PDA

View Full Version : AWD 4 Brushless motors & smarter electronics?


Ridin24z
2010.10.02, 08:21 PM
Hi all, this is my first post here. Very excited about sharing new ideas here.

I have a hefty project in mind that involves a complete over haul of my AWD Mini-z.

Currently, I've owned this MA-010 for about a year and it has a few great modifications. Li-po, higher current FETs, atomic AWD motor and an awesome MR2 body and wheel set.

I was recently looking at a few companies that are engineering cars with brushless motors in each wheel. That means, no power loss in drive train, better braking control, no diffs, even battery recharging from braking!

I want to do something similar to my mini z.

With everything but the wheels and steering ripped of the car, there's plenty of space for batteries and control/power systems in there and I'm sure the wheels can be fitted with magnets and a tiny stator be built within.

The first question is, can commercially available RC brushless motors go slowly and smoothly enough for accurate control of the car? (I'm talking about having the car slowly creep while lightly touching the throttle).

As far as I know, trying to achieve this with such motors may result in a sort of stepper motor action. Is this correct? Would a finer motor be required?

On the electronics side of things, I was thinking about using a dsPIC or FPGA system (only ones I'm familiar with). Any alternatives? Bearing in mind that the system will control the following:

-Battery management
-Drive of all 4 motors (24 digital signals to FETs and 4 analogue feedback lines!!!)
-Steering servo control
-RC Demodulation
-Adjusting drive from MEMS??
-Data logging/relaying??
-Adjustable/Programmable settings for all these systems

I want all the electronics on one or two small board that can fit into the shell (SMD of course)

I have machining facilities and am comfortable with making a more adequate chassis.

I know it sounds like a tall order but I what to turn this great little car into an engineering masterpiece!

Some may argue that it wont be a mini-z any more and it'll ruin the car, but personally, I'd like to open the car to be so much better =]

All in that great GTR shell ^_^

Thanks, all comments and ideas welcome.

Raj.

pedrocamp
2010.10.02, 11:05 PM
Quite a tall order as you say but I would like to see what you come up with. It is quite a small scale to be attempting a project like this. I would also say there wouldn't be a MiniZ part left... except the body.

cowboysir
2010.10.03, 02:42 AM
i can visualize what you describe but I dont see a purpose other than an imaginative engineering exercise. I'd be curious how you manipulate motor control to develop a differential type function (like mcLarens MP12 4C throttle steer system?).

I'd seriously be keen on seeing a drawing at least...schematics always make for fun reading.

color01
2010.10.03, 07:23 AM
The thing about wheel motor cars though is that they don't tend to turn well, you can't speed up a wheel motor electronically a whole lot more than max speed. I realize that BL motors can be run with different timing to modify speed, so perhaps that is what you'd have to do to achieve "proper" differential action with a wheel-motor car.

Currently there are some small brushless outrunners that can fit snugly in a Mini-Z wheel, and we (GSR) have toyed before with these kinds of motors in building a speed run car. For us, we only had to have two ESC's communicating with each other, and that's where we got stuck. If you want AWD you are going to have to deal with two more ESC's on top of that, and they each need to communicate with each other (L/R, F/B) for the car not to handle like crap.

This is a very tall order indeed, you get any headway with it it'll be quite a splash. Good luck. :)

Ridin24z
2010.10.03, 09:40 AM
Well I'm planning to build the esc myself. Want to get all the control on one chip. I have a nice FPGA board with lots of pins and a USB port =]. It's got plenty of processing power and 1Mbit boot rom. That way the feedback from all the motors can go all into the same chip and process the signals out to the FETs from there. Simple enough where hardware is concerned. Gonna be a pain coding it though. I'm tempted to machine the motors from scratch. May need to sneak into uni to use the lathe :P. Can you link me the motors you speak of?

This is the fpga board I own: http://www.knjn.com/shop.html?pg=img&src=5101

reaper
2010.10.03, 02:37 PM
as colour0 said there are some very small out-runner bl motors hitting the market right now some around 2g

i think if your going to get this working you need to be a little more realistic , by this i mean scale up your build
I'd go a least 1/10 get that to work then think about miniaturization

Ridin24z
2010.10.05, 06:57 AM
I think I should be able to get this working. I've ordered a nice little outrunner small enough to fit in the wheel.

I'm currently redesigning the chassis. The stock chassis wont allow an organised fit for everything. Gonna get it CNC milled out of 2mm carbon fibre (prob something cheaper to prototype with). It's gonna be a very simple design so it can house as much as possible. I'll post a 3D model hopefully later today.

In terms of control systems, PIC32MX460 looks promising. Found a nice breakout board with the chip and USB port on board. 30ish from Sparkfun.

Anyone know of any good MOSFETs to use with the BLMs? The Max current of the motor is 12A and is 8A Continuous. Bear in mind, I'll need 24 of them (6 per motor) :P. Rather not have to stack them if possible.