Motor interface electronics

I think you might want to check the above statement for ... ah... ah... a typo.

"This is a stupid discussion. My statement about not wanting to use microcontrollers because I think they are too limited..."

"It is a stupid argument made for the sake arguing. "

"This is why there are always arguments, everyone here is so excited to critisize or find error, and then assume great ignorance or stupidity, rather than assume there was a typo. Geez."

It is an interface that may go away, but none of the concepts become invalidated.

Sorry, USB devices are typically more expensive as compared to parallel port and serial devices.

Cool, I didn't know that, what boards have them?

Well, the reason why I would assume is that, it is presumed that someone building a robot would have some sort of I/O card.

really?

In this particular thread, I asked a (what I thought was) simple question about a specific requirement, and by and large, the response has been to say I'm doing it wrong. This isn't an answer to the question.

When I try to defend my approach, people say I am argumentative. Sheesh.

I don't know why people can't accept that I wish to do something differently than they do.

How the hell do you know what my experience is?

This is the thing that bothers me the most. I am constantly bombarded by the absurd assertion that "many people with far more experience," and when I assert that I am not without experience, and that I may, in fact, know exactly what I am doing, I am called a grandstander. Sheesh.

It has become clear to me that at least one person "with far more experience than me" is a little confused about some very basic electrical theory.

I don't want to do this, and it is not nessisary.

Sure they have, maybe you haven't read all the responses.

As I have said OVER and OVER, latency is not really an issue for me.

One I/O card and I'm good to go.

Sure.

Because it is not a requirement, and I don't think it is very much easier or cheaper.

There are a couple links in a few of the threads.

In the late 1970s I built my first robot using an RCA 1802 and a mattel big track.

One of my first jobs was Denning Mobile Robotics, and I loved every minute of it.

I have spent the last 20 years on the periphery, sometimes doing things that directly relate: power and control systems, data acquisition and device drivers. Sometimes doing things the broaden my exposure to wider computing: massively parallel coding (clusters), learning pattern analysis, graphical device drivers, applications programming, and high speed server programming. A number of other projects as well.

I love robotics.

What's to understand about my motives?

At this risk of this being a reference to me, let me say I don't pretend to be an expert in all matters related to robotics. No one can be. I know just about motor control to get by. I know more about some things, and less about others.

But in my defense, you and I (and, to be honest, you and many people here) tend to talk at cross-purposes, so there are many confusions. Debate seems to follow you. You obliquely mentioned a circuit design using DACs for your motor control, and it took me some time and many re-readings of your posts to get a better idea of what you were doing. I think I understand now, but this is why I asked for some external references to your choice of circuits and approach. Sometimes when someone else explains a thing, it quickly becomes clear.

I think your plans will garner less anguish when you get to the point of actually posting some circuits, code, and pictures, as indeed you have started doing. Actual examples are a wonderful thing!

-- Gordon

:-)

Sorry about the confusion.

A lot of times I assume a level of knowlege or familiarity with a topic that isn't there.

It's an interface that _will_ go away, so what good are concepts without physical connectors?

parallel port

I disagree, USB is plenty cheap now. FTDI chips are not much more expensive than a max232. Even the cabling is cheaper to manufacture since it uses fewer conductors.

I have two ITX boards (a C3 800 and an M10000) The M10000 has an SMBUS connector, but I don't think that the C3 800 does.

I guess I don't fully understand your goal. I thought your intention was to design the whole thing, therefore you would pretty much pick the required i/o card to work with your programming.

including

When you say that peoples statements are "stupid", you are generally going to be considered as being argumentative.

They are only trying to help you. They, like myself, think you are not going to be able to complete all the subsystems without the assistance of a micro without allot of undue difficulty.

I have a fair idea, based upon your own statements. In the past years, I have read very many of your posts in COLA. You are about the same age as myself.

People who "know _exactly_ what they are doing" don't usually ask questions. ;-)

Are you going to have any ultrasonic ranging going on?

I think I read them all. I don't recall anyone saying that it absolutely could not be done with a single ITX board.

I'd really like a better description of said i/o card. Is it a PCI card?

We disagree again.

Could you point me to something matching what you had in mind for your particular project?

There are PCI parallel cards and the time-frame for parallel ports disappearing (totally) is probably measured in years.

I'll have to double check that. Last time I looked, they were considerably more expensive.

There are a few goals, some may actually conflict slightly, so it, like all design work, is a balancing act.

If it costs $30 in discrete parts to build something, but a ready made device is of comparable capability and price, you choose the ready made device because it is pprobably easier.

I was talking to a friend yesterday, and while I've been saying the system has this requirement or that requirement, my unstated or "audience" goal is a junor or senior in high school. The assumption is some metal shop, some basic computer programming classes. The only stretch would be Linux, but my son and nephews don't seem to have a problem with it.

I don't think I've ever said something was "stupid." That has a insulting quality that I don't like to use and it is hard to make such a statement without it being an ad hominem.

Yes and no. Yes, in that they want to help me use a microcontroller. No, in that any approach that involves not using a microcontroller is met with great resistance.

What "undue" difficulty would that be? I've already posted some code that maintains velocity using the mouse encoders and my motor amplifier.

Maybe so, but it 42 isn't that old, and it has a lot to do with the projects I've accomplished.

That's clearly wrong. I know "exactly" how I want something to work, now I have to find the best most cost effective and efficient way of doing it. That's called engineering.

Perhaps, and the interface to ultrasonic rangers is typically a pulse width output that represents the time between transmit and response.

There are a number of ways to do this:

(1) Uses a CTC chip to time the duration. (2) Use an interupt driven software routine. (If latency is a problem, I can hook the interupt directly and NOT let Linux handle it.)

When I find the one I want, or if I have to build it, I will post the info. Right now I am using a velleman board, but I am not happy with it.

Yes, and that is OK.

I currently have a velleman K8000 board. It was *always* considered a temporary device. One could build the circuits for about $25~30 dollars or less.

I am always making typos, letters in the wrong order, wrong letters, etc.

OK, clearly I did use the word stupid, and for that I am wrong and appologies to all who may have been insulted.

However, at these points the discussions had already crossed the ad hominem point, but you're right I did say it and that was wrong.

Most motherboards. SMBus (System Management Bus) was an attempt from Intel to reinvent I2C without calling it I2C. You can run most I2C devices from this. I've heard good things abour running a Deventech SFR08 sonar unit directly from the motherboard. I haven't tried it myself yet.

Google on "SMBus I2C SRF08"

-- D. Jay Newman

reply canceled.

The best way to win an argument is to shut up, and know that you are correct.

The best way to prove you are correct is to build it and show how you did it.

Rich

I already posted the mouse encoder PID code.