On Sunday, June 03, 2012 03:06:54 PM dave did opine: > On Sun, 3 Jun 2012 11:39:30 -0400 > > gene heskett <ghesk...@wdtv.com> wrote: > > On Sunday, June 03, 2012 10:37:11 AM cogoman did opine: > > > I reworked the postscript program so you can specify encoder wheel > > > size, and index pulse width for an encoder, then you can call the > > > subroutine (done twice at the bottom) at different places on the > > > paper. Hopefully the instructions in another message for using GIMP > > > can get you the resolution you need. > > > > > > On 06/02/2012 03:58 PM, gene heskett wrote: > > > > Thanks cogoman. This would be useful if I had room for a 6" OD > > > > wheel, but my limit in the space I have is about 2.7" in diameter. > > > > > > %! Postscript utility for printing multiple encoder wheels > > > > > > /inch {72 mul} def % #points/inch (don't change me) > > > /sizeo 2 inch def % outside radius of encoder wheel > > > /sizei 1.8 inch def % inside radius of encoder wedges > > > /segments 256 def % number of segments (black and > > > white) /ipwidth 5.5 def % width of index pulse bar > > > /moveout 4 def % how far from left side to put encoder > > > wheel /moveup 4 def % how far from the bottom to put the > > > encoder wheel /angle 360 segments div def > > > > > > /wedge > > > > > > { /radius exch def > > > > > > /angle_s exch def > > > /angle_e exch def > > > newpath 0 0 moveto > > > 0 0 radius angle_s angle_e arc > > > closepath > > > > > > } def > > > > > > /encoder > > > { > > > > > > % ADD IN THE ENCODER WHEEL AT THE SPECIFIED DIAMETER > > > gsave > > > > > > 0 1 segments { > > > > > > 360 segments div rotate > > > angle 0 sizeo wedge > > > 2 mod 0 eq {1} {0} ifelse > > > setgray fill > > > > > > } for > > > > > > grestore > > > > > > %THIS ROUTINE FILLS THE CENTER WITH WHITE, AND SOMEHOW > > > > > > PUTS A SOLID CIRCLE > > > > > > % INSIDE THE GROUP OF WEDGES > > > %THIS LINE SPECIFIES WHERE TO PUT THE CENTER > > > > > > 0 0 sizei 0 360 arc > > > gsave > > > > > > 0 setgray stroke > > > > > > grestore > > > > > > 9 setgray fill > > > > > > ipwidth 0 sizei wedge > > > 0 setgray fill > > > > > > %THIS ROUTINE PUTS A TINY DOT IN THE CENTER TO AID IN > > > > > > MOUNTING THE DISC > > > > > > %THIS LINE SPECIFIES WHERE TO PUT THE CENTER > > > > > > 0.0 inch 0.0 inch .01 inch 0 360 arc > > > gsave > > > > > > stroke > > > > > > grestore > > > 0 setgray fill > > > > > > } def > > > > > > gsave > > > > > > 2.5 inch 2.5 inch translate > > > encoder > > > > > > grestore > > > > > > 2.5 inch 7.5 inch translate > > > encoder > > > > > > showpage > > > > With your hints on the import settings, I may have something usable > > if I can get it cut out accurately enough. As for slot widths, this > > 50% duty cycle won't cut it at all, but if occurs to me that I can > > probably adjust not only the width of the index, but the angle too > > such that two of them could be mounted with the image sides to each > > other, and adjust the timing relative to each other to get the slot > > width needed. I haven't re-added the center dot since I re-sized > > that snippet to be the spindle hole yet. That dot could be the > > registration dot for cutting them out of the sheet with an etch bit > > in the mill. > > > > Would it be possible to adjust the 'duty cycle' of these by > > multiplying the slots but sequencing the black and white such that 3 > > out of a sequence of 4 are black? or 4 out of 5, 5 out of 6 etc etc? > > > > My brass one when it was done about 23:00 last night, I put on and > > was able to make it work at least as well with the encoder module in > > mode 0 as I was able to make the 39 cycle wheel work in mode 1. But > > its duty cycle was about 57 on to 43 off, with an actual slot width > > setting of .31 for the duty cycle. These opto units are great, 10 ns > > response times and rail to rail outputs, but their aperture, probably > > being expanded by the polished interior surfaces of these slots, is > > eating my lunch. > > > > I'd try blackening the brass, but blackening is not very effective at > > the low incidence angles involved. Before that, I