Trouble with Z-Zero for PCB making.

More
16 Jul 2021 15:56 - 16 Jul 2021 15:57 #215065 by Jocman
Hi all.
I'm going to mill some PCBs, but I'm getting an annoying behaviour.....
My job steps are:
- Surface mapping with z-touch probe.
- gcode adjusting by GCodeRipper
- PCB bit Zero
- run the adjusted gcode.
Well, it happens that sometimes (1 of 3 times...) the bit doesn't carve the surface. Or even on some points the bit carvs, on other points doesn't....
I tried several time to re-map the surface ad re-adjust the gcode, but no change: same, identical behaviour....
I tried to cheat the system by going to Z-zero and set a new (fake) offset at 0.01mm, but nothing changed (well, thinking about, that's normal....)
I don't understand why it happens....I checked the PCB but it's well fixed on the surface....
The procedure I follow is always the same, and when it works, the PCBs are piece of cake.....
This drives me very crazy.....

Andrea
Last edit: 16 Jul 2021 15:57 by Jocman.

Please Log in or Create an account to join the conversation.

More
16 Jul 2021 21:45 #215097 by tommylight
What drives and how are the motors moving the parts ?

Please Log in or Create an account to join the conversation.

More
17 Jul 2021 08:20 #215123 by Jocman
I'm using hybrid drivers and lead screw for the mechanics.
(Hope this is the information you need....)
But i don't think (or i hope so) is something related to electronic and mechanic.
I tried a "suicide" test: after zeroing Z,i reset the bit so to actually being a little bit lower than real zero: of course i lost the pcb because it carved very deep this time,but it worked "fine"....
I'm pretty sure it could be related to my z probe (is self made),but the fact i re-made the mapping 3 times and for 3 times got the same behaviour make me think the probe is working..
Maybe the alignment drills made the pcb raise up a little bit,but the next remappings should consider that.....

Please Log in or Create an account to join the conversation.

More
24 Jul 2021 11:42 #215764 by andypugh
Maybe X and Y are getting swapped in the heightmap? or maybe the corrections are in the wrong direction?

Please Log in or Create an account to join the conversation.

More
25 Jul 2021 19:06 #215899 by robertspark
could it be that something mechanical is slipping in the z-axis when under vertical load?

some bits don't like a vertical plunge and prefer a ramped plunge

given the distances involved are minute, could it be something in the z axis is slipping under load?

Please Log in or Create an account to join the conversation.

More
26 Jul 2021 10:49 #215950 by Jocman
Honestly I don't know....
Every time I re-zero the machine, detach the PCB from the working plan, re-attach it, redo the surface probing and retry with the PCB milling.
Sometime it works at the first attempt, sometimes I've to do it a couple of times.
Basically I do not change anything in my working routine....
It seems something happening ramdomly....

Please Log in or Create an account to join the conversation.

More
26 Jul 2021 20:09 #215982 by robertspark
I have been trying to think how you can test this, and the best that I can come up with is running a series of stepped cuts in parallel.

​​​​​say 5x stepped cuts over 50mm / 2 inch (10mm / 3/8" steps)
then offset the cut say 3mm / 1/8" and run the stepped series again,
offset again and repeat and then inspect the depths of cut (eyeball)... they should be all the same.... if not... maybe the tool is slipping in the collet or the z axis spindle is slipping in the drive nut / pulley / plumb drive 

​​​​​to load the tool evenly you will need to run the cut full length at 1st depth, retract, move back 40mm, plunge to second cut depth, run for 40mm, retract, move back 30mm, plunge third depth of cut, cut for 30mm, retract, move back 20mm, plunge forth depth of cut, cut for 20mm, retract, move back 10mm, plunge 5th depth of cut, cut for 10mm, retract, move back 50mm, offset 3mm, repeat sequence of cuts.

​​​​​​what tool dia do you run?
what spindle speed?
what plunge rate?
what cutting federate?
what depth of cut and I can try to hand ball write you a test gcode.
(do you work in mm / in?)

Please Log in or Create an account to join the conversation.

More
28 Jul 2021 07:16 #216110 by Jocman
I work in millimeters
For my PCB I use a vbit 0.1mm / 20°
The spindle (Kress) is set about to 12.500rpm
Plunge rate100
Cutting feedrate 350
Deept of cut: 0.08

If I pinpoint that:
- the gcode is the same
- the bit is well tight in the collet so as the spindle on the Z axis
- the probe (so far as I can see...) is working fine
Nevertheless, sometimes I get the problem....
The only "different" thing is: when I detach the PCB from the wooden spoilboard, I have to re-locate the fixing screws (of course I don't put them in the same holes). Can it be related to the fixing system??? Is very basic: small screws with washers along the PCB perimeter....
I'm starting considering this aspect.....
Maybe the screws, getting deep into the wood, will raise inappreciably the wood around the holes; inappreciably, but maybe enough to trick the probe / bit??? So when I "change the holes" this not happens or is fix....
In the next days I'll make such ...fixing gigs with the 3D printer, so to keep the screws little bit far away from the PCB and at the same time let me press the PCB on the spoilboard

Please Log in or Create an account to join the conversation.

More
28 Jul 2021 07:54 - 28 Jul 2021 08:03 #216111 by billykid
I often make printed circuit boards without any compensation. I have a plastic base that I smooth with the cutter after having fixed it the first time then I fix the printed circuits with the double-sided adhesive and cut with 0.2 vbit,
Mauro
Attachments:
Last edit: 28 Jul 2021 08:03 by billykid.

Please Log in or Create an account to join the conversation.

More
29 Jul 2021 06:53 #216192 by Jocman
Very excellent results (far away from mine....)
I tried with the double-sided tape, but it leaves to many sticky debris.....
What brand of tape do you use?

Please Log in or Create an account to join the conversation.

Time to create page: 0.287 seconds
Powered by Kunena Forum