Author Topic: Magic Pie 4 (Vector Control)  (Read 37980 times)

Offline Barbados

  • Confirmed
  • New Member
  • *
  • Posts: 5
Re: Magic Pie 4 (Vector Control)
« Reply #15 on: August 15, 2014, 01:04:03 PM »
is the controller in the Magic Pie 4 the same as in the Smart Pie 4?
how many and what FET's?

Offline Bikemad

  • Global Moderator
  • Professor
  • PhD. Magic
  • ******
  • Posts: 5,494
Re: Magic Pie 4 (Vector Control)
« Reply #16 on: August 15, 2014, 02:39:29 PM »
There is no capacitor precharge function in the vector controller so you will still get a nice big fat spark if you are using a battery without BMS and a much reduced spark if you connect to a BMS equipped battery that is already turned on before you connect. ;)

The controllers appear to be physically identical but they are factory loaded with different firmware to suit the particular motor.

The FETs are sealed underneath the circuit board inside the aluminium casing and I am not about to ruin a perfectly good controller just to find out what's inside it.
As these controllers are so compact, I would assume they only have 6 FETs to do the switching just like all the previous GM internal controllers.

Alan

 

Offline Barbados

  • Confirmed
  • New Member
  • *
  • Posts: 5
Re: Magic Pie 4 (Vector Control)
« Reply #17 on: August 15, 2014, 02:45:24 PM »
hello Alan, thanks for your answers!
6 fet isn't that much :-(  the fan really must help something. 6 fet is the minimum used in any
bike controller I know about.

and the missing spark protection doesn't sound good either

Offline Bikemad

  • Global Moderator
  • Professor
  • PhD. Magic
  • ******
  • Posts: 5,494
Re: Magic Pie 4 (Vector Control)
« Reply #18 on: August 16, 2014, 08:41:39 AM »
Well I've just did a test ride on proof of concept front MP III fitted with Vector with some mods on winding  (I was not told what exactly was done - phase shuffle?)
All is working fine. I do not see any ON/OFF behavior however torque is low, especially from full stop. It feels more like 2-3A on MP III however CA readings shows around 20-25A on full throttle. 
Actually I've tried to hold bike with rear brake and do WOT from full stop so it even does not spin front wheel while CA shows around 10-15A draw. Wheel just stays on the ground and after few seconds controller shuts the power to the motor due protection.
I'm not saying it has difficulties to start spinning but it's slow to reach it's RPM even when wheel is lifted. I can ride it from full stop but nothing to compare with original mp3...

Any ideas on why torque is so low? Could it be HAL sensors connected in wrong order or motor would not work in that case at all?


Who modified the windings and how did they know what was required?

The low speed torque from my MP4 appears to be a lot more than my MPIII so I would guess that your low torque is because the coil windings are still not optimised correctly to suit to the controller, perhaps your stator was simply rewired from Star to Delta configuration? ???

Is it any faster than it was with the MPIII controller?

I have had a good look at the hall sensor positions on both my original MPIII and the new MP4 and they appear to be in exactly the same position on the stator, so I know that my "all or nothing" throttle problem must be due to the windings.

I don't know exactly what the difference in the windings of the stator is, but it definitely makes a big difference to how the motor performs. ;)

Alan
 

Offline rkts

  • Confirmed
  • New Member
  • *
  • Posts: 5
Re: Magic Pie 4 (Vector Control)
« Reply #19 on: August 16, 2014, 04:12:25 PM »
Hi Alan,

Guys who modified it are working with GM products for a long time however it's  their initial experience with Vector controller.
This is rather experiment that I've volunteered as my original MP3 controller needed some fixing anyway.

Delta wiring might be exactly what they did. They were having issues to make it work correctly for few days and were unwilling to tell me what they did to make it work :) I will try to ask again.

While speaking about HAL sensors I was guessing not about position but rather incorrect sequence of hall sensors connected. I'm not sure what would be behavior if you change sequence of HAL sensors with this controller? Motor will go in reverse or will loose it's power?

