manded by their users. Specifications such as NRC's 12GP8 and the Sigma Specification contain requirements which measure barrier qualities, weathering resistance, and fogging properties because they are recognized problems by the people who are in the business of making and using insulated units.

The laboratory tests which are most meaningful to us in sealant development are quite simple in principle. We use test applications of the sealants involved on the substrates the fabricator is using. Exposure of these test applications to long term high temperature conditions, ultraviolet energy, both dry and in the presence of moisture, is a very empirical and reliable test. What adheres to one aluminum alloy will not necessarily adhere to another. A sealant may have excellent resistance to ultraviolet energy but as soon as moisture is introduced along with the ultraviolet energy, the sealant can fail adhesively.

Resistance to pressure buildups in the interior of an insulating unit can be simulated with a standardized test chamber where pressures can be controlled and exerted on the sealant which is unsupported other than by its own adhesion and cohesive forces.

The barrier qualities can most reliably be tested by use of the ASTM Test Method E-96. Procedure E of this method is most severe. Results from this test must be tempered with the sample size involved and the conditioning of the sample prior to the test. A sealant tested at a 20 mil thickness may have an MVT rating of 2 grams (H2O)/24 hr/m2 but when tested at different thickness or aged in a weatherometer prior to testing, the MVT value will vary. One study made with a two-part polysulfide, which was tested at various thicknesses before and after weatherometer exposure, showed the barrier properties degrading from 3 to 20 at 20 mils thickness.

I would like to go into a little more detail on the individual laboratory test that we use in evaluating our sealants before we think they are ready to go into a unit and then be tested in the unit itself. Adhesion and cohesion is rated by the normal method using overlap shears made up of adherends of the aluminum or stainless steel or glass involved. A two part polysulfide should typically give an overlap shear value of over 100 pounds per square inch. A butyl-polyisobutylene tape will give a very low value, and it should fail cohesively. Actually, it does not contribute anything to the structural rigidity and the value that it gives is not really important. Barrier properties, as mentioned, are measured by the ASTM E-96 Method and a good polysulfide underneath the conditions stated in Procedure E (0 to 90 percent relative humidity) at a 20 mil thickness un-aged, should give values between two and five grams of moisture through a square meter area in 24 hours. Using the same set of

conditions, a butyl-polyisobutylene tape will give a value of .1 to 2 gram.

Resistance to ultraviolet is best tested on glass and, in this case, polysulfides should give at least a 60 day resistance to ultraviolet in a dry condition and 30 days in a wet condition. The manner of testing is to simply put a casting of the sealant down onto the glass, expose it 12 inches away from the sun lamps, periodically peeling off the polysulfide. A cohesive failure should result. Fogging resistance can be tested in a chamber which contains the polysulfide or other test material. The air temperature is kept at 160 °F. A glass plate on top of the vessel is cooled at 50 °F. by circulating water. Volatiles that are inside the compound which could come out after it is in the unit, will be condensed on the glass surface.

Two-package sealants and pre-extruded tapes both have advantages and a portion of the qualities necessary. It is necessary to use both types of materials in a single unit and obtain the optimum performance for each property or quality desired. If the butyl-polyisobutylene sealants are not used, it is not possible to obtain the optimum combination of service performance properties of low permeability, non-fogging, and long term resistance to degradation.

Because of the variations in field use and conditions, and the elusiveness of reliable data concerning field failures, it is necessary to rely primarily on laboratory tests which measure the qualities required for good, long-term, perform

Mr. McKinley: Could we now have questions from the floor for our four speakers? Mr. Beavers?

Mr. Beavers: I have a question for Mr. Amstock. If a manufacturer uses an accelerated cure of a polysulfide sealing system, is the performance of the seal impaired in any way?

Mr. McKinley: Thank you. May we have a question for Mr. Gwyn?

Participant: Can Mr. Gwyn give us any information on the effect of using reflectivelycoated glass in sealed double-glazed units?

Mr. McKinley: Thank you. A question for Mr. Mazzoni ?

Participant: Has Mr. Mazzoni's company conducted tests recently on polysulfide sealed doubleglazed units?

Mr. McKinley: Thank you. One for Mr. Box? Participant: Could Mr. Box expand his previous remarks to include discussion of other types of sealants than those made by his company for insulating-glass units?

Mr. McKinley: Thank you. Sir?
Participant: (Inaudible)

Mr. McKinley: I belive the question is, how can we cope with glazing problems, and it is directed to all four speakers. Thank you. Let's go and get a cup of coffee and be back here at 4:00 o'clock. On your way, if you would like to pick up an extension to your registration list Mr. Cook has these at the registration desk.

