Successfully reported this slideshow.
We use your LinkedIn profile and activity data to personalize ads and to show you more relevant ads. You can change your ad preferences anytime.
Troubleshooting the Batch Vacuum Metallized PartK. Anetsberger and J. Stava, Midwest Tungsten Service, Inc., 7101 S. Adams...
vapor lay flat on a perpendicular surface, but stand on edge on          parts for the batch vacuum metallizer. Burned par...
throughout the chamber. Filaments should light as evenly as                racks and fixtures can sometimes leach out cont...
basecoat is properly cured. If it is not, raise bake time or tem-       ting is oxygen trapped in the aluminum expanding r...
REFERENCES1. S. R. Prance, et al., “Vacuum Coating,” Metals HandbookVolume 2, 518 (1964)2. J. M. Donovan, “Design and Vacu...
Upcoming SlideShare
Loading in …5



Published on

  • Be the first to comment

  • Be the first to like this


  1. 1. Troubleshooting the Batch Vacuum Metallized PartK. Anetsberger and J. Stava, Midwest Tungsten Service, Inc., 7101 S. Adams St. #6, Willowbrook, IL 60521 1-800-626-0226 fax: 630-325-3571 http://www.tungsten.comABSTRACT one of the most difficult gasses to pump from the chamber. The primary problems created by high humidity are long pumpdownTroubleshooting coating defects on batch vacuum metallized cycles and burned shots. The simplest way to reduce prob-parts can be frustrating and time consuming. Productivity lems from humidity is to air condition or dehumidify the metal-declines and costs rise as time is spent chasing problems. A lizing area.systematic approach which identifies metallizing problemsand proposes solutions can reduce the time spent trouble- The other major environmental concern which contributes toshooting. metallizing defects is dust. The two most common types of dust that metallizers have to deal with is dust from the out-INTRODUCTION doors and dust from cardboard. Many metallizers run with all doors open to the outside, particularly during warm weather.Over the years, substantial resources have been allocated by Often dust from outdoors simply blows in. It can also be car-batch vacuum metallizers attempting to solve problems with ried in on shoes of employees. As for cardboard, it is used astheir vacuum metallized products. Time and money which packaging for finished parts and can be blown through thecould have been used to produce parts were allocated to pin- shop, but the larger problem arises when cardboard is usedpoint that elusive problem which made the finished product while transporting parts around the shop. Plastic parts canunacceptable. Although it is unrealistic to believe that a batch carry a static charge which will attract dust particles, particu-metallizer will never produce any scrap or that he will never larly when humidity is low. Placing parts on cardboard sheetshave to put any time into troubleshooting, much more time or in cardboard boxes is inviting trouble. Plastic bins or metaland effort is spent on such non-productive activities than is cages are usually better for this purpose. To minimize dust,necessary in most cases. Familiarity with the metallizing pro- the metallizing area should be located in a separate roomcess and its limitations as well as a systematic approach to which can be positively pressurized with filtered make-up air.troubleshooting can increase the productivity of most batch This causes air to flow out from, rather than in to the metal-metallizing operations. lizing area, preventing dust from the rest of the building from entering. To remove the dust which does settle on parts, partsProblems with the batch metallizing operation are usually due can be blown off with a destatic air gun prior to racking. Otherto problems in one of the following areas: substrates, ap- possible causes of dust are air makeup systems which drawplied coatings, equipment operation, fixturing, pumping/fir- outside air into the metallizing area, dirt which has not beening procedure, and environment. While these groupings are filtered out of coatings, or unfiltered air in air lines.logical, from a troubleshooting point of view, it is usuallyeasier to categorize according to problem, as some metalliz- PART DESIGNing defects can have several different potential causes. Another prime consideration when attempting to minimizeTHE METALLIZING ENVIRONMENT metallizing defects is part design. Certain parts lend themselves better to metallizing. What are the qualities of a good part forThe potential for metallizing defects begins well before parts metallizing? First, avoid large flat areas. There is great visualare being fired in the chamber. Let’s begin by examining the appeal in a mirror-like surface, but in most metallizing opera-environment the metallizing system operates in. The two big- tions, such a large, flat area will be a showcase for problemsgest environmental causes of metallizing defects are humid- such as dust, coating defects, and surface imperfections [2].ity and dust. Humidity is trapped on the surface of (absorbed) Molding defects such as sink marks will also be more notice-and is trapped within (adsorbed) substrates [1], chamber walls, able. Deep cavities or recesses should also be avoided. It isfixtures, carriages, and evaporants. In the vacuum chamber very difficult to coat the sides of them properly. Surfaces per-this moisture is released from these surfaces into the space pendicular to the path of the evaporant travel coat better thanbeing evacuated. It must then be pumped away. Water vapor is those parallel to this path. This is because platelets of 1
