Locke's description of the two types of metal crowns are as follows: "You will notice....that the flux has been controlled by applying a metal cap provided with top wire and tie wire grooves. When the higher voltages, that is to say 44 Kv and up, are encountered, such a cap becomes considerably less effective and while some improvement is noticed, this improvement is not sufficient to completely prevent flux concentrations at critical points under voltage conditions only slightly in excess of normal operating voltages. Accordingly, insulators for these higher voltages are made with a metal top insert which...so retards the formation of corona that the interference point is raised safely beyond any voltages that will be encountered under regular operating conditions." Note the metal cap was used on insulators 23521, 23522, and 23523 and the metal insert was used on insulators 23524, 23525, and 23526. The metal insert was shown in the 1941 catalog on insulators all five multipart styles indicating they abandoned the metal cap design. No Locke insulators have ever been reported with the metal cap design.

I know many of you will find it interesting to read Locke's 1932 discussion of RFI causes, solutions, and changes in insulator designs to combat RFI, so below is a shortened version of most of the report. I received several comments from collectors after the January PIN on the Brown patent thanking me for discussing RFI. I think this report will help answer other questions some of you may have as well as give you insight into the early efforts to solve the growing RFI problem.

RADIO INTERFERENCE
(taken from a Locke report dated September 1, 1932)

With the rapid increase in the use of radio receiving sets and the equally rapid improvements in sensitivity and selectivity of such sets, radio interference has assumed the dimensions of a major problem in public relations. It is found that near the large cities where there are a number of broadcasting stations there is very little demand on the part of the engineers of the power companies for the so called "nostatic" insulators. Such demand comes almost exclusively from the isolated places where radio sets must be forced to bring in distant stations.

Radio interference is caused by current flow or, more properly, sudden current flow and interruption, and not by voltage. All kinds of electrical apparatus and appliances can cause radio interference and equally bad interference can arise from 110 volt lighting circuits as from high voltage lines and equipment. Broken wires in lamp cords, loose fuse plugs, and many similar things can wreck reception. The hardest part of eliminating the offender here is finding it.

The sources of radio interference from high voltage lines are many, and before any recommendations can be made to eliminate the interference all such sources must be considered. Current leakage around distribution transformers, fuse cutouts, loose fuses in the cutouts, are a fertile source of complaint especially as they are generally near residential districts. The remedy here is self-evident.

At relatively high voltages shielding of clamps is necessary. Corona appears at the clamp tips as much as or more than at the bolt ends. The rounding of the comer of the nuts alone is far from sufficient. Locke grading shields are a real answer in this case. Arcing horns on suspension clamps have been found to be serious offenders. They should in every case of complaint be considered and if necessary either be removed and replaced with Locke grading shields or with Locke spherical tipped arcing horns.

On wood pole construction loose hardware has been a source of trouble due to charging current sparks jumping across small gaps. Hardware should be tight and small gaps avoided. Frequently with the conventional cross arm construction the bolt of the insulator pin and the arm brace bolt are too close together. This design should be watched carefully. The brace bolt should be midway between insulators. This is an important detail also in protecting the insulator against lightning damage. 

Until the above conditions, if existing, have been remedied, there is no need to consider the insulator. Trouble from insulators is a very small percentage of the total radio interference. When every other potential source of trouble has been considered is the time to consider the insulators unless, of course, the condition is obviously caused by defective insulators. It must be remembered that dirty insulators will invariably cause radio interference when the insulators are wet by fog or rain.

The cap and pin type suspension insulator is not a source of radio interference at operating voltages when used in accordance with standard practice. There have been a few complaints from the clevis type insulator where cotter keys have been too long. The Locke cotter key is short and causes no radio interference. For the higher voltages the socket type insulator is looked upon more favorably. 

There have been many attempts to design pin type insulators which should be free from radio interference but to the present none of these have been successful. The metal coated hood, while eminently satisfactory, for a short time, will not withstand the action of the elements. The coating wears unevenly leaving small isolated islands of metal which cause more and worse interference than the original uncoated insulator. 

