# $URL: svn+ssh://mrwhat@ssh.boim.com/home/mrwhat/svn/Walker/trunk/JansenDefs.pl $ # $Id: JansenDefs.pl 427 2014-08-23 20:42:39Z mrwhat $ # # Definitions for a Jansen linkage, with parameter settings to include # in a perl script to generate laser-cut drawings ######################################################################## # (A) This one had longer, rounder steps, but not quite as wide(stable) of a stance. # This yielded the initial 3D printed prototype robot. # Althogh a locked linkage was found about .6mm from this one, this did # not exhibit much stiff-near-locked issues when initially built # circa June 2103 #$AC = 17.5; #$Ay=6.1; $Bx=48; #$BH=43.2; $EF=46.35; #$CD=59; $CH=73.2; #$BD=48.6; $BE=46.7; $DE=65.4; #$FG=77; $FH=41.3; $GH=56.2; # configuration (B) circa August 2013. Although my optimizer code # ranked this as farther from "lock" than (A), it exhibited locking # behavior when printed $AC = 17.5; #$Ay = 5.53; $Bx = 51.17; #$BD = 48.41; $BE = 45.76; $BH = 42.87; #$CD = 59.19; $CH = 74.37; $DE = 65.34; #$EF = 46.13; $FG = 77.29; #$FH = 41.77; $GH = 54.44; $config = 'E'; # configuration (D) circa 130909. Added factor to favor configurations # where the CH linkage stays away from the B axle. # smoother motion. Seems likely to be more robust to small manufacturing errors # Opt17 run, plot 237: # 1.38 -51.33 48 46.23 44.2 59.79 73.6 67.47 47.47 77.51 41.63 54.86 # had a little bit of a bump when cranks were vertical, and feet were a mm or two # high on the step cycle (robot dropped down a mm or 2), point of low PE that needed # to be lifted up as cranks went past vertical. if ($config eq 'D') { print STDERR "setting Jansen configuration D\n"; $Ay = 1.38; $Bx = 51.33; $BD = 48 ; $BE = 46.23; $BH = 44.2 ; $CD = 59.79; $CH = 73.6 ; $DE = 67.47; $EF = 47.47; $FG = 77.51; $FH = 41.63; $GH = 54.86; } else { #if ($config == 'E') { # configuration (E) circa 140227. Tried to flatten stride, in the hopes # that the little dip when cranks are vertical will be less pronounced than # in cofiguration (D). #3.318 -53.833 48.020 46.977 40.359 59.635 71.441 66.437 47.467 75.792 40.843 55.790 print STDERR "setting Jansen configuration E\n"; $Ay = 3.318; $Bx = 53.833; $BD = 48.020; $BE = 46.977; $BH = 40.359; $CD = 59.635; $CH = 71.441; $DE = 66.437; $EF = 47.467; $FG = 75.792; $FH = 40.843; $GH = 55.790; } $cutWidth=$preview?.2:.01; # suggested .01 for real (ponoko) laser cut, but we can't see that in pixel display # hardware size definitions $mmi = 25.4; # millimeters/inch (I think this is EXACT. Actually 25.40000) $diamQ = $mmi/4; # 1/4" $diam316 = $mmi*3/16; #$Bdiam = 4.76; # diameter of B axle 3/16" .. ideal 4.7625 $Bdiam = 6.35; # move to 1/4" OD axle, same as standoff? $Adiam = 3.18; # diam of A (crank) axle, 1/8" ... ideal 3.175 #$diamS = 4.744; # 3/16" OD nylon spacer, measured $diamS = $diamQ; # 1/4" OD Al standoff, measured $diam4 = 2.76; #4-40 screw, OD of thread $diam2 = 2.2; #2-56 screw, OD of thread # hole radii, with fuzz for laser width #$fuzz = 0.1; # subtract fuzz from holes in wood. In plastic, fuzz seems much smaller due to melt filling # Quelab's laser seems to have a