the following:

* If you are running the batteries in parallel use one fuse per cell. This is a must have when running parallel cells since if one cell goes bad for any reason a fuse will save the entire pack from a fire.
* Please make sure to use a BMS with a tight tolerance on Over Voltage limit. Maximum allowed voltage should not exceed 4.25V including tolerance errors. Best option is to have a variable voltage setting available in some of the "Smart BMS" systems.
* Make sure the BMS never allows peak current exceeding 80% of the cells specification. 80% because cells always vary in their quality and some fall outside the specified current rating.

I am not interested in you subjective calculations, they are not correct since they lack in facts.
To back any claim up about you specific controller you need a schematic, a BOM and the voltage/current limits the controller has.
You killed the RC pack the way you built it, it had nothing to do with degradation at all.
The NICD controller probably had a normal NICD cut off voltage of 16V (800mV/cell) or 18V (900mV/cell) which would be 2.28V or 2.57V per LiPo pack.
This in turn delt a two puch kill to the pack when current spiked during periods of far to low voltage.

Also, making assumptions on people's knowledge is counter productive.
It's a habit you should un-learn.
 
Pomst said:
The law is not followed at all, we have the same laws here and these laws are toothless.
All the motors I have measured used much more then 200w under load and some even peaked at 1000W while all being specified at 200W. The specs are just for show.

Good luck with the project.
Please, for your sake make sure of the following:

* If you are running the batteries in parallel use one fuse per cell. This is a must have when running parallel cells since if one cell goes bad for any reason a fuse will save the entire pack from a fire.
* Please make sure to use a BMS with a tight tolerance on Over Voltage limit. Maximum allowed voltage should not exceed 4.25V including tolerance errors. Best option is to have a variable voltage setting available in some of the "Smart BMS" systems.
* Make sure the BMS never allows peak current exceeding 80% of the cells specification. 80% because cells always vary in their quality and some fall outside the specified current rating.

I am not interested in you subjective calculations, they are not correct since they lack in facts.
To back any claim up about you specific controller you need a schematic, a BOM and the voltage/current limits the controller has.
You killed the RC pack the way you built it, it had nothing to do with degradation at all.
The NICD controller probably had a normal NICD cut off voltage of 16V (800mV/cell) or 18V (900mV/cell) which would be 2.28V or 2.57V per LiPo pack.
This in turn delt a two puch kill to the pack when current spiked during periods of far to low voltage.

Also, making assumptions on people's knowledge is counter productive.
It's a habit you should un-learn.

You are the one making the fantasy assumptions.

Germans adhere very tightly to the law. At any time the police could ask to have a vehicle tested by the TÜV and the company Sachs would have a huge reputation to lose if they were selling illegal bikes designed and sold for license free use. This bike is about 25 years old, so maybe you might not be aware of these things with all the China crap available these days who can not be held responsible for anything.

...and here in Australia you definitely don't want to be caught riding something over 25km/h with out a motorbike license, the fines are much higher than Germany. Things are so tight here that you will be heavily fined and could lose you license even if passengers in your car are using mobile phones. So don't tell me anything about toothless laws.
Even riding an un-motorised pushbike without helmet, smack $250 and I actually agree with that.

Yeah with the fuses, I might have been a bit lazy in the bike back then.
The batteries in the bike were charged with an external LiPo balance charger suitable for up to 8s, was not that easy to find back then.
I made an extension for the balance cable to reach the bike. I regularly checked cell balance.
Discharge management was me watching the volt meter when I rode. When it dropped to 21V, it would be me pedalling all the way back, but that hardly happened because I always had the batteries fully charged and still at least 30% left when I came home.
After degradation the range did not reduce that much, it would just quickly drop to 21V on hills so I was forced to do a lot of pedal assist to keep it up or I just pushed it up by foot. After the hill, voltage would be back to 27V and it worked OK even riding on bumpy grass routes.
The motor never exceed a 1C load for the battery and at that load I can not kill a battery.
That is a fact proven by actual real life tests and maybe one day you might learn how these speed controllers work.

