Motor interface electronics

I started with an apprenticeship with TAFE course as an Aircraft Maintenance Engineer Avionics sort of a cross between electrical , instrumernt and radio trades, fitter and technican.

Had to do anything from install new systems sat nav and comms, GPS , entertainment systems , to replacing and refilling oxygen , new fuel quantity sensors and harnesses, reconnecting aircraft engines and generators , ignition systems , uploading new system software, replacing light bulbs.

Worked as an AME for ten years.

Currently doing a Physics and Electrical Engineering degree. With a bit of luck should finally finish the Physics degree this semester then probably another two years to finish the Electrical Eng. Its a six year full time degree, which I'm doing sort of part time.

I like Programming robot controllers by Mike Predko uses pic16f627 but is easily used on other 16f pics and easy to get working on 18f pics.

The other fun thing is helping people fault find and fix faults via a online subject forum.

About half the problems can be related to software problems, things like installing mplab with anti-virus softwware running (has a tendancy to stop the debug arrow from being installed) or xilinx webpack diabling all usb ports.

Or installing software in the wrong order.

Then majority of the rest of the problems are caused by bad soldering,

Best way to learn I reckon , build your own circuits on a bread board, also gives a good feeling when you get something working.

Unlike a premade board , using a breadboard you have to learn a bit more about electronics, things like reading chip pin numbering or how to read resistor values and how transistors work and how to actually connect things together.

Alex

Wouldn't say its all hardware. More of a focus on smaller more easily buildable stuff.

Or doesn't even work (seen more than enough PHD type projects) that should have stayed on the drawing board.

Also know quite a few academics who think that building practically working circuits and building hardware is only for technicans not something that engineers should dirty their hands with.

Want to get a CMU type line following contest going at the uni I'm at (I'm secretary of the Engineering society 700+ members) but hit a brickwall in the mechantronics faculty guys. Still working towards it but going to take a good bit longer.

I agree with you but personally prefer a distributed controller approach. Basically having smart sensors etc so have a micro at the sensor that processes the raw data then sends it to main controller in the format you want.

Doesn't just have to be a micro at the sensor or motor can also stick in analog circuits to smooth out rough edges due to the digital or other problems.

If I had the time, I'd like to have a good look at the psoc from cypress gives you a mixture of programmable analog and digital blocks with a micro in there as well.

Also helps with timing issues and simplifying the main controller. On power use I prefer not working with 240V, to easy to screw up. With small battery stuff it doesn't matter if I'm half asleep or freshly awake, if I screw up the worst that can happen is having to replace a blown chip or similar.

Got all the bits for a larger bot using electric bike / scooter motors and medium gell cells (Oatley electronics) but need the time to get it built.

About another seven weeks until the end of semester and having enough of a struggle with electromagnetics and optics (wave packets and so on). Also doing a High Performance Computing subject parallel programming in fortran 90 / 95 using openmp and mpi.

Alex

You can buy them from quite a few places around but not always that cheaply.

a good resource for where to buy things is oz over clockers

wheres the smiley ??

Dons very good.

No don't buy any Dick Smith pic programmers. Total waste of money.

One of the kitsrus.com kits from dontronics for pic programmer and get one with a zif socket.

Or build a cheap avr programmer or if you can afford it an atmel stk500.

Alex

Thanks - some of the points might be a little subtle if you are not used to the language typically used to talk about this stuff. I appreciate that you appreciated it :-) I guess my main point was that the term "real time" is actually very precise, within the circles that use the term in the way that it is being used here, and to give one example that I think most everyone can relate to - CD burning. "Real time traffic updates" - Puleeeaze.

Exactly.

I agree. I really want to see him do it without any specialized support hardware. PC MOBO only, as he has said! I'm wondering how he's going to generate a 20KHz PWM signal using his general purpose I/O card, continuously, without interruption - not just for short bursts at a time, as required to drive a simple h-bridge. Let alone multiples of those, each at a differing duty cycle. And still be able to do his vision processing, etc, etc. Microcontrollers have built-in hardware to do this - just set up the facility and plug the duty cycle into a register. Set and forget.

Heck - even R/C servos which are the dregs of motion control and if your signal is off by just a few microseconds you get jitter. I just don't see you having this kind of control using a PC parallel port where system response is highly dependent on load.

Oh, nevermind. I just remembered - he's using a DAC to drive some op amps. I'll take a good reliable h-bridge with high current MOSFETs anyday, though. Check this one out:

"Good judgment comes from experience, experience comes from bad judgment"

I love that! It's so true. More so for some than others :-)

I seriously hope MLW doesn't give up, though - keep at it. Persistance in this field is key to success! You never know, maybe the end result of all this is that he will design a 40mA Athlon based PC motherboard with all the bells and whistles that we need for our robots that runs on 4 AA's and make a fortune selling it to us for $99.

-Brian

I think a bare-basics robot using a mini-ITX can be done for under $500, and less if you assume you're going to scrounge. It all depends on the "extras," but here are some prices for readily available parts:

Mini-ITX board (VIA ML5000): $135

Total: $463

We're assuming a free OS of some type. I intentionally didn't include some things like a mouse for hacked encoders as you can get those for free out of a dumpster, or for very little at Fry's ($3).

