Hello all,

I have to come up with an idea of a project that must use multiple micro-controllers (16 to about 20).

Project is for educational purposes and is targeted for a group of people with intermediate skills in electronics and embedded programming.

Any sensors, actuators, displays can be used. The only restriction is that the CAN communication protocol must be used as the main communication between the micro-controllers (perhaps not between all of them).

Any ideas would be greatly appreciated

 

CAN is just a physical layer/link layer protocol. Was your project going to use an existing higher level protocol on top of CAN or were you going to invent your own new new thing?

 

Is it safe to assume that this is a laboratory project, and would not be put to an application?

When I first read your post, I thought an automated garden that could measure soil moisture levels and selectively water necessary portions of the garden. However, this would probably only use a few microcontrollers (depending on garden size, of course).

If this is for a lab setting, it might be interesting to have the room riddled with sensors and controllers attached to the bus. Then display all of the information on a computer, where you can set preferences of what are acceptable parameters and possible solutions.

This is off the top of my head. You neglected to mention how much time is allotted for said project...

 

AFAIK isn't CAN aimed at vehicle type communication bus, wouldn't Modbus be more universal, with more available resources?
Max.

 

Hello,

CAN bus is also used in our mearurement equipment.
It is used for setting and controlling temperature, controlling the modulator (frequency, pulse width and pulse amplitude) and receiver settings.

The following page of the EDUCYPEDIA has a lot of links with information on the canbus:
http://educypedia.karadimov.info/computer/datacommunicationbuscan.htm

Bertus

 

CAN is used for automation, but it was developed for use in automotive applications. It was designed for noise tolerance and safety-critical data integrity because those were needed for use in vehicles.

If you want to use that many micros then you should make it easier on yourself. Pick one or two types of applications that can be broken up and tiled onto multiple micros. Needing only one or two circuit board designs that can be made in multiple will keep the project easier to manage.

For the educational aspect I would think that two or three different micros would be the sweet spot with two actual circuit board designs.

For sake of making the project exciting and relevant I would go with the hot topic and plan on a 3D printer.

To justify so many processors, make it impressive, and multiply the excitement I would make it a huge 3D Printer.

An idea I have had that I will give to you to research and develop.

Think of it as a big InkJet printer but spraying a liquid ceramic cement.

So first ganged boards will control a preheater element, the water/fluid line flow valve, the powders/binders measures and mixes, the application nozzles, and the bake heaters. Each one will have a set of sensors that are used for feedback and go/nogo reporting. Each one might work a 20 to 50 cm swath so with 10 of these boards you should get at least 2 meters

The other board will interface to a computer, parcel data out to the workgang, control the advance of their gantry position, and monitor their reports, plus position/limits, and emergency stop.

The micros are a big enough job, but so much depends on the ceramic mud. I know there are a wide variety of grouts and muds out there so that there must be something out there that could be made to work.

Interesting enough question that I felt like trying to offer a suggestion.

Another way to do the same sort of project but at higher cost and more valuable material might be to spatter weld.

A simpler way would be to make a sandcastle. Just sand and a plasticizer like Carrageenan (Seaweed pulp).

 

I'm surprised to see how many replied. Thank you all for this. I will study the proposals carefully.

The project will take place in a laboratory and the allocated time is about 2 months. The level of complexity should be intermediate.

Unfortunately, CAN has to be used in this project.

Any more ideas, suggestions are welcome

 

A water sculpture fountain.

 

How about a car board computer? You can get pretty much anything out of that and the CAN will be at it's home.