Calibrating Magic Controller to suit non-GM kit

[mrcarl]

Hi Everyone,

This is my first post but I have read a lot before posting myself

I have been searching the forums trying to find the answer to my problem but so far have had no luck.  I have a non-GM 36V 500W kit and the matching controller died.  I then purchased the Magic Controller and USB cable to suit.

Before I purchased the Magic Controller though, I purchased two other controllers and had to swap the colors around like so:

Motor side:        B|Y|G
                          -------
Controller side: Y|G|B

This (on the non-GM controller) gave me almost correct results - correct direction of travel (clockwise) etc. but a vibration through the bike until I reach full speed then the vibration stops (I assume this is due to less load being on the motor once up to full speed or it's the phase angle is different).

This why I got the Magic Controller.

I know the matching controller for my kit (dead one) says it's Phase angle is 120 degrees.

So when I attached the GM Magic controller, I had to use the same configuration as above before I got smooth operation.  The trouble is with the above configuration is that the wheel is rotating anti-clockwise.  Not a problem I thought, I'll just use the programming interface and USB cable to change the direction.....

After many times programming the controller and then fitting it up to the bike, changing the direction parameter from forward to reverse etc. No matter which parameter I used, the direction is always anti-clockwise.

So I shorted the reverse pin on the wiring loom and now the direction of travel is correct.  However, the torque and RPMs are much lower than the original or other two non-GM controllers.

This raises two immediate questions for me:
1. Why does it not change direction after updating the config via the USB cable and software, but a physical short does the trick?
2. Why are the torque and RPMs significantly lower using the GM controller?

So, I tried to calibrate the phase angle as per the instructions for the Magic Controller like so:

Motor phase Calibrating (Use it only if your motor is not from
GoldenMotor):
The controller can work brushless motors with different phase angles other than
defaulted 120 degree. Below are procedures to allow the controller to match the
motor phases:
    Cut-off power to controller and motor
    Grip and hold power brake lever, turn throttle to maximum speed position,
    press and hold the cruise control button, turn on the power switch (using
    battery key)
    Connect battery power to the system, release throttle, release brake lever,
    release cruise control button.
    Listen for two beep sounds, then turn the motor wheel forward and reverse
    once.
    When you hear a beep sound, motor phase calibrating is successful.
    If it fails, swap any two of the three motor phase wires, and repeat the above
    steps.

To achieve the above (because I did not have a cruise control button, brake switch, or battery key) I installed two momentry buttons for the cruise and electric brake, and installed a switch from the positive battery terminal to 'G' on the 12 pin connector on the controller.

No matter what I do to achieve the correct sequence to hear two beeps from the controller which signifies you are in calibration mode.  I never hear two beeps

This is driving me mad.  I have tried so many times, but can never get the controller into the correct mode, and this mode is not achieveable with the USB cable and software.

I wish GM had a series of dip switches that you could just use to enter the mode.

Please help guys...I am going crazy here....

Kind regards,

Carl  

[Bikemad]

Before I purchased the Magic Controller though, I purchased two other controllers and had to swap the colors around like so:

Motor side:        B|Y|G
                          -------
Controller side: Y|G|B


After many times programming the controller and then fitting it up to the bike, changing the direction parameter from forward to reverse etc. No matter which parameter I used, the direction is always anti-clockwise.

So I shorted the reverse pin on the wiring loom and now the direction of travel is correct.  However, the torque and RPMs are much lower than the original or other two non-GM controllers.

Hi andto the forum Carl.

When you say you "swapped the colors around", did you swap just the thick phase wires, the thin hall sensor wires, or both?

If you swap the two phase wires over on the green and yellow controller connections (and do the same with the two hall sensor wires) it should reverse the wheel direction without having to short the reverse connection.

Have you tried running it in sensorless mode by leaving the hall sensor wires unplugged?