  2. 2. vapor lay flat on a perpendicular surface, but stand on edge on parts for the batch vacuum metallizer. Burned parts have aa parallel surface. As a cavity becomes deeper in relation to its straw, blue, or black tint to them instead of the usual brightwidth, the aluminum that gets down deepest into the cavity is silver. The dark shot is caused by an interaction between thetraveling in a parallel direction to the side walls of the cavity. evaporant and a contaminant. In the batch vacuum metallizer,The result is a poor dusty gray coating on the side walls deep such contaminants come from leaks, volatiles in plastic partsin the cavity. The solution to this problem is to avoid recesses or coatings, or lubricants [4]. Troubleshooting the burned partthat are deeper than they are wide, and add more filaments to will show the importance of leaving reject parts on the car-increase the chances of a more perpendicular deposition path riage to troubleshoot.[3]. Where possible, a second surface coating of the oppositeside of a deep well coats much better. If the burned parts are always located near the front of the chamber, the problem is a leak or source of outgassing nearGATHERING INFORMATION the front of the chamber. Check for a dirty, over-greased, or cut door seal. A dirty seal may be preventing proper seatingNow that we’ve discussed how to avoid problems before of the door gasket. An over-greased gasket will attract dirtmetallizing, let’s discuss what to do about metallizing prob- and excess grease may be a source of outgassing. Use onlylems that occur during metallizing. The most important rule vacuum grease. A cut in the door gasket may be allowing airin troubleshooting the batch vacuum metallized part is trouble- to enter the chamber. Correct any of these conditions. If theshoot while parts are still on the carriage. This cannot be discoloration persists, check the porthole for leaks. Somestressed strongly enough. Some of the most valuable infor- chambers have gas feedthroughs or emergency vacuum breaksmation about the cause of a metallizing defect is tied to its in or around the door, check these next.location in the chamber. Far too often, some poor soul sits ina room with a lone part trying to figure out what’s gone wrong If the burned parts are near the rear of the chamber, the causein his metallizing department. This is futile. Train shop per- is usually found within the pumping system. Check for sig-sonnel to leave bad parts on the carriage, or to note their spe- nificant amounts of oil at the back of the chamber. If found,cific location before removing them. If you remember noth- wipe clean. Check periodically for return of oil. If oil re-ing else from this paper, remember this. The second rule of turns, check valves for correct operation. Also check valvetroubleshooting is to have histories available. Keep records seals for dirt which may have been drawn in and is preventingon pumpdown times, cleaning schedules, oil changes, num- valves from completely closing. Make certain oil is not fouledber of rejected parts, weather, substrate lot, or any variable with water. This will reduce its effectiveness. Check to see ifyou can think of which affects your metallizing operation. pumpdown times have risen. Check that diffusion pump isThe reason for this is simple. When you start making bad operating at recommended temperature. Lower than normalparts, you need to know what you are doing differently than temperature may indicate a burned out heater or improperyou were doing before. The more information you have about coolant flow. Higher than normal temperature may indicateyour operation, the easier it will be to pinpoint potential inadequate coolant flow or blocked coolant line. Also checkcauses. Realize that this information comes at a cost. It takes planetary drive feedthrough for leaks.time and effort and therefore money to collect this informa-tion. But gathering information on a regular basis soon be- If discoloration is on all surfaces of all parts, check vacuumcomes habit and is almost always cheaper than trying to make pressure to be sure it is in the proper range. If discoloration isup for the lack of it. Third, be