US1425645A - Process of purifying and dehydrating transformer oil - Google Patents

Description

dehydrated. In use,

" NITED ST orrlca.

CYRUS HOWARD HAZEQOOD, OF NUTLEY, JERSEY, ASSIGNOR TO THE DE LAVAL SEPARATOR COMPANY, OF NEW YORK, N. Y., A CORPORATION OF NEW JERSEY.

PROCESS OF PURIFYING AND DEHYDRATING TRANSFORMER OIL.

No Drawing. I Application filed May 3,

To all whomit may concern."

Be it known that .I, CYRUS HOWARD HAP- GOOD, a citizen of the United States, and resident of Nutley, county of Essex, and State of New Jersey, have invented a new and. useful Improvement in Processes of Purifying and Dehydrating Transformer Oil, of which the following is a full, clear, and exact description.

'The flow of current through a static electric transformer causes so rapid a rise in temperature that the heat cannotbe radiated by air cooling. Hence it is customary to immerse the transformer in oil, which absorbs the heat as it is generated. The oil is contained in a casing having a relatively large surface exposed to the air, which absorbs the heat from the oil as rapidly as the oil absorbs heat from the transformer. This oil is known as transformer oil.

It is essential that this transformeroil shall be of very high dielectric strength, to secure .which it must be almost absolutely the oil gradually absorbs moisture, which, although it accumulates slowly and at no time reaches more than a very small proportion, nevertheless so far lowers the dielective stren th of the oil as to necessitate its re-dehydration.

' Any oil containing asubstantial percentage of free Water may be purified, to the ex-' tent of removing the larger part of the water, by prolonged settlement by gravity.

The most prolonged settlement, however, will still leave in the supernatant 011 a pro-' portion of water not substantially less than one-quarter of one per cent or 2500 parts to the million. For many purposes this degree of purification suflices. Where a still further purification is desired, it may be effected by centrifugation in any machine of the cream separator type, wherein water will escape from one outlet and oil from another. However, the percentage of moisture even in used and contaminated transformer oil is far below the percentage of water that will resist elimination by gravity treatment; while the specific centrifugal treatment when conducted in, the most effective type of centrifuge specified, yields an oil of scarcely greater dielectric strength than transformer oil that has been contaminated by prolonged use, and therefore not approximating the dielectric strength demanded by electric Specification of Letters Patent.

Patented Aug. 15,1922. 1921. Serial No. 466,495.

other words, used transformer oil requiring dehydration contains far less moisture than oil after it has been dehydrated to the greatest extent possible by, gravity, while the specified centrifugal treatment is ineffective to reduce the proportion of water to the extent required to meet the requirements for fresh transformer oil.

Resort, therefore, has been had to filtration. The filter generally in use co'm rises a large number-say IOOO-sheets of blotting paper. The oil to be dehydrated is forced under pressure through the composite blotter. It is usually necessary to pass the oil successively through a plurality of blotter presses. The dehydrating efi'ort is remarkable, the percentage of water being reduced .to (say) nine or ten parts of water to the million parts of oil. The dielectric strength of the dehydrated oil will vary, but may reach 4:7,000 volts, or substantially in excess of the actual requirements. The process is fairly expeditious, the blotter presses being capable of dehydrating about seven hundred gallons of oil in five or six hours. Where av plant contains but, one blotter press, it is usually necessary to run the oil through the same press a plurality of times, which multiplies the time required for complete dehydration.

However, this process is open to serious objections. Certain minute shreds and particles of blotting paper are found suspended in the dehydrated oil. These particles collect on the transformer coils and form a coating retarding the transfer of heat. Moreover, in each operation, a certain percentage of the paper units of the blotter press tear or break, due to their moisture content. These must be frequently replaced at considerable cost for both material and of centrifugal force, has been dehydrated, and confining the water from escape so that the flow of water that forms at the periphery gradually increases in thickness as the process proceeds.

In the ordinary single chamber open bowl centrifugal clarifier, the object is to carry off all the liquid, only the solid or semisolid ingredients remaining in the bowl. These solids, which accumulate on the bowl. wall, are, fromtime to time, removed by interrupting the clarifying process and cleaning them out. a If such a bowl be used for carryin out my process, the water will tend to be orced out to the periphery and remain there, but a very small percentage of the water will flow out with the .oil and the maximum dehydration obtainable will scarcely exceed that represented by a dielectric strength of 35,000 volts, which, in some cases, would be a permissible degree of dehydration, but would not meet the standard exacted in many cases.

