GMAC Regen Problematic

[xango]

Hi,

i have a GMAC 10T with baserunner and CA3. Problem is that this Combination GMAC BR + CA3 start automatictly Regen at around 42-44km/h independ of the brake switch.
I made my settings that the GAMC starts Regen at 55km/h. And Grin coudln't help me with this case. I tried also with diffent settings but without solution.
Maybe there is one or more Guys here which have the same and found a solution. Here my settings in attachment:

 

[E-HP]

i have a GMAC 10T with baserunner and CA3. Problem is that this Combination GMAC BR + CA3 start automatictly Regen at around 42-44km/h independ of the brake switch.

Are you trying to implement variable regen via your throttle, or separate throttle lever (is that what you mean by “switch”?)?

 

[Epithemeus]

What's the unloaded motor speed? All motors without field weakening regen above unloaded speed.

 

[xango]

Are you trying to implement variable regen via your throttle, or separate throttle lever (is that what you mean by “switch”?)?

via my normal throttle. No extra throttle! And downhill i use the throttle only in curves and normaly not. I let the bike speed up as posible but with max speed of 55km/h. At that speed the Regen should start.

 

[xango]

What's the unloaded motor speed? All motors without field weakening regen above unloaded speed.

but field weakening is not activated. unloaded speed? What does that mean?

 

[E-HP]

via my normal throttle. No extra throttle!

What voltage and wheel size are you running? Below is based on a 52V battery, and 26" wheel size, using the motor Kv detected by the L10. No load speed is 42.3kph, assuming 52V, a little faster on a fully charged pack (58V). If you were going downhill, the back EMF would naturally cause the motor to regen if the speed exceeded 42.3kph (as Epithemus pointed out).

Motor Simulator - Tools

Our ebike motor simulator allows you to easily simulate the different performance characteristics of different ebike setups - with a wide selection of hub motors modeled, and the ability to add custom batteries and controllers and set a wide variety of vehicle parameters you'll be able to see...

ebikes.ca

 

[xango]

But downhill the grade is about -10% and therefore the no load speed is higher

 

[E-HP]

The no load speed is if you applied full throttle with the wheel of the ground. Going downhill will make the motor attempt to exceed that speed and therefore will begin regenning.

 

[xango]

The no load speed is if you applied full throttle with the wheel of the ground. Going downhill will make the motor attempt to exceed that speed and therefore will begin regenning.

Ahh, than the no load speed or the max speed at even ground is about 40km/h. Is that what you mean? If so the setting Regen speed max 55km/h makes no sense? my batt Voltage is 33-46V (11S batt)

 

[amberwolf]

unloaded speed? What does that mean?

Lift wheel off ground. Use full throttle. Wheel now spins at unloaded speed.

Check speedometer; if you see a speed lower than that you get when you have the regen happening, that's why your regen happens.

All controllers are a 3-phase rectifier even when unpowered. If the mtoor spins fast enough to generate a voltage higher than the battery voltage, then this will cause current to flow back to the battery (regen) no matter what the controller does, even if it's turned off.

 

[E-HP]

Ahh, than the no load speed or the max speed at even ground is about 40km/h. Is that what you mean? If so the setting Regen speed max 55km/h makes no sense? my batt Voltage is 33-46V (11S batt)

I made an error in the graph above. No, max speed on ground tends to be 10% slower due to rolling resistance and wind resistance. Eliminating those loads, the no load speed is 53kph for the example I provided earlier.

Motor Simulator - Tools

Our ebike motor simulator allows you to easily simulate the different performance characteristics of different ebike setups - with a wide selection of hub motors modeled, and the ability to add custom batteries and controllers and set a wide variety of vehicle parameters you'll be able to see...

ebikes.ca

With the 48V battery you used in your graph, the no load speed is 49.7kph, as modeled.

Motor Simulator - Tools

Our ebike motor simulator allows you to easily simulate the different performance characteristics of different ebike setups - with a wide selection of hub motors modeled, and the ability to add custom batteries and controllers and set a wide variety of vehicle parameters you'll be able to see...

ebikes.ca

 

[xango]

As i understand now: Regen is always activated when the speed downhill exceeds the no load speed . Is that true?
If so than higher start of Regen need more batt voltage.

 

[amberwolf]

There are two ways to help prevent the "auto regen".

One is to use a higher battery voltage, one that is *always* above the generated voltage, even when it is empty. That's the simplest option. Since the *runner series is a current-control throttle, then it will still behave about the same. (vs a speed-control throttle where it would be more difficult to control at a higher voltage for small throttle amounts).



Another is to use a different wheel size, though this is not as practical; it has to be proportionally different enough to avoid generating that higher voltage.

You *might* be able to use Field Weakening setup for that high a speed to actively drive the motor during the downhill, but this will make you go even faster and at some point you'll go faster than the system can deal with, as the voltage generated would be higher than the controller FETs/etc can handle, and they'll fail.


Which brings up another situation with downhill regen: If you start out with a battery near or at full charge, the regen will continue to charge the battery until the BMS deactivates it to protect it from overcharge, disconnecting the battery from the controller.

If the BMS is a common-port type, that will mean the controller has nowhere to dump this current now, and it becomes a voltage that builds up (not like static electricity) very fast and can quickly exceed the voltage the controller FETs and other parts can handle, destroying them.

If the BMS is a separate-port type, then it can't stop the regen current, and the cells will begin overcharging. How serious this problem is depends on the length of time the regen goes on, and the amount of current flow there is. If it's a very low current, less than 50mA, it's likely that the BMS balancing shunts (if it is a balancing type) will handle it. Above that (whatever the shunts can do) the cells will charge.

