Hello everyone,

I am in a project with a CPU 1512SP F-1 PN and 5 Servos. The controllers are SINAMICS S210 PN (6SL3210-5HE10-8xfx). They are working well but the only thing is that one of them stays with the Homing_Done bit in ON all the time, even if I shut down the machine and turn on again it stays with that bit in ON. I already did a factory reset 5 times and remake the parametrization and downloaded the parameter.

May anyone can help me to solve this problem? (I am really struggling whit this and that bit is very important for my application)

Hi there,

basically the only possibility why the technology object always shows that it is homed, even after power-off/on, is that this one TO is parametrized and used with an absolute encoder, and you did the absolute encoder adjustment once. Then the behavior would be alright, since the whole purpose of absolute encoders and absolute encoder adjustment is, that the axis can define its absolute position correctly when the PLC runs up, and therefor the TO then shows that it is homed already.

The only other alternative I can think of: You do a homing somewhere in the program by accident. Depending on the homing mode, there does not even have to be a movement, so mechanically you might not even see anything happening (e.g. with MC_Home.Mode=0, which is the default value, the actual value gets set, but no movement happens with this, and therefor this also is done quite fast).

If you don't have an absolute encoder and the TO parametrized with absolute encoder type, then doing trace measurements would help for sure. Tracing the statusword of the TO can help you find out if a homing command is triggered, you might also directly record the calls of MC_Home. I'd then set the sampling of the trace to OB MC-Servo, which is where the TOs get calculated and actualized. So with this you'd ensure that you record every value-change at the TO.

Regards,
daviso

Hey Daviso,

Yes, actually I think that is what is happening I did a homing with absolute encoder by mistake with that TO, do you know how can I go back from this in the TO?

Because the rest of them are working well and it's just with this one that i did a homing with absolute encoder.

I already deleted the TO and created again but i think somewhere in the PLC has that configuration for that DB. Do you know what i can do to solve?

Kind Regards

Pedro

Hi there,

to be honest, I do not really understand why you want to do this. If this is really caused by the TO using and being parametrized with an absolute encoder and you having done an absolute encoder adjustment, then having the TO in homed-state is really what all this (absolute encoder and adjustment) is about. So why would you want to loose this?

What leads to the homed-state being reset with absolute encoders was already posted by Technology Team above. One of these things will reset the homed-state. The only other thing I can think of would be an encoder error at the TO, while this is active (MC_Power.Enable/Status=TRUE). But this also is not really an option, since it would mean to produce an encoder error by purpose, just to get rid of the homed-state. For one I don't really have an idea how to automatically get the TO in such a state (without manually interfering, e.g. by unplugging the encoder while the TO is active). Also this is not really a workable thing to do. And, again: Having the TO in homed-state after absolute encoder adjustement was done once is the whole purpose of this.

What's the use-case behind your question? Why do you want to get the TO out of the homed-state? Perhaps looking at it from this point of view might lead to a solution...

Regards,
daviso

Hi Daviso,

Ok probably I have to explain you better the situation, I didn't want to home the axis by absolute encoder, I did by mistake. What is happening right now is always that I start with the machine (after it is power off) that axis always have homed_done in ON state, but the position of the encoder is unknow let say 270000 even if you are in 0 position. So the controller already doesn't know where is the axis but the TO thinks that is already homed but it does not.

This is a big problem because my initial conditions looks for that bit in order to Initialize my app, so if I tell to the axis to go to 130 (for example), it is going to be a mess and possible I will break something on the way.

That's why the others axis are working properly (they were configurate for Homing not as absolute encoder) but this particular axis have this problem because I did at the beginning homing by absolute encoder and then i changed, so I would like to lose that bit for make the homing not as absolute encoder but I cannot accomplish that.

If you need further information please let me know 

Hi,

but if you have an absolute encoder, and the position the TO determines from this is just currently wrong, why not do another absolute encoder adjustment, and then it fits again?

Anyway, if you want to get rid of the homed-state from the absolute encoder adjustment for now, change some encoder parameter of the TO temporarily (e.g. steps per revolution or something like this in the TO), compile and download it. Then the homed-state should show FALSE. Change the value back to the right value, compile and download again, so that the parametrization is correct again.

If you then don't do another absolute encoder adjustment (MC_Home with Mode 6 or 7), then you'll probably have what you want.

Regards,
daviso

Hi Daviso,

I reset the memory of the CPU and solve it.

Thank you very much