Printed Circuit Board Assembly & PCB Design SMT Electronics Assembly Manufacturing Forum

Printed Circuit Board Assembly & PCB Design Forum

SMT electronics assembly manufacturing forum.


CMMs (Coordinate Measuring Machines)

Daniel Woon

#19793

CMMs (Coordinate Measuring Machines) | 7 May, 2002

Hi,

We are looking for some measuring machines to evaluate the ability of a SMT placement machine to produce quality.

We will mount some dummy components on a glass plate and then insert the assembly into the measuring system which can measure the positional placement accuracy. The machine will display the the Cp, Cpk and other related information.

I also understand that some machines can produce detailed information as to where problems occured, allowing better indication of the root causes.

Can anyone share with me the manufacturers of this measuring machine?

Many thanks in advance,

Daniel

reply »

Sir Alexandr

#19801

CMMs (Coordinate Measuring Machines) | 7 May, 2002

Sorry if info is useless, but placement machines must have own system programs for completly ajustment of placement in a gap about 0.01 mm . Also headblock camera has enought increment to found correct place and fiducial setups utility. Maybe your lots has corrupt herberfile ?

This message was posted Add this forum to your site! Click to learn more. the Electronics Forum @

reply »


JAX

#19804

CMMs (Coordinate Measuring Machines) | 7 May, 2002

What?

reply »


JAX

#19805

CMMs (Coordinate Measuring Machines) | 7 May, 2002

Daniel, If you just plan on mounting chip components don't worry about getting an accurate measurement. The parts will vary in size, corrupting your accuracy measurements. If you are going to place leaded components you can probably get away with using an AOI machhine, unless it is a visual comparator. If you want to use a true measuring system why not use glass slugs instead of dummy components? Just a thought.. Anyway, you should be able to find multiple manufacturers just by doing a quick search. Off the top of my head you might want to look at Olympus America, but I'm sure you will get plenty others to look at.

reply »

#19813

CMMs (Coordinate Measuring Machines) | 7 May, 2002

http://www.cyberoptics.com/flashIndex.htm

See model KS200 Post Placement Measuring System

reply »

#19816

CMMs (Coordinate Measuring Machines) | 7 May, 2002

I believe Mitutoyo makes a CMM for that application. They were soliciting the idea at APEX show a couple of years ago. I believe they were going to develop some software to do just what you are asking for (following the IPC-9850 format). A CMM, however, is an expensive solution. Not to mention the cost of training personnel to use it. It is not the type of machine you often find in an electronic manufacturing facility. They sometimes require special rooms to be built to maintain temperature stabilty for accuracy reasons. Also, in order for the data you collect to be valid, gage R & R studies need to be performed on the CMM.

Another idea...why not just do glass part placements and let the pick and place machine evaluate itself. There are several manufacturers of this type of kit and they work quite well. With a kit from Accuspec you verify the placements with a microscope (something you probably already have). With the kit from Applied Images (Rochester)you can make the determination with the camera of the placement machine or a microscope. These kits will give you the X, Y, and Theta offsets.

Another possibility is a company called Cybertron (Germany). They make a machine that evaluates placement machines using their glass parts kit. It is basically a gantry machine with a camera on it. Using a glass plate with fiducials on it, parts are placed and the part fiducials are compared to the plate fiducials to determine accuracy.

reply »

#19817

CMMs (Coordinate Measuring Machines) | 7 May, 2002

Most machine supplier have CMM's and proove their accuracy with attached protocol of CPK. We used to have a machine from View Engineering. I can not locate their website, I only found a used machine. http://www.equiptechonline.com/PDetail.cfm?pid=1798&catID=73&subID=38&sid=6 However, here is a site of other vendors: http://www.mmsonline.com/dp/mtg/byzone/Detail.cfm?ID=3640%20&zone=I If you don't use the machine every day, you may be better off using the mentioned glass plate and glass slugs from Accuspec and others.

reply »

Daniel Woon

#19820

CMMs (Coordinate Measuring Machines) | 8 May, 2002

Hi guys,

Thanks for the valueable inputs.

Daniel

reply »

Jones

#19833

CMMs (Coordinate Measuring Machines) | 8 May, 2002

We currently use the Mitutoyo vision measuring machine Model #qv202-pr05f. We use it as a tool for measuring placement accuracy on both our chipshooters and glue machines. It also has the ability to check the glue dots for accurate sizing. As far as telling you what is wrong with your machine...not exactly. It does provide you with X,Y and theta readings which you can use to isolate a placement problem.

Jones

reply »

ianchan

#19836

CMMs (Coordinate Measuring Machines) | 9 May, 2002

Hi mate,

think "GSI Lumonics" a Canada based mafg of laser orientated product-machines may be able to help you.

we saw their SVS8200/8300(machine model) theory based demo. looks good enough to warrant the time and resources to evaluate their machine with a practical test run. that's what we are in discussion now.

check out : http://www.gsilumonics.com

reply »

andyy

#19896

CMMs (Coordinate Measuring Machines) | 13 May, 2002

I agree with gregp on this one. I'm not going to discuss AOI as my company manufactures AOI systems, but I will comment on the CMM question as I use/have used these to qualify calibration targets and to characterize our systems. We actually have a Mitutoyo system.

Firstly, the two most common CMMs are the Mitutoyo and the View Engineering system. Both are fairly expensive and both are tricky to program (which you will need to do if you wish to use them effectively).

I have had numerous conversations with equipment suppliers (placement machine manufacturers) and customers about the capabilities of CMMs when measuring real components. Some maintain that the CMMs are sufficiently accurate and others have proved to themselves that they aren't.

When we purchased our Mitutoyo, they made no claims about accuracy when dealing with real components on real boards (which is unsurprising - warp etc will inevitably degenerate measurement capabilities unless adequately accounted for). That said, whether the measurements are accurate or not, they are repeatable (1 standard deviation on an 0201 is about 1 micron) - but beware, this depends upon the skill of the user and the method by which centroid offsets are calculated.

The only sure fire way to ensure that you get accurate numbers with a CMM is to place glass slugs on glass plates. This is what many of the placement machine manufacturers do repeatedly before shipping new equipment.

Good luck with whatever you decide to do.

Andy

reply »

Boundary Scan

SMT Machines