Why does it not change direction after updating the config via the USB cable and software, but a physical short does the trick?
From what I've seen, the software can only enable or disable Reverse, it does not actually select it.
When reverse is enabled using the software, it basically means that the reverse switch is able to operate the "Reverse" function within the controller.
The "mode" setting determines whether reverse can be operated while still moving, or whether the wheel must be completely stopped before reverse can be operated.

The Forward and Reverse parameters that you refer to are actually located under the PAS (Pedal Assist) heading, and I therefore suggest that this only relates to the direction of the magnetic disc in relation to the sensor, which only seems to work when rotated in one direction.

Why are the torque and RPMs significantly lower using the GM controller?
The torque and revs might be lower purely because it's running in reverse.
I'm assuming that you've already set the "Reverse Scale Speed" to Max.
Swapping the phase and hall wires (as mentioned above) might make a difference, but we won't know until you've tried it out.

It will be interesting to know the outcome.

Alan
 

 

[mrcarl]

Hi Alan,

Thank you for your insight and warm welcome to the forum

To answer your first question, yes I only changed the thick phase wires around and not the thin hall wires.

I have tried your suggestions and have found the following (after removing the reverse direction short so I had the wheel running smooth but anti-clockwise):

Suggestion of running sensorless mode:

I connected only the + and - of the hall sensor terminal block.  When I powered up the controller I found that the wheel did not turn without spinning the wheel a little first to get it going (as is a known symptom of running sensorless), but it still did not change anything except now I had to spin the wheel first.  Nothing else changed.

Suggestion of changing the thin hall wires as I did with the thick phase wires:

By disconnecting and reconnecting the thin hall wires to match the thick phase wires as per my original post for the thick wires:

Motor side:        B|Y|G
                          -------
Controller side: Y|G|B

The motor did turn clockwise, but it was not right (very low RPMs) and made a grinding sound.  This is the reason I changed the thick phase wires around from their matching colors in the first place.  So the effect was it brought me back to square one by negating what had already been done.

For example:

Thick phase wires:            Thin Hall wires:          Thick phase wires:            Thin Hall wires:
Motor side:        B|Y|G           B|Y|G                                              Y|G|B           Y|G|B
                          -------   +     -------                  =                         -------           -------
Controller side: Y|G|B           Y|G|B                                              Y|G|B           Y|G|B


Regarding the RPMs and torque being less because it's in reverse mode:

The first thing I did after reading your post was to remove the physical reverse short I had put in to make the wheel go clockwise and there is no noticeable change in RPMs or torque in either direction.  Also, yes I had set the "Reverse Scale Speed" to MAX.

I am learning a lot though through this process.  Thanks for pointing out that the rotation direction in the software is under the pedal assist section.  I did not notice that...doh!

Alan, have you ever tried to get the Cruise controller in to "motor phase calibration" mode by doing this (as per the GM controller manual).

8. Motor phase Calibrating (Do not use it unless your motor is not from GoldenMotor):
The controller can work brushless motors with different phase angles other than defaulted 120 degree.
Below are procedures to let controller match the motor phases:
        Cut-off power to controller and motor
        Grip and hold power brake lever, turn throttle to maximum speed position, depress the cruise control
        button, turn on the power switch (using battery key)
        Connect battery power to the system, release throttle, release brake lever, release cruise control
        button.
        Listen 2 beep sound, turn the motor wheel forward and reverse once
        If hear 1 beep sound, motor phase calibrating successfully completed.
        Otherwise, swapping any two of three motor phase wires, and repeat above steps.
 
As I can not seem to get this to work.  No matter what I do, or how many times I try the controller never enters this mode (signified by hearing two beeps).

I would love for someone to try that may have purchased a kit, that way you would have a Brake switch, Cruise control switch, and battery key.  Because I can never get the mode to trigger by simulating two momentary switches and a rocker switch.

I would love to know the results as it may validate or disprove what I'm thinking; "Does this function actually work on the Cruiser controller?"