systematic in your approach to dark black, has the chamber or carriage recently been strippedtroubleshooting. Follow a set of guidelines such as the ones with caustic? A failure to properly rinse and dry after causticwhich will be shown in this paper. This keeps you from over- cleaning often results in a chamber full of black parts. If discol-looking a possible solution and gives you and others working oration is randomly distributed about the chamber, check thesewith you a frame of reference. When a possibility is elimi- possibilities. The chamber and fixtures may need to be strippednated, everyone can be confident that it has been thoroughly clean. Vacuum levels may be higher than is normal (lower pres-considered and everyone can move forward together. Time sure). This can draw volatiles from chambers walls, carriages,and efforts will not be lost chasing down blind alleys and dead and parts which may not outgas at higher pressures. Theseends, or in disagreement over what the next step should be. volatiles react with the depositing aluminum and cause it toAs many people as possible should be trained in this process. discolor. Has the basecoat been properly cured? UndercuredIt is better for an operator to solve a problem than for him to basecoats are often a cause of discoloration. Oven tempera-have to call his supervisor over and have the two of them tures and bake times should be closely monitored. Also checkworking on it together. that coating has not been applied more heavily than is usual. If so, it may have failed to cure adequately. If the burned partBURNED PARTS seems to consistently occur in one area of the chamber, check the filaments near that area. Are they properly connected? DoThe first problem we will deal with is the problem of burned they burn brighter or more dimly than others in the chamber?parts. This is probably the single biggest cause of rejected Many burned parts are the result of poor current distribution 2
  3. 3. throughout the chamber. Filaments should light as evenly as racks and fixtures can sometimes leach out contaminants topossible across the carriage. Here is the reason. If some fila- parts under vacuum. The contaminant film will deposit on thements burn more brightly than others, it will be difficult or parts and prevent proper adhesion. A more subtle problemimpossible to determine the proper firing point which should can arise from the prevaporization of aluminum. If preheatbe used. If one uses the brightest filaments as a determiner, stage is too hot or too long, a thin film of aluminum can de-then filaments which are not getting as hot will not fire off all of posit prior to the regular evaporation. This thin film will ruintheir evaporant, or might drip evaporant off the filament onto the adhesion of the heavier layer which is then deposited overparts. Chances are that the coating will be inadequate and the it. Give consideration to how the part was handled prior tofilaments which are lighting improperly will have a diminished metallizing. Are all personnel wearing clean cotton gloveslife. If the decision is made to use the dimmer filaments as a when handling parts? Oils from the hands can prevent properguideline, by the time they have gotten hot enough, long adhesion. Also check for backstreaming in the metallizer. Thisenough for them to evaporate their charge, the hotter fila- is especially likely if the problem occurs in one chamber andments will have already fired and will have been pouring ex- not in another. Check pumps for proper oil levels, for cor-cess heat into the chamber, probably resulting in burned parts. rect operation of the valves, water in pump oil, and properAs you can see, even lighting of filaments is a prerequisite pump operating temperature [6].for good coatings. Also, if a filament burns out during a fir-ing cycle, it will sometimes cause a dark area. If the burn Finally, we have the adhesion of the topcoat to the metallizedalways appears on the same area of the part, check the layer. This can be a result of an undercured topcoat, an im-fixturing. Does the burned area pass close to the filaments at proper coating for the job, overexposure to atmosphere priorany point during the rotation? As a rule, parts should never to topcoating or subsequent finishing. Most people canpass closer than six inches from the filaments. troubleshoot the coating problems quickly using techniques discussed earlier in relation to basecoats. It is usually morePOOR ADHESION difficult to trace defects which are a result of atmospheric ex- posure. Parts which are not immediately topcoated can be ex-The next most common problem in batch vacuum metallizing posed to oil vapor, humidity, or dust. The farther that uncoatedis poor adhesion. Poor adhesion can be divided into three parts travel through the plant, or the longer they remain un-areas and it is important to know which phenomenon you