Who want to make a touch probe?

[10bulls]

that's why i think a 3point setup is the best ( and i made my drawings just for that setup ), the slightest motion wil cause any of the 3 switches to break circuit, and thus giving the control a sign "hey!!.. something's touching!"

if you increase the diameter of the disc that operates the switch, it might even do "side probing" but i'm not sure it will be very accurate..

Hello Arie,
I was re-reading your post and got confused. You picture looks like a rennishaw type probe with 3 pairs of balls, but your last post said '3point setup' and disc. Can you elaborate a little please?

Also Arie, I would be interested to hear how you are getting on with your probe. You too Giacco!

...and icupeter, how is your optical system getting on?

I have a box of surface mount bits and bobs ready for my mk II probe, just need a couple of extra days in the week.

Here is a close up photo of my prototype which shows the disc resting on the pins. ideally I would have that bead of silicon around the shaft and the dayglo perspex to keep dust out and restrict the disc from wandering (the photo shows the shaft needs recentering).
The rough prototype performed well enough, I can't wait to see what a well put together one can do. I want to measure the sideways deflection distance in various direction as well...

[arie kabaalstra]

with a 3-piont setup, i meant that the switches always rest on 3 sets of contacts.. just like my probe, and about any other one.. the "one switch" probe that's also posted in this thread doesn't have that setup, so it will be very hard, if not impossible to set it up right.. and have it fuctioning well..

recently i've designed a probe that's both easy to build, and accurate, it uses a disc-shaped piece of PCB, 6 copper or brass studs, and 3 steel bearing needles( or similar material, like shafts of drills )

http://i12.photobucket.com/albums/a2...e-1/probe3.jpg

[10bulls]

with a 3-piont setup, i meant that the switches always rest on 3 sets of contacts.. just like my probe, and about any other one.. the "one switch" probe that's also posted in this thread doesn't have that setup, so it will be very hard, if not impossible to set it up right.. and have it fuctioning well..

Ah I see, that clears things up thanks. You are obviously quite a skilled machinist and the 6 ball setup is no problem for you. Do you have any scanning results Arie?
Great work...nice drawings!

[miljnor]

would you be willing to give up the cad files for your probe or post them hear?

[arie kabaalstra]

maybe.. but i'll have to "tidy-up" my drawings first..i'm also thinking about selling the complete documentation, milling machine, along with all "accessories" like baby-toeclamps, T-nuts, Touchprobe, basemaster..

[camcompco]

i don't want to sound like a dope here, but what is the funtion of the ball bearings?

thanks much

[Fifty_ohm]

The ball bearings are contacts so there is electrical continuity through the contacts and the post when the probe is in a normal position (not touching anything). As soon as the probe tip touches something the post lifts off the ball bearings and breaks the circuit. Hope that makes sense.

[camcompco]

it's like a lightbulb going off...got it.

Thanks for the help

John

[captainkirk]

Mr. Bean and 10 bulls have put up some great information.

My question is (this is a very old thread) is there anyone who has taken all of the information and created a probe that would be willing to share drawings, files, etc..

Has there been any advancements in the probe desing that are not noted here that would be worth incorporating??

I am looking to build a touch probe and want to build the best one I can without a lot of trial and error.

Thanks,

Kirk

[10bulls]

My question is (this is a very old thread) is there anyone who has taken all of the information and created a probe that would be willing to share drawings, files, etc..

Has there been any advancements in the probe desing that are not noted here that would be worth incorporating??

Well this year my busy-o-meter has gone well and truly off scale, but I am finishing my 'day job' in just over a week and more touch probe development is planned. My rough as guts prototype is still working well so I will mostly be working on making it more mechanically accurate and adjustable.

ERRATA!
There is an error on the circuit diagram I posted on my web page.
R1-6 are actually 1.8kR and the R7 (from the collector of the transistor) is 12k. My apologies for this and I will try and update the web pages soon.

I recently had some fantastic feedback from someone who replicated the design using hand tools and pieces of scrap. He did a fantastic write up he plans to put on his website so I won't steal his thunder. He also alerted me to the incorrect circuit diagram (many thanks John!).

Here is a link to a guy called nophead who has turned the 1 penny touch probe upside down into a tool height adjuster for his fantastic rapid prototyping machine. He also mentions ways to possibly simplify the circuit.

http://hydraraptor.blogspot.com/2007_05_01_archive.html

Good luck captainkirk! May you boldly probe where no probe has probed before!

[lerman]

The ball bearings are contacts so there is electrical continuity through the contacts and the post when the probe is in a normal position (not touching anything). As soon as the probe tip touches something the post lifts off the ball bearings and breaks the circuit. Hope that makes sense.

Yes, the ball bearings complete the circuit, but there is a reason that they are balls. The six spheres plus the three rods form a kinematic mount that serves to precisely and repeatedly position the probe in three dimensions.

Motion in any direction will break the circuit.

Ken

[arie kabaalstra]

i made my probe with needles from a needlebearing, and 6 copper dome-pins, tested it multiple times.

Accuracy is well within 0.002 mm, now for a second parallel port to put the thing in action..

[m1911bldr]

I like the simplicity of the mechanics in 10Bulls' probe. I am thinking of a little mod to enhance the re-centering of the probe after contact is made. Why not put a 1.25 x 45 deg chamfer/taper on the bottom of the moving disc and locate "ball head pins" of some sort to interact with the chamfer/taper? They would re-center the disc upon its return from a contact, alleviate the need to do all that drilling and positioning of multiple balls and keep the electrical simplicity/sensitivity of the original design. A diaphragm of "inner tube" type rubber sheet on the bottom of the outer body, positioned to center the moving disc and apply downward tension on it would replace the spring, keep the disc centered and seal the probe body.

[arie kabaalstra]

the best way to make a probe recenter again is simply providing 3 resting points//

the ballbearing setup is the only way to get it right.. accuracy is far less important as it seems.. a tripod will never "wiggle", just make sure you can align the whole probe, to get the probe-needle centered.

[arie kabaalstra]

http://i12.photobucket.com/albums/a2...1/SCREENIE.jpg

Yipeee!!.. I've got mine working!!

[arie kabaalstra]

[ame]http://www.youtube.com/watch?v=dzRdQG5nEpg

Yippeee.. Got my probe working...

[arie kabaalstra]

Done some more work on the scanning program.. i scanned a model of a Mercedes CLK with a " profile following " routine..

first the probe moves down, until it touches, retracts and probes sideways.
if it doesn't touch within a give distance, it moves down until touching, if it touches something while going sideways, it will retract again..

saves a lot of time

[markus_detroy]

Im a newbie here ...sorry...if this was asked or answer in another post,
but what soft ware are you using... in your below post? to capture the probe
xyz? data

I was wonder how 3d articuating arm work? eg for example...in the old days
the granite shop would send someone out to from the shop to make up a template of your wall for granite counter tops... not any more the have guy with his laptop and a articulating arm / probe and they make a soft ware template..... have anyone seen this or know of a way to make a diy probe like this?

[arie kabaalstra]

Articulated arm probes "know" their position by rotary encoders, every joint has one, and by knowing the angles of the arm sections, and their lengths, the software is able to calculate the x, y, and z coordinates

I capture my coordinates with USBCNC, a PC-based CNC controler, which is basically the only professional control for an affordable price,
www.usbcnc.com

[PEU]

http://i12.photobucket.com/albums/a2...e-1/probe3.jpg

Hi, can you share the CAD models, they look great. Thanks!!


Pablo