Bad scan??

[[Ca...]]

Below is a snip of a scanned ID. I can't explain what's going on with the points gathered. Everything looks normal during the run itself. The probe isn't touching anything but the (clean) ID, but that's not what the points show. I use this probe for several other features, and there are no problems with anything else. There are a series of ID's all at a similar height, and they all look like this...

Thoughts/ideas?

Dirty scales? Faulty probe head?

Scan.JPG

[[Am...]]

You are 100% sure that the shaft isn't touching somewhere? Does this look the same no matter where the part is located?

[[Ca...]]

I've rerun the ID with a couple of different probes. Very similar results. It's not shanking out. The ID pictured is a through hole, there's nothing to hit.

[[Ow...]]

I believe I've read that XXT passive scanning heads will show the vectors incorrectly on the cad screen.
Are you using a passive head?

[[Ia...]]

Have you tried constructing different features/strategies to see if you get the same results?

[[Ca...]]

Passive? I'm not sure what you're asking, or how to answer you. It's a VAST XXT TL3. The head is 2-3 years old. I've never seen scans like this, not even close. I've propped the fixture up about 2 inches, and reran the ID's and got similar readings....

[[Gr...]]

Does this only happen on this plan? Or do similar holes in the same direction show the same issue?

[[Ke...]]

Possibly a broken "collision protection plate" inside the XXT. I have seen normal scans show this, due to a broken plate interfering with the movement of the probe. Simply removing the XXT and shaking it slightly should tell you if there is one "swimming around" in there!

[[Ca...]]

There are 10 diameters, of various sizes and heights, but all oriented left to right.

They all look similar.

I feel like an idiot child having to ask, but how hard do I shake it? I shut it all down, took the XXT off, and shook it a bit harder than I'd shake a bottle of OJ. I didn't heard anything.

[[Ke...]]

Lol. No worries, I doubt anyone has ever told you to do this! I have found that these plates can get stuck in the spring mechanism, so a little shake may not produce a rattle if that is the case. Otherwise, a slight shake should produce a rattle, which is the tell tale sign. If still unsure, you can actually see these plates if you look inside the end of the XXT.

[[Br...]]

Is it really close to touching the bottom of the holes? If it comes into contact with the bottom during the scan it will look like that.

[[Ca...]]

The ID pictured is a through hole, with two scan paths, there is no bottom.

[[sh...]]

We had a severely worn probe tip that made scans like that about a year ago.

[[Go...]]

Please sign in to view this quote.

Move the sensor by placing it near your ear and if you hear something inside the sensor then very surely you will be damaged by an insurance or retainer of the restraint system, this type of sensors already in two occasions have given me that problem of damaging that suspension system by say it in a way

[[Jo...]]

Is it possible that you are out of travel ? For example, if a part is not in the nominal position, the defined path cannot maintain contact with part that is grossly out of position.

[[Ca...]]

The bore is fairly close to it's correct size and location. The probe makes full contact with the feature.

[[Lo...]]

Is this a machined hole or a drawn (formed) extruded ID? How was the hole manufactured? What probe size are you using?

[[Ca...]]

It's a 20mm hole, I'm using a 3mm probe. It's a machined hole, circle interpolated. It's near a mirror finish.

I'm totally stumped.

[[Ch...]]

Do you have a gage ring that you could run and check your equipment with?

[[Cl...]]

I'm with Chad. Try and replicate the measurement scenario with a ring gage.
If it's showing bad that way, you've eliminated the part as a variable.

[[Ca...]]

I check a ring gage, with the same probe, and in the same orientation as the feature, yesterday. The ring looked gorgeous. Another detail is that the "bad" spots are always at the top of the scans. I can rotate the part and/or the probe, and the "bad" spots are always at the top.

Also, today, the machine is fumbling probe changes. Smaller (lighter) probes aren't staying in the probe rack, the magnetic head is pulling them forward, out of the rack when it drives away.

A coworker (unhelpfully) suggested the CMM could be on a ley line...

[[Ch...]]

That fact that the gage ring measured good would indicate to me that it is either the part or the measurement strategy.

The fact that the probe head is acting funny would indicate that it is the probe head or machine.

Something is missing because these facts contradict each other.

[[Ca...]]

It's extremely frustrating Chad.

I'm also trying to work on several other things, so this problem doesn't really have my full attention, which it SHOULD have, if I have a hardware problem. I'll tweak the program, and see how single points, and partial scans look....

[[Fa...]]

Hello
if you have done the verification on a ring and the defect is not present you don't have hardware problems, if re-doing the measurements with the piece (and the probe) rotated on another axis and the defect follows the rotation then the problem is the piece.

[[Se...]]

Has the CMM been crashed? Or depending on your set up, one of the length or segmented measurement systems been compromised? Sometimes the segmented measurement systems are strips that are easily accessible. And little things such as cleaning, etc can even compromise them. Leaving you still able to calibrate and run parts correctly, but those areas compromised could cause trouble in various ways depending on how bad it is.