Need help figuring out which xmos chip(s) I need

Non-technical related questions should go here.
MeltManBob
New User
Posts: 3
Joined: Fri Nov 18, 2011 3:56 pm

Need help figuring out which xmos chip(s) I need

Postby MeltManBob » Fri Nov 18, 2011 4:20 pm

Hey everyone, I have no experience with micro-controllers, processors etc. I have a ton of project ideas that could really benefit from electronic control but I am unsure of where to start. I'm not sure what the limitations are of an XMOS chip or how many projects could be handled by a single chip. I'm on a pretty tight budget so any suggestions that save money are welcome. Essentially so far I am envisioning each project being created with it's own controller, possibly combining very similar projects onto one. Then I would like a main controller to interface all of them. This seems a bit similar to one of the layouts I saw here where an L1 was used for the normal functions and an L2 was added for connectivity and interfacing with other devices.

All of these projects are centered around a van that I am converting into a small camper. For example, one project is to use stepper motors to roll up and down shades over the back windows. Another would be the task of controlling a water pump to control the water flow from a small faucet. I would also like to interface the vehicle alarm, power windows, locks, radio, etc. There would be an interface panel inside the van but I would also like to create a wireless handheld device that could control everything like the panel inside the van.

Another project that is fairly important is the power control. I have six deep cycle batteries that I need to be able to recharge. The plan is to build a generator from a small motor and a vehicle alternator. I would like to be able to control the speed of the motor and adjust the voltage and current coming out of the alternator. This would also include monitoring the temperature, voltage, and current of each battery and logging this data. Down the road I would like to add 2-3 solar panels which will need a charge controller also. I would like to mimic an MPPT style charge controller for efficiency.

Some other projects that are less important would be a DC-DC power supply for a desktop computer so that I'm not converting 12DC to fake 120AC then into a standard power supply. Also I would like to build a small table top CNC machine.

Obviously this is a decent list and is biting off more than I could chew right now. The point of illustrating these different projects is to show how much they can vary and how simple or involved they can be. I realize that it is probably unlikely that one chip or even just a few will suffice. I was thinking of designing around the single core XMOS chip with the largest number of I/O's. In the end I want a nice user interface to control all of these things which leads me to think that I might need to implement a micro processor like an ARM to handle the interfacing between all sub-systems, GUI, and interfacing with the wireless controller.

So where do you guys think I should start? What things will dictate how much can be handled by one XMOS single core chip?

I have a general priority list of which projects should be developed first but I don't want to develop something that is hard to integrate with future systems. Right now the more important project would be the battery charge controller.
User avatar
Bianco
XCore Expert
Posts: 754
Joined: Thu Dec 10, 2009 6:56 pm

Postby Bianco » Fri Nov 18, 2011 11:48 pm

Even though nothing of this is out of league for a few XMOS chips I have to ask why you don't go for a handful of 8-bit microcontrollers in an easy to solder DIP package. Most of the tasks you want to implement aren't very CPU intensive.
MeltManBob
New User
Posts: 3
Joined: Fri Nov 18, 2011 3:56 pm

Postby MeltManBob » Sat Nov 19, 2011 12:21 am

Would I be able to incorporate those micro-controllers into one system later without a ton of work? The goal is similar to wanting a smart home that had all of the potential for smart systems built in at the beginning so that if something extra was desired later, it's just a matter of getting the device and hooking it into the existing system.
User avatar
leon_heller
XCore Expert
Posts: 546
Joined: Thu Dec 10, 2009 10:41 pm
Location: St. Leonards-on-Sea, E. Sussex, UK.

Postby leon_heller » Sat Nov 19, 2011 1:40 pm

It could be done using, say, SPI for inter-chip comms between 8-bit devices such as PICs or AVRs. Another option would be the Parallax Propeller, which contains eight cores on the same chip with 32 I/Os, any I/O can be controlled by any core. These devices are suitable for a beginner, as they are available in DIL packages, and are relatively simple.
User avatar
Bianco
XCore Expert
Posts: 754
Joined: Thu Dec 10, 2009 6:56 pm

Postby Bianco » Sat Nov 19, 2011 1:52 pm

You can use different chip architectures in one system and connect them together as long as they can talk the same protocol. I would problably use RS-485 to interconnect different chips. The transceivers are cheap and it uses a standard UART to send and receive data from the bus. Not to forget it is very noise immune due to the differential signaling. You will have to figure out a protocol on top of RS-485 though (master/slave etc). This can be daunting if you do not have prior experience with microcontrollers or embedded programming (and using XMOS will not make that any easier). One has to take small steps.
MeltManBob
New User
Posts: 3
Joined: Fri Nov 18, 2011 3:56 pm

Postby MeltManBob » Sat Nov 19, 2011 4:38 pm

I figured that starting with something like XMOS might be overkill for the beginning projects that I will start to learn on but I'm pretty certain that it won't take long for me to have a need for it's capabilities. Also I like that it can be programmed in a language that is not proprietary. I will be taking a C programming class in the spring as part of my electrical engineering prep so I'm already headed towards that language.

I guess for now what is really keeping me stuck is my inability to gauge what my projects are in terms of hardware and software complexity and requirements.

Is there a place where I can basically find a mentor to help me learn how to develop my ideas into real circuits?

I looked at the propeller stuff and that seems like it would work fine, could you compare them with the xmos? I would like to know what their limitations, usefulness, strengths and weaknesses are with respect to each other.

As for cost, I was originally looking at making my own board from the L1. Would making the starting board be cheaper than just buying one of the simple developer boards? Just curious in case I still want to go this route.

Thanks for the advice so far, although quite a few of the terms I have to look up to start to understand what you're talking about.

Who is online

Users browsing this forum: No registered users and 1 guest