slightly larger kerf. $fuzz = 0.15; $Brad = $Bdiam/2 - $fuzz; # snug, critical fit $Arad = $Adiam/2 - $fuzz/1.15; # tight, for press-fit $rad2 = $diam2/2 - $fuzz; # sufficient for screw pass-through (NOT for tapping) $radS = $diamS/2 + $fuzz*0.35; # loose, for free-moving hinges # try to standardize, all 1/4" standoff, axles, no. 4 screws $rad4 = $diam4/2 - $fuzz*0.01; # sufficient for screw pass-through (NOT for tapping) on acrylic $radQ = $diamS/2 + $fuzz*0.5; # loose, for free-moving hinges $dHole = 10; # mm between holes on braces #$Frad = 7; # braced knee joint radius $Frad = 9;#10; # 7 is actually thinner than default for 1/4" standoffs $linkWidth = $dHole; #$deg2rad = 0.01745329252; $endPart = "'>\n"; # endpath and location(translation) group to end each part ############ Set standard parameters for node hole radii (ir ... inner radius) # and node radii (or ... outer radius) $irS = $diam4 /2 - 0*$fuzz; # free passthrough, #4-40 screws $irA = $diam316/2 + 0.2*$fuzz; # for free main-bar bearing surface $irB = $irA; # free rotating for BED, BH -- moved to 3/16 steel main bar $irC = $irA; # for C-side of crankLinks, CD, CH #$irD = $diamQ /2 + 0.2*$fuzz; # for D side of CD $irD = $irC; # using 3/16" OD standoffs, for D side of CD $irE = $irD; # for BED E hole, free rotating over 1/4" standoff, for now $irF = $irD; $irH = $irC; # using 3/16" OD 1/2" long standoff $orA = 6; # for crank-arms $orB = 7.5; # B of BED and BH $orC = 7; # C-side of CD, CH $orD = 6; # D-forks on BED $orE = 7; $orF = 9; $orG = 6; # foot (Ground) $orH = 7.5; # for around 3/16 standoff hinge # 7.5 seems good for thick foot, forks could be smaller ################################################## Final re-scale for drawings $DesiredCrankLen = 20; # I think I'll try this size for an eventual 384mm sq full robot print #scale things up a bit for a "standard" 20mm crank ($AC,$Ay,$Bx,$BD,$BE,$BH,$CD,$CH,$DE,$EF,$FG,$FH,$GH) = &vScale($DesiredCrankLen/$AC, $AC,$Ay,$Bx,$BD,$BE,$BH,$CD,$CH,$DE,$EF,$FG,$FH,$GH); # derived Jansen linkage dimensions $DEleft = ($DE*$DE + $BD*$BD - $BE*$BE)/(2*$DE); $FGleft = ($FG*$FG + $FH*$FH - $GH*$GH)/(2*$FG); $DEperp = sqrt($BD*$BD - $DEleft*$DEleft); $FGperp = sqrt($FH*$FH - $FGleft*$FGleft); ### This distance varies with each configuration, and various hard to analyse # factors like linkage widths, motor size... # It is too hard for me to compute, so I'm just setting it by hand. $dA_Motor = 40;# 40 OK if spacers/braces BELOW motor 45; # distance from A axis to drive motor axis # At this seperation, here is a gear pair for a slow motor (feet period FASTER than motor axle) : $gearPitch = 10.01;#8.54; $teethDrive = 8; # OD=16*2 ##13; # OD = 2*20.377 $teethMotor = 17;#20; $drivePlateOR = 18.5;# <-- smaller for 2mm plates20; #25; # gear radius, 20.377 $rShaft = 1.5 - 0.10*$fuzz; # motor drive shaft dimensions $dShaft = 1 - 0.05*$fuzz; # Assuming circular head on gearhead motor $motorRad = 7.75; # no fuzz needed on Acrylic $motorType = "16GA1627"; # isreali motor slightly bigger/longer #$motorType = "mitsumi"; # fasttech 16GA motor $gearHeadRad = ($motorType eq "16GA1627") ? 8 : $motorRad; 1;