Historical Articles

 

Trouble Shooting In the Plating Room

RICHARD B. SALTONSTALL

Technical Director, The Udylite Corporation, Detroit, Michigan

December, 1949

Trouble shooting in plating presents quite different problems today than it did a number of years ago. No longer are generators used that are so unstable in their electrical characteristics that they reverse their polarity with no apparent reason. The other auxiliary equipment found in the plating room--pumps, motors, filters, agitators, rheostats, temperature controls, water deionizers, plating barrels, semiautomatic and full automatic conveyor machines--have over the years been improved to the point that they seldom give trouble provided they are properly used and taken care of. Many proven proprietary processes are now furnished by a number of suppliers, with details regarding the operation and formulation made available to users. Chemicals for preparing processing solutions and metals for anodes are of high purity. Good cleaning compounds and organic solvents are at hand for a wide variety of jobs and for use in several types of equipment such as soak and electrocleaning tanks, spray-washing machines, tumble cleaners and various degreaser combinations. Modern welded steel tanks with inert nonmetallic linings have gone far in preventing solution contamination and stray currents.

Taking all these factors into consideration one might think that there should be no troubles and hence no need for trouble shooting. However, such is not the case; other factors have complicated the picture. Higher quality is continually being demanded. This involves use of heavier coatings, which are always more difficult to plate than thin coatings. Economics and competition require faster plating, and there is always the controversy of cost vs. finish or brightness to trouble the plater.

Quality of base metals is usually poor in a period of shortages and this at times creates additional problems for the plater. Lack of knowledge or interest on the part of designers regarding the inherent limitations of plating processes makes the platers more difficult and often increases finishing costs unnecessarily.

Difficulties In New Installations

It is not possible in this article to discuss in detail each problem which may arise, and all its possible causes. Trouble is often experienced when starting up a new hand-line or automatic-machine cycle, especially when sample parts have not been available when the equipment was designed. The-only sure way t~ determine the best method of racking each part is the famous "cut-and-try" method. It is almost beyond human ability to provide for each contingency which may arise after operations have commenced. For example, one operator prepared a cyanide copper solution very carefully, following printed instructions. When all was ready, a rack of work was placed in the tank and nothing happened but gas evolution at the surface of the work. A hurried telephone call to the supplier resulted in the logical answer that the power source was connected backwards. Careful checking, both by tracing the leads and by reversing the connections proved this answer to be incorrect---still only gassing. But investigation of the materials used in making up the solution revealed that powdered pumice had been used instead of copper cyanide. This is, of course, an extreme case, and the solution of most starting troubles is not so simple.

Occasionally a number of things go wrong simultaneously and considerable time and patience are required to get production running smoothly. For example, a full automatic machine designed for copper-nickel-chromium plating on zinc-base diecastings can present a very troublesome and puzzling combination of problems. It might turn out a few hundred racks of excellent work the first time it is put into operation, and then-trouble! On one such occasion with which the author is familiar the following took place: The diecastings were from new dies and many of them were not of plating quality. The polishing and buffing was poor and no provision was made for inspection of the castings prior to plating. The racking was satisfactory for copper and nickel but had to be changed for chromium. All ventilating ducts led to a common duct and spray from the chromium tank condensed and ran back into the other solutions. The rack fingers had to be bent to change the position of some of the work and in the bending the rack coatings were damaged in such a way that the damage was not detected. 3:his resulted in a second and unobserved source of chromium contamination which continued to cause trouble after the first source had been eliminated. On top of all this, one item which was an appreciable part of the total production was so designed as to be extremely difficult to plate even in a hand-operated line. One can easily imagine the results of such a combination of factors, especially when the assembly line of a motor-car manufacturer was practically shut down for lack of the most difficult part. Bare zinc showed in deep recesses, bare copper in recesses not so deep, and there were pitting, roughness, blistering, mischromes and chrome burning. One difficulty after another would be corrected but the benefits were almost completely camouflaged by those still existing. Only by the persistent efforts of several thoroughly experienced plating men over a considerable period of time were all the faults eliminated and satisfactory work produced. This, of course, is another extreme case, which fortunately does not occur very frequently.

