Gen3 Cupcake with 3G5D Shield and adding a z-min stop

[Dan Laskowski]

I love my two old Cupcakes -- they are well tuned and getting great results, but I thought I would finally add a Z endstop for auto height adjustment.

I have read through lots of posts and checked through all the stuff I have collected over the years about what I have, but I can't find schematics with enough info to decipher what to do to get an endstop to work.

Does anyone know what needs to be done?  I see the two wire CD type connector for the Z axis.  I have soldered the two pads together.  I don't have any chips installed for splitting out the min and max endstops.  I know about telling the firmware I have endstops.  A min endstop is enough for me.  I won't be installing X or Y endstops.

I'm guessing that one side of the switch needs to be grounded and one held to 5 volts through a 10k resistor with a mechanical switch opening and closing to trigger the z stop action?  If so, which of the two wires is ground and which is pulled up?  Maybe I have the wrong picture about what needs done?

Thanks for any insights someone might have...

Dan

[Dan Laskowski]

I found a bit of time to play and after carefully looking at the 3G5D board and making a couple of measurements, it looks like the pin on the left [I'll call it #1] is the sense pin, the middle two are grounded and the right pin is 5 volts.

I used a 10k resister to pull the sense pin up to 5 volts and then shorting the sense pin to ground made the z min stop work!

I am playing a bit more, but it looks like I am on my way...

[Dan Laskowski]

I'm happy to report that Zmin switches are on both of my Cupcakes and working!  There is one little hiccup though.  

The G161 Z F750 command to zero the Z axis is weird.  Sometimes, it "locks" the Z-axis into quivering like it wants to go up and down at the same time.  One one machine it happens about 50% of the time I start a print.  Cancel and start the print again and it will seek zero and run like a champ.  The second machine is more like 100% failing to zero.

I can issue a G0 Z-50 command in place of a G161 and it will find zero, but the G161 command [when it works] is the better way to start printing.

Any thoughts on the G161 behavior? 

[Dave Slater]

Add a temporary jumper across the zmin switch if it is normally closed or

Add a temporary disconnect into the zmin switch if it is normally open

when the zaxis starts quivering operate the temporary to observe if the quiver stops.

--
You received this message because you are subscribed to the Google Groups "Vintage Makerbot" group.
To unsubscribe from this group and stop receiving emails from it, send an email to vintage-makerb...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Dan Laskowski]

I tried to turn the Z-axis when it was quivering and it could be turned and stopped when you got to the switch, but it was fighting you the whole way down...

[Dan Laskowski]

Just for grins I changed the start script to this:

M116 ; wait for everything to heat up

G0 Z5 ; move up five just for grins

G0 Z-5 ; move back down as a test

G161 Z F750 ; Seek z-min quickly

It heats up, moves up five, moves down five and then goes crazy when it hits the G161 command...

If it helps, I am using Simplify3D as my slicer feeding GPX 2.4 writing to an SD card which I hand carry to the printer for printing.

[James McCracken]

My cupcake has chattered like that before, usually by combination of a partially bound z rod and too fast feedrate.

I use the twotimes wobble arrestors from before twotimes yanked his designs, if you have a copy or similar design, I found it really helped with bonding

--

[Dan Laskowski]

Thanks for that thought -- my Z-axis is my own design and works quite differently than most of the other systems out there.  It makes sense that there is a hardware / software interaction of some sort with regard to the G161 command, but I'm struggling to understand what it might be.

The printers have been running with my Z-axis design for hundreds of hours and I have never seen this behavior before.  It only showed up when I added the G161 command to the start script AND as noted above, G0 up followed by G0 down works, but when you get to the G161, it locks up.  That is weird.

I will slow down the feed rate and maybe even disassemble the z-axis a bit so that the rod is really free to turn and see if the problem goes away...

[Dan Laskowski]

Mr. McCracken -- you nailed it.  I slowed the feed rate down to 150 [normal speed] and it worked like a champ!  And this was on the printer that was failing 100% before.

And as always -- something had to be different between the printers.  The one failing 100% still has one of the old MK6 big heavy steppers on it while the second one has a lighter QU-BD extruder meaning the weight is different on the two hence the different failure rates...

[James McCracken]

Good to know!!! Glad to help, Dan. 

On Mon, Nov 5, 2018, 7:27 PM Dan Laskowski <dklas...@gmail.com wrote:

Mr. McCracken -- you nailed it.  I slowed the feed rate down to 150 [normal speed] and it worked like a champ!  And this was on the printer that was failing 100% before.

And as always -- something had to be different between the printers.  The one failing 100% still has one of the old MK6 big heavy steppers on it while the second one has a lighter QU-BD extruder meaning the weight is different on the two hence the different failure rates...

--