Speed of Motor will be controlled by gain

The LM741 is a very old op-amp and is most likely a bad choice.

This is a news group where we discuss designs. You need to make a start on figuring out what the circuit must do and how to do it before you are likely to get much help here.

Brush up on your Ohms law. You're going to need it.

How much power does the motor need? Usually an OpAmp doesn't output the required power for a motor, you'll need to add a power transistor or a MOSFET. It depends on your motor type, but if you build a PWM circuit, you could save big heat sinks. With a MOSFET you don't need an OpAmp any more and with one or two more transistors you can build the PWM.

LM324 comes to mind but there should be a transistor between that and the motor. Possibly a rather big one. Plus some resistors.

Yes, and best to post in the s.e.basics group. Also, I'd suggest to get a copy of "The Art of Electronics", it's all in there.

And do it before someone tries to have Ohm's law declared unconstitutional ;-)

Or just use a CD40106 as PWM oscillator and drive a FET with that. If it's a big one with lots of Cgs use a pair of pnp/npn as buffer. But I think the OP is just beginning with all this stuff so maybe he should get his DC circuit working first. Then move on to PWM. Then maybe to a Blackfin DSP with a nifty PID algorithm in there ;-)

They already did. The new law is only in effect only on February 29 on odd numbered years. ;-)

As Ken says the 741 is so old, and not really suitable. You could try the second circuit on this site

What test equipment do you have?

martin

A Blackfin DSP would be idle 99.9% of the time with a running PID algorithm. But it should be fast enough to drive a VGA output and to run a Linux system. Then you could implement it with Java and a nice GUI, which should use the Blackfin to full capacity :-)

Isn't that what modern operating systems are supposed to do? Generate tons of work for the processors in order to support sales at the semiconductor plants ;-)

Oh, and then he'd probably also have to upgrade to 1GB or RAM ...

LT1636 may be a better way to go. LM324s don't pull up very well or very far and its inputs aren't as robust. When power is removed suddenly, teh inductive spike from the motor will need some snubbing.

At work we often refer to "Ohms opinion". Ohms law often does not apply in the marketing department.

Use a LM555. You can get the PWM and buckets of base drive.

snipped-for-privacy@frank-buss.dehttp://www.frank-buss.de,http://www.it4-systems.de

It may not be. The OP may want to sample the tach with a 24 bit converter at 44KHz so that he can handle rapid changes in the load. He may also want to sample the motors voltage and current and linearize the torque and move the gain cross over frequency of the servo loop up to about 5KHz so that he can resond quickly and accurately to those load changes.

We know that some motors cog as they rotate slowly, at low speeds he could modulate the PWM operation to remove the torque variations and produce a smoother rotation.

If there is a CPU, you can bet someone will think of a way to use it up.

Yes a Blackfin does that nicely.

No, actually there would still be time left over. You could install a copy of something like "boch" and install XP on the virtual machine it makes then all the CPU time would be taken if the user opens one application.

snipped-for-privacy@frank-buss.dehttp://www.frank-buss.de,http://www.it4-systems.de

I don't like that circuit. I would at least add a glitch catching diode on the motor with its cathode on the positive side and a E-B resistor on the transistor.

I still think a LM555 would be a better way to go.

Actually, it's reality that doesn't apply. The sign on their door should say "Welcome to the Twilight Zone". :(

What are you willing to pay to have your homework done for you?

Good Luck! Rich