are coated, the greater the chances of this problem. To preventdealing with. Adhesion loss can be between basecoat and sub- problems of this type, utilize the shortest route and the short-strate, aluminum and basecoat and topcoat and aluminum [5]. est exposure time for metallized parts to be topcoated. ToEach has different causes. troubleshoot this problem, trace the route the parts travel look- ing for potential sources of contamination. Remember that someLoss of adhesion between basecoat and substrate is usually contaminants, such as silicone sprays can travel throughout adue to a problem with the substrate. Check these things. Does plant.the material have an internal lubricant? Was a mold releaseof any kind used during molding? Often adhesion problems IRIDESCENCEcaused by a mold release will show up in the same isolatedarea again and again. Is the plastic filled with fillers or rein- The next problem encountered by batch metallizers is irides-forcing materials? Was reground or recycled material used? cence. A similar phenomenon, known as Newton fringe, isIf not, check that no surface contamination has occurred. Is often confused with iridescence. The distinction is an im-there any possibility that the surface of the part has become portant one. Newton fringe is composed of colors such ascontaminated? Wash several parts to verify. Use the stron- pink, green, and purple. It is sometimes said that Newton fringegest solvent available that will not damage the substrate. Met- has the same colors as are seen in the Aurora Borealis. Newtonallize again and check results. fringe has only two causes. First, the topcoat could be too thin. Correct this by reducing the coating less to raise theLoss of adhesion between aluminum and basecoat is some- viscosity, or by applying the topcoat more heavily (for spraytimes also called delamination. This is the most common of operations). The second cause could be too thin of an alumi-the three types of adhesion problems. As above, the cause num shot. This is corrected by depositing more aluminum orcould be with the substrate. The same tests apply here too. altering fixturing for better coverage [7].The coating may be contaminated. Try a different batch ofcoating. Sometimes mold release will wash off parts and build True iridescence contains the colors of the rainbow. Irides-up in a flow coating system. Try applying the coating by spray- cence is usually the result of a physical shift of the 3 layersing it. The basecoat may be undercured. If so, adjust baking (basecoat, metallizing, topcoat) in relation to one another. Thisparameters or check for proper film thickness. Sometimes shift results in a prism-type effect and thus the colors. Whatthe metallizing process itself is the cause. Was this part double can cause coating layers to shift? Very often the answer isfired to salvage it from a previous rejection? This will some- improper thickness or improper curing of one of the coatingtimes be problematic. If so, consider the part scrap. Dirty layers. Let’s deal with the basecoat first. Check to see that 3
  4. 4. basecoat is properly cured. If it is not, raise bake time or tem- ting is oxygen trapped in the aluminum expanding rapidly, likeperature as required. Check for temperature variation and inad- water in hot grease, or it is because the aluminum itself be-equate airflow in ovens. The basecoat film thickness may be comes vapor so rapidly that it projects some molten alumi-excessive or inadequate. Measure basecoat thickness if pos- num. By incorporating an appropriately timed preheat thesible, and check with coating supplier for recommended thick- spitting can largely be eliminated. The second cause of alu-ness. If your basecoat was cured at a low temperature and minum drops on parts is molten aluminum dripping off ofyour topcoat at a high temperature, you may get iridescence. filaments. If preheats are too long or too hot, molten alumi-This results from the basecoat moving as it finishes curing num will run down a filament and collect on the last turn. Ifduring the topcoat bakeout. If your basecoat is thin and your enough collects, it will drip off. The jarring action of the car-topcoat is thick, the underlying coatings will move as the top- riage planetary drive can worsen this problem. A drip troughcoat cures. This can also happen if your basecoat is normal placed six inches or so below the filament can help to reducebut the topcoat is excessive. The important lesson here is watch this problem. Be cautious. The closer the drip trough is to theyour film thickness. Also, if the part is not allowed adequate air filaments, the more it will shadow the evaporant from the parts.dry time before baking, the coating may shrink rapidly as it Sometimes molten metal will bounce out of the trough and landdries and again cause iridescence. Also watch baking tem- on parts anyway. It is important to