By Mike Fitzpatrick and Ken Goldstein, Ph.D.
Hank Hogan recently wrote a news article for CleanRooms exploring the relatively slow pace of the changeover from Federal Standard (FS) 209E to the newer ISO 14644-1. This month, we will follow up on Hogan's theme and coolly suggest that there's no reason to panic if you've yet to take the leap.
While this topic appears to have more lives than the proverbial cat, hopefully, this column will help everyone involved in the cleanrooms industry move on to more pressing issues.
Think back to just two years ago and the publicity surrounding the release of the new ISO standards. Remember, we were told that 209E would be forced out of existence just as soon as the new standards were released; and when they were, there would be a wholesale stampede to ISO 14644-1.
Why hasn't this happened? Well, simply put, the old standard and its replacement are not that different. In fact, FS 209E and ISO 14644-1 are remarkably similar.
Aside from some terminology and notation, there are few differences:
Terminology: FS 209E defined cleanliness classes 1 through 100,000, with each being a multiple of 10 times the previous class. ISO defines the classes as ranging from ISO Class 1 through ISO Class 9, and because the scale is logarithmic, each class remains a 10¥ multiple of the previous one.
Three new cleanliness categories: Two of the new categories, ISO Classes 1 and 2, are cleaner than anything found in FS 209E; while the other, ISO Class 9, is “dirtier” than anything in FS 209E.
Number of sampling locations: The new ISO standard permits using considerably fewer sampling locations.
Treatment of outliers: What do we do with an “obviously incorrect” particle count reading? FS 209E does not address this issue. ISO 14644-1 provides some guidance.
Dissecting the details
These are the only significant differences. The rest are relatively minor. Now, let's look at the messy details:
- Terminology: While the FS 209E cleanliness classes increased logarithmically, our description of them was arithmetic—1, 10, 100 and so on. The new ISO standard moves directly to a fully logarithmic scale. While engineers, scientists and technicians should be comfortable with this, management may find this slightly challenging in the beginning.
- The three new cleanliness categories: The two “cleanest” ISO Classes, 1 and 2, will be useful for people in microelectronics where cleanrooms have been operating off-scale on the clean end for years. Until recently, people in this line of work knew they were cleaner than FS 209E class 1, but they lacked the common terminology to describe just what they had, as well as the procedures to prove it to others. The ISO Class 9 classification will be useful to those who are just beginning to establish contamination control procedures in such industries as food and beverage preparation and packaging, paint spray applications and similar areas where airborne cleanliness is increasingly important but just getting started.
- The number of sampling locations: To determine the number of sampling locations using the new ISO standard, simply take the square root of the total area of the clean zone (in square meters) and then round up to the next integer. In the older FS 209E, things were a bit more complicated. For non-unidirectional flow rooms (classes 1,000 through 100,000), we divided the area (in square feet) by the square root of the cleanliness class and then rounded up to obtain the number of sampling locations. For the unidirectional flow rooms (classes 1, 10 and 100), our minimum number of sampling locations was the area (in square feet) divided by 25 and then rounded up. For small rooms—10 square meters and smaller—there is not much difference between the number of sampling locations required by the two standards. But for larger rooms, the new ISO standard permits significantly fewer samples. As the size of a room increases, this difference between the numbers of sampling locations increases. There have been and will continue to be long involved discussions as to which is the preferred approach. To arrive at a temporary truce, let us note that the ISO requirement is only the minimum number of sampling locations, and users are free to use a higher number.
- The treatment of outliers: Statisticians might describe an “outlier” as a data point that belongs to a different population. In other words, it is a data point that just doesn't belong. Consider the following: You are taking particle counts in an ISO Class 6 (FS 209E class 1,000) room. Everything seems normal until we obtain a reading of 127,650 particles in one of your samples. What do you do? The first thing, of course, is to take another sample and look carefully at the results. If you get a similar (high) number, in all likelihood, there is a real problem causing the high counts.But if the second count is back in the “normal” range, you can breathe a small sigh of relief. But what do you do with the “erroneous” data point? FS 209E was silent on this question. ISO 14644-1 refers to this piece of data as an “outlier” and permits you to discard it under certain carefully defined conditions—and that is certainly a step in the right direction. Any statistician with a lick of sense will quickly agree to throw the silly thing out before it contaminates the data. But there is a catch. The new ISO standard only permits users to discard one outlier. In a large room with many sampling locations, there is a significant probability that users will have to discard more than one.
Getting on the same page
Since the issues listed above are the only real differences between FS 209E and ISO 14644-1, it's easy to understand the reticence of some in the U.S. to jump on the ISO bandwagon. Yes, the new is an improvement over the old, but not to the extent that those still using the old are hurting themselves in any significant way.
Yes, we should make the transition from FS 209E to ISO 14644-1 and help put the world on the same page. But the timing will depend strongly on commercial and local factors. And if you and your company have not yet made the change, don't sweat it. It really isn't that big of a deal.
Mike Fitzpatrick |
Michael A. FitzpatricK has participated in the design and construction of semiconductor facilities for over 24 years and is a Senior Member of the Institute of Environmental Sciences and Technology (IEST). Mike can be reached at: [email protected].
Ken Goldstein |
Ken Goldstein is principal of Cleanroom Consultants Inc. in Phoenix, Ariz., and is a member of the CleanRooms Editorial Advisory Board. He can be reached at: [email protected]