Kind regards,

Carl

[muzza.au]

G'day Carl,

You should try to rearrange your phase and hall sensor wires to a different combination. As you have 3 phase wires and 3 hall wires (minus the red and black as they don't change), then you have 36 different combinations. You will find that there will be 2 that work for one direction and 2 that will work for the opposite direction. Note: when testing a combination, do not use full throttle as you may damage your controller. I've done this myself when I've paired up different brand motors and controllers with no problems.

Attached is an excell spreadsheet that will help you find the right combination.

Muzza.au

[Leslie]

Does this help...



I think you need to swap the outside wires green and blue.

[mrcarl]

G'day muzza,

Thanks for the excel spreadsheet!  It's helped me to cover every scenario, which is fantastic!  I tested all combinations last night with a controller I got from Perth (not the Cruise controller) and I have found a happy medium with it.  It's still got a vibration through the bike at certain revs but at least my phase wires all look matched

Because I bit the bullet and ordered a full kit the other day from GM, I'll just leave the Cruiser controller as a spare for the new kit.  Saves mucking about with a mismatched kit / controller combination.

By the way, I'm in sunny Queensland.  Where abouts are you?

Kind regards,

Carl

[mrcarl]

G'day 317537,

Thanks for the picture and suggestion.  I've noticed you're quite active on this forum and that's fantastic.
Thank goodness for all of you generous people sharing all this info.  We (us noobs) would be floundering and lost without you guys and your sage like advice.

Keep up the great work!

Kind regards,

Carl

[muzza.au]

By the way, I'm in sunny Queensland.  Where abouts are you?

Kind regards,

Carl

I'm in cold dreary Melbourne, all my life except I did live in Maroochydore for 3.5 years, about 5 years ago though, so I always like to go back to Queensland at least once a year for holidays. Melbourne's not all that bad. We have great restaurants and culture and we are also the sporting capital of Australia.

I believe Leslie (317537, look at the numbers upside down is says Leslie) is from Queensland too.

Muzza.au

[MonkeyMagic]

I'm in Melb too Muzzatron we should totally race eachother

[muzza.au]

I'm in Melb too Muzzatron we should totally race eachother

Something I've been thinking about for a while is that there should be something like "swap meets" or some sort of get together of electric bike enthusiasts to just hang and compare bikes, race, or give technical support, repairs etc and even swap parts etc. Each city would have to organise there own etc. It could be like we meet once a month in a predetermined location etc. And it shouldn't matter what brand electric bike you have.

Just a thought

Muzza.au

[Leslie]

G'day 317537,

Thanks for the picture and suggestion.  I've noticed you're quite active on this forum and that's fantastic.
Thank goodness for all of you generous people sharing all this info.  We (us noobs) would be floundering and lost without you guys and your sage like advice.

Keep up the great work!

Kind regards,

Carl

Hi Carl.Thank you

I'm only doing what other have done before.  I don't usually need to post help threads as most of information most would need is buried here somewhere but I do like to post just the same.  Yep I'm in QLD too.

This ebike thing can.certainly require you to give up your left side of your brain to understand.better what is going on.  I love the electronics as it pushes me to take on good challenges, it would be hard to learn anything if I ddnt love doing it in the first place.

There is a lot more I'd like to do,  Id like to design my own controller and possibly build an ironless axial motor like the CSRIO or one based on iron powed cores or discover something entirely new and beter again.


Id like to work on magnetic suspension as I feel this is another place to regain wasted energy and a place to start looking at reducing bearing friction as well.

Like when you join all your phase wires on your motor you get resistance and can not move the motor.  The same theory can apply to magnetic shockies.  You could do up a great swaybar replacement system for tilting recumbents that actually lean to the amount you need.

[Leslie]

Actually I think there is a lot more energy to be milked from your shockies than I originally thought.  

Think of how many watts you would need to waste to make all your shockies behave the way they do when you use your vehicle.  You would need a pretty good motor and plenty of watts to similate this.

Then takie the energy it would take to simulate a bumpy road and put as much back into your pack as possible.  Looks to me like a pursuit worthy of a looksy.