GB1578286A - Anaerobic curing insulation resin system - Google Patents

Abstract

A resin which does not cure in the presence of oxygen is applied to an object in the presence of oxygen. The object is then brought into a gaseous atmosphere which does not contain oxygen. In this manner, resins can be hardened at room temperature. No problems occur during this, such as the resin running off during hardening and the formation of mechanical stresses in the hardened resin. The object has, for example, a conductor (1) covered with an insulation (2, 3). The insulation (2, 3) is impregnated with a hardened resin (4). <IMAGE>

Description

(54) ANAEROBIC CURING INSULATION RESIN SYSTEM (71) We, WESTINGHOUSE ELECTRIC CORPORATION, of Westinghouse Building, Gateway Center, Pittsburgh, Pennsylvania, United States of America, a company organised and existing under the laws of the Commonwealth of Pennsylvania, United States of America, do hereby declare the invention, for which we pray that a patent may be granted to us, and the method by which it is to be performed, to be particularly described in and by the following statement:- The is invention relates to the formation of a cured, resinous coating on an article.

Conductors for use as coils in generators and motors are insulated using the VPI (vacuum-pressure impregnation) process by wrapping with mica tape, then with a glass binding tape to hold the brittle mica tape on. The wrapped conductor is placed in a vacuum then in a resin under pressure. It is removed and heated in an oven to cure the resin.

While this system is used commercially, it requires a great deal of energy for the baking ovens because the copper in the coils must be heated along with the resin. Run-off of the resin during cure can occur, and mechanical stresses in the resin can be induced by expansion of the copper during cure and its subsequent contraction during cooling.

Anaerobic resins are resins which will not cure in the presence of oxygen, but will cure when placed between two oxygen-impervious surfaces, and therefore they are commonly used as adhesives.

According to the present invention a method of forming a cured, resinous coating on an article comprises applying an anaerobic resin to the article in the presence of oxygen, and placing said article in a gaseous atmosphere which does not contain oxygen.

The invention also includes an article comprising a conductor covered with insulation which is impregnated with a cured anaerobic resin.

It has been found that insulated conductors impregnated with anaerobic resins will not cure in a vacuum, which is an advantage in the VPI process, but can be cured by contact with a gaseous atmosphere which does not contain oxygen. This means that the resins can be cured at room temperature, thus avoiding the problems previously encountered with heatcured resins. Electrical properties are comparable to the heatcured resins now in use, except at high voltages. Unlike electron beam and UV cured resins, anaerobic resins cured under a gas cure in depth, down to 6 inches or more.

In order that the invention can be more clearly understood, convenient embodiments thereof will now be described, by way of example, with reference to the accompanying drawings in which: Figure 1 is an isomeric view in section of an insulated conductor impregnated with an anaerobic resin.

Figure 2 is a diagram of an apparatus for continuously coating a wire using the curing system of the invention.

In Figure 1, a conductor 1 is covered with several layers of mica tape insulation 2 and one layer of woven organic tape insulation 3, which holds the mica insulation in place. An anaerobic resin 4 impregnates the insulation and forms an outer coating.

In Figure 2, a conductor 5 passes from capstan 6 into bath 7 of an anaerobic resin 8. The wire then passes over sheave 9 into closed tank 10. An inert gas flows into the tank from conduit 11. As the wire passes over sheaves 12 in the tank it is cured, then leaves the tank and is wound on capstan 13.

An anaerobic resin is a resin which will not cure in the presence of oxygen, but will cure at room temperature when placed between oxygen-impervious surfaces.

Solventless anaerobic resins are required for the VPI process. Most anaerobic resins are acrylics, such as diacrylates, which polymerize by addition through a double bond. Commonly used acrylic anaerobic resins include tetraethylene glycol dimethacrylate and tetraethylene glycol diacrylate. An acrylic anaerobic resin may contain a reactive comonomer such as ethyl methacrylate, styrene, or 2-ethylhexyl acrylate. An organic peroxide free radical initiator such as cumene hydroperoxide or t-butyl perbenzoate is often used to initiate cure. An accelerator, usually a tertiary amine such as N,N-dimethyl-p-toluidine, and a coaccelerator, usually an organic sulfimide such as benzoic sulfimide, may be present to reduce curing time.

