how to measure surface profile

[jmccann]

I have a microprocessor and a surface profile call out pointing to the bottom surface of one of the pins. There are 48 total pins.

I understand the theory of what surface profile is (somewhat), but I'm unsure on how to measure it.

We do have a microscope scanner that can produce 3d height map renderings of the leads.

The way I see it, it can be one of two things.

I can create a plane on each individual pin, and then create one much larger plane out of all those planes. Then, I can measure the z difference between the large plane and each individual small plane.

Or, I can measure the surface roughness (min and max height) of each individual pin bottom.

 

[djhurayt]

kind of depends on the drawing callout, can you post the drawing or that portion of it?

 

[jmccann]

Attached the image with the surface profile call out

 

[djhurayt]

My interpretation is that this profile is being used to tolerance coplanar surfaces (your 48 solder contact pads). This being what appears to be a surface mount electronic component, my guess is that the desired condition is that when this thing is placed on a circuit board they want < or = 0.10 between the board and the contact sitting farthest off the board.

So, no I don’t think you are looking for surface roughness (which is a different symbol) of each of the 48 pads.

And that you have the correct philosophy of first measuring the z location of each pad. Then create a plane out of the three tallest pad. These would be the three that if placed on the circuit board would be the three point definition of a plane. Then all the other points must be within 0.10 of this plane. This could be done on some of the digital 3D microscope or even a CMM by laying the part on what would be the top surface when mounted and then touching off on each pad . . .

 

[powerhound]

jmccann,

This drawing is severely lacking. djhurayt is attempting to interpret the intent but it's all just a best GUESS as he already acknowledges.

I have a feeling this is why you haven't gotten a response from any of the heavy hitters on this forum. There's just no way to really know since the drawing is so bad.

Keep in mind that whatever you wind up doing, if your customer is satisfied, it in no way means that the drawing is sufficient. It just means you GUESSED correctly.

Good luck.

John Acosta, GDTP Senior Level
Manufacturing Engineering Tech
SSG, U.S. Army
Taji, Iraq OIF II

 

[flash3780]

I don't see any datums referenced on the profile callout, so I suppose that it can float around radially. I assume that the features are curved?

Anyhow, there are a couple of ways to measure profile on curved surfaces. You could take impressions of the features and inspect them on a comparator, or you could use a CMM. It really can be a challenge to inspect curved surfaces manually regardless of the GD&T used on the print.

 

[3DDave]

flash - the legs on an IC package are typically flat and are intended to lay flt on the circuit board. This is to limit the amount that the legs might be lifted above the board and with it limit the solder to fill the gap.

 

[djhurayt]

that was my interpretation 3DD

 

[jmccann]

This part is not doable on the CMM. The features are far too small for the probe, and the pins will permanently bend under the pressure of the probe.

 

[djhurayt]

Ya, just went out an felt the pressure it takes to get a reading on our CMM. Never realized it was as much as it is. Agree would certaily flex the pins if not permanently deform them.

Is the resolution of your 3D scopes Z-direction accurate enough?

other verification options
* a granite/flat table and go-nogo feeler gauges, 0.10 goes, 0.10+ does not
* realy trick would be a series of indicators (if it is even possible to get them that close) or maybe some lazer depth gauges that you zero out with something very flat, then set the part on it, no gauge should read greater than 0.10, how much money do you have

 

[jmccann]

Our 3D scopes should be accurate enough. The measurement will not be recurrent enough to warrant the set up of a gauge system.