Duplomatic BSV N 160 Werkzeugwechsler / Turret

Hi Bruno,
I tried to compile the component on a 2.8-rtpreempt system and it does not compile. Well it does, but only creates duplo file in /usr/bin and not a .so in /usr/lib/linuxcnc/modules/ like other comps.
I see in your output that you have a rtai setup. Did you ever try with rtpreempt?
Is there a reason that you use the option "option userspace yes", it doesn't seem to be so common anymore.

BR
Tom

In the meantime I changed to comp using the reatime function FUNCTION(_) and reduced the tool number to 8.
Let's see how the tests from Daniel work.

BR
Tom

Is there a reason that you use the option "option userspace yes", it doesn't seem to be so common anymore.

Hi,
when a component has a lot of time-delayed logic, it is not reliable in realtime, code parts are not taken into account.
In userspace, we are free to do what we want.

sorry for my google english ...

Hi
First I've tried the dulo.comp of Bruno. Didn't work as it should. I think it is a problem of the locking switch. Then I tried the realtime Comp of Thomas. No reaction of the Turret.
So I will move a little bit the placing of the locking switch, so it will change the state a little bit earlier. If I understand well the sequence: After the turret is in position, then the motor should for 200ms turn CCW. Does it matter, if the locking switch changes the state in between these 200ms? After these 200ms the indexing solenoid should open, motor should stop and motor brake should tighten. Is that so?
BR and thanks for all your help!
Daniel

Hi
I tried again the duplo,comp. But it is true, the MESA card is too weak to connect directly the indexing solenoid and the motor brake. I must connect these components to a relais.

Until now: Sometimes the turret moves nearly correct. But then, about 30s after locking, the turret begins to turn again, locks again and so on. Strange...

This Duplomatic ist really not so easy to configure.

Daniel

Hi,
Sorry for my silence, I'm very busy right now.
You should use halscope with all the turret entrances and exits to figure out what's going on. You can post screenshots.
The difficulty comes from the fact that everything goes very fast.
The rotation is reversed until it locks. The delay of 200ms is after this locking.
It takes time to fully understand the time cycle of your document (page 16).
There is a screw behind the motor to turn it manually. By pressing the indexer by hand, you can check if you can manually lock the turret. You have to feel by turning the motor that there is a spring inside, it takes several turns to feel the lock.
Is the VDI disc present? Its inertia is important for locking.

Good luck!