It is not faster, speed is pretty much same but it's harder to reach max speed (due lower torque?) But I'm able to reach around 45kph (5x something while pedaling)
But for delta wiring I could explain it with 2 factors:
1. Speed It is SW that controls it and it's set to default 380RPM. I might hit that limit. I don't have Vector cable to test it.
2. In theory delta wiring is considered to be low torque at low RPMs but can reach higher speeds on BLDC motors. Are that 380RPM considered to be high enough RPM? For some reason I would say that 10KRPM is high and not ~400.

Offline GM Canada

  • Super Gary
  • Confirmed
  • PhD. Magic
  • ******
  • Posts: 1,544
Re: Magic Pie 4 (Vector Control)
« Reply #20 on: August 22, 2014, 03:04:01 AM »
Well I've just did a test ride on proof of concept front MP III fitted with Vector with some mods on winding  (I was not told what exactly was done - phase shuffle?)
All is working fine. I do not see any ON/OFF behavior however torque is low, especially from full stop. It feels more like 2-3A on MP III however CA readings shows around 20-25A on full throttle. 
Actually I've tried to hold bike with rear brake and do WOT from full stop so it even does not spin front wheel while CA shows around 10-15A draw. Wheel just stays on the ground and after few seconds controller shuts the power to the motor due protection.
I'm not saying it has difficulties to start spinning but it's slow to reach it's RPM even when wheel is lifted. I can ride it from full stop but nothing to compare with original mp3...

Any ideas on why torque is so low? Could it be HAL sensors connected in wrong order or motor would not work in that case at all?


None of this makes any sense to me. What you tested in my opinion is not possible to be the new vector controller. I have checked it out and see no possibility it would even fit. I would love to see so e pictures of this controller and how it fit in the hole and over the axle.

Gary

Offline rkts

  • Confirmed
  • New Member
  • *
  • Posts: 5
Re: Magic Pie 4 (Vector Control)
« Reply #21 on: September 15, 2014, 08:12:13 AM »
Hi Gary
Apologies for late reply. I was only able to get my hands on that this weekend.

What is the purpose for me to lie? :) I'm just decribing issues we've faced converting MP3 to MP4.

I think there is main difference between MP3 revs. So called MP3.1 and MP3.2. MP3.2 has a bit different controller?, different winding and, possibly, narrower axle.
Mine is MP3.2
I did take some pics of the controller but did not remove it as it is silicone sealed including axle. See attached.
It will be now converted back to MP3 but before that we'll try to understand a difference and what exactly needs to be done for it to work with this controller.

The official answer from GM is that it does not work but for some reason I don't believe it's not possible to make it work in worst case rewinding wires just for matter of science :)

I did swap my motor with Vector on weekend and I can confirm that performance boost is incredible.
I was not able to stop wheel from spinning (front) all the time on unpaved road with 100% Accel setting. Last time I was not able to spin wheel no matter how hard I tried.
It runs something like 5 times better in terms of torque and acceleration than MP3 with Vec and better than original MP3 with stock controller.

Offline Mgizen

  • Confirmed
  • Junior Member
  • **
  • Posts: 36
Re: Magic Pie 4 (Vector Control)
« Reply #22 on: January 11, 2015, 11:30:56 PM »
The new LCD panel (BAC-601) for the vector system can you limit wattage, and speed etc? Like the MG cycle analyst? Is GM Canada carry this product?
thanks MG

Offline Bikemad

  • Global Moderator
  • Professor
  • PhD. Magic
  • ******
  • Posts: 5,494
Re: Magic Pie 4 (Vector Control)
« Reply #23 on: January 12, 2015, 12:01:03 PM »
Unfortunately, it does not affect the power available from the throttle, it only alters the amount of power provided by the pedelec system, which is ultimately limited by the maximum settings in the controller.

The mode settings available during use will vary the power available from the pedelec which is produced in relation to the rider's pedal speed (cadence) and the "PAS Ratio", "Battery drawn current", "Rated phase current" and "Acceleration" settings programmed into the controller.

Alan
 

Offline Mgizen

  • Confirmed
  • Junior Member
  • **
  • Posts: 36
Re: Magic Pie 4 (Vector Control)
« Reply #24 on: January 13, 2015, 02:49:28 AM »
Oh ok so the cycle analyst is maybe worth the money in that respect. thanks for info Alan.