think I'd cut into > > a trace on the board and dim the leds as I fed them with a current > > that was mid- range of the spec sheet, s/b around 12.5 ma if my math > > is correct. I think at the 25 ma max, these things could see right > > through a sheet of .03125 brass! > > > > I figured this encoder was gonna be a piece of cake when I started > > it, but I think Murphy must have eaten the cake too, he sure drinks > > enough of my beer. :) > > > > As for the various methods of blackening brass, discussed here > > previously, most were butt ugly, giving all the colors of a good skin > > bruise rather than a nice flat black. I believe that was because of > > my inability to get all the cutting oil washed back out of the brass > > with clean acetone or boiling in dish soap. I can't go enough > > narrower to fix it with a 1/32" mill as I can't trim the slot width > > more than another thou, so today I will try dimming the leds. There > > has to be a fix someplace, I just have to find it... > > > > Someone posted a link to a PID tuning procedure that has by now > > expired from this mailbox. Could I ask whomever that was to please > > post it again? > > > > What I had last night, piddling around in the hal config screen but > > didn't save, seemed to be pretty stable, but a speed change was > > immediately to about 80% of the diff, a small pause and then it > > ramped to the final speed over the next 1.5 seconds. That would be a > > sign of which gain is too high? > > > > And even at 2 rps, my attempts to stall it by hand weren't > > successful, so I believe this one will be usable once the leds are > > dimmed a bit. I need to see if there is a 100 ohm trimpot in my junk > > box. > > Any chance of collimating the LED's by putting a slit in front of them? > > Dave > Placing the slit would be an exercise in precision I haven't mastered. One thing at a time so I can see results of an individual adjustment. ATM it looks as if I'll need to make another psb, I blindly cut a trace about 15 thou wide leading the the leds series R's, without groking that it was also the - rail for the rest of the circuit, so I need to fire up eagle & see if I can somehow incorporate a 500 ohm trimpot per led, in the space I have.
If I can find some 2 legged ones with a suitable lead spacing at 500 ohms, I should be able to make that work without remaking the board. Eagle is bing a pita about upgrading, I have several copies of libpng14.so.14 on the system, but the installer can't find them even when added to ld.so.config.d. Grump! Radio shack I guess to see if perchance they have 3 of them. Hopefully I won't have to make another board, these are, shall we say, fragile? > > Cheers, Gene > > ------------------------------------------------------------------------ > ------ Live Security Virtual Conference > Exclusive live event will cover all the ways today's security and > threat landscape has changed and how IT managers can respond. > Discussions will include endpoint security, mobile security and the > latest in malware threats. > http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/ > _______________________________________________ > Emc-users mailing list > Emc-users@lists.sourceforge.net > https://lists.sourceforge.net/lists/listinfo/emc-users Cheers, Gene -- "There are four boxes to be used in defense of liberty: soap, ballot, jury, and ammo. Please use in that order." -Ed Howdershelt (Author) My web page: <http://coyoteden.dyndns-free.com:85/gene> Hello. Just walk along and try NOT to think about your INTESTINES being almost FORTY YARDS LONG!! ------------------------------------------------------------------------------ Live Security Virtual Conference Exclusive live event will cover all the ways today's security and threat landscape has changed and how IT managers can respond. Discussions will include endpoint security, mobile security and the latest in malware threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/ _______________________________________________ Emc-users mailing list Emc-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/emc-users