Re: [CAD_CAM_EDM_DRO] Coordinate Measuring Machine

The Renishaw-type 3-arms-resting-on-6-balls circuit is VERY sensitive
and detects any touching in X, Y, or Z. It is not simply a "z-touch".
Really the probe sensitivity is limited by the flex in the system, that
the probe itself will flex some minute amount before the touch force
overcomes the spring retention and lifts an arm off one of the ball
contacts. It's not much. In many cases the deflection of the work
material itself is more.

The accuracy of a leadscrew itself is only like 0.010"/ft for "standard"
and 0.003"/ft for "precision".

But the tip geometry creates more substantial limitations. The ball-tip
Renishaw probes seem to go down to 1mm, and cannot consistently map
features smaller than this, and inaccurately report cloud points in a
predictable way. That is, a perfectly square edge where we go from
horizontal to vertical will scan as a radiused corner in the raw data
because of the probe tip's diameter and spherical shape.

One could write a program to undo such things but AFAIK you'd need to
start from assumptions of what the shape is, because artifacts from the
probe's geometry on 3D surfaces the probe will register a touch
prematurely because of touching the side of the ball, but you don't know
for sure what direction that touch came from. Seems like this has a
risk of introducing more artifacts than it removes.

Danny

On 12/4/2011 12:31 AM, "hanermo" - CNC 6-axis Designs wrote:
> Unlike some of the other replies, in my mind a CMM involves "real"
> resolution similar to, but not necessary equal, to those found on
> commercial machines.
>
> As such, at the low end, you would need to think in terms of microns for
> resolution.
> It is normally seen as a given that your mechanical resolution, be it
> movement or measurement, must be 10 x whatever you are trying to do.
> At a very low minimum, a cmm should have a 0.01 mm resolution and
> repeatability, both well within any amateurs abilities, and within
> victorian engineering and manual tools.
>
> Almost all probe designs I have seen are not, in my mind, adquate.
> Basically, because their premise is that creating a "touch" is
> sufficient without any regard to the plane, direction, etc of the contact.
> Thus they make a point cloud in a fairly repeatable, but by no means
> accurate manner.
>
> The basic problem is lack of adjustment of the make or break point in
> all plains, x,y,z.
> Likewise, they are typically not assembled on a flat surface (lapped
> plane, optical flat, surface plate) and no effort is made to measure
> bend of the probe or repetability.
> The most important problem, in my mind, seems to be that they make no
> effort to measure errors.
> If an erro is, say, x:0.002, y,0,004, z:0.003 all +/- 2 microns, this
> would then make an excellent base for a high quality solution.
> All above are easily solved problems.
>
> Screw-type mechanisms, can easily resolve and repeat to 1 micron, think
> of microscopes, optical stages, astronomy, sine bars etc.
> The keys will be repeatability,adjustment, etc.
>
> Some ideas can come from differential screws, and or special fine pitch
> screws.
>
> Making a mechnical machanism that is repeatable to few microns is not hard.
> I proved this with my 12x cnc lathe.
> Basic repeatability is about 2-3 microns, to an optical home switch,
> with lathe mechanical resolution of 1 micron.
> This was achieved by use of 0.1 micron resolution mechanical motion
> train (30.000 counts/turn servos).
>
> But this needs a build based on quality as a premise (ie the optical
> switch triggers an electronic gate for repeatbility) rather than a
> simple switch-makes-contact.
> A 2-sensor optical switch assy cost about 40$, with real-world micron
> resolution, +/- 2-3. Plenty good enough, imo.
> As one manufacturer of machine tooling stated "we cannot build stuff to
> 1 micron, but can adjust stuff to it".
>
> I would make a probe similar to those suggested, but would try to
> measure it for error.
> Also, I would make it adjustable, as far as possible, for near-zero error.
> A fine (0.25 mm) screw offsetting a 1 m rod of drill rod can easily
> adjust a point to better than micron levels, with 40:1 mechanical advantage.
>
> Then measure a mic std, eg 1 inch, using say 0.1 micron resolution.
> Adjust until your results are as close as possible to 25.400 mm.
> Use screws to fix mounts, with a 0.1 mm oversized mount hole. Adjust
> x,y,z separately.
>
> Maybe 50 hours to make a good one, and results probably about 50x better
> than the suggested probe examples.
> Error stack is x * y * z, ro 3 microns = 27 microns error.
> Typical suggested probes are likely about 10 * 15 * 10 = 1500 microns.
>
> Accuracy, resolution and repeatability are not the same !
>> I m making a CMM trainer as my final year engineering project... i m
>> facing problems.. reinshaw probes are very expensive and not
>> economically feasible for a trainer..
>>
>> any idea?
>>
>
>
> [Non-text portions of this message have been removed]
>
>
>
> ------------------------------------
>
> Addresses:
> FAQ: http://www.ktmarketing.com/faq.html
> FILES: http://groups.yahoo.com/group/CAD_CAM_EDM_DRO/files/
> Post Messages: CAD_CAM_EDM_DRO@yahoogroups.com
>
> Subscribe: CAD_CAM_EDM_DRO-subscribe@yahoogroups.com
> Unsubscribe: CAD_CAM_EDM_DRO-unsubscribe@yahoogroups.com
> List owner: CAD_CAM_EDM_DRO-owner@yahoogroups.com, wanliker@..., timg@...
> Moderator: pentam@... indigo_red@... davemucha@... [Moderators]
> URL to this group: http://groups.yahoo.com/group/CAD_CAM_EDM_DRO
>
> OFF Topic POSTS: General Machining
> If you wish to post on unlimited OT subjects goto: aol://5863:126/rec.crafts.metalworking or go thru Google.com to reach it if you have trouble.
> http://www.metalworking.com/news_servers.html
>
> http://groups.yahoo.com/group/jobshophomeshop I consider this to be a sister site to the CCED group, as many of the same members are there, for OT subjects, that are not allowed on the CCED list.
>
> NOTICE: ALL POSTINGS TO THIS GROUP BECOME PUBLIC DOMAIN BY POSTING THEM. DON'T POST IF YOU CAN NOT ACCEPT THIS.....NO EXCEPTIONS........
> bill
> List Mom
> List Owner
>
> Yahoo! Groups Links
>
>
>
>