To make it cheaper you'll have to scrounge, and you'll need to in order to have money left over for sensors. You can probably get a used/refurbishbed hard drive at the swap meet for $25, a refurbished motorcycle battery for no more than $35, etc. You can build an IRPD for a couple bucks in parts from Radio Shack. A cheap USB Webcam would be a nice addition. I've seen them, new, at Pomona swap for $20. Not bad quality, either.

It *is* possible to *build* (not buy ready-made) a robot for under $500 using commercially-available parts, but none of this is new or unique. It's what you do with this off-the-shelf stuff that makes it special.

-- Gordon

Yes. I'm impugning your ability to listen, and benefit from it. To wit:

No, your point wasn't missed, and is repeated again here, so you'll know it was heard. "Less than $500 dollars." Response? There's no challenge in it. It's already passé. Use a modern micro for the motion control, the rest is easy.

Supporting argument::

The "$250 robot anyone can build" is already built, and running. Several copies are now being cloned at universities and within clubs.

It does PID control of two differential motors with profiled control. It does Steering, so it goes in a straight line. It does Odometry, so it knows where it is in a grid. It does drive-to-a-point Navigation, so told to go to a point in a grid it gets there. It does sonar ranging. It does IR ranging. It does obstacle avoidance, so if something is in its way, it goes around it. And it can do more.

The universities are doing remote monitoring via cheap RF serial cables. BeerBot uses the serial to a cheap speech synthesis module, btw. Anyway its an unwise bet against a DSP talking, if it needed to. It can also read CMUcam's at 115,200 in the background. Vision, Navigation, Remote Monitoring, already in practice.

("But wait, there's more!" :)

It can do GPS decoding and drive to a designated GPS point, It can do networking via RS-232, CAN, SPI, and reads I2C from Devantech Compass, and SRF-08 rangers. It has left over A/D's, it has left over PWM, it has left over timers. etc. And what's amazing it does this in a high level language, within a real time OS, is interactive, and has left over processing power.

So actual running hardware is sufficient support to suggestons given.

Robot base with micro? proven reality $250. Remaining $250? Buy a Mini-ITX PC-whatever if you must. Money left over? Go splurge. For everything else, there's MasterCharge.

Yup, just about sums it up.

interrupt driven I2C DAC off the parallel port to an LM12.

I think that this is the time for him to maike a new BIOS, and write an OS in ASM.

Those look like 18201's on that board. Trying to make a toaster ;^)

Try 16 LM12's and you can cook a pork roast.

You have motors, but you need motor drivers like h-bridges or similar. I guess relays can work but even those aren't all that cheap. Control would be rather hideous, though.

Also, don't forget his $50 I/O card.

-Brian

Of course they are small. I generally prefer to use Linux boards for the brain of my robot.

However, because the microprocessor is small, it can handle one task very quickly. For example, the motor driver I'm using uses an AVR microcontroller to handle the PID and so forth. I just sent it serial commands via the RS485 port (or RS232, or I2C, or other means I've forgotten about).

I use others to control one or more sensors.

I have both a VIA M10000 that runs around 25-30 watts, and an iBase 890c with a Pentium M that runs at least 60 watts.

And I *still* use microcontrollers to do the nitty-gritty stuff.

I would love to see you make it. I don't think you can do this without microcontrollers controlling the motors and sensors. And I think that you'll need them.

I would love to see a robot base that has all of the basics:

- motors and their controller - a sensor controller (or more than one) - a common bus to hang this all together

All would be (IMHO) controlled via a serial bus (I like RS485), but USB would also be useful. And most major microcontroller manufacturers are producing controllers with USB.

Gack! Frankly I am very disappointed in the low-level Mini-ITX motherboards. I would prefer a base that allows you to put whatever Linux box you want on it.

For some problems it is a good solution. For others, not so much so.

-- D. Jay Newman

Out of five computers , only 1 computer here has a ps2 port , only two have serial and parallel ports , three have a serial port and the other two here are usb only with no pci slots.

From pc98 onwards microsoft has stipulated that new pcs should not have parallel ports and more recently no serial ports.

Luckily for us , there are still manufacturers who produce pcs with serial and parallel ports but these are steadily decreasing.

Looks like even the new Via ITX boards are have dropped parallel ports.

Other problem is a lot of the pci parallel port cards are 3.3V not 5V and quite a few are "windows" ports and not full hardware so won't work for all parallel port uses.

usb -> parallel are fine for printers but not much else.

For one of my bots for vision I've used a cmu cam

I'm going with a distributed approach as it lets me add features without having to rewire and lets me keep size down. My problem is lack of storage and work space.

A blackfin stamp board(can run uclinux)

Alex

H-bridges are among the things you can cobble together for very little, or (as I noted in another thread), get for free from semi companies. Outfits like BG Micro sell PNP and NPN power transistors for like a quarter each, so if you're willing you can put an H-bridge together for a buck or two.

I also did not include a couple of quad decoder chips from US Digital at $3.05 each, which I'd prefer to use rather than the PS2 interface. These would be personal choice items.

