qi-hardware/wernermisc
1
0
Fork 0
mirror of git://projects.qi-hardware.com/wernermisc.git synced 2024-11-22 07:37:12 +02:00
Code
355614bb34
wernermisc/vialtray/doit

25 lines
554 B
Plaintext
Raw Normal View History

vialtray: make a tray for vials containing 0402/0603 components (in progress) This doesn't work yet. In particular, going through "cameo" doesn't seem the most convenient choice, since "cameo" requires paths to be closed, which is something the paths we use here aren't. (Conceptually, thet are closed, but the implementation skips some small segments.) This commit is just a snapshot before I experiment with changing the algorithm.
2010-11-28 01:17:25 +02:00
#!/bin/sh
DIR=/home/moko/svn.openmoko.org/developers/werner/cncmap
RECT=$DIR/rect/rect
ALIGN=$DIR/align/align
ZMAP=$DIR/zmap/zmap
GP2RML=$DIR/gp2rml/gp2rml
CAMEO=../../../cae-tools/cameo/cameo
vialtray: ready for first test run (on-going) - doit (GEN): corrected name of toolpath generator - doit (rdata): fixed typo ("rdate" vs. "rdata") - doit (Z): set Z depth here, to avoid manual depth adjustment - doit: don't use cameo - we compensate for the tool size in tray.pl - doit: increased feed rate from 2 mm/s to 4 mm/s - doit: clearance is above top layer, so 5 mm are more than enough - tray.pl (zf): increase floor depth from very conservative 15 mm to 17 mm - tray.pl (zs): increase depth step from 2 mm to 2.5 mm to accelerate milling - tray.pl (mw): increase wall thickness from 2 mm to 2.5 mm in order to enlarge gap between sides (was smaller than tool diameter) - tray.pl (r1): stop 2 steps before the edge, since we have the outline as another radius-like step - tray.pl (do_col): account for tool radius - tray.pl (do_circles): lift the tool after each disc, not only after each column - tray.pl (do_cols, do_circles): temporarily skip 1st column - tray.pl (main): do circles before the outline, to reduce chatter when cutting the outline - tray.pl (main): since we're no longer using "cameo" here, don't print directives for it
2010-11-28 03:41:45 +02:00
GEN=${1:-./tray.pl}
vialtray: make a tray for vials containing 0402/0603 components (in progress) This doesn't work yet. In particular, going through "cameo" doesn't seem the most convenient choice, since "cameo" requires paths to be closed, which is something the paths we use here aren't. (Conceptually, thet are closed, but the implementation skips some small segments.) This commit is just a snapshot before I experiment with changing the algorithm.
2010-11-28 01:17:25 +02:00
vialtray: ready for first test run (on-going) - doit (GEN): corrected name of toolpath generator - doit (rdata): fixed typo ("rdate" vs. "rdata") - doit (Z): set Z depth here, to avoid manual depth adjustment - doit: don't use cameo - we compensate for the tool size in tray.pl - doit: increased feed rate from 2 mm/s to 4 mm/s - doit: clearance is above top layer, so 5 mm are more than enough - tray.pl (zf): increase floor depth from very conservative 15 mm to 17 mm - tray.pl (zs): increase depth step from 2 mm to 2.5 mm to accelerate milling - tray.pl (mw): increase wall thickness from 2 mm to 2.5 mm in order to enlarge gap between sides (was smaller than tool diameter) - tray.pl (r1): stop 2 steps before the edge, since we have the outline as another radius-like step - tray.pl (do_col): account for tool radius - tray.pl (do_circles): lift the tool after each disc, not only after each column - tray.pl (do_cols, do_circles): temporarily skip 1st column - tray.pl (main): do circles before the outline, to reduce chatter when cutting the outline - tray.pl (main): since we're no longer using "cameo" here, don't print directives for it
2010-11-28 03:41:45 +02:00
rdata="0 0 0 150 0 0 0 100 0"
Z=-27.6
vialtray: make a tray for vials containing 0402/0603 components (in progress) This doesn't work yet. In particular, going through "cameo" doesn't seem the most convenient choice, since "cameo" requires paths to be closed, which is something the paths we use here aren't. (Conceptually, thet are closed, but the implementation skips some small segments.) This commit is just a snapshot before I experiment with changing the algorithm.
2010-11-28 01:17:25 +02:00
rect=`$RECT $rdata | awk '{$3 = ""; print}'`
# cameo: we inline the tool diameter with the data
$GEN |
awk '{ if ($3 != "") $3 += '$Z'; print $0; }' |
$ALIGN 0 1 $rect |
# angle, reference (lower left corner), rect
vialtray: ready for first test run (on-going) - doit (GEN): corrected name of toolpath generator - doit (rdata): fixed typo ("rdate" vs. "rdata") - doit (Z): set Z depth here, to avoid manual depth adjustment - doit: don't use cameo - we compensate for the tool size in tray.pl - doit: increased feed rate from 2 mm/s to 4 mm/s - doit: clearance is above top layer, so 5 mm are more than enough - tray.pl (zf): increase floor depth from very conservative 15 mm to 17 mm - tray.pl (zs): increase depth step from 2 mm to 2.5 mm to accelerate milling - tray.pl (mw): increase wall thickness from 2 mm to 2.5 mm in order to enlarge gap between sides (was smaller than tool diameter) - tray.pl (r1): stop 2 steps before the edge, since we have the outline as another radius-like step - tray.pl (do_col): account for tool radius - tray.pl (do_circles): lift the tool after each disc, not only after each column - tray.pl (do_cols, do_circles): temporarily skip 1st column - tray.pl (main): do circles before the outline, to reduce chatter when cutting the outline - tray.pl (main): since we're no longer using "cameo" here, don't print directives for it
2010-11-28 03:41:45 +02:00
tee check |
$GP2RML 5 4 4
vialtray: make a tray for vials containing 0402/0603 components (in progress) This doesn't work yet. In particular, going through "cameo" doesn't seem the most convenient choice, since "cameo" requires paths to be closed, which is something the paths we use here aren't. (Conceptually, thet are closed, but the implementation skips some small segments.) This commit is just a snapshot before I experiment with changing the algorithm.
2010-11-28 01:17:25 +02:00
# clearance, xy speed, z speed
Copy Permalink