The ordinary cream separator of the known multiple disc type is not available 'to secure the degree of dehydration desired.

If a water seal for the heavy liquid discharge outlet is initially established, a large percentage of water is removable, but the desired degree of dehydration is not obtainable, because the oil is not initially distributed around the'periphery so as to insure the subjection of all the oil to the maximum centrifugal force.

While a single chamber open bowl clarifier is permisslbly usable in some cases, I prefer to take a bowl of the type shown in the Snyder Patent No. 1,283,343, October 29, 1918 (which is especially adapted to separate solid ingredients and a heavy liquid from a light liquid) and plug up the outlet for the heavier ingredients. In this separator, provision is made for initially distributing the oil throughout the peripheral part of the bowl, and the cone disc liner divides the bowl into a multiplicity of separating compartments, wherein the .oil is subjected to multiple centrifugations,

insuring the elimination of the maximum percenta e of water before the oil is free to escape. oreover, these discs afford a large area for catching and holding any impurities, such, for. instance, as particles of dirt, or fine shreds of blotting paper that have been carried off by oil that has previously beendehydrated by filtration through a fil-' ter of the blotter press type.

In operation, the oil to be dehydrated is inflown continuously. The water is carried to the periphery of the bowl, from which escape is impossible, owing to the plugging up of the hea liquid discharge. During the first part 0 the process-the fihn of liquid which builds up around the peripheral wall of the bowl is of negligible thickness.

process proceeds; but owing to the very small proportion of water in the oil, a very This thickness gradually increases as the I have also,found it possible to utilize a centrifugal clarifier of the multiple chamber type, for example,-.a clarifier in accordance with the Leitch Patent No. 1,37 0,326, March 1, 1921, which is utilized for the purpose of removing solid impurities from a liquid. Such a bowl affords an inner chamber for the collection of the heavier sediment. From this inner chamber there is an escape near the axis of the bowl for the liquid that has been partially desedimentized. If transformer oil is run into such a bowl, most of the water will remain in the inner compartment, while the dehydrated oil will outflow over the central weir and escape into the outer compartment. In this compartment most of the remainder of the water is eliminated. The degree of dehydration in such a bowl is superior to that obtainable in an ordinary clarifier or in an ordinary cream separator but is not equal tothat obtainable in a bowl of the Snyder type with the outlet for the heavyingredient closedunless it is equipped with a multiple disc liner.

- By the use of my improved process, if carried out in the preferred character of separator, I am able to dehydrate seven hundred gallons of transformer oil in about two and one-half hours or less than half the time required to dehydrate the same quantity of oil by forcing the same through a blotter filter. Careful tests show the dielectric strength of the dehydrated oil to be about 53,000 volts or in excess of that obtainable by the old method. Moreover, my process eliminates all solid particles, including dirt, and also any paper particles that may have been taken up by transformer oil that has been dehydrated by the described filtration process. The power required to operate the certrifuge is less than that requiredto force the oil through a blotter filter. The expense of upkeep is entirely negligible. I

Havin now fully described my invention what I c aim and desire to protect by Letters Patent is:

1. The process of continuouslyand practically completely dehydrating 011, such as a in will be carried to the periphery, confining the water'from escape throughout the process so as to build up as the process proceeds a surrounding film or envelope of water of gradually increasing thickness but of a maximum thickness relatively small in comparison with the thickness of the remaining body of liquid that at any given time during the process is being subjected to said force, and while the oil is inflowing and is being subjected to said force floating off the oil from which the water has been eliminated at a point substantially closer to the axis of rotation than the inner wall of the water when the latter is of maximum thickness.

2. The process of continuously and practically completely separating a mixture of a relatively great proportion of relatively.

light liquid and a relatively small proportion ,of relatively heavy liquid, which comprises subjecting an inflowing mixture of such liquids to centrifugal force whereby the heavier liquid will be carried to the periphery, confining the heavier liquid from escape throughout the process so as to build up as the operation proceeds a surrounding film or envelope of heavier liquid of gradually increasing thickness but of a maximum 28th day of A ril, 1921.

CYRU HOWARD HAPGOOD. -Witnesses:

JOHN S. PAUL, HERBERT D. WILLIAMS.