Controlling stepper motor with atteched encoder

[camcompco]

Hello Al,

I just wanted to say that i am still working on this, didn't want to come of as rude for not following up. As it turns out I really needed to re-re-re read the picbasic pro manual, i was relying on the basic i learned back when i used Lotus 123, much has changed :-)

I am experimenting (trying) on the 18f4431, it has built in Quadrature feedback, sounds like it would be good for my app??? thoughts??

any who, will make sure to follow up soon,

All the best
John

[aratti]

Hi John, yes PBP manual is of great help.
As far as QUI is concerned, this technique increases by a factor of four the resolution of your encoder, so an encoder with 500 pulses per turn will result in a 2000 pulses per turn. This at a first glance could seems great but using a stepper could create some problem. Let me give you a numeric example.

You have a stepper with a resolution of 200 steps per turn, using half step driving you will obtain 400 steps per turn. An encoder of 100 pulses per turn will be your choise, using quadrature, otherwise compensating for lost steps could be a serious problem since the step movement of your motor is a fraction of the encoder quadrature count. (In case of an encoder of 500 pulses per turn your motor step will generate 5 encoder counts per motor step)
Just bear in mind this complication in choosing the encoder if you will use the quadrature.

PS. Is your stepper motor 200 steps per turn or higher?

Cheers

Al.

[camcompco]

Hi Al,
yes, i have a boatload of new stepper motors, and they are all 200 steps or higher. Most of my encoders are 400 cpr.

I have many steppers that are gear reduced giving actual 0.18 degree per step translation at the shaft. My applications require relatively low speed on the steppers,60-100 rpm range , it's more a matter of being accurate, than fast.

I do not need the accuracy of 2000 pulse per turn, i just thought (without really knowing what the heck i am talking about ;-) that the PIC18f4431 QUI features would make hardware more straight forward as the functionality would be handles at the chip "hardware" level rather than at the motor controller level.

Again, total newbie here, just trying to figure this out so I can actually build a prototype.

At the end of the day, all i need it to get this darn thing up and running, as it turns out I then have to spend untold hours trying to just understand that I can do before I can actually do anything

i have been trying to get the combination of :

• Program in PBP
• schematic using parts I have So i understand how to actually connect things on a breadboard
• using motors and encoders that I have (almost exclusively HEDS-5545 or HEDS-5645 on double shaft VEXTA steppers)

So far, a simple, totally newbie, easy to understand without tons of research, etc. . . .has eluded me

Your fine assistance is well appreciated but was a little less than "turn-key" LOL and is still being worked on by yours truly :-)

thanks again for the continued help . . .man do I need it.


Lost in New Jersey

John

[HenrikOlsson]

Hi,
The QEI module in the 18F4431 is great. It can be set to decode in 2x or 4x mode so your 400 cycles per revolution encoder will give a count of either 800 or 1600. I don't see that as a problem even if you're running the motor in full step mode. More resolution is almost never a bad thing :-) In theory it should allow you to measure how much the shaft is actually lagging the commanded position and allow you to slow down when the lag start aproaching 2 full steps. Because the instant the lag equals 2 fullsteps the rotor desynchronizes and position is lost.

/Henrik.