Actually I disregarded it! (Mike -- blueeuedpop -- will get the reference ) I prefer a USB adapter ($25) to one of those mythical $5 microcontrollers! Or if the budget would allow, I'd FAR prefer to shave off some bucks by buying refurbished hard drive, battery, etc., and use what's left over for a MAVRIC II, and subprocess with it. No shortage of I/O there. Randy's PlugaPod, at 90 bucks, also would save some headache in interfacing motors and quadrature feedback, among other things. In any case, this type of hardware is up to individual choice, as would be sensors.

-- Gordon

Not a kosher robot, I take it...

-- Gordon

I was waiting for someone to do this joke! (I almost did it myself once or twice.)

-- Gordon

OK, so we agree.

Nonsense. A 20MHZ CPU (8/16 bit) is not faster than a 2GHZ (32/64) bit CPU.

That is cool, I never said microcontrollers do not have a place, but if you don't need one, why have one?

Ditto

OK, my bad, yes some ITX boards are low power.

ok.

Why do you think that? Why does everyone think controling a motor is any more complicated than playing a video or buring a CD? It isn't, it is simple math applied at fairly regular periods to adjust the output power. If I was *really* concerned about "real time," I'd put the routine in an interrupt handler.

Got that built, working on PID software.

The number and type are to be determined.

That's an interesting requirement, what for? Yes, there will be documented ways to handle the I/O but why make up some sort of bus, screw terminals will work. For switches and such, and USB, parallel port, and maybe I2C.

RS485? No thanks.

What are you basing this statement on? I.e. which version of Linux is known not to run on which mini-ITX board. I have had no problems.

I have yet to see a "problem" for which it is the best solution available. If you need something simple and small, SQLite is better and smaller, if you need a "real" SQL database, PostgreSQL.

You are the one who makes the ad-hominem attacks.

I'm not talking about a simple line walker or remote control scooter. I'm talking about a fully programable robot with a complete tool set that anyone can build.

Yes, I've seen the "beer bot."

OK.

OK, the design criteria has a few things that may not have been obvious:

"Total cost" is $500, that is OS, hardware, development system, interfacing, charger, everything. "Everything."

The development model is not micro-controllers simply because I want to keep one tool chain. I want people to be able to develop software on their PC and run it on the robot. When you have a micro-controller, you almost always need two, one on which to test, and one to run, unless you only use the one, but that sucks because you have to kill the robot to try something new.

The development model is meant to share code and expertise. A million roll your own microcontroller hackers aren't going to do anyone any good. Using Linux (or a BSD), it is possible to use more normalized API and build on open source.

lastly, while it is funny, I'm a bit curious about everyone so adamant about me using microcontrollers. Everyone is taking it so personally. I have simply said that (1) a PC can do everything the robot needs and I don't need any microcontrollers, and (2) that I wouldn't base my robot on a microcontroller because they are too limited.

Neither one of these statements is technically flawed. The first one is factual and rather hard to dispute unless you put artificial constraints on it like that buying an I/O board invalidates the assertion. The second one is clearly a subjective statement, and there are plenty of facts in support of my position and while my position is not one with which you may agree, it is certainly understandable from the design criteria I have stated.

Yet, you have resorted to insults, said I don't know something, called me ignorant, I think you have a problem.

Actually $99.

I've not seen $40, seen $65.

Fire alarm/instrument batteries. Deep cycle gell cells. $35.

Kids toy with two motors and steel axil, $65~120

$60G Microcenter $29

Linksys or Belkin wireless router $30 (on-line or microcenter with rebates). EPIA board comes with ethernet jack. The router actually has a secret, you can install a custom Linux on it.

My total: $355

$5.00 or less.

I have almost $150 to play with on my budget. one or USB cameras, one or two ultrasonic sensors, and of course, bump sensors.

Using a PC based development model is very important. The tool chain you use to create applications should be usable for the robot as well.

Hardware vs software. Hardware support is a lot easier to program for.

Also what latency or delay between the cpu and sensor or motor or motor controller ? Going to end up a similar time for both solutions just to get the data and again to get a control value back to the motor or sensor.

Depending on the way the solution is implemented, processing time is almost irrelevant of course the way you would do it on a pic and on a P4 would be different.

Because the cd burner or dvd burner have microcontrollers that run them the cpu just sends the data to the device in the correct format and at a rate it can hand via the drivers (its a bit more compilcated than that)

I think he means bus protocol or bus type like RS485 or can etc D.J , Brian , Randy and others came up with

Also good for keeping wire runs short use hardware to help with error checking of signals.

How does this compare with your project?

- John

Just curious on what language are you using ?

Thinking linux quite a few people would assume c for development.

c , c++ or something else ? asci c or c++ or a subset like embedded c++? standardlibrary or threadsafe ?

Going to have any backup for the hdd ? Or will the code fit on cf or mmc / sd ? or bot be able to be booted from cf etc ?

Are you going to have any or many failsafes ?

Also are you going to make it self hosting ? so you could halt the bot, adjust the code as necessary , recompile then restart ?

Are you going to have a console via bluettooth or rf so you can login and adjust things ?

Seen a few people use python for robotics