Shank Probe

[Scott Brown]

Good Morning

I am using a shank probe in a program and have adjusted the nominal depth of each hit by 1.5mm so that I can ensure the part will hit on the shank. My question is, what value do I use for the shank offset during calibration(1.5mm or 3.0mm)??? The shank probe is 1.5mm in diameter. Also looking at the calibration results (vectors) how do you determine what is acceptable.

Thanks
Scott

[Matthew D. Hoedeman]

SHANK BAD, BALL GOOD

Oh, you should search for shank probe issues on this board, the old board and the other board. IMO, you should NOT be doing what you are doing with that probe. Use a ball probe and use a sample hit to set the depth. What you are doing will give you BAD data if the probe is not absolutely, 100% perpendicular to the surface, period.

[rebeldude]

I agree with Matt, but if I understand you right, you are taking a sample hit on top of the part , then coming down 1.5 mm and taking a perpendicular hit on the part. As far as the shank offset, that will depend on how long the shank is of course. I have a 1mm by 35.5mm shank and check it 5mm down the shank.

[Scott Brown]

Shank Probe

Hi Matt

I agree with you that shank probes are trouble, but I don't really have an option here. I have to check the profile of a wire that is welded to an assembly. The wire has a profile of 3.0mm back to ABC. I have made an effort to have the probe angle as square as possible to the wire and the results match up with what we are seeing on the fixture. I'm just not sure about the calibration process. The help file says the offset value is the distance from the tip but that would make the offset 3mm which seems incorrect to me.

Scott

[Bill **********]

If you're asking about SHANK QUAL depth in the probe qual routine, set it at 1.5 so that you are qualifying the shank at the same depth that you will be taking hits on your part.

Bill

[Paul Sarrach]

What type of part are you inspecting? Why the need for a shank? What type of edge is it? I think there is a place for cylinder probes. Thats why they make them. Make sure your Probe head is square, thats where most of the errors come from. But if at all possibly stick with a ball


Typed this while you where answering. Do as BIll stated. it will work fine.

[Matthew D. Hoedeman]

I have yet to see ANYTHING in Pcdmis for you to indicate that you are using the SHANK area of a probe for the measurement you are taking. I am not saying it isn't there, I never looked, but if you can't tell Pcdmis that you are using the shank, it will not know that you are using it and will automatically use the value for the center of the ball. The shank calibration is for cases when the 'height' does not come into play, it is only giving a diameter of the shank at the height you used for the calibration. Now, you say you are checking a wire welded on an assembly, I have to assume that this 'wire' isn't like one of the strands you have in speaker wire, but more along the lines of the wire in a wire welder. If that is the case, what I would do is use the 0,0 angle and find and use strictly 2-d points/vector to measure on the profile of the wire, in this case, no Z vector. Keep it as close as you can to the ball of the probe. The more off-angle the shank is, the worse your reading will be. In a 'perfect' world, you can use either the fixture or a feature on the part to set the XY alignment to, using that same depth of touch to ensure that the relationship between the 'datum' and the wire touches stay correct. What I mean is, if you are hitting the probe 3mm deep on the wire, then you should be hitting the 4-way locator hole 3mm deep. I would only do this for the wire check points, for all the rest, I would still use the 'ball' of the probe. As for the depth, I think that when they say the distance from the 'tip', that they are talking about from the end of the cylinder, which would be the center of the ball at the end of the shank. However, you should be able to watch it calibrate and 'see' where and how deep it is taking the touches, at least close enough to see if it means from the center of the tip-ball or the end of the ball itself.

[Paul Sarrach]

When I use a shank, Its for edges ( plastics pointy or rounded edges) and I will chage the vectors of cad so that I am hitting them in the x and or y machine axis. With great results. I think Matt is the best in the world when it comes to PCDMIS but I disagree with him on this. We always have. I think the error of the shank will be less than the error of trying to get a good hit on a radius to radius hits. And more repeatable. But that is Just my opion. And I have been dead wrong before.

[Scott Brown]

Shank Probe

Thanks for the quick responses everyone. I just recieved an email from tech support confirming the use of a 1.5mm shank qual offset. Looks like the help file is refering to the center of the tip and not the end. The program is set up simular to Matt's suggestion (just 2D vector points). The rest of the part is checked with a seperate program and a 3mm ball stylus.

Thanks again
Scott

[Matthew D. Hoedeman]

Yep, we just have to agree to disagree. Fortunately, I don't have to deal with plastics, strictly metal and sheet metal so I don't have to pull my hiar out wondering if the numbers are really as good as they say for those nasty plastic edges.

[Paul Sarrach]

Thats why I like hard gages for plastic trim. I will certify it, then check my cmm results to the gage. Seems backwards, but its the only way I feel confident in my numbers. Maybe I can get some parts outsourced to you Matt, so you can feel the fun, first hand.

[Bill **********]

An additional note about using shank probes. Pay attention to what Matt said about being perpindicular to the edge. If your shank is skewed at all your data wil be total garbage.

Bill

[Matthew D. Hoedeman]

Thanks Paul! Thanks too much! I have enough fun as it is, I don't need any more. in fact, I am really going to have fun tomorrow, going out to the range and shot up some rounds. Got my laser bore sighter this week, gonna get ALL my guns zeroed in. If you send me any of those plastic parts, I'll be ready for 'em!

[tking]

Thanks Paul! Thanks too much! I have enough fun as it is, I don't need any more. in fact, I am really going to have fun tomorrow, going out to the range and shot up some rounds. Got my laser bore sighter this week, gonna get ALL my guns zeroed in. If you send me any of those plastic parts, I'll be ready for 'em!

If they're shaped right they might fit in the trap machine, you could blast 'em like a clay pidgeon, Yahoo!!

TK

[Matthew D. Hoedeman]

Oh yeah, trap is fun (I have 3 pitchers myself). Just to freak other shooters out, I'll pull out the 20ga, put a scope on it and shoot trap. I hit about 60% with it too!