Thank you MisterBill2.
Let me make my project a little more learn.
I am trying to modify a old Power wheelchair as a RC control car. I want to use the motors and the control unit which employ a Hall effect joystick to control the chair speed and direction.
Here what I want to do is to replace the Hall effect joystick by a Arduino joystick as the link above. But the unit did not recognize the new joystick.

the link from the controller to the drive package is a 6 line cable use CAN BUS. I am using 433MHz RC control pair.

OK, that makes sense about the link, and it also means that you should be able to establish a radio link using the CAN BUS communications protocol. THAT part may be an interesting challenge. One caution is that the 433 MHz link is located within the amateur radio assignment and so you will need an FCC issued amateur radio license to use that frequency band. There are license free CAN BUS communication modules available in some of the license free bands, so that may be an option. The benefit is that you would have the control head already, not needing to do any redesign effort. So I suggest investigating the requirements for using the 433 band before transmitting, at least in areas that the FCC has authority over.

 

OK, that makes sense about the link, and it also means that you should be able to establish a radio link using the CAN BUS communications protocol. THAT part may be an interesting challenge. One caution is that the 433 MHz link is located within the amateur radio assignment and so you will need an FCC issued amateur radio license to use that frequency band. There are license free CAN BUS communication modules available in some of the license free bands, so that may be an option. The benefit is that you would have the control head already, not needing to do any redesign effort. So I suggest investigating the requirements for using the 433 band before transmitting, at least in areas that the FCC has authority over.

Thank you MisterBill2,
The CAN BUS used between the PAD and driver module only. The 433MHz communicate from my hand holder to the PAD.

 

Thank you MisterBill2,
The CAN BUS used between the PAD and driver module only. The 433MHz communicate from my hand holder to the PAD.

OK, it was not clear to me that the existing pad was still part of the system. So this will be a multi-stage remote control, with a separate hand held control package having an RF link to a receiver interfacing with the present control PAD that runs the motor drivers and other features. Some motorized wheel chairs also have a tilt function and even a seat elevate function, so those may also need controlling.
Duplicating a reed switch is simple, duplicating a hall effect sensor to control the speed is a lot more complex, because it requires duplicating the effect of a variable resistance, and possibly not being referenced to a system common point. So this project may become fairly complicated.

 

I wonder?

Reading through the posts, I am left with the impression that the wheelchair control unit (WCU) is expecting a series of on/off pulses for control. I think this because of the identification of the microswitches and the TS description of the operation of the original WCU.

It looks like the Arduino joystick module provides continuous position data (continuous ad in providing intermediate values; not necessarily continuous in time). Meaning more than just pulses counted by the WCU.

This may explain the inability of the WCU to recognize the Arduino joystick module. It is expecting pulses and is receiving continuous position data.

The Arduino can be programmed to convert the format of the joystick module to a series of pulses expected by the WCU. Check for the joystick return to home and only send a pulse when it is moved again. Or calculate the position difference and send a pulse every defined position delta.

Does this approach seem reasonable?

 

I wonder?

Reading through the posts, I am left with the impression that the wheelchair control unit (WCU) is expecting a series of on/off pulses for control. I think this because of the identification of the microswitches and the TS description of the operation of the original WCU.

It looks like the Arduino joystick module provides continuous position data (continuous ad in providing intermediate values; not necessarily continuous in time). Meaning more than just pulses counted by the WCU.

This may explain the inability of the WCU to recognize the Arduino joystick module. It is expecting pulses and is receiving continuous position data.

The Arduino can be programmed to convert the format of the joystick module to a series of pulses expected by the WCU. Check for the joystick return to home and only send a pulse when it is moved again. Or calculate the position difference and send a pulse every defined position delta.

Does this approach seem reasonable?

The TS finally mentioned that between the motor driver module and the control pod is a CANBUS link, which explains all of those big ICs in the early photos.So it is not simply a string of pulses, but rather a Computer Area Network sort of communication system. The project is growing much more complex as the whole thing gets explained.

 

The TS finally mentioned that between the motor driver module and the control pod is a CANBUS link, which explains all of those big ICs in the early photos.So it is not simply a string of pulses, but rather a Computer Area Network sort of communication system. The project is growing much more complex as the whole thing gets explained.

Ok, I got that. But is the CANBUS necessary? Without his Arduino remote, is CANBUS even used in the WCU? My point is that I think it is a series of pulses that need to be formatted for the CANBUS at the ARC and extracted at the WCU.

I thought that CANBUS was NOT in the original WCY. As I interpreted the design, CANBUS was an addition to allow communication between the (original) WCU and his Arduino remote control (ARC).

On a functional level, the addition of the ARC with a joystick module functions like this (speculative at best):

1. The ARC reads the joystick position
2. It translates that data to the commands expected by the WCU
3. This translated data is sent via CANBUS to the WCU
4. The WCU reads the data from the CANBUS (perhaps requiring additional interface circuitry on the WCU)
5. Something (?) interprets this data and presents it in the proper format to the original WCU

I don’t know if the WCU has CANBUS. I’m speculating that it doesn’t. Why would it when the control switches are on the WCU board?

