another project....

geniusintraining · May 16, 2008

I need to measure the thickness of a copper clad laminate, currently I use a linear probe (Mitutoyo) on a Festo cylinder, they work OK and I have made several, the cylinder strokes and the probe measures the distance of the stroke, its built on a granite slab with out anything it measures zero on the surface of the slab, so place the laminate under the cylinder and stroke the cylinder, then the distance off the granite is the thickness of the material, +/- .0005 inch, measuring from .009 to .09

I would like to come up with something different, non contact and faster, my cycle time is just under 3 sec, stroke down, pause (let the probe settle) decide go/no-go, stroke back up. I can not fire the cylinder fast or hard or it will mark the surface of the laminate, its about 8-6psi on a 50mm bore, the stroke is 200mm but only using about 25mm of travel (using positive stops)

So my thoughts were an analog proximity switch anyone ever use them for positioning or anything similar? Do you think it would work? I can mount it just above the surface of the table, my concerns are the resolution of the prox switch, can it decipher .0005”? consistently?

Mark

Steve Bailey · May 16, 2008

How large is each pice of laminate? Could you have a wheel riding on the granite surface and slide the laminate under the wheel? It eliminates the time to stroke the probe forward and back. If the piece of laminate is large enough or moves slowly enough there will be settling time for the signal from the probe to stabilize.

I haven't used any non-contact analog proximity sensors myself. There are both inductive and ultrasonic models available. Can you be sure the laminate is lying flat on the granite without applying some force to hold it down?

CharlesM · May 16, 2008

Laser Mic?

http://www.keyence.com/products/vision/micrometer/micrometer.php

http://www.keyence.com/products/vision/laser/laser.php

geniusintraining · May 16, 2008

Steve Bailey said:
How large is each pice of laminate? Could you have a wheel riding on the granite surface and slide the laminate under the wheel? It eliminates the time to stroke the probe forward and back. If the piece of laminate is large enough or moves slowly enough there will be settling time for the signal from the probe to stabilize.

I have tried this but I could not find a wheel that would not mark the laminate or when I did it had to much slop in it and could not give me a good reading

Steve Bailey said:
Can you be sure the laminate is lying flat on the granite without applying some force to hold it down?

This is the reason I am using the cylinder now (rev 2387324)

back when I started doing this we had a lot of 'bow and twist' we are getting better now, I was thinking about using vacuum this would get me the non-contact

CharlesM said:
Laser Mic?

I have tried lasers before but due to the reflectivity of the copper it was terrible at best, unless I calibrated them every measurement, but I will look into it again...

Thanks guys

leitmotif · May 17, 2008

As you mentioned copper is reflective.
I wonder WHAT IF
you took advantage of the reflection

You coulld ensure the material is flat against the granite ie "suck it down" with vacuum pump. Granted the material would have to be absolutely flat and may require a high polish.
Then aim a laser at a known angle and short distance ie X
Read the height (?) of the reflected beam at a longer distance ie multiples of X such that proxes spaced at say 1/2 inch intervals would represent a change in thickness of 1/4 of your tolerance value.

Dan Bentler

geniusintraining · May 17, 2008

leitmotif said:
Then aim a laser at a known angle and short distance ie X
Read the height (?) of the reflected beam at a longer distance ie multiples of X such that proxes spaced at say 1/2 inch intervals would represent a change in thickness of 1/4 of your tolerance value.

Hi Dan,

I thought about it but I am not getting it....

Two lasers?, one as a reference the other for the actual measurement? subtracting the difference...

L D[AR2P#0.0] · May 18, 2008

How do the copper clad items you are measuring get onto the granite slab ?

geniusintraining · May 18, 2008

By the operator, before they load the laminate into the machine, they will slide it under the mic/laser/prox...

Its a smooth motion they pick it off one stack and place it on the measuring bed before loading it, I may do it with a pick and place if I can get all the other pieces working

We have several different cutting methods this is just one.

radsett · May 18, 2008

geniusintraining said:
Hi Dan,

I thought about it but I am not getting it....

Two lasers?, one as a reference the other for the actual measurement? subtracting the difference...

It's a cute idea. Think back to your basic physics classes on reflection. The incoming and outgoing angles to the surface will be the same. The angle to a fixed laser will depend on the height of the mirror it is reflecting off of. If the laser is x above the holding surface and there will be an easily calulated angle from the laser to the surface. As the surface rises the nagle decreases. The sensitivity to height/angle will depend on how far away you detect the deflection provided you have a perfectly flat mirror. Flatness might end up being your limiting factor.

The detector could be a linear array for measuring or a simple point/line target for pass/fail.

If you have a visible laser you could do a simple test using a piece of paper for the target.

