I was wondering how to meet the AS9102 FAI for a printed circuit board that has hundreds of holes. Circuit boards typically contain many groups of holes. Each group is of the same size. Do we have to measure each hole and document it on the form or can we take a sample? If we must measure each hole, can we document the range of the measurement?

Do we have to measure each hole and document it on the form? Yes, Unfortunately you will have to measure each hole.

If we must measure each hole, can we document the range of the measurement? Yes, I typically just document the range.

One of the problems with printed circuits is plating consistency across the board. This can cause holes to be plated more in some areas, less in others (I'm assuming you're talking of plated thru holes).

Some, by design may just be 'via' holes, others are component holes. One way to quickly assess size is to have a pin jig made, rather than individually measure all holes - since the only practical way is with a tapered hole gauge which isn't the most accurate. Once you've got the gauge you could drop a board on it periodically just to check process control.

We design and manufacture to customer spec. We might build 6 to 12 pieces and may not build any more. It doesn't make much sense to do it under these conditions. Can we take exception in the case of a small lot?

We might build 6 to 12 pieces and may not build any more. I know its fustrating, I do FAI's on pieces we only make one or two of.:mad: It doesn't make much sense to do it under these conditions. Can we take exception in the case of a small lot? You can take exception only if your customer agrees. Unfortunately you will still have to show compliance to 8.2.4.2 of AS9100. If your an AS registered organization.

I know its fustrating, I do FAI's on pieces we only make one or two of.:mad: You can take exception only if your customer agrees. Unfortunately you will still have to show compliance to 8.2.4.2 of AS9100.We have a process in place and can do it, but at what cost to the customer? There is the law of diminishing return.

We have a process in place and can do it, but at what cost to the customer? There is the law of diminishing return.Do you not include the cost of performing the FAI, prior to accepting the contract?

Do you not include the cost of performing the FAI, prior to accepting the contract?Yes, but unfortunately someone failed to include the cost in the original quote.

Yes, but unfortunately someone failed to include the cost in the original quote.

That is unfortunate indeed. To the best of my knowledge though, AS9102 requires that every feature be verified without exception. I wouldn't buy off a FAIR from a supplier otherwise, and I'm sure I wouldn't hear the end of it if I did.

FWIW: A neighboring company in the industrial park where my business was located came to me with a similar query about a thin metal plate which had 625 holes of varying sizes.

It so happened I had visited another neighbor who had just installed a closed circuit TV with magnifying, measuring lenses, which compared and measured very small products with a three decimal accuracy. I walked the one neighbor over to the other to see the system. They struck a bargain where the TV guy performed the first article inspection with video-tape record of each of the 625 holes with exact[?] measurement for 3 first pieces (one from each of three production machines.) The whole process was completed in less than one day from their first meeting.

The system was capable of being programmed to perform a rapid test of either EVERY product (100%), or of the samples in SPC charting. It was very cool.

Perhaps you have an organization near you with similar equipment? (I'd go to manufacturers and distributors of such equipment to find "satisfied customers" where you can go "see it in action.")

Yes, but unfortunately someone failed to include the cost in the original quote. Have you tried asking your customer for an amendment to the contract, as you forgot to include the price of the FAI.
That is unfortunate indeed. To the best of my knowledge though, AS9102 requires that every feature be verified without exception. Unfortunately, domingue is correct.
FWIW: A neighboring company in the industrial park where my business was located came to me with a similar query about a thin metal plate which had 625 holes of varying sizes...

...Perhaps you have an organization near you with similar equipment? (I'd go to manufacturers and distributors of such equipment to find "satisfied customers" where you can go "see it in action.") I have actually done this. Might be something to consider.

FWIW: A neighboring company in the industrial park where my business was located came to me with a similar query about a thin metal plate which had 625 holes of varying sizes.

It so happened I had visited another neighbor who had just installed a closed circuit TV with magnifying, measuring lenses, which compared and measured very small products with a three decimal accuracy. I walked the one neighbor over to the other to see the system. They struck a bargain where the TV guy performed the first article inspection with video-tape record of each of the 625 holes with exact[?] measurement for 3 first pieces (one from each of three production machines.) The whole process was completed in less than one day from their first meeting.

The system was capable of being programmed to perform a rapid test of either EVERY product (100%), or of the samples in SPC charting. It was very cool.

Perhaps you have an organization near you with similar equipment? (I'd go to manufacturers and distributors of such equipment to find "satisfied customers" where you can go "see it in action.")We have something that we can use to take the measurements. It can be programmed, but in this case we'll do it manually.

Late reply, but I hope that my comments add some value.

I've worked in the PCB, PWB industry for some years, and there are a couple of things to consider.

One is that a large number of holes on any 'plated through' circuit board are there for only one reason, and that is to conduct power/signals from one layer to another.

As a result of this, the physical size of these "through holes" (or vias) is not critical - ie. they are not in any way used to fit, affix or attach.

Nevertheless, the thickness of the plating is critical - for example the holes can fail under thermal shock if the amount of the deposited metal is sub-standard - as such, the plating thickness on these should be verified - microsectioning is usually the chosen method, either from a test coupon manufactured at the same time, or (more prefereable in my mind) from sacrificial samples.

There are devices which can check the plating thickness in a non destructive manner, but unfortunately these devices are limited by the size of the probe, and as such, cannot usually verify the small (< 0.6mm via holes).

In terms of other holes, ones which take components, those that are interference fit (push or press fit connectors for example) should be verified, but others where the connection wolll be by soldering, I'd suggest that statistical sampling of the holes would be adequate.

Note that due to the characteristics of electrolytic plating, certain areas may be subjected to higher or lower current flow, and as such, the sampling of the holes should take place at a minimum towards all four corners and in the center - Any decent plating technician should be able to suggest the danger areas (high aspect holes in high or low current density areas as an example).

Given that modern multi layer circuit boards can often contain more than 10,000 holes, including buried and blind vias which cannot be physically verified without destructive testing, my suggestion would be to apply the known tools and techniques to verify to the satisfaction of the cutomer and registrar without overkill.