However, since most readers ere not often called upon to start up new operations, the balance of this article will be devoted to a discussion of troubles which may occur in established operations and of various method of diagnosing and eliminating their causes.

POOR ADHESION

Probably the most serious fault commonly encountered is blistering, peeling, or in extreme cases misplating or "skipping" in recessed areas, because when it takes place, there is no choice but to strip and rework or discard the object on which the defect has occurred. The first point to establish in the case of composite coatings is where the faulty bond appears and whether it is general or local. Is the entire plate peeling from the base metal? Does the surface of the base metal come away with the plate as sometimes happens with zinc diecastings? Is nickel peeling off copper or off itself? These questions can be satisfactorily answered by a careful examination of the defect. If the faulty bond is between the base metal and the plate, it is logical to look for cleaning trouble and to consider a change in the treatment or composition of the base metal before it reaches the plating cycle. If the condition is general on all or most of the work, it is advisable to stop operations until the situation has been remedied. In such cases, the trouble is usually easier to locate than when the defect occurs only occasionally.

A good way to check on the cleaning is to hand scrub a few pieces with a stiff brush and pumice, then introduce them directly into the first plating solution, bypassing the preparatory cycle. If satisfactory adhesion is obtained, surfaces of acid dips and rinses in the preparatory cycle should be checked for grease or oil, and the cleaners strengthened or changed. If adhesion is poor on the hand-cleaned work, or if misplating persists, contamination of one or more of the plating solutions is indicated and they should be tested by means of small-scale beaker or Hull Cell tests in the laboratory for possible need for purification.

It should be remembered that a wide variety of alloy steels are being used today, containing varying quantities of alloying metals. As the content of chromium, nickel and molybdenum increases, good bond becomes increasingly difficult to obtain; and in the case of stainless steels a low-pH nickel strike must be used on the base metal before good bond of other plated metals can be secured.

In the case of zinc-base diecastings, the plate sometimes peels off and a layer of zinc comes with it. This can be caused by cold dies. The first metal which enters the die solidifies quickly, is kept solid by the cold die and does not fuse into the metal making up the bulk of the casting. At times deeper acid etching will compensate for such conditions if the poorly adherent skin is not too thick, but the difficulty should be corrected in the diecasting room. .

If nickel is peeling from unbuffed copper, insufficient rinsing (or cleaning in the case of some bright coppers) may be responsible: Sometimes in a severe adhesion test the copper deposit "splits", i. e., is visible both on the back of the peeled plate and on the base metal underneath. This is not necessarily a fault of the copper deposit, but may have taken place simply because something had to give and the copper had the lower tensile of the three materials. There is a pronounced tendency today to substitute a low-pH nickel strike for the copper strike on steel which eliminates the "split copper" problem. Present practice is to go directly from the nickel strike into the nickel plating bath without intermediate rinse or acid dip.

This change, however, does not necessarily solve all bond problems. Plating nickel on nickel, while it has been done successfully in production for a number of years, is rightfully thought by many to be rather touchy operation. The transfer should be made very quickly to a "live" work rod or the work should enter the nickel plating tank with "live" contact if the operation is to be completely successful. This also applies to a lesser extent when bright nickel is plated on copper and other metals.

When peeling results between the nickel strike coating and the nickel plate it is usually severe. Sometimes the nickel plate itself is laminated and a portion of the deposit peels off when chromium plated. The cause is almost without exception electrical: either contact has been broken for an appreciable period of time, or intermediate electrode effects are present. An occasional failure of this type is most annoying and is often difficult to correct. In one case it occurred in one of two semiautomatics operating side by side on the same type of work from tile same cleaning line. In desperation, the operator interchanged the solutions, but the defect persisted in the same machine as before. It could be duplicated at will by lifting a rack off the cathode rail for 10 seconds, but it was impossible to catch a rack breaking contact. The trouble finally disappeared--we never knew why.

One frequent cause of peeling on reworked parts is failure to remove all the nickel before replating. The use of the nickel strike is very helpful in such case.

Aside from faulty cleaning, probably the most common cause of blistering, peeling, or skipping is contamination of cleaners, acids, rinses, and plating solutions with chromic acid. Where danger of such contamination exists, it is wise to make regular small additions of sodium hydrosulfite to all alkaline solutions, and sodium bisulfite to the nickel when necessary. Excesses of the latter will cause dullness in some bright nickels.