The LiFePO4 batteries in the tractor are a total different story, they have a dedicated BMS and I have similar setups running for a few of years very successfully in off-grid house power systems on a 90%-10% cycle but it rarely gets that low. Even in winter there is plenty of sun here to charge them.
The tractor is a mini version of that.
Can turn on the spot(rear and front steering) and even move sort of sideways, will eventually autonomously mow around trees in an orchard, pick fruit with the robotic arm, that is also a digger, fetch me a beer from the esky, have a 'ludicrous mode' that can only enabled with a pass key: illegal 35km/h ... a fun project that will be always work in progress.
 
jaapv said:
The Lidl is selling a nice eBike for 839,-- Euro :roll:

Unfortunately they did increase the price ... I did paid almost the same for this , 1 year ago:
https://www.bavarianstrongbikes.de/product-page/bavarian-mtb-adventure

I also got other 2 eBike for less .. one for 750, and another one for 550 ... all new eBike

Do you have a link of this Lidl ?

Just for get an idea how is evolving the market

PS: I think the power of an eBike is limited by the controller ... it is not a motor issue .. an eMotor that has a high load can produce much more power then its original specs .. but it will be normally also outside the ideal efficiency curve
 
elm70 said:
jaapv said:
The Lidl is selling a nice eBike for 839,-- Euro :roll:

Unfortunately they did increase the price ... I did paid almost the same for this , 1 year ago:
https://www.bavarianstrongbikes.de/product-page/bavarian-mtb-adventure

I also got other 2 eBike for less .. one for 750, and another one for 550 ... all new eBike

Do you have a link of this Lidl ?

Just for get an idea how is evolving the market

PS: I think the power of an eBike is limited by the controller ... it is not a motor issue .. an eMotor that has a high load can produce much more power then its original specs .. but it will be normally also outside the ideal efficiency curve
That's correct.
Also, controllers ability to quickly limit current is utterly useless on all cheap controllers. It's one of the reasons they overheat and release the magic smoke.
Personally I recommend http://lyen.com.

As a comparison 200w is 0.27hp. That's not even enough to keep speed on a flat surface for a 90kg person with a normal crmo frame.
Even a measly Hoverboard has Two 250w-rated motors.
 
Pomst said:
As a comparison 200w is 0.27hp. That's not even enough to keep speed on a flat surface for a 90kg person with a normal crmo frame.

That statement alone proves that you obviously skipped classes when you had science at school.

I weigh over 80kg and both of my LEGAL E-bikes do fine on level roads at legal speed.
Also that you now suddenly agree that speed controllers DO limit the power after elm70 mentioned it, but vehemently argue against that fact with me with several posts.

You seem to be some kind of troll with the unsubstantiated rubbish you are posting.
 
MadTechNutter said:
Pomst said:
As a comparison 200w is 0.27hp. That's not even enough to keep speed on a flat surface for a 90kg person with a normal crmo frame.

That statement alone proves that you obviously skipped classes when you had science at school.

I weigh over 80kg and both of my LEGAL E-bikes do fine on level roads at legal speed.
Also that you now suddenly agree that speed controllers DO limit the power after elm70 mentioned it, but vehemently argue against that fact with me with several posts.

You seem to be some kind of troll with the unsubstantiated rubbish you are posting.

Either you didn't read what I wrote or you don't understand how it works.
The controllers do limit but not to 200w as they are slow and inaccurate.
This in turn allows for a high in-rush current pushing far more power to the motor then legally allowed.

Please get out the calculation for friction analysis on a deformed rubber inflated wheel on a surface with small pebbles on it.
I would love to see it.
 
Pomst said:
MadTechNutter said:
Pomst said:
As a comparison 200w is 0.27hp. That's not even enough to keep speed on a flat surface for a 90kg person with a normal crmo frame.

That statement alone proves that you obviously skipped classes when you had science at school.

I weigh over 80kg and both of my LEGAL E-bikes do fine on level roads at legal speed.
Also that you now suddenly agree that speed controllers DO limit the power after elm70 mentioned it, but vehemently argue against that fact with me with several posts.

You seem to be some kind of troll with the unsubstantiated rubbish you are posting.

Either you didn't read what I wrote or you don't understand how it works.
The controllers do limit but not to 200w as they are slow and inaccurate.
This in turn allows for a high in-rush current pushing far more power to the motor then legally allowed.

Please get out the calculation for friction analysis on a deformed rubber inflated wheel on a surface with small pebbles on it.
I would love to see it.

Geez, get a grip.
If 200W can move my fat ass at 20km/h with properly inflated tyres on standard asphalt I don't have to calculate shyte.


