Making EUCs faster: from Y to Δ connection

[Boris Lämpel]

Hey people,

I just met a guy who reconnected the motor wires of his scooter from Y to Δ connection to make it run faster. Because the relation of speed and torque is inversely proportional he sacrificed the possibility of climbing hills.
Did anyone of You figure out or even test if changing the connection would reform the riding characteristics of the wheel?
I just think of a switch where You could go fast on a Gotway HT model or even faster on a Monster pro.
I do not want to lose my motherboard by testing it out without knowing exactly what could happen. Any thougts?
Maybe the wheel electronics already switch the connection at a specific speed automaticly - I don´t know...

[mrelwood]

I have no idea what this means:

6 hours ago, Boris Lämpel said:

from Y to Δ connection

But I have a hunch that it’s something that the EUCs can’t utilize. EUCs are generally speed limited by software, so unless you somehow manage to modify the mechanism that the mainboard uses to sense the speed, the top speed remains the same no matter what you do to the main motor wires.

 If you remove the top speed limit from a GW though, then a raw electromechanical modificarion could affect the top speed.

6 hours ago, Boris Lämpel said:

Because the relation of speed and torque is inversely proportional he sacrificed the possibility of climbing hills.

What happens on a scooter when you try to accelerate faster than the available power allows? You accelerate a bit slower.

 What happens on an EUC in the same situation? You fall face first to the ground, at speed.

 Which is why it would require a very self-destructional person to try a modification like that on an EUC.

[Boris Lämpel]

It should (in my imagination) behave the same as the difference of the HS and HT models by GW. With a HS model You can accelerate, but there is not enough torque to do as fast as with HT models. You will only have a cutoff if You go far beyond the limit.

Do GWs really have a fixed speed limit? The output current is the factor limiting the wheel and starting beeping. But with Δ connection it would need less power for the same speed, so it should be able to go faster...

Our motor has 3 phases and they can be connected differently to the motherboard: like a Y (every phase connected only to motherboard) and  Δ (phase wires are connected also together - it is another variant of managing a motor).

Edited July 17, 2021 by Boris Lämpel

[RockyTop]

If it was possible, it would reduce low end torque to less than half. This would not be a big deal on a scooter. 

Edited July 18, 2021 by RockyTop

[mrelwood]

10 hours ago, Boris Lämpel said:

It should (in my imagination) behave the same as the difference of the HS and HT models by GW.

HS and HT motors have a different magnet width and number of wraps in the motor coils (and/or wire thickness in the coils). If the same could be achieved with a few additional connections, it would’ve been much more cost effective for them to do so. But they didn’t, so I’m quite sure that your idea would not work the way you imagine.

Quote

You will only have a cutoff if You go far beyond the limit.

I might not completely understand your thought behind this comment, but the way I read it, no. The wheel’s power has a changing maximum limit, but it’sa hard limit. If you go over the limit even a little, the wheel is no longer capable of keeping the pedals flat. They start to dip. If the overlean continues, the motor will turn off when the pedals reach an angle of 45 degrees. At close to top speed it happens quite suddenly, and is usually not recoverable, eventually causing a cutoff.

When the pedals go soft, you loose the solid platform to support adjusting your lean angle very quickly. If you try to lean less forward, the pedals will just give out from under you.

Quote

Do GWs really have a fixed speed limit?

No, they don’t. If the speed limit is removed, you can accelerate any GW or the Sherman even just slowly beyond the motor’s maximum rotational speed, causing a cutoff.

Quote

The output current is the factor limiting the wheel and starting beeping.

On Gotways, unless things have changed since the MSX, the beeping starts at a preprogrammed speed limit curve, that varies based solely on the battery voltage. GW published the chart back in the day, and the beeps that were misleadingly called “80% beeps” started on the 84V MSX at 58km/h at 100% battery, linearly down to 45km/h at 10% battery. Each wheel model had its own predetermined speeds for the beeps.

KSs and Inmotions do monitor the phase current and the PWM duty cycle, and have a much more complicated logic behind the “overpower” beeps.

Quote

But with Δ connection it would need less power for the same speed, so it should be able to go faster...

If it needs less power at any speed, why was the modified scooter not able to climb hills? And why aren’t all EUCs connected that way?

 

Quote

it is another variant of managing a motor).

Shouldn’t it then need a different management system at the control board as well?

Edited July 18, 2021 by mrelwood

[RockyTop]

Yes the change is about 2:1

Edited July 17, 2021 by RockyTop

[yuweng]

 

[Mark13i]

17 hours ago, RockyTop said:

Yes the change is about 2:1

or maybe 1.732:1 (√3) - theoretically? 

Edited July 18, 2021 by Mark13i

[yuweng]

Did abit of research, it seems easy but from what i hear it doesn't feel safe on an EUC, just like the variable field motor i posted above, may be its best suited for e-scooters or ebikes instead 

[Eucnhtusiast]

On 7/17/2021 at 2:52 PM, Boris Lämpel said:

Hey people,

I just met a guy who reconnected the motor wires of his scooter from Y to Δ connection to make it run faster. Because the relation of speed and torque is inversely proportional he sacrificed the possibility of climbing hills.
Did anyone of You figure out or even test if changing the connection would reform the riding characteristics of the wheel?
I just think of a switch where You could go fast on a Gotway HT model or even faster on a Monster pro.
I do not want to lose my motherboard by testing it out without knowing exactly what could happen. Any thougts?
Maybe the wheel electronics already switch the connection at a specific speed automaticly - I don´t know...