The judicious use of suitable wetting agents hi processing solutions can be helpful, but their promiscuous use can do much more harm than good.

Pitting

Pitting, the second major defect, is a rather complex phenomenon and is almost as serious as peeling or blistering, since plate which is appreciably pitted is not acceptable and nrast be removed if the plated object is to be salvaged. It is often difficult to distinguish pitting from roughness, but this can usually be accomplished by careful examination at various angles to the surface under good magnification. Pitting may be defined, for the purpose of this discussion, as small indentations in the deposit visible to the naked eye which may or may not be associated with similar indentations ill the base metal.

"Gas pitting" is one type which was very troublesome in nickel solutions before wetting agents were employed to prevent it. For a few years, it was heard of only occasionally, but now it is being encountered again, principally in high-efficiency alkaline solutions. Gas pits are usually recognized quite easily by the "tails" which project vertically upward or in the direction of agitation. Frequently. small bright circular areas surround such pits. These pits appear to be caused by an adhering hydrogen bubble, or by a stream of hydrogen bubbles coming from a focal point, which prevents local deposition of metal. Clean solutions tend to pit less than dirty ones, but this is only part of the story. Wetting agents are being used in some alkaline plating solutions and do a creditable job of preventing gas pits, but their use has its disadvantages. Sometimes the addition of a small quantity of sodium cyanide to lower the cathode efficiency slightly will correct the gas pitting, or the addition of an organic material other than a wetting agent will change the polarization characteristics in such a way that pitting stops. Air agitation is helpful in preventing gas pitting when its use is not objectionable for other reasons.

Another type of pitting, encountered in bright nickel solutions, is called "grebe etching". This may occur in localized areas at various current densities and when observed under magnification appears to be a mass of small irregular indentations in the surface of the plate, sometimes extending to the base metal. Increasing the wetting-agent concentration to a value higher than normal will frequently overcome this type of pitting, but the relief is usually only temporary and the bath must be purified to remove the cause of the defect.

Still another type of pitting seen occasionally in bright nickel solutions containing wetting agents has been classified as "foam pitting". This highly objectionable phenomenon occurs on the under side of surfaces which are horizontal or nearly so. It is caused by pumps sucking air into the solution at leaky packing glands or other leaks in the suction line. This air is dispersed by the pump impeller into many tiny bubbles which are partially stabilized by the wetting agent. These tiny bubbles tend to rise in the solution and seem to come up against horizontal surfaces faster than they can escape around its edges. The result varies from black pits to large black wormy-looking areas. The obvious remedy is to prevent air from being sucked into the pump.

Pits in base-metal surfaces are often mistaken for pits caused by faulty plating. A good bright-plating solution will tend to minimize or hide small imperfections such as polishing lines, but when the base metal is pitted the bright plates tend to accentuate the pits as far as appearance is concerned. A case in point recently occurred in a shop where grille sections of fair-quality cold-rolled steel were being prepared for bright nickel plating by only a single wheel operation on a sisalin buff. The plated finish passed inspection for some time, then all of a sudden the plate was not bright enough. Careful examination showed that the surface of the steel from a new lot was more deeply pitted than previously and that the sisalin buffing operation no longer was sufficient to eliminate the pits, which were very fine and close together.

Another example of base-metal pits is cold-shot areas in diecasting. A certain large casting was giving much trouble, always in the same location on the casting. The pitting was so bad that black streaks, characteristic of zinc, were emanating from the pits. By increasing the thickness of the copper strike coating prior to bright copper plating it was possible to hide the defect almost completely, although it is probable that this area would still be vulnerable to corrosion. The real remedy, of course, would be to improve the casting.

Roughness

A third major fault which at some time or another plagues most plating operations is roughness of deposits. From a visual standpoint, this is more serious now that so much bright plating is used than earlier when many electrodeposits were heavily buffed to brighten them. From the standpoint of corrosion resistance, roughness is always detrimental. If the plate is buffed to eliminate the roughness, particles arc almost certain to be torn out, greatly reducing resistance to corrosion and often creating a pitted appearance.

Roughness, like pitting, may be so fine and widely distributed as to be confused with lack of brightness.