I see the root problem is sussing the command protocol. From the TS description, I strongly suspect that he’s not going to obtain the proper protocol with that Arduino joystick module and will require some processing at the ARC. And if using CSNBUS, will require additional processing at the WCU.

This is consistent with the TS original question.

 

In the original configuration as a wheel chair of some kind the control head communicated with the motor control module via CAN bus, so as to avoid having a large bundle of wires between the control head and the motors drivers box. And that makes a lot of sense because then there are only a limited number of conductors and it is a lot easier to make a very durable cable with only six low current signal wires. So then, somehow, a receiver was intended to interface with the existing control head to create the signals to drive the revised package as an RC car. So the arduino is intended to translate, or something like that. So the whole thing has become quite complex indeed, because standard RC controlling is done with a very specific format of multiple pulse width modulation signals all in an analog modulated digital string.

 

In the original configuration as a wheel chair of some kind the control head communicated with the motor control module via CAN bus, so as to avoid having a large bundle of wires between the control head and the motors drivers box. And that makes a lot of sense because then there are only a limited number of conductors and it is a lot easier to make a very durable cable with only six low current signal wires. So then, somehow, a receiver was intended to interface with the existing control head to create the signals to drive the revised package as an RC car. So the arduino is intended to translate, or something like that. So the whole thing has become quite complex indeed, because standard RC controlling is done with a very specific format of multiple pulse width modulation signals all in an analog modulated digital string.

Thanks for the clarification. So many of my points are lessened in importance. But the remaining issue I see is the TS must discover how to translate the data from the joystick module, to the protocol used on the CANBUS. IMHO, the TS has insufficient information to program this protocol. Without (a lot) additional information, this becomes an impossible project.

 

Thanks for everyone's enthusiastic participation, let me try my best to explain some thing please.

1. The power wheelchair [PW], has two ~400W DC motors that were driven by a power module [PM], the PM was controlled by a pod which employed a joystick, shown at the beginning picture.

2. I want to change the PW as RC controlled, first of all, how to power on the pod, it's the topic up from. #6 sghioto helped me figured out the power on switch was build up by reeds and trigger by a small magnet.

3. #6 MisterBill2 and some body inquire about the reason to dig, the whole story gradually surfaced. Actually in my opinion, the key point is how the pod communicated with power module. There is a Six-core cable between the pod and PM, I'll attached the picture of it. there are 2 core for power charging, the other 4 cores voltage changed from power on to joystick tilt indicated. If possibly as my guessed the 2 cores at the middle for MOSFET trigger and the 2 cores right for PWM signal, thing easy , but I was told that the communication between pod and PM is CAN BUS, I confused because there do have a interface board with a infineon saf-c167cr-lm inside the PM(see picture).

 

Thanks for everyone's enthusiastic participation, let me try my best to explain some thing please.

1. The power wheelchair [PW], has two ~400W DC motors that were driven by a power module [PM], the PM was controlled by a pod which employed a joystick, shown at the beginning picture.

2. I want to change the PW as RC controlled, first of all, how to power on the pod, it's the topic up from. #6 sghioto helped me figured out the power on switch was build up by reeds and trigger by a small magnet.

3. #6 MisterBill2 and some body inquire about the reason to dig, the whole story gradually surfaced. Actually in my opinion, the key point is how the pod communicated with power module. There is a Six-core cable between the pod and PM, I'll attached the picture of it. there are 2 core for power charging, the other 4 cores voltage changed from power on to joystick tilt indicated. If possibly as my guessed the 2 cores at the middle for MOSFET trigger and the 2 cores right for PWM signal, thing easy , but I was told that the communication between pod and PM is CAN BUS, I confused because there do have a interface board with a infineon saf-c167cr-lm inside the PM(see picture).

View attachment 171916

The communication between he control pod and the motor driver module is a lot more than just PWM signals, that much is fairly certain. Looking at the circuit board in the photo of the drive module shows that there is a large section with digital logic, far more than would be for only a PWM drive arrangement. Have you powered up the whole wheelchair and seen how that control pod actually functions? Notice that there is what looks like a display screen at the top. And the one on/off control also says "mode". meaning that some sort of additional functions are available. So really the whole thing is far more than just a driver that could be controlled by an arduino board.
At the edge of the picture of the driver board is something that may be additional connectors, but the picture is not large enough to see what that part is. But it certainly is a useful picture. The different voltages seen on the connector are what the digital signals look like to a voltmeter, they are not simply PWM or analog. So the whole system is quite complex and probably able to deliver far more control than just driving the two motors. Does the wheel chair have a tilt function? Or a raise/lower function. The one person who I know has a chair that not only drives, it also can raise and tilt to stand them up. And it has a horn to warn people, and some running lights as well. So it may work far better for you to consider using that 433 MHz link to connect the CAN bus and not use an arduino as part of the system. And a 2-way communication system is a lot simpler than a whole RC arrangement.