The free radical initiator can be stabilized with a free radical stabilizer such as hydroquinone.

The conductor is preferably copper because it is most widely used for electrical insulation and it is known to accelerate the cure of anaerobic resins, though other metals can also be used.

The primary insulation is preferably mica, especially for high voltages, as it has excellent electrical properties. Glass, asbestos, Nomex (Registered Trade Mark) (a polyamide believed to be made from meta phenylene diamine and isophthaloyl chloride, sold by Dupont), and other types of insulation could also be used, either alone, in mixtures, or in mixtures with mica. Mica insulation is usually made with a polyester backing to hold the mica together. The insulation may be a tape which is wrapped around the conductor, the amount of insulation depending upon the voltage drop across the insulation. Mica insulation is preferably impregnated with 3 to 30% (preferably 5 to 12%, by weight based on the mica weight), of a polymer which is coreactive with the anaerobic resin, in order to ensure a better bond. Polyesters, acrylics, polybutadienes, or other unsaturated monomers may be used as coreactive resins.

The VPI process is the preferred method of insulating a conductor because it leaves very few air gaps in the insulation, but other methods may also be used. The wrapped conductor is placed in a tank which is then evacuated. The anaerobic resin is admitted under pressure, usually at least 46 psi, although 90 to 100 psi is preferred. The resin should saturate the insulation. Typically, the insulation will contain about 5 to 35% (by weight based on the insulation weight) of the anaerobic resin, although 20 to 30% is preferred. The resin is permitted to drain from the wrapped conductor and is cured by contact with a gas which does not contain any significant amount of oxygen. This may be accomplished in the same tank or the wrapped conductor may be cured in a separate tank. Nitrogen, carbon dioxide, or mixtures of these two gases is preferred as it is inexpensive, safe, and easy to handle, but other inert gases (other than oxygen) may also be used. It has been found that if nitrogen is used to cure an acrylic resin the rates of cure are optimum at a nitrogen flow rate of 6 to 201pm (liters per minute).

The invention will now be illustrated by the following Examples: Example I.

The following table gives various anaerobic resin compositions which were prepared and tested for gel time and storage stability.

Difunctional Catalyst, Accelerator, Coaccelerator, Gel Time Storage Acrylic Vinyl Cumene N,N-Dimethyl- Benzoic Inhibitor, Viscosity in N2 Stability in air Resin Monomer Monomer Hydroperoxide p-Toluidine Sulfimide Hydroquinone at 25 C at 25 C at 25 C Number (75 parts) (25 parts) (parts) (parts) (parts) (parts) (cps) (hours) (days) 1 Tetraethylene Styrene 2.0 0.4 0.30 0.04 3.0 8-16 > 100 Glycol Dimethacrylate 2 Same Styrene 2.0 0.4 1.00 0.04 3.0 5-6 48 3 Same Styrene 2.0 0.4 3.00 0.04 3.0 3-4 > 100 4 Same Styrene 2.0 1.0 0.30 0.04 3.0 8-16 56 5 Tetraethylene Styrene 2.0 0.4 0.30 0.04 3.0 2-3 15 Glycol Diacrylate 6 Same 2-Ethylhexyl 2.0 0.4 0.30 0.04 0.5 3-4 > 100 Acrylate 7 Same 0.4 0.30 0.04 0.5 4-5 > 100 8 Same Ethyl 2.0 0.4 0.30 0.04 0.5 < 1.0 > 100 Methacrylate 9 Same Styrene 2.0 0.4 0.30 0.08 3.0 4-5 > 100 10 Same Ethyl 2.0 0.4 0.30 0.08 0.5 < 1.0 100 Methacrylate 11 Same Ethyl 2.0 0.4 0.30 0.16 0.5 1-2 > 100 Methacrylate Samples of the above resins were limited to about 50 gms due to the high exotherm produced during cure. The storage stability tests, which determine storage time until visual gelling occurs, were carried out on samples of about 50 g of resin stored in 4 oz. glass jars or polyethylene containers. Because of the uncertain effects of the fluorescent lighting in the laboratory on the resin stabilities, the samples were stored in the dark. Periodically, gel times were run (under N2 flow) on the samples to check for retention of cure reactivity during storage.

