Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP Path: utzoo!mnetor!uunet!husc6!bbn!uwmcsd1!marque!ddsw1!dino From: dino@ddsw1.UUCP (Laura Watson) Newsgroups: sci.bio,sci.physics Subject: Re: Enzyme action Message-ID: <255@ddsw1.UUCP> Date: Tue, 13-Oct-87 08:43:27 EDT Article-I.D.: ddsw1.255 Posted: Tue Oct 13 08:43:27 1987 Date-Received: Thu, 15-Oct-87 06:32:04 EDT References: <240@ddsw1.UUCP> <2084@kitty.UUCP> <249@ddsw1.UUCP> <2100@kitty.UUCP> Reply-To: dino@ddsw1.UUCP (Laura Watson) Organization: Traveller's Aid, Mundelein, IL Lines: 133 Keywords: More info... Xref: mnetor sci.bio:736 sci.physics:2490 In article <2100@kitty.UUCP>, larry@kitty.UUCP (Larry Lippman) writes: > I now understand your problem, but don't understand why you can >find no solution [no pun intended]. Because I want to invent a better solution than the best of the existing ones available; they are just not quite good enough. As Hofstadter would put it, I want to "jump out of the loop." (This is sci.physics, isn't it?) > I have seen SMD production lines, and most of them use a vapor >cleaning machine to remove solder flux and solder mask (if the removeable >variety). Vapor cleaning machines generally use a chlorinated aliphatic >hydrocarbon solvent, similar to perchlorethylene. The best of this gamut I know of is Dow Chemical's Prelete (tm) which contains 1,1,1-trichlorethane and aliphatic alcohols. There are unsolvable (insoluble?) problems with it, however. It works, but not very well. It takes a cleaning cycle of >30 minutes for our application, and still leaves a lot of flux and residues behind (for reasons I will try to explain below). The second best of this type of cleaning material is the Alpha Metals 1003 (tm) azeotrope containing tetrafluorodichloroethane and n-propyl alcohol. But it has problems with acid formation and tends to corrode the cleaning equipment if you don't watch it like a hawk and replenish its acid inhibitors frequently. >Such a heated vapor >will penetrate areas around SMD pins quite nicely. What is wrong with >vapor cleaning? This is not true in our situation. The heated vapor cannot penetrate areas around the SMD solder legs (no pins) for one very simple reason: those areas are already filled with liquid solvent which does not run out because of the capillary effect! I am talking about components mounted from 1 to 8 mils off the board, averaging somewhere between 2 and 4 mils (1 mil = 0.001"). I am talking about square, flat, Leadless Chip Carrier (LCC) packages up to 1"x 1", with solder pads (no leads) on 25 and 50 mil centers. Most SMD applications have substantially more space than this; they don't have quite the same problem we have. When you put a room-temperature object into a heated solvent vapor at, say, 130 degrees C, the higher-temperature vapor starts to condense on the surface of the room-tempreature object. Condensation continues until the board reaches the same temperature as that of the vapor. The surfaces are covered with a film of liquid until that time. Don't forget, vapor and liquid coexist at the liquid's boiling point. And this liquid quickly fills the capillaries under the components. And the liquid stays put in there as long as you are at atmospheric pressure in solvent-saturated conditions. And even afterwards. What we find is that maybe a week after the boards are cleaned, enough of the solvent will evaporate to break the capillary. Boards that were ostensibly clean suddenly have puddles of flux/solvent mixture appear around the components. It was underneath all the time. > Other SMD and standard PCB production facilities use non-rosin >fluxes which are water-soluble (with detergent), and _literally_ wash the >soldered boards in a dishwasher following soldering. Obviously, a detergent >acts as a surfactant, lowering surface tension, and allowing penetration of >the detergent and water into all areas of the PCB. What is wrong with >using a non-rosin flux which is water-soluble? It is conceivaly possible that we could clean the rosin flux using an alkaline detergent solution and a dishwasher, but my experience has been that surfactants have a nasty habit of sticking to surfaces and not rinsing off completely. Which could cause problems with the adhesion of coatings applied to the board later. There are couple of very good reasons for sticking with rosin flux: The most important is probably the fact that we have a substantial bit of expertise built up around using a certain type of solder paste which contains rosin flux. We went through testing a lot of different paste solder compositions, including a few non-rosin ones, to even find one that would do the job without spewing massive quantities of solder balls all over the place. If we changed the material we are using, we'd have to go through a steep learning curve all over again, probably affecting relearning the paste application, reworking tooling, tightening sampling inspection criteria, etc. Besides that, I think that the rosin flux paste does a better soldering job -- fewer voids in the solder joints upon X-raying. Another big reason is that we are making this stuff for the military. It would be politically difficult, though probably not impossible, to get an agreement to deviate from all those military specifications which contractually require us to use rosin flux. Also, we seem to have a very conservative design team who are a little superstitious about flux. (And maybe I'm a smidge superstitious about it too.) Rosin flux has a lot of long-term studies behind it's use in high- reliabiliy applications. Non-rosin fluxes, the last time I checked, did not have 5 or 10-year long studies of the long term effects of their use, since they have only come out in the last 5 or so years. Water-soluble fluxes are generally thought to leave ionic residues which can degrade insulation resistance under harsh atmospheric conditions, such as at sea, on ships or in submarines. Rosin is unique because it is acid only at soldering temperatures. At room temperatures the abietic, pimaric, & etc. acid molecules actually change to a non-acidic configuration. So if you do have a few traces of it left somewhere, it doesn't hurt anything electrically. (But I think it may possibly present some heat transfer problems where heat transfer is very critical.) But all of this application-specific reasoning notwithstanding, I think the idea of enzymatic or other biotechnological cleaning agents would probably be worthwhile even on regular industrial cleaning of various sorts. Do you know how much money is spent in industry on exhaust hoods and ventilating equipment to keep the solvent vapors away from personnel? Do you know how much money is spent in industry on disposing of liquid waste solvents in hazardous waste landfills? Do you know what the maximum stack limit currently is for the emission of photochemically reactive solvents into the air? Do you how many salaries in industry are directly tied to educating workers on the safe handling of chemicals? Among those chemicals in the widest general use are cleaning solvents. Do you know that almost every time NIOSH tests a solvent-type chemical for carcinogenicity on rats, it ends up *tightening* the exposure limits for that chemical? Necessitating more and better ventilation? Solvents used to be cheap and easy to clean things with, but they really aren't any more. The people working on the detergent cleaners of various sorts are on the right track, they just aren't going anywhere. Wouldn't it be neat if I could get a tankful of those microorganisms that eat oil spills on the ocean cheaply enough to remove industrial oils from metal parts? Wouldn't it be biodegradable? Couldn't I wash it down the drain? Wouldn't it be less dangerous to people's health? Now that I have writ much more than I intended to on this subject, I guess I'll get back to my favorite salt mine. -- Laura Watson ...ihnp4!ddsw1!dino Contentment is the smother of invention.