Radio interference as it applies to insulators is caused by corona which in turn is caused by over stressing the air by dense electrostatic flux. The picture of the action of the flux through an insulator can probably be a little more easily understood it we think of the insulator as a conductor of electrostatic stress lines or flux. The flux originates at one terminal and must find its way to the other, the terminals being the live parts. On one side is the tie wire and conductor and on the other generally the pin. The resistance of the metal parts and the cement joints is zero. Flux distribution may therefore be entirely different on one side of a cement joint than on the other. The greatest amount of flux will be along the shortest path between terminals. Some, however, will flow by the most devious routes. This sometimes means that great flux density may occur near the junction between shells causing the flux to jump across the air gap. Since the resistance to flux of air is seven times that of porcelain the voltage drop in the air at these points is great in proportion causing over-stressing of the air and corona and consequent radio interference. On an ordinary pin type insulator the flux resistance of the porcelain heart is so low that a large amount of flux flows. This all must start at the tie wires and conductor causing a great flux density at this point both through the porcelain and through the air. The result is that the voltage drop in the air, being proportional to flux density, is great enough to cause corona.

In other parts of the insulator and in other insulator designs, extending porcelain masses may offer a relatively low resistance to the flux which can then find its way back to the conducting parts only by jumping through air with the resultant over stressing of the air. Corona is not continuous. It does not form until a certain voltage is reached and therefore goes out and forms suddenly again between each half cycle. Since insulators are primarily condensers, this sudden formation is accomplished by a sudden current rush to fill the condenser. This rush broadcasts a shock wave which will find its way through any radio set regardless of its tuning.

The obvious cure is to produce insulator designs in which the principal pathway for the flux will be through porcelain -- designs wherein insufficient flux passes through the air to cause overstressing. The very old type elementary multipart insulators may cause some trouble because of their poor voltage distribution. Such insulators which are usually of the conical cup shape may possibly be improved by partly filling the space between shells with pitch compound or some similar material. These old insulators cannot be improved without careful consideration of tie wires, pin holes, and spacing between shells. On this type of insulator no action should be taken until samples have been submitted to the manufacturer for examination.

No treatment with conducting paints will give permanent benefit. At first such treatment may show an improvement but this improvement will rapidly disappear as the paints lose their conducting properties through rapid digesting and weathering. As pointed out earlier, this will leave isolated islands of conducting surfaces to draw current charging sparks. This can in a small measure be overcome by wrapping the tie wire several times around the head of the insulator so that there will be portions of it which will maintain control with the coating for as long a time as possible but ultimately the interference will become considerably worse than it was before the treatment was applied. In favorable localities certain conducting paints will have a maximum efficient life of about two years. 

Old style insulators with porcelain threads when mounted upon lead tipped pins may be the source of trouble. Coating the pin hole with a conducting paint which will short out the gap between pin and porcelain will tend to relieve this situation. This must not be confused with painting the heads of the insulators. In this case paint will be shielded from the weather and will be practically permanent. 

Old style pins with wood cob and metal bolt will cause bad interference unless the pin hole surface is treated to make it conducting and the bolt positively seated on such conducting surfaces. Probably as good a way as any to overcome radio interference with pins of this kind is to partially fill the pin hole with Portland cement and then screw in the pin. Wood pins in porcelain pin holes normally cause no radio interference.

Insulators with cracked shells will almost invariably cause interference due to sparking through the crack to the next shell or pin. These, of course, should be removed regardless. The present designs of standard pin types are better balanced than the old type insulators and lend themselves readily to further improvement from the radio interference standpoint.

The Locke Insulator Corporation made a big step in this direction by standardizing on the metal threaded pin hole. Even with this, certain care must be exercised. The space between the metal threads and the porcelain must be filled with cement. Narrow open spaces between metal and porcelain will invariably cause trouble. Porcelain threaded insulators for the higher voltages should never be considered unless mounted on wood pins. Where metal pins are to be used some measure of protection can be obtained, as mentioned before, by coating the inside of the pin hole with a conducting material. 