It looks like a very simple PWM speed controller tutorial is required:

A PWM speed controller is not some basic DC switch that lets all the energy freely flow at once.
PWM stands for pulse width modulation.
That means it is constantly switching on and off, for example every 0.1milliseconds for a certain duty cycle.
A controller determines this duty cycle.
A current sensor measures the current instantly.
If this exceeds a defined value the controller can respond straight away in the next 0.1millisecond cycle with a lower duty cycle.

Any hobbyist can build such a controller with a simple PIC12F683. It costs $1.
It's A/D converter responds within 10µsec and two lines of code later I have set it's CCP (Capture Compare PWM) module to a new duty cycle

A power MOSFET can change it's on or off state within 50 nanoseconds, that is certainly not slow and can not create any inaccuracy.
It is important that they switch fast because that determines their thermal loss, not just the on-resistance.
If they would switch slow they would burn out.

I would literally have to purposely program in a delay to make such a PWM controller laggy to allow higher initial inrush current.

None of the Chinese controllers I have here go past their specified limit but some do not even reach that limit.

BTW I also have one of those iMortor wheels. I did mount it on a fairly light but standard Alu bike. Maybe that is where you might have got that 200W won't move 90kg rubbish.
The speed controller is so piss weak it doesn't reach 200W but that is by design because the included battery is tiny.
Nevertheless I have reached speeds over 25km/h with it, perfectly flat road, zero wind. Just needs a lot of patience, takes maybe a minute to get to that speed.
It sucks up hills, 12-13km/h fully charged on a light slope.

Then I found a YT video of a guy who ripped it all apart and built his own controller on top of the bike, huge, nasty with fat heatsink obviously current limit was not part of the deal.
He then displayed what 500W looks like on a 200W motor: he had wheel spin!
 
MadTechNutter said:
Pomst said:
MadTechNutter said:
That statement alone proves that you obviously skipped classes when you had science at school.

I weigh over 80kg and both of my LEGAL E-bikes do fine on level roads at legal speed.
Also that you now suddenly agree that speed controllers DO limit the power after elm70 mentioned it, but vehemently argue against that fact with me with several posts.

You seem to be some kind of troll with the unsubstantiated rubbish you are posting.

Either you didn't read what I wrote or you don't understand how it works.
The controllers do limit but not to 200w as they are slow and inaccurate.
This in turn allows for a high in-rush current pushing far more power to the motor then legally allowed.

Please get out the calculation for friction analysis on a deformed rubber inflated wheel on a surface with small pebbles on it.
I would love to see it.

Geez, get a grip.
If 200W can move my fat ass at 20km/h with properly inflated tyres on standard asphalt I don't have to calculate shyte.


It looks like a very simple PWM speed controller tutorial is required:

A PWM speed controller is not some basic DC switch that lets all the energy freely flow at once.
PWM stands for pulse width modulation.
That means it is constantly switching on and off, for example every 0.1milliseconds for a certain duty cycle.
A controller determines this duty cycle.
A current sensor measures the current instantly.
If this exceeds a defined value the controller can respond straight away in the next 0.1millisecond cycle with a lower duty cycle.

Any hobbyist can build such a controller with a simple PIC12F683. It costs $1.
It's A/D converter responds within 10µsec and two lines of code later I have set it's CCP (Capture Compare PWM) module to a new duty cycle

A power MOSFET can change it's on or off state within 50 nanoseconds, that is certainly not slow and can not create any inaccuracy.
It is important that they switch fast because that determines their thermal loss, not just the on-resistance.
If they would switch slow they would burn out.

I would literally have to purposely program in a delay to make such a PWM controller laggy to allow higher initial inrush current.

None of the Chinese controllers I have here go past their specified limit but some do not even reach that limit.

BTW I also have one of those iMortor wheels. I did mount it on a fairly light but standard Alu bike. Maybe that is where you might have got that 200W won't move 90kg rubbish.
The speed controller is so piss weak it doesn't reach 200W but that is by design because the included battery is tiny.
Nevertheless I have reached speeds over 25km/h with it, perfectly flat road, zero wind. Just needs a lot of patience, takes maybe a minute to get to that speed.
It sucks up hills, 12-13km/h fully charged on a light slope.

Then I found a YT video of a guy who ripped it all apart and built his own controller on top of the bike, huge, nasty with fat heatsink obviously current limit was not part of the deal.
He then displayed what 500W looks like on a 200W motor: he had wheel spin!
That's a lot of assumptions.

Good luck with your wood tractor.
 
Back
Top