As long as the battery is never near full, there's no worry. But if you live at the top of a hill and start at a full charge...that can be a problem. :/

 

[xango]

Thankyou amberwolf. This now i understand.
This thread can now be closed

 

[E-HP]

The consolation is that with the slow 10T wind, you can climb that 10% for a good 12 minutes at 32kph before overheating, so that's like 6km of climbing.

Motor Simulator - Tools

Our ebike motor simulator allows you to easily simulate the different performance characteristics of different ebike setups - with a wide selection of hub motors modeled, and the ability to add custom batteries and controllers and set a wide variety of vehicle parameters you'll be able to see...

ebikes.ca

 

[xango]

All controllers are a 3-phase rectifier even when unpowered. If the mtoor spins fast enough to generate a voltage higher than the battery voltage, then this will cause current to flow back to the battery (regen) no matter what the controller does, even if it's turned off.

One question more: If the controller is off how can the generated voltage charge the battery? i thought the Regen always happens thrue the controller

 

[amberwolf]

As I said before; all controllers are a 3-phase rectifier even when unpowered. FETs have an intrinsic diode that lets current flow in that direction even when not powered or controlled.

Basically like this, with the motor being the three inputs at the side and hte battery being the "resistor" at the right.



This thread has some discussion of possible solutions, but there are unresolved issues of switchover times, etc.
Alternate Regen 'dump' - Soliciting Suggestions | Endless Sphere DIY EV Forum

 

[E-HP]

An open circuit, by disconnecting the phase wires from the controller, would stop any auto regen. Just not practical, or just use regular brakes to keep the speed down, which is more practical.

 

[amberwolf]

An open circuit, by disconnecting the phase wires from the controller, would stop any auto regen. Just not practical, or just use regular brakes to keep the speed down, which is more practical.

The downside of disconnecting the phases from controller is that you have to stop to disconnnect, then stop to reconnect. If you don't stop, the unloaded voltage on the phases could be high enough to damage the controller.

Slowing down might be enough, you could test and see what voltages are generated at what sped to find the point at which it is too risky.

Additionally, you'd need the switch or contactor to handle the voltage during connect and disconnect, if not done at a stop. Otherwise the contacts could be arc-damaged or even weld shut.

 

[E-HP]

The downside of disconnecting the phases from controller is that you have to stop to disconnnect, then stop to reconnect.

On the rare instance when stuck at the top of the hill with a full battery, disconnecting, then descending quickly and reconnecting may take less time than crawling all the way down the hill to avoid overcharging. Depends on how easily accessible the phase wire connectors are. If it's a single connector like an L10 or Z910, it would be pretty easy.

 

[Epithemeus]

You *might* be able to use Field Weakening setup for that high a speed to actively drive the motor during the downhill, but this will make you go even faster and at some point you'll go faster than the system can deal with, as the voltage generated would be higher than the controller FETs/etc can handle, and they'll fail.

I'm not entirely sure how field weakening works, does the motor have to actually be "driving" the wheel for field weakening to work? Apologies if I bungle my explanation here but by driving I mean adding kinetic energy to the vehicle beyond what gravity is providing. Essentially could you use just enough field weakening to cancel out the regen drag? Kind of like electronic freewheeling just beyond the back emf limit of the motor. Thus raising your downhill speed ceiling from that of the motor back emf/battery voltage to that of the FET voltage limits?

If yes, does the phase runner/baserunner support this behaviour?

 

[amberwolf]

I'm not entirely sure how field weakening works, does the motor have to actually be "driving" the wheel for field weakening to work?

Yes. FW "simply" changes the relationships of the current being sent to the motor phases, to "simulate" the way a field winding in a brushed motor (equivalent in this case to rotor magnets in a brushless) would have been driven with a weaker field to allow the motor to spin faster.

(I don't understand the specifics of exactly what FW does; it's a math thing and I don't really do those)

Apologies if I bungle my explanation here but by driving I mean adding kinetic energy to the vehicle beyond what gravity is providing. Essentially could you use just enough field weakening to cancel out the regen drag?

If you could automate it to remain balanced that way, that would be nice...but I don't know how easy that would be.

If yes, does the phase runner/baserunner support this behaviour?

It doesn't have it built in to automate this, but it supports FW in general. You'd need some external computer with the right sensors to monitor all the required conditions and tell the *runner what to do.

 

[xango]

Next Ride, probaly on next Saturday i will try with small amounts of Field weakening and to see what happens.
In thread no 7 i added an screenshot of a simulation with the grade of -10%. I guess that the simulator is not good for the Regen Case. Is that right or other settings are nessesary?

 

[E-HP]

Yes. FW "simply" changes the relationships of the current being sent to the motor phases, to "simulate" the way a field winding in a brushed motor (equivalent in this case to rotor magnets in a brushless) would have been driven with a weaker field to allow the motor to spin faster.

Just trying to see how this works in practice. Does it require riding downhill on full throttle in order for field weakening to compensate for auto regen?

 

[neptronix]

MAC and phaserunner seem to have a lot of various issues at high voltages/speeds.

MAC is a particularly hard motor to drive due to it's very high eRPM. Infineon controllers had to be changed to support a higher commutation rate to get MACs working smoothly.. phaserunners, like non reworked infineon controllers, hit a limit at high speeds with this motor.. which happen to be right about where you're having problems ( ~55kmh ).

In this case you will see erratic behavior with the phaserunner no matter how well you tune it.. it's more of a matter of what kind of erratic behavior you want, since you're out of tuning range. And the CAv3, being a piggy back controller, can complicate this further. Try eliminating the CAv3 from the equation and see if that helps. Try also turning virtual freewheeling off, if you have it on.

If all attempts of tuning fail, you might want to consider a controller intended for high RPM motors like a VESC. That thing will drive anything from a DD hub motor to a 15k RPM RC motor.