Causes of roughness may be divided into two general classifications--those existing on the work when it is placed in the plating bath, and those existing in the plating bath itself. Both may be present at the same time, which makes the problem more confusing. A good preliminary test to locate the source of the roughness is to apply a heavy copper plate to a few pieces, buff them, and introduce them at various positions in the cycle.

One of the most common causes of roughness when plating on steel is faulty polishing. On many occasions the addition of a single operation with finer emery and grease stick or with a tampico wheel makes the difference between severe roughness and acceptably smooth work. To perform the final polishing operation with a dry wheel is to invite roughness. Slivers of steel adhere to the work, which is usually more or less magnetic, and act as starting points for nodules. When these nodules are removed by buffing, they are either torn out, or cut down so that the end of the steel sliver is exposed, and in either case rusting may begin very soon.

Insufficient cleaning may also cause roughness. A striking example was the elimination of roughness by the addition of an anodic cleaner immediately prior to the spray washer. This roughness had been so severe that a heavy sisalin buffing operation on a semiautomatic lathe had been necessary to remove it, undoubtedly at a tremendous sacrifice of corrosion resistance.

Removal of soluble polishing and huffing compounds by means of a vapor degreaser can leave insoluble solids on buffed brass, copper, or zinc which will result in roughness. The use of a solvent spray will in many cases correct this condition.

Roughness which occurs in the plating solutions themselves may be divided into two types--settling roughness which occurs on horizontal surfaces or nearly horizontal surfaces facing upward, and general roughness which occurs on any or all surfaces regardless of their position.

The first type, settling roughness, is usually not difficult to cope with. It. is caused by the presence of solid particles in the solution, which under the influence of gravity settle on the work and become included in the plate. A good batch filtration, or in many cases continuous filtration, will correct the trouble by removing the particles, provided they do not enter the solution or form too rapidly. The source of these particles is often difficult to find. Faulty anode corrosion and attendant sludging is one source. In cyanide solutions this can often be corrected by adjusting the anode current density, provided the anodes are of satisfactory purity. Bags can be used in acid solutions, but fine metallic particles can get through bags. Bags cause trouble in alkaline solutions, and quite recently there has been much interest in building semipermeable diaphragms for anodes in bright copper solutions which have caused considerable roughness troubles. There is almost always some sludge on the bottom of a plating tank, and if the solution is stirred to the bottom, intentionally or accidentally, this sludge will cause settling roughness. Iron precipitated as ferric hydroxide in the solution or falling into the solution as rust from overhead structures is another source. Powdered metal from incorrectly designed sliding contacts on an automatic machine has caused baffling cases of roughness. Any solids falling into the plating solution from the air can cause not only roughness but chemical contamination; hence the practice of separating plating from unavoidably dusty polishing operations.

Calcium sulfate has the peculiar property of becoming less soluble as the temperature is increased. Thus, if a sulfate solution is treated with lime and filtered at a low temperature, calcium sulfate will precipitate when the temperature is raised. Depending on particle size, either settling or general roughness may result.

Two rather obvious causes of roughness are leaky filter diaphragms and anode bags with holes in them.

A less common source is the formation of nickel hydrate in the bath caused by burning on unoccupied rack contacts. When burning on high-current-density areas of work forms rough areas, faulty solution balance or too high a current density is indicated. Certain impurities in plating solutions such as carbonates in copper and phosphates in nickel baths may cause burning and rough deposits at normally safe current densities.

Pigmented rack coatings may disintegrate slowly and the pigment slough off in the plating solution. Steel parts may be magnetized in polishing operations or under the influence of direct-current fields in plating. Even though adhering metallic slivers have been removed, such magnetic parts will attract magnetic particles in the plating solution and roughness results. Bumper bars which were apparently nonmagnetic when they entered the plating cycle have come out so strongly magnetic that they would cause au open pocket knife balanced on the finger to dip sharply when the knife blade was brought close to the bumper. Bags were not being used on rather poor cast nickel anodes and, needless to say, the plate was exceedingly rough.

There are probably other possible causes of roughness but those discussed are believed to be the most common.

STREAKS AND BLOTCHY AREAS

A fourth major defect which is peculiar to bright plating is streaking and irregular blotchy areas which are less bright than the surrounding area. The tendency is always to attribute this defect to a faulty condition of the bright-plating solution, but there may be other causes.