Of the various coatings which have been tried on insulator heads metal applied by the Metalaver process is undoubtedly the best. Unlike plating, this gives an appreciable thickness of metal but even so the effective life of such a coating is doubtful, and unless the metal is definitely bonded to the porcelain sparking will occur across the tiny air gap. For insulators which are to be exposed to the weather we do not recommend any form of coating whatsoever. 

In this connection it might be well to sound a note of warning against pin type insulators having the tops covered with a metal spray all the way to the outer edge. There has been some demand for this type of insulator but it not only throws away a large part of the insulation value of the porcelain, but defeats its own purpose by causing flux concentration at other parts of the insulator which increases rather than lessens the interference. Cap insulators have already been offered but our tests have shown them to be so poorly balanced that freedom from complaint can not be expected from them. 

A long series of investigations in the Locke laboratory has shown definitely that in addition to the metal cap, very slight but very important modifications in external design are necessary if a permanent relief was to be obtained. These improvements are so slight that little if any difference can be noted in the contour of the new interference proof insulators. Many insulators with open joints show corona dangerously near the operating voltage whereas insulators of the old abutting type were free from corona at these points. Consequently the joints of these insulators have been so designed that they exhibit the same corona characteristics as the old abutting joint insulators with none of the mechanical hazards of this type.

Porcelain threaded pin holes induce corona unless treated as outlined in the foregoing pages. The insulators we recommend as the remedy for radio interference consequently have metal threaded pin holes. In the larger pin type insulators corona is visible in the pin hole at too low a voltage for safety. This is so near in fact that the use of the larger pin type insulators at slightly higher altitudes than is usual would definitely provide a source of radio interference. To overcome this the lipped type pin hole has been adopted as standard.

HARDWARE: 
Inserts [for the three larger insulator styles] and caps [for the three smaller insulator styles] are high grade gray iron castings, properly finished and smoothly galvanized. Those are entirely adequate for the purpose and are less expensive than malleable. The caps are adaptations of switch type caps and are attached to the porcelain in exactly the same manner. 

The inserts used on the higher voltage insulator are similar except in externals to the inserts which have been successfully used for many years on various supporting insulators. The section projecting into the insulator is practically a split ring. This splitting adds a flexibility which, while it may never be needed, is an additional safeguard.

ASSEMBLY:
In assembling these insulators all the usual precautions are followed. Each insulator is provided with expansion joints, cement is kept to a minimum thickness, steam cured and made definitely water repellent. On the insert type insulators the top hole is sanded and the inserts are provided with corrugations which will satisfactorily resist the torsion and bending actions which may be expected on pin type insulators. In the upper part of the cap a pool of paraffin is provided. This pool will slowly melt in the sun and will keep any spaces filled, at the same time insuring that the cement will permanently have the quality of repelling water;

DESIGN AND CHARACTERISTICS: 
Early in our experimental work fire found that the old style abutting joint insulators were very free from corona at the joints -- a necessary condition if the insulators are to be free from radio interference. This led us to seriously consider an improved form of abutting joint but fortunately before this had progressed very far we found that slight modification of the open joint would have the same desired effect. This modification has been applied not only to the noiseless insulators but also to all cemented pin type and switch type insulators. This means that all such insulators manufactured by the Locke Company now have joints with corona points comfortably above the operating voltages. 

The shape of the outer edge of the pin hole and the position of the porcelain outside of the center shell all affect corona within the pin hole. We have carefully balanced this up on all our insulators although only on the largest insulator has it been found necessary to go to the lipped center shell. Smaller insulators are entirely satisfactory without this ripping feature. That is to say, smaller Locke insulators are. We have not had either the time or the opportunity to thoroughly test all competitive insulators but we believe that much of the radio trouble caused by them is the result of improper design at this point. Puncture values will be just as high if not higher than that of the regular line insulators. The leakage distance is practically the same as on corresponding standard pin types.