Mr. Amstock: I will attempt to answer Bill Beavers' question first. His question was: if a manufacturer accelerates the cure of the polysulfide system, does he get a degradation of the seal? On the contrary, most good polysulfide seals manufactured will gain properties such as adhesion, resistance to UV and moisture and combinations of them because of this accelerated cure by heat. Again, most manufacturers have the capability of adjusting application time or cure rate by means of different retarders in their system and this by no means will detract from the ultimate properties of the cured sealant system, or the performance of the sealed unit. And if I may answer my glazier friend back there, who claims that he has quite a bit of breakage or failures because of glazing techniques.

Participant: Almost everybody in the industry

has.

Mr. Amstock: Well, I can't agree with you fully on that one, I'm sure there are a lot of problems in the field because of this condition. I don't know if you were there, but yesterday the SIGMA organization had a lengthy meeting on glazing techniques for installing lights into the sash or building as well as a discussion on the various types of compounds that are used. I would say that with the type of potential speci

fication that was discussed, which is a very difficult one to develop, and the experience that we've had from the two major glass suppliers, I think this will be resolved. Again many glaziers, and I've seen it happen in the field, just slap the unit in as though they could hardly care less about clearances or setting blocks. This is the responsibility of the glazier himself or the foreman or supervisor on his job, and of no one else.

I would think that breakages due to poor glazing practices should not be charged to the manufacturer of the glass, or to the component manufacturer, because I personally couldn't accept that type of logic in such cases.

Mr. McKinley: I think you've uncovered a very practical side of our problem and I believe when we get to the wind-up this is one of the items that ought to be on our list. Thank you for the question, and thank you, Joe.

Dan Gwyn: The question that was asked me has to do with a reflective copper film and really doesn't pertain to this meeting. I talked to the young man back there. . . . I think he is satisfied with the answer. It has nothing to do with the edge seal of our unit.

R. W. McKinley: Very good. Do you care to comment on the glazing question?

J. D. Gwyn: I certainly am not the best one to discuss glazing but I have the strong impression that my company, as well as PPG, has, and offers, detailed information on glazing. I do know that we have a department which looks into all large buildings that are under consideration, as far as sash is concerned, to see the glazing matters appear satisfactory.

Mr. McKinley: Perhaps it is time to move along. Mr. Mazzoni.

Mr. Mazzoni: The question put to me was, "Have we done any recent testing on polysulfide double glazed units?" I would like to point out that I purposely did not identify any type of unit that we have in there, at least I tried not to. The main purpose of the paper was to try to correlate the accelerated test data with the field data and, once having a correlation, to use this as a basis for projecting, from the results of the accelerated test of any new material that may be used, an estimate of how well the material will do in the field, because you can't very well have field data come to you all the time. In other words, if you have made a variation you can't practicably put this into the field and wait five years or ten years. So the prime purpose of the paper was to show that there is some kind of a correlation that can be made with a PPG type accelerated test and the service type test.

Mr. McKinley: Thank you, Ray. Mr. Box.

J. A. Box: The general question to me was, "Why didn't I include discussion of other types of sealants in my remarks? I hesitate to say

specifically why in many cases because we don't manufacture these other types of sealants for insulating units so I would be knocking someone else's line of products. The reason we don't use them is because we think that what we're using for insulating glass is the best.

There are many sealants that are excellent for various specific applications, but they are not suitable for sealed double-glazed units because of the combination of properties required for this application. Among essential characteristics needed are satisfactory rate of cure, very low vapor permeability, adhesion, strength, low creep, and resistance to uv. For one reason or another, many otherwise good sealants are therefore screened out as regards use for insulated glass units.

As far as the glazing question is concerned, I'd like to say that the way the unit is glazed is also a source of a lot of problems. Mr. McKinley: Yes, Mr. Orbeson.

Mr. Orbeson: I would like to comment in part on this glazing problem. Concerning the information that we have published, let's say that in a general way, we will work with any manufacturer of glazing materials or sash to develop his methods to fit our units, but why should we publish a book of lengthy practices on some 203 installation methods to help our competitors.

Mr. McKinley: I think our panel deserves a real hand. . . I want to thank each of you and I'm sure your questions will not end as you step back into the audience.

Mr. McKinley: Very early in the game, one of the reasons for focusing our attention on the National Bureau of Standards was the possibility that the next paper might be available. The subject is "An Exploratory Study of Laboratory Testing Methods and Standards for FactorySealed Double-Glazed Window Units." Our speaker is Henry E. Robinson, Chief, Environmental Engineering Section, National Bureau of Standards.