Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[DerKrawallkeks]

Hello,
I am using a progressive regen lever, and I would like to change the 4500rpm limit, above which regen is turned off. It feels quite weird, especially since the regen lever looses all function.
Now my question: is there a specific reason this limit is there and what bad could happen if I make it higher? It seems weird they'd just put a limit there for no reason.
I've searched on the forum, and a couple of guys set it to 6000rpm, with no apparent downside.

Thanks for your help guys!
DerKrawallkeks

[mryan]

Most likely will have negative affects on the life of a controller. An open source controller called vesc would start frying chips if you exceeded a certain rpm and this could be similar. The controller has to convert DC to ac, and the exact opposite happens with Regen, so perhaps it has something to do with not wanting to fry internal components.

The other option is that it will overcharge the battery. As the rpms get higher, the Regen voltage that happens will get higher. With the vesc it would act as a brake but not actually be generating any power up until a certain speed, but after a certain point it could spike the voltage.

Just some ideas

[DerKrawallkeks]

Thanks mryan!
I've used a vesc in my hybrid honda dax. They have flyback diodes on the mosfets (like any esc and the Sevcon probably as well). So above a certain rpm it starts feeding into the battery uncontrollably.
Mmh maybe they are afraid of some scenario of going downhill with an empty battery..

[rgutt]

According to some who have used the Sevcon's DVT software to access the controller's setup, the recharge limit in the controller is set to 91 amps or less. I don't have the exact C rating for the battery at my fingertips, but that recharge rate should put it somewhere slightly less than 0.8C, which is a walk in the park for a lithium battery. I can't speak to Zero's logic behind the 4500 RPM regen limit, but I doubt it had anything to do with the capability of the controller or battery. More importantly, you can increase that number as much as you want (assuming you have access), and as long as the Sevcon is setup correctly, you won't hurt a thing.

The true limitations of regen are more complex involving many parameters that can be monitored by the Sevcon (and MBB), but that's probably beyond the intent of this inquiry.

[victor6.7y]

I'm not complete sure, but as a rule of tumb I model an electric engine as the following:

• Current - torque
• Voltage - RPM

Besides that you have magnets turing around coils which give you a U = L  dI/dt and having a magnet passing the magnetic field of a magnet will induce a current in the coil.

Anyway, high RPM usually tries to produce high voltage.
Even if you are able to reduce the voltage by sort of trying to suck the current out of the engine. You still have to be able to deal with the voltage properly. Our zero's work with a ~100 Volt system. I can imagine you want to set a voltage limit (or RPM limit) at which point you say; Hey lets ensure our electronics don't get fried.

A flyback diode needs to be able to handle a certain power (voltage * current). When you drive fast, you have a lot of momentum that needs to be stopped through regen. That might be more power/voltage/current than the system can handle.

[rgutt]

At any speed to which the controller has driven a PMAC motor, the resulting back EMF of the motor cannot be greater than the voltage that was fed it in the first place.

[victor6.7y]

Not sure if that is completely true due to temperatures and capacitor stack

[rgutt]

Not sure if that is completely true due to temperatures and capacitor stack

I'm not sure where you're going with either of those. Capacitors have nothing to do with the voltage a motor can generate. They can't amplify the AC current the motor generates natively, and they can only boost DC voltage when connected and switched for the purpose. Though it is likely the Sevcon has that function to facilitate regen at low RPM, those circuits are active only by necessity and thus wouldn't be active when the voltage generated by the motor was already high enough to generate the target recharge current to the battery.

As for temperature coefficients, since an increase in temperature both increases the resistance of copper and reduces the strength of a magnet, if you want to get down to the last few millivolts, the back EMF of a motor immediately after removing the driving voltage can only be ever so slightly less than the voltage that was originally feeding it. In context of our bikes, that would technically only apply on a level grade or a downgrade if the speed is high enough that wind resistance prevents further acceleration of the bike after removing the drive.

Was there some other angle you were considering? It has been quite a long time since I had a physics class, so I could be missing something. I'm just not seeing it. I hope the guy who created this thread finds this interesting.