The gel time measurements (i.e., time required for visual gelling to occur) were made on lOg samples in 2 inch diameter aluminium dishes. It was found that the ordinary laboratory vacuum was insufficient to gel the samples within a reasonable time. However, rapid gelation was achieved by placing the samples in a desiccator and passing nitrogen through the desiccator. The following table gives the results of testing resin number 10.

Cure Condition Gel Time at 250C Air 48 hours N2 stream (13 Ipm) 30 to 60 minutes Vacuum 24 hours At the end of 24 hours, a very small amount of gel (about 7% of the resin) was observed at the bottom of the aluminum dish of the sample placed in the vacuum.

The following table gives the results of similar test performed on resin number 9.

Cure Condition Gel Time at 250C CO2 (13 Ipm) 20 minutes N2 (50 psi static) 7 to 8 hours Air(l3lpm) > 2 days O2(l3lpm) > 2 days Pressurized CO2 at 50 psi also produced a rapid cure, Static nitrogen or carbon dioxide under pressure results in a slower cure than does a flow of nitrogen or carbon dioxide at atmospheric pressure, but requires smaller quantities of gas.

Using a similar procedure, the effect of the nitrogen flow rate on resin number 9 was determined. The following table gives the results.

Flow Rate (Ipm) Gel Time (hours) 1.62 7.5 3.25 4.0 6.50 2.0 13.0 1.5 20.0 3.0 26.0 4.0 The above table shows that a flow rate of about 6 to about 20 Ipm is critical to obtaining a rapid cure at pressure of about one atmosphere.

Example II.

Power factor data were obtained on mica composites impregnated with the most promising anaerobic resins, resin numbers 8 and 9. Two types of composite were prepared, one (sample A) was made by brushing an anaerobic resin on a piece of "raw" mica paper ("Cogemica" sold by Cogebi Co.), (4 in. x 4 in.) about 20 mils thick, and the other type of sample (sample B) was fabricated by brushing the anaerobics over a polyester-impregnated mica tape wrapped 6 layers thick (i.e.; 3 wrappings half-lapped) on copper tubes (8 in. long, 0.5 in. od). Because of the very fluid nature ( < 3.0 cps) of these anaerobic resins it was observed that rapid and thorough penetration of these mica products occurred.

The samples were gelled under a stream of N2 (13.0liters-min.) at room temperature and it was noted that the copper tube samples exihibited rapid gel under these conditions (i.e. < 30 min.). To establish the extent of cure of the anaerobic resins in these composites and the necessity (or not) for a heat post-treatment of these materials. it was decided to bake one set of these samples for 4 hours at 135"C after the initial room temperature cure. It was felt that a comparison of the electrical data obtained from these two sets of samples would determine whether or not an additional heat treatment is necessary to obtain full cure with these.anaerobic materials.

Power factor measurements were recorded at 250C and 150"C. The copper tube samples were also measured at 1, 1.5, and 2 kV at both temperature to evaluate the effect of voltage on the power factors of these mica composites.

Power factor data obtained with copper tubes wrapped with the Sample B mica tape and impregnated with two of the anaerobic resins are shown in the following table.

% % Power Factor (100 x tank Heat Cure? Applied Voltage Temperature (4 hours Resin (OC) at 1350C) 1.0 1.5 2.0 25 No 2.10 2.16 2.71 Yes 2.3 2.32 2.59 9 150 No 31.4 32.4 35.8 Yes 28.3 28.9 31.9 25 No 2.32 2.32 2.5 Yes 2.48 2.49 2.52 8 150 No 27.9 28.9 31.1 Yes 34.2 35.7 37.7 The results for resin number 9 at 1500C show that the sample, which did not have any heat treatment after the initial cure, has slightly higher power factor values than the sample which had heat treatment. However, both samples showed similar power factor values at 250 C.