On 7/17/2021 at 11:14 PM, Boris Lämpel said:

It should (in my imagination) behave the same as the difference of the HS and HT models by GW. With a HS model You can accelerate, but there is not enough torque to do as fast as with HT models. You will only have a cutoff if You go far beyond the limit.

Do GWs really have a fixed speed limit? The output current is the factor limiting the wheel and starting beeping. But with Δ connection it would need less power for the same speed, so it should be able to go faster...

Our motor has 3 phases and they can be connected differently to the motherboard: like a Y (every phase connected only to motherboard) and  Δ (phase wires are connected also together - it is another variant of managing a motor).

Plugging the motor wires in the wrong order makes the wheel unusable, the control board knows only one order for the motor windings. And good luck changing that one order in firmware to accept different configs. If you still plug the leads the wrong way you'll get an out of control wheel that will flip itself over because it activates the wrong winding regarding to the position indicated by the gyro. This can be reversed by plugging the wires correctly.

"But with Δ connection it would need less power for the same speed, so it should be able to go faster..." completely wrong, delta uses 1.73x current all the time at every speed, in every situation because the torque constant is lower than in wye. Wye is much more battery friendly, 1.73x the range than on a delta wound motor.

The motors in EUCs are strictly wye, delta isn't even possible without completely reworking the wiring. Also 1.73x current draw(at every speed, in every situation) would burn out many control boards and possibly motor windings. 

The limiting factor regarding speed is not output current, but rather the lack of voltage to push that current through the motor. The back emf rises with rpm and once high enough, denies any current flow into the motor, thus resulting in a fall. Yeah, sure the KV goes up by 1.73x so theoretically a higher top speed is possible, but because the boards can't handle much more current continuously the torque would be much lower, leading to reduced performance at high speeds(read, lower than predicted top speed because of the limited current) while consuming the same amount of power. Consider this purely theoretical example below:

To get to 70kmh and reach the point of cut out you would need 3kW of power out in a wye configuration(~36A from the battery, consider this max safe continuous output before board burns up).

All else being equal a delta motor could reach a theoretical top speed of 121kmh before cut out, but would need 5.56kW (62.3A) to produce enough torque to keep the rider up. Now limiting the current to 36A on the delta motor would reduce the cut off speed to exactly 70kmh.

Here we have 2 different but still the same types of cut outs, both because the motor ran out of torque, but different because the wye motor ran up against the EMF and the delta motor couldn't get enough current to produce torque. So unless you are willing to engineer a whole new control board that can withstand higher currents and give up almost half your range, you won't be going any faster on a delta motor compared to a wye motor.

 

" Δ (phase wires are connected also together - it is another variant of managing a motor)." This would require rewiring the connections inside the motor, if you were to connect the motor outputs all together, you would short the board.

"You will only have a cutoff if You go far beyond the limit." Well, yes, but actually no. Going beyond the motors capabilities even just by 10s of watts makes you faceplant, especially at high speeds.

[Arbolest]

10 minutes ago, Eucnhtusiast said:

Plugging the motor wires in the wrong order makes the wheel unusable, the control board knows only one order for the motor windings. And good luck changing that one order in firmware to accept different configs. If you still plug the leads the wrong way you'll get an out of control wheel that will flip itself over because it activates the wrong winding regarding to the position indicated by the gyro. This can be reversed by plugging the wires correctly.

"But with Δ connection it would need less power for the same speed, so it should be able to go faster..." completely wrong, delta uses 1.73x current all the time at every speed, in every situation because the torque constant is lower than in wye. Wye is much more battery friendly, 1.73x the range than on a delta wound motor.

The motors in EUCs are strictly wye, delta isn't even possible without completely reworking the wiring. Also 1.73x current draw(at every speed, in every situation) would burn out many control boards and possibly motor windings. 

The limiting factor regarding speed is not output current, but rather the lack of voltage to push that current through the motor. The back emf rises with rpm and once high enough, denies any current flow into the motor, thus resulting in a fall. Yeah, sure the KV goes up by 1.73x so theoretically a higher top speed is possible, but because the boards can't handle much more current continuously the torque would be much lower, leading to reduced performance at high speeds(read, lower than predicted top speed because of the limited current) while consuming the same amount of power. Consider this purely theoretical example below:

To get to 70kmh and reach the point of cut out you would need 3kW of power out in a wye configuration(~36A from the battery, consider this max safe continuous output before board burns up).

All else being equal a delta motor could reach a theoretical top speed of 121kmh before cut out, but would need 5.56kW (62.3A) to produce enough torque to keep the rider up. Now limiting the current to 36A on the delta motor would reduce the cut off speed to exactly 70kmh.

Here we have 2 different but still the same types of cut outs, both because the motor ran out of torque, but different because the wye motor ran up against the EMF and the delta motor couldn't get enough current to produce torque. So unless you are willing to engineer a whole new control board that can withstand higher currents and give up almost half your range, you won't be going any faster on a delta motor compared to a wye motor.

 

" Δ (phase wires are connected also together - it is another variant of managing a motor)." This would require rewiring the connections inside the motor, if you were to connect the motor outputs all together, you would short the board.

"You will only have a cutoff if You go far beyond the limit." Well, yes, but actually no. Going beyond the motors capabilities even just by 10s of watts makes you faceplant, especially at high speeds.

 

*applause*

Very nice explanation!