[DerKrawallkeks]

Hey everybody,
I would like to add my own non-confirmed semi-knowledge to the discussion:

As rgutt wrote correctly, if a motor is driven with full battery voltage (100% PWM duty cycle, so full throttle), it will go a certain rpm.
It will produce exactly battery voltage, if you're not actively driving it.

So it will always produce a voltage less or equal than battery voltage. It can never put current back into the battery like this, since the generated voltage is always lower than the battery voltage (unless it's mechanically spun at even higher speed).

The way regen works is by using a special mode of operation, that works like this:

1. the coil is shorted, both ends are connected
2. current will start to rise quickly
3. the coil is now connected to the battery, and the inductance of the coil wants to keep the current going
4. current flows into the battery
5. the current drops quickly, now the process repeats

This circuit is called a switch mode power supply. Only in this case, the coil is a motor coils, and we're using it for regen.
As I said, this is always required even if you're going at very high speed, since the produced voltage is always lower than the battery voltage.

@rgutt, what you hinted at with the capacitor boost circuit, that switches capacitors around exists, but it is not used for regen braking as far as I know. I think it is only used in very simple and cheap boost converters.

[rgutt]

That seems like a logical way to get the voltage necessary from the motor to recharge the battery. I hadn't really thought about how it would actually be implemented, but that makes sense. It strikes me as far simpler than switching capacitors.

I was going to add to my previous post that some might note that field weakening can be used to spin a motor beyond its design speed which if then put into regen could actually generate a voltage higher than that of the battery. However, that is not only an unlikely situation on our bikes (unless you're coasting down Pike's Peak and capable of doing so at 110 mph), but also only indicative of a setting to cut regen at max motor RPM, not at 4500 RPM.

[DerKrawallkeks]

The field weakoning, isn't that only a thing with electrically excited motors?
Or how would that work on our BLDCs?

[rgutt]

I never went that far into electrical engineering, but from my 30-second Google PHD, you can basically use the currents in the stator to reduce the back EMF that would have otherwise been created by the passing magnet if the current weren't there. Since the total current available to drive the motor is fixed, doing this takes a corresponding amount away from the current to drive the motor. I'm sure there are some true electrical guys here who go really dive into the details and math.

[DerKrawallkeks]

Ah I just graduated as well. Seems like you're right, even in a BLDC motor the magnetic flux can be lowered by some trickery, allowing higher rpm due to lower back EMF.
that doesn't have anything to do with regen, but anyway:D I learned something today

[victor6.7y]

Hey all. First thank you for checking an challenging me. I haven't gone to this subject for a long time.


To answer the first questions

Not sure if that is completely true due to temperatures and capacitor stack

I'm not sure where you're going with either of those.

I was looking at it in the wrong way. I forgot to consider that the vehicle will "in generally" not be able to drive faster than the engine is able to make it go.
So if the bike is able to go 6000 rpm by using its engine, it won't be able to go 7000 rpm.  Thank you all for reminding me.

With that I agree. I see no reason for the charge rpm to be limited to 4500 rpm when the bike is able to drive 6000 rpm. This since the Voltage will not get higher than the voltage applied.


With the quote:

Not sure if that is completely true due to temperatures and capacitor stack

I was referring to the amount of current the MOSFET / "flyback diodes" (not sure if they are called that) are able to handle. This has nothing to do with the voltage. It is related to the force with which the bike is stopped. (Should have elaborated on that properly)
When thinking of pushing your electronics to the limit. I would try to allow as much current as possible trough the MOSFETs (Did that before, and burnt a few, trying to build my diy BLDC driver  ). The capacitor stack will most likely be able to deliver that current. When accelerating maximally and then accelerating the MOSFETS will propably heat up twice.

But then smart people, like Sevcon and Zero will probably ensure that the maximum power going through the system is lower than the Sevcon is able to handle.
Having the maximum power determined by the Capacitor stack and Battery.


For some indept info the following link is a good staring place.
https://en.wikipedia.org/wiki/Gyrator%E2%80%93capacitor_model#Magnetic_inductance




In summery:
I think when not going faster than the bike is able to take you; And not de-accelerate, using "regen breaking", faster than the bike is able to accelerate you; there should be no problem for the bike.
*There is a small thingy with the capacitor stack (when the capacitor stack is charged,,, you want to deaccelerate as much as the battery is able to accelerate you.)