Robin2:
That makes it hard to give useful advice.
Sorry about that, copy paste error. Mea culpa.
Have you tried the examples in my link?
I checked it out, it gives me some info, but I am still looking for something with encoder feedback, failure state and position correction. Thanks though, I appreciate it.
My german is very poor but I doubt if the motor comes with an absolute-position encoder and if it does not then you will have to home the motor to identify the ZERO position.
Yes, you will of course have to home the machine to calibrate, but as I say, not every time you start up. Once you have homed it, it can keep track of it's location and you can save the location to the EEPROM which will persist through power cycles, and if the device gets moved or misses steps, it doesn't need to rehome. This can be a time saver depending on the application.
There are other ways to avoid damage - a big RED switch or its equivalent. IMHO choosing a motor that is only marginally capable is building up trouble for yourself.
You want your stepper to be properly suited to the project, no bigger. Powerful enough to do the job it's supposed to do without missing steps but also not so powerful that it powers through a problem that it shouldn't be powering through. Obviously, you want to give it a bit of a margin of error, but not more than is reasonable. This really depends on the application.
For instance, one of the things I am making is an automatic moving saw stop to measure miter cuts that would run up and down the workbench of my miter saw. I should use a fairly weak motor for this. One that is more than powerful enough to move the stop back and forth, but if it bangs into something substantial on the bench, it should pause, beep and need to be resumed after the obstruction is cleared. I can only do this with a relatively weak motor and an encoder to know I missed steps. If I used a giant stepper that wouldn't stop for anything, not only is it overkill and a waste of money, it could knock all kinds of stuff around that it shouldn't, potentially into the saw operator, god forbid while he was operating the saw, which could be quite dangerous.
This kind of feedback based control is required, among other things, for any automated machinery that is not physically locked out by a cage or some such to prevent human contact with the machinery in industrial settings.
E-stops are certainly a must for any device that can do damage, but 99% of the time, even if you are watching it like a hawk, by the time you react and slap that button its already too late. And really, who watches their machine alertly for hours on end as it runs?
If the machine stops as soon as it knows it missed a step, it's nearly instant and has only moved an mm or less into trouble, and can save a lot of heartache. The problem can also be corrected and the machine can resume without losing it's work or the machine. I have had to replace machinery in factories without feedback control because it lost it's position and destroyed itself for want of a simple encoder. Needless to say, the new machine was closed loop.
I really don't understand why everyone here is so against encoders. I see it in other posts too. It doesn't cost that much extra, is not that much more difficult to program and only adds benefits to your project, is safer, and I can't think of a single instance where it harms. So what if you never miss steps and never need the encoder? What did you really lose?
My reservation is that it may not be able to time the steps consistently if it is distracted by WiFi. Using an ESP8266 in conjunction with an Arduino gets around that problem as the Arduino could choose when it takes time out to communicate with the ESP8266.
By all means do some tests and post your results to allay my doubts about the ESP8266 behaviour. It would be useful information for others.
That's what I was thinking. I don't doubt you are correct. I really have no experience with this hardware and software. Though I could disable the WiFi too. I don't really need it. I'll probably play around with both methods.
Any thoughts on the ESP32? Powerful enough to have my cake and eat it too?
Thanks again for the feedback.