Optical Sensor Advice

[[Ke...]]

Factors to consider:
Most of the parts that we inspect range in size from very small to very large, and are *highly* reflective.
Keyence equipment - too small of a volue for our parger parts & most parts need to be coated
ATOS-Q - Can accomodate larger parts, but part reflectivity is still an issue.

Ideally, we want to be able to inspect parts without resorting to using an anti reflectivity coating on the parts.

Parts inspected include:
High Precision Bearings (very reflective & tightly toleranced)
Threads & Splines (internal & external)
Plastic Molded - Black and White plastics - often small and flexible

These are the Zeiss options that are currently inder consideration:
LineScan, ViScan, & DotScan

Some of my initial thoughts:
LineScan: I am likely to see some of the same issues as other systems regarding highly reflective parts?
ViScan: "Spotlight on two dimensions"...better for stampings & such, but may not be the best choice for 3D geometries??
DotScan: Looks like it may be the best option for 3D components with highly reflective surfaces?

I would appreciate any feedback from people with experience any or all of these systems.

[[Ri...]]

LineScan - it's a red-light laser, so reflectivity will be an issue.

ViScan - 2.5d Camera, not suitable for 3d geometry

DotScan - great for 3d geometry, as long as you are able to get within the working distance of the sensor.

[[Ke...]]

Please sign in to view this quote.

Thanks for your feedback.
A couple of follow up questions:

Resolution vs Accuracy - Resolution 60 / 36 / 28 nm: Do you think this resolution statement would be adequate for a part with a max form deviation of 5 microns?

Line Width/Spot Size: 16 µm / 9 µm / 8 µm: Do you think this would be acceptable for a roundness/diameter measurement of a bearing raceway?

[[Ri...]]

Please sign in to view this quote.

I don't know what machine you are talking about, but on a small Prismo:

A DotScan 1mm has an accuracy spec of 1.6µm +L/350
A DotScan 3mm has an accuracy spec of 1.9µm +L/350
A DotScan 10mm has an accuracy spec of 2.9µm +L/350

A LineScan doesn't have a traditional error spec, but it depends on your working distance:
8-32mm working distance is 2.9µm
25-63mm working distance is 12µm
50-94mm working distance is 20µm
100-200mm working distance is 50µm

Regarding the LineScan, I am not a huge proponent of measuring form deviations with a Laser. The biggest challenge with Laser scanning is reflectivity. The more reflective, the more uncertainty. This can be improved with a coating, but that is a band-aid.

I do not personally have much experience with the DotScan, I don't think that reflectivity is much of an issue with this sensor, but I haven't seen results on bearing surfaces. I have seen very impressive flatness measurements using a RT-AB and a DotScan, I would think that roundness would be similar.

Having said all of that, 5µm is not a lot. Most CMMs cannot handle that tight of a form tolerance. Typically I would suggest something like the Rondcom which would be an excellent choice, but there has been great strides in measuring tight form tolerances on CMMs, but it would require something like a high level Prismo or better.

[[Ke...]]

Please sign in to view this quote.

Thanks for the detailed response, it is greatly appreciated.
I have a Accura II (9/16/9) with an RT-AB that I use for most measurements.
The tighter tolerances are usually for smaller bearings which are difficult to fixture & inspect on the MarForm or TallyRond.

[[Ri...]]

Please sign in to view this quote.

It looks like the specs on the DotScan are the same from a Prismo, and that size Accura II.

I'd reach out to your local Zeiss rep, and see about getting a demo/application study on your parts. Zeiss will have a machine (probably not an Accura II) that can perform the demo/study.

Good luck.

[[Cl...]]

Have you looked at QVI (OGP) Zone 3? We just bought one. Camera and tactile.
Imports CAD as well. Not sure if they come large enough for your bigger parts?