keep the trough clean toperatures in regard to the substrate. High baking temperatures prevent outgassing from the buildup. The best way to combatcan distort plastic substrates and cause iridescence or even the spatter problem is to eliminate it by adjusting preheat andsevere part warping. Some lesser causes of iridescence are firing cycles.backstreaming of pump oil and water absorbtion by parts.Nylon absorbs a high amount of water. Check with your resin OTHER DEFECTSsupplier if you are unsure about the water affinity of yoursubstrate material. Backstreaming very small amount of oil will There are, of course, other causes of metallizing defects.cause iridescence. Look for the problem to be most severe Many of them have to do with the coatings applied. Althoughtoward the rear of the chamber. Large amounts of oil coating problems are not the primary focus of this paper, herebackstreaming will cause a dark shot or poor adhesion. are some of the more common coating problems and some proposed solutions. Blistering of finishes can be caused bySHADOWING trapped organics in either substrate or coatings. Make cer- tain parts are clean and dry before coating and that coatingsShadowing is a problem which results from not all surfaces are completely cured. Moisture in coatings can cause blush-being metallized in an even manner. Causes include improper ing of the coatings. Common sources of moisture are waterfixturing of parts, poor filament positioning, firing time/car- in air lines, high absolute humidity in atmosphere, water inriage rotation mismatches, or poor part design for metalliz- solvent-based coatings, or water in reducing agents. Fisheyesing. All surfaces to be metallized need to be in a direct line are coating defects that look like little eyes or rings. Theyof sight with one or more filaments during the firing cycle in can be caused by surface contaminants, particularly silicones.order to achieve proper coverage. Problems occur when some- Keep all silicones away from the painting and metallizing ar-thing obstructs that direct line of sight or when a part is not eas. Crazing of the finish is usually caused by coating soak-designed or positioned properly to achieve the proper expo- ing in to low density areas of a part or from coatings whichsure. Be aware that standoff posts can have a shadowing ef- have very strong solvents in them. If a coating is excessivelyfect if they are too large or are too close together. The car- thick or insufficiently atomized when sprayed, this can causeriage needs to complete a minimum of one complete rota- crazing as well. These same factors can sometimes cause ation during the actual firing of the filaments. Several rota- similar phenomenon which looks like cobwebs. A dull finishtions are best. Parts that have recesses which are deeper than can be caused by inadequate basecoating, by harsh solvents,they are wide will be metallized with difficulty. Fixture ob- or by low density areas of the substrate. Many problems suchlong cavities for maximum exposure time. Sometimes the as orange peel, runs, wrinkles, or bubbles in the finish areonly way to address this issue is to fire longer, add more fila- caused by improper use of thinners and solvents, or by im-ments, or increase evaporant charge. proper application of coatings [8]. It is best to contact your coating supplier for these types of problems.SPATTERS CONCLUSIONSpatters of aluminum sometimes show up on parts. There aretwo primary causes of this. The first is spitting. When firing, No metallizing operation can hope to completely avoid metal-aluminum will pop or spit off the filament and land on parts, lizing defects. A systematic approach to troubleshooting com-usually causing a small burn mark and perhaps sticking on to bined with good historical data on the metallizing operationthe part, or perhaps bouncing off. Spitting usually occurs in can significantly reduce the cost and aggravation such de-direct fire operations or when preheat is too short or tem- fects will cause.perature is increased too rapidly. The cause of the actual spit- 4
  5. 5. REFERENCES1. S. R. Prance, et al., “Vacuum Coating,” Metals HandbookVolume 2, 518 (1964)2. J. M. Donovan, “Design and Vacuum Metallizing,” Pro-ceedings of Society Of Vacuum Coaters, 10 (1962)3. F. Adams, “Taking the Mystery Out of Metallizing,” VacuumTechnology, November (1975)4. L, Winters, “Vacuum Chamber Problems In DecorativeMetallizing,” Proceedings of the Society of Vacuum Coaters,11 (1960)5. M. Self, “Laquering Techniques for Vacuum Metallizingon Metals and Thermosetting Plastics,” Proceedings of theSociety of Vacuum Coaters, 26 (1957)6. R. Schneider, “Vacuum Metallizing Problems, Causes, andCorrections,” Proceedings of the Society of Vacuum Coaters,130 (1975)7. Ibid., 1328. A. Schwartz, “Appearance of Decorative Vacuum Coat-ings,” Proceedings of the Society of Vacuum Coaters, 25-29(1960) 5