OpenPnP with LinuxCNC via linuxcncrsh?

tommylight wrote:

-Why would anyone assume the Z axis home to be down? It is up, always and on every machine i came in contact with.
-Even if using pneumatic Z axis, it should have homing disabled if no air pressure
Or am i missing something?
I have yet to see a machine, any machine that does not home Z first.
I am sure there might be some strange machine that does that, but i can not think of one.
 

Yeah you are missing something. We are talking about homing 1 motor that controls 2 heads opposing each other. On a Mill it's pretty much guaranteed that the direction Z will move during homing is the safe direction. That is not the case here.

 

The pnuematic thing does not equate. Think of a see-saw where home is center Advancing positive retracts 1 person and advances the other person. Homing negative does the opposite.

If you have switches at the extremes of both sides then one head is fully extended when the other is fully retracted. If XY is in some willy nilly position how do you home Z and guarantee 1 nozzle does not hit something on the table?

If your switch is in the center, how does Z  take the shortest path to center if the machine does not know where center is because it has not been homed yet? If a home switch is in the center how does it find it without hitting limits first and going back to my last point?

In the above cases, I think it's safer to home X and Y to a neutral position first, maybe that Z center switch dissallows XY homing unless it is tripped, that is another safe option but you would still want to bring X and Y to a neutral corner first. The only time I think it is best to home Z on this type of head first is if you do have absolute position feedback on it.

So I mentioned the Potentiometer for absolute position feedback. In that case it actually might be best to home Z first because you can know exactly which direction it takes the shortest path to center, there's no posibillity of extending one head during Z homing unless the pot broke. If the pot breaking or misreading is a worry, than you can get into all kinds of complicated scenarios with both a pot and a center switch but I'm sure you get the point.

I think I would still home it to the top and then use your work offsets to select which extruder was in use. Any height difference could be accounted for then 
And yes home Z first.

rodw wrote:

I think I would still home it to the top and then use your work offsets to select which extruder was in use. Any height difference could be accounted for then 
And yes home Z first.
 

Extruder? I think you completely missed the point.

It's like an echo chamber in here. Understand the issue first, then offer suggestions

phillc54 wrote:

LinuxCNC would know which direction to home dependent on whether the home switch was activated or not when homing commences.

 

Assuming you had a home switch that stayed in contact the whole time (or at least most of) it was beyond one side of home.  The picture I posted was just an example of the way one side extends while the other retracts. Most of them are belt loops and are hard to add home switches to (like mine)

Definately getting away from the point of the post......lets bring it back now.
 

I had previously mentioned that I was looking for an alternative to the see-saw approach because of the homing problem. With any system, you can't move anything unless you know it is safe to do so and driving Z down on a PNP without knowing where the axes are is unsafe.

plhsystems wrote:

I had previously mentioned that I was looking for an alternative to the see-saw approach because of the homing problem. With any system, you can't move anything unless you know it is safe to do so and driving Z down on a PNP without knowing where the axes are is unsafe.
 

Yeah I spent the last 5 posts trying to explain that. Anyway, the potentiometer approach seemed like such a good idea I changed 2 of the analog inputs on the PCB I'm working on from vacuum sensors to  connectors for a potentiometer.

I think I like your previous idea of having a switch permanently made when the axes is not at home and someway down. That way, you could use two switches (one for each side) and have the centre point where neither switch is made and hence home. So homing would be simply:

• See which switch is made
• Move that side so that it goes up
• Stop when the switch in no longer made

And as long as the switches are at the same height, finding the true centre point wouldn't be too hard, just halve the distance from when the switch on one side releases to when the switch on the other side makes.

Excuse my lack of knowledge, but I'm not sure what you mean by home_offset, unless you mean you move to the switch and then offset from the switch, but this won't work. In this instance the home switch is in the middle of the axis, if you don't know where you are on that axis, how can you know which direction to move to get to the middle? You can't have the switch at either end of the axis because this risks plunging one of the nozzles into the bed.