Robert

TimWilborne · May 18, 2008

I'm trying to figure out if you could take an edge guide like you probably use on your rolls of copper center the rolls through the machines. If you could turn one of them sideways you could measure the thickness. I'm thinking of the capacitance type that pretty much takes the top and bottom reading and determines the centerline of the material. Pretty much the same thing as the laser idea.

leitmotif · May 18, 2008

radsett said:
It's a cute idea. Think back to your basic physics classes on reflection. The incoming and outgoing angles to the surface will be the same. The angle to a fixed laser will depend on the height of the mirror it is reflecting off of. If the laser is x above the holding surface and there will be an easily calulated angle from the laser to the surface. As the surface rises the nagle decreases. The sensitivity to height/angle will depend on how far away you detect the deflection provided you have a perfectly flat mirror. Flatness might end up being your limiting factor.

The detector could be a linear array for measuring or a simple point/line target for pass/fail.

If you have a visible laser you could do a simple test using a piece of paper for the target. Robert

GIT
Take a laser pointer and or narrow beam flashlight and clamp at downward pointing angle onto a table. Place mirror flat on table. Observe where light ends up on wall. Now place something 1/2" or so thick between mirror and table. Note different position on wall.

IF you go with this idea two lasers would be ideal. One would be standard and would be affected by heating and expansion wobble etc of granite block. The other would also be affected so I think the idea of subratcting the two would work and eliminated errors.

Dan Bentler

geniusintraining · May 18, 2008

leitmotif said:
GIT
Take a laser pointer and or narrow beam flashlight and clamp at downward pointing angle onto a table. Place mirror flat on table. Observe where light ends up on wall. Now place something 1/2" or so thick between mirror and table. Note different position on wall.

IF you go with this idea two lasers would be ideal. One would be standard and would be affected by heating and expansion wobble etc of granite block. The other would also be affected so I think the idea of subratcting the two would work and eliminated errors.

Dan Bentler

What a cool idea!....

So how can I get the feed back?, I understand the concept but how can I decide the value ("read the height"), if the angle changes how can I move the receiver? or do I have too?

I have a few lasers around so I will give it a try next week... this week I am booked up (installing a new press, CompactLogix

)

Thanks to all for the ideas

CharlesM · May 18, 2008

I have tried lasers before but due to the reflectivity of the copper it was terrible at best, unless I calibrated them every measurement, but I will look into it again...

A laser sensor is one thing but a laser mic is different. We used them to read diameter of carbide circuit board drills coming off a centerless grinder. I don't think copper wire would be any more polished than that. I don't know how much the keyence ones are but the "Lasermic" brand was around 10K. If you need 0.0005 then smoke and mirrors are not going to work.

Maybe look at this example
http://www.laser-view.com/wire_laser_micrometers.htm
http://www.freedomlaser.com/Product%20OV.pdf
http://www.boyuttechnic.com/index/page-app2-en.htm

geniusintraining · May 19, 2008

Hi Charles,

The system we had was a 'laser mic', it was similar to the ones you link to, but we had 3 heads (send and receive) that trended the thickness as the laminate was feed into the machine, it had its own conveyer section.... it would work good under the ideal conditions, very seldom were the conditions ideal.

Thank you for the links.

I like the angle idea, just trying to think how to capture the change....

TConnolly · May 19, 2008

C-Scan.
http://www.sonix.com/products/c-scan.php3

Can be integrated into production systems.
An ultrasonic emitter and receiver are mounted on a scanning head above the materail. Every material boundray will produce an ultrasound reflection, you configure it to pick up the reflection from the top surface of the copper and the back surface of the copper. Time of flight is calibrated to thickness. The optical properties of the material won't affect the measurement, and it is 100% non-contact. The C-scan is very accurate and very fast. Not only can it tell you the thickness of one layer of material, it can also simutaneously measure the thickness of a laminated assemby and the thickness of each layer in the cladded assembly. Additionally, it will find internal anomolies and delaminations.

another project....

geniusintraining

Lifetime Supporting Member + Moderator

Steve Bailey

Lifetime Supporting Member + Moderator

CharlesM

Member

geniusintraining

Lifetime Supporting Member + Moderator

leitmotif

Member

geniusintraining

Lifetime Supporting Member + Moderator

L D[AR2P#0.0]

Lifetime Supporting Member

geniusintraining

Lifetime Supporting Member + Moderator

radsett

Member

TimWilborne

Lifetime Supporting Member

leitmotif

Member

geniusintraining

Lifetime Supporting Member + Moderator

CharlesM

Member

geniusintraining

Lifetime Supporting Member + Moderator

TConnolly

Lifetime Supporting Member

Similar Topics