On Tuesday 24 April 2007, Dave Engvall wrote:
>Hi Gene,
>
>IIRC emc2 has a very nice probing routine that will do what you want
>and log the results.
>The problem, of course, is a decent probe. I would recommend trying
>to make a Renishaw type probe ... ie. 3 points in a triangle
>where the probe can lift off (disconnect) any single node to trigger
>contact. This will give you 3D which can be nice.
As an electronics tech, the 3 points throws me a little. Is the contact disk
segmented somehow so one can connect them in series or ?? I do rather like
the idea that it could also be used as an edge finder in addition to a
contour follower though. That could be used to re-zero the table after
re-clamping a second such part, but these things tend toward onesies you
know. :-)
>I'm thinking about making one from PVC but have not
>gotten around to it. (round toits are hard to chase down ;-) ).
I had one once, in the early 60's, made out of some white wood, about the size
of a silver dollar. NDI whatever became of it though & haven't seen any for
sale since. Now there's another project, make some more of them for sale at
the local redneck flea markets, but out of something more durable. I can see
the sale card now.
Are you fresh out of round toits?
Get your new ones here,
only $4.95 in brass,
or $3.95 in aluminum.
:)
>Make it with a screw in probe so you can make your own or buy one
>of the well dimensioned (and not cheap) sapphire ones.
My mill wouldn't be able to work at that sort of accuracy anyway, I was
checking repeatability today and .001" is asking a lot of that little beast.
Too much backlash, it needs error mapped in all 3 axis's & the head tilts
about a thou just reversing the z axis travel. Veddy poor design. For that
reason alone, all carving would have to start at the top centerline of the
work, plunging down and outward to the edge, then retract, recenter, and
plunge in the other direction, otherwise there will be an artifact at the
centerline due to imperfect backlash comp in the z axis alone.
><http://wiki.linuxcnc.org/cgi-bin/emcinfo.pl?Probing_With_EMC>
>
>I don't think there is any how-to to go with the above link.
>
>I think that all that is required is to connect to a pin that HAL
>exports.
>I'm certain there will be someone out there with more specific
>knowledge.
>Good luck and tell me how it goes as it will be awhile before I get
>to it.
>
>On the other side of things synergy has a routine that will define
>text and wrap it around a solid.
I've seen that in video, dunno if we could make gcode out of that or not, but
its a heck of an idea, possibly saving the effort to re-invent the wheel if
it can.
>This is way beyond my level of
>competence
>so I don't even know which level of synergy (2.5 D, 3D wire,
>parasolids) is needed to make it work. Just thought I'd mention it
>so you can drool.
Damn, I lose more keyboards that way. :-)
But not tonight, I got tripped and fell into a 4x6 trailer with expanded metal
flooring this afternoon, bent my glasses and scratched a lens, and about 6 sq
inches of gauze, tape & triple antibiotic on a kneecap that was previously
road rashed pretty good about 18 years ago when I threw away a Suzuki
GS-1000G while missing a deer. I missed the deer, but not the road. How it
feels isn't for discussion in mixed company right now.
I'd bought a tiller 2 weeks ago, and now the missus wants to put a garden in
in the back yard of her mothers old place, so I had to get a trailer to move
the tiller on as its 400+ pounds and now sports a wide track axle setup so it
won't tip over near as easily as the one you can see at Tractor Supply for
$750 will. It was while trying to get around in a sea of small trailers when
the tongue of the one I was standing in the back of went airborn. My "Well
DUH!" moment for the day.
>HTH
>
>Dave
Thanks Dave.
>On Apr 24, 2007, at 5:59 PM, Gene Heskett wrote:
>> Greetings;
>>
>> Has anyone drawn up a probe of the sort one could mount in the
>> spindle collet,
>> hook up to emc so that when contact has been achieved, emc records the
>> location?
>>
>> And, assuming this has been done, has the scanner routine to do
>> this already
>> been written too?
>>
>> I'm thinking in terms of the engraving done on a bolt rifles floor
>> plate, or
>> trigger guard, so that if an interesting pattern could be scanned with
>> potrace, and then the z axis treatment added so as to carve, say
>> with a small
>> diamond burr moving at call me next week speeds, to a consistent
>> depth of
>> say .005", the pattern obtained with potrace, but on a curved
>> surface such as
>> the bottom of a trigger guard or floor plate.
>>
>> But I'd need to be able to probe the contour of the item, and do it
>> without
>> doing any preliminary damage to the blueing, which is not a very good
>> electrical contact surface. So I'm thinking in terms of a guided
>> needle
>> whose upper end can be set against a single button of a conductive
>> elastomer
>> keyboard switch, the effect of which would be rather like a
>> potentiometer as
>> the pressure increases, and we could trip and stop the z axis
>> descent when a
>> logic 0 was detected, record all 3 axis's positions, then reverse
>> the z
>> drive, move to the next position .005 away in the scan, and repeat.
>>
>> It seems like we ought to be able to do that if the contact
>> pressure sensor is
>> already a working item.
>>
>> Anybody care to share insights/drawings for such a beast/tool?
--
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)
He who laughs last is probably your boss.
-------------------------------------------------------------------------
This SF.net email is sponsored by DB2 Express
Download DB2 Express C - the FREE version of DB2 express and take
control of your XML. No limits. Just data. Click to get it now.
http://sourceforge.net/powerbar/db2/
_______________________________________________
Emc-users mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/emc-users
