Rolls-Royce Archives
         « Prev  Box Series  Next »        

From the Rolls-Royce experimental archive: a quarter of a million communications from Rolls-Royce, 1906 to 1960's. Documents from the Sir Henry Royce Memorial Foundation (SHRMF).
Page from 'The ELECTRICAL REVIEW' detailing tests for high-grade insulating oils.

Identifier  ExFiles\Box 32\4\  Scan276
Date  5th October 1928
  
X.1588. File?
OCTOBER 5, 1928. THE ELECTRICAL REVIEW. 553

Insulating Oils.

Details of the Tests necessary to ensure the Essential Requirements of high-grade Insulating Oils.

By ARTHUR GILES.

WITH the growth of high-voltage systems and the increased size of electrical plant, oils for insulation and cooling have become commonplace. In the early days of oil-immersed gear, considerable trouble appears to have been experienced; the attention of oil refiners had been chiefly concerned with the production of lubricating oils, and the problem of producing an oil whose chief merit was electrical insulation was not readily understood. To-day, when oil is used in enormous quantities, the problem is only partially solved. In consequence of the above it behoves all users of oil-immersed gear to have some data regarding the oils they are using, but it seems doubtful if many engineers actually test their oils, either when purchased or at intervals during service. With regard to the purchase of oils, the British Engineering Standards Association has drawn up specifications for the various grades of transformer and switch oils (Specification 148—1927), and buyers of oil would do well to quote the B.E.S.A. Specifications when ordering.

While there are several oil refiners whose products are beyond doubt, the writer is inclined to advocate the testing of all oils on purchase, and at intervals depending on the nature of the service. In the early days of oil-immersed gear, it was common practice to use linseed oil in which a quantity of resin had been dissolved, the mixture being termed resin oil. Resin is now, however, banned on account of its tendency to accelerate sludging. The oils now used are obtained from crude petroleum by fractional distillation, and vary considerably in quality. Naturally the more refined oils are the most expensive, but since reliability is the first consideration in high-voltage transformer work, only the first grade oils should be used, leaving the lower grades for the less exacting equipment such as circuit breakers, in which a certain amount of sludging is permissible. The chief requirements for good insulating oils are: high electrical insulating qualities; freedom from corrosive acids, either animal, vegetable, or mineral; freedom from all materials likely to produce oxidisation and consequent sludging; and high flash point.

The following tests are necessary to ensure these qualities:—

Dielectric Strength Test.—The results of this test are, of course, the primary indication of good or bad oil; the object is to find the voltage required to completely break down a layer of oil between standard test electrodes. Samples of oil should first be subjected to the dielectric strength test, when, if they are satisfactory, they can then be submitted to the more detailed chemical and physical tests. In order to attain uniformity, the B.E.S.A. has suggested a standard test arrangement (B.E.S.A. Specification 148—1927). The test pressure should be applied between two 13-mm. diameter polished brass spheres arranged horizontally with a 4-mm. gap. A suitable cell to contain this arrangement can be left to individual ingenuity and resources, although the B.E.S.A. has approved a design for a standard test cell. The chief requirements are that the cell be insulated, and give at least a 2-in. head of oil above the electrodes. The test should be made at 60-68 deg. F.{Mr Friese} With the above arrangement the maximum stress in the oil would be

Rmax = (V/0.15) x 1.208 volts per inch,

where V=maximum volts between the electrodes, provided that the middle point of the h.p. winding of the test transformer is earthed.

In most cases one terminal of the test transformer would be earthed; then

Rmax = (V/0.15) x 1.253 volts per inch.

Using standard ½-in. (13-mm.) electrodes, good oil would require a minimum of 22 kV (r.m.s.) for complete breakdown. The test is made with a gradually increased voltage until an arc-over occurs. If the test is observed closely, it will be seen that, preliminary to the formation of an arc, there is a slight intermittent sparking, accompanied by the formation of fine hair-like threads of carbon between the electrodes. This does not, however, appear to influence the dielectric strength to any marked extent. The test voltage is raised until a complete arc-over occurs, which is identified by its characteristic sound, apart from the visual effects. It is important that the preliminary sparking be ignored, since the establishment of an arc is the only reliable indication of dielectric strength. Insulation resistance tests are not a safe guide to dielectric strength, oil having the descending temperature coefficient characteristic of an electrolyte.

Tests for Moisture.—Water in insulating oils naturally spells disaster. It has been estimated that .03 per cent. of moisture will lower the dielectric strength by as much as 25 per cent. To actually detect moisture is fairly easy, and the well-known “poker” or “needle” test is entirely reliable. A knitting needle or piece of wire is heated to redness, and just as the glow dies out it is plunged into a small sample of oil. If any moisture is present it will be indicated by a series of pronounced cracking sounds, while if the oil is dry there is simply a hiss and a puff of smoke. This test appears rough and ready, but is surprisingly effective; the writer has succeeded in readily detecting one part in 50,000, or 0.002 per cent. content. Moisture may also be detected by boiling a sample of oil, when the oil will froth and steam come off, which may be seen by holding a cold piece of glass over the boiling oil. Trouble with moisture is unlikely in newly purchased oil, but may sometimes be suspected in used oil, or in oil which has been in stock for some time. It is important when taking test samples to draw always from the bottom of the transformer and storage tanks. Oil should always be stored in steel drums and never in wooden casks, in order that the moisture risks may be minimised.

Detection of Sulphur.—Sulphur is objectionable since it attacks copper windings. It may be detected as follows: About 25 c.c. of oil, in which is placed a piece of polished copper 1 in. square, is heated at about 100 deg. C. for 12 hours. If a dangerous proportion of sulphur is present, the copper will, on examination, be found to be discoloured with black copper sulphide.

Detection of Inorganic Acids.—Insulating oils should contain no inorganic acid. This may be detected by shaking a small quantity of oil, say, 50 c.c., with an equal volume of tepid distilled water to which has been added a few drops of methyl orange indicator. The water layer, after settling, will appear red in the presence of acid. In another test which has been recommended a sample of oil is shaken with roughly twice its volume of warm distilled water. The mixture is allowed to separate, and the oil is poured off and again shaken with twice its volume of warm distilled water and separated. The water solutions are then mixed, and the mixture can be tested for acidity with any good indicator, e.g., tincture of cochineal: a few drops when added give a red indication in neutral solutions, yellow in acid and purple-violet in alkali.
  
  


Copyright Sustain 2025, All Rights Reserved.    whatever is rightly done, however humble, is noble
An unhandled error has occurred. Reload 🗙