In full automatic and hoist-operated cycles, where the transfer time is usually longer than in hand-operated cycles, there is always a tendency for hot cleaners to dry on the work before the transfer into the subsequent rinse tank is completed The presence of wetting agents in cleaners may aggravate this condition since drainage is better and the thinner film of alkali will dry more quickly. The exact outline of bubbles from dried cleaner foam may appear on the finished work. Elimination of wetting agents, lowering of cleaner temperature, and use of mist sprays to keep the work wet in transfer will usually solve such problems.

Some bright cold-rolled steels which are formed into shapes and bright plated without polishing have organic materials rolled into the surface and are extremely difficult to clean. Failure to remove these organics can cause unevenness in brightness.. There is no general remedy for this condition. Sometimes degreasing prior to the usual cleaning will solve the problem. In unusually stubborn cases it is necessary to tumble the parts wet with a fine abrasive. Insufficient Cleaning is frequently the cause for such defects. Pickling is retarded on unclean areas, and this may cause splotchy bright plate. Over-pickling and failure to remove smut causes dullness which is really fine roughness.

However, clouds, streaks, and lack of uniformity of brightness are in many cases caused by solutions being out of balance or contaminated, and then small-scale beaker or Hull Cell tests will usually duplicate the difficulty.

White areas in chromium plate are easily caused by intermediate-electrode effects, and general whiteness by broken electrical contact.

POOR Corrosion RESISTANCE

The last common fault to be discussed is poor corrosion resistance. Corrosion-resistance requirements in most plating specifications for cathodic coatings are based on thickness and salt-spray resistance--not one or the other, but both. The importance of this in specification plating cannot be overemphasized. While correlation between salt-spray hours and various conditions of atmospheric exposure has not been established, the salt-spray test will pick out unsatisfactory deposits. In-other words, coatings which fail in the salt spray will very probably fail in service, but coatings which meet salt-spray specifications will not necessarily stand up well in service.

A specification which requires 0.001 inch of nickel and 100 hours salt spray does not guarantee that 0.001 inch of nickel will withstand the salt-spray test for 100 · hours without breakdown. On a given base metal it may be necessary to deposit twice or several times the specified thickness of nickel to meet the salt-spray requirement. Recent results of A. E.S. Research Project No. 5 indicate that the presence Of as little as 50 mg/I of copper in the nickel bath can cut salt-spray life in half. This should be confirmed by further experience.

Much difficulty is being experienced today in meet-lng salt-spray specifications on nickel plated steel. Some attribute this to poor quality of plate from proprietary solutions, others to poor quality of base metals, and others to occlusions and impurities in the plate. Certainly any one of these factors, or even worse, a combination of two or more, will mitigate against satisfactory resistance to corrosion.

However, data based on surveys made by a large motorcar manufacturer indicate that in spite of the handicaps we are progressing in the right direction. The quality of bright plate on automobiles is definitely better now than ever before.

The previous remarks refer principally to cathodic coatings. In the case of cadmium and zinc, which are anodic to steel, the purpose of the salt-spray test, when used, is to check for bare areas or areas of thin plate.

Some readers are familiar with the atmospheric exposure tests of copper-nickel-chromium and nickel-chromium coatings on high-carbon steel now in progress under the auspices of Committee B-8 of American Society for Testing Materials. A recent unofficial interpretation of the data so far collected reveals two rather striking facts. Assuming good steel, well polished and cleaned, the corrosion protection is dependent principally on (1) the thickness of nickel (whether or not copper is used as an undercoat) and (2) the type of atmosphere to which the coatings are exposed. This data in the opinion of the author shows quite conclusively that copper cannot be substituted for nickel. The performance of the so-called "Victory" plate on automotive bumpers early in the War also indicated this fact.

There have been some failures caused by cracking of nickel-chromium deposits. It is a known fact that bright nickel plate is not as ductile as white nickel plated from a good Watts-type bath, and may not be sufficiently ductile at present to meet all requirements; but progress is also being made in this respect. While fully bright nickel may never equal white nickel in ductility, recent developments have progressed sufficiently that one may anticipate appreciable improvement in the near future.