For resin number 8, the sample with the additional heat treatment exhibits somewhat higher power factor values at 1500C than the sample that did not have the additional cure. The room temperature values for both samples, however, remain the same. It is difficult, on the basis of these data, to say whether an additional heat postcure is necessary for these anaerobic materials to achieve full cure. Nevertheless, it may be significant that all four samples showed very similar room temperature power factor values.

Although no attempt was made to identify the optimum mica tape for these anaerobic impregnants, the power factor values in these composites using the Sample B mica tape are not considered to be excessively high. Although the values (i.e. 32-37% at 1500 and 2kV) would be unacceptably high for high voltage applications, they would be suitable for low voltage equipment insulation (e.g., < 13.8kV).

Dielectric strength measurements were carried out on mica composites made with Sample A impregnated with resin number 8. The samples were tested under an aliphatic hydrocarbon transformer oil sold by Westinghouse Electric Corporation under the trademark "WEMCO C' at room temperature using a voltage rise of l kV/sec. Power factor and dielectric constant data obtained with the Sample A composites using resin number 10 are summarized in the following table.

Test Power Dielectric Temperature Cure Capacitance Factor % Constant (OC) Conditions (pF) (100 x tans) (E') 16 hrs at 250C 36.3 0.89 3.4 under N2 25 Same + 4 hrs at 58.4 1.6 5.2 1350C in air 16 hrs at 250C 138.6 4.8 12.9 under N2 150 Same + 4 hrs at 94.2 23.0 8.4 1350C in air Once again there appears to be an interesting difference between the sample which had the additional heat treatment and the one that did not. The power factor of the room temperature cured sample, measured at 1500C was much lower than that of the other one. The reasons for these differences are not very clear at this time; however, both samples would be considered as having acceptably low power factor and dielectric constant values for insulation up to a voltage of about 13.8 kV.

The following table shows the dielectric strength for the two samples used in the previous table. The measurements were made at an N2 flow rate of 13.01pm under "WEMCO C" oil at 250C using a voltage rise of I kV/sec. Both composite samples were used in the previous power factor measurements and probably experienced I hour at l500C before being tested for dielectric strength.

Composite Breakdown Dielectric Thickness Voltage Strength Cure Condition (mils) (kV; rms) (voltstmil) Overnight (16 hours) 21 16.0 765 at 250C in N2 stream Same + 4 hours at 20 16.5 830 1350C in air Same + 4 hours at 20 18.5 935 1350C in air The sample which had the additional elevated temperature cure would appear to have somewhat high dielectric strength than the room temperature cured sample. However, both composites would seem to have high dielectric strengths than those of epoxy resin impregnated mica composites of similar thicknesses.

Typically, the epoxymica composites show values of 400-600 volts/mil compared with 700-900 volts/mil for the anaerobic resin samples.

Example III.

To illustrate the importance of curing these resins under a stream of N2, the following experiment was carried out. Four copper tubes were wrapped with Sample B mica tape as described previously. One set (2) of these samples was impregnated with resin number 5 and the other with resin number 8. One of each set was allowed to cure in an N2 stream (flow rate 13 Ipm) for 32 hours, then both sets were left in air for two weeks. The cure conditions and resin retention for each of these samples is shown in the following table.

Copper Wrapped Tube Copper Tube Resin % Resin Sample Resin Weight Weight Cure Weight After Retention Retention No. Number (g) (g) Conditions Impregnation (g) On Mica B1 5 58.97 69.38 Air 70.25 0.87 8.35 B2 5 56.56 67.13 N2 68.18 1.05 10.00 B3 8 57.79 67.93 N2 69.06 1.13 11.15 B4 8 56.90 67.23 Air 68.36 1.13 10.92 The resins had been stored two months at room temperature in polyethylene bottles prior to being used. The resin "run-off" from the samples cured under N2 was found to be negligible.

Power factor measurements at 1 kV were then carried out on these four samples at 25 C. The results are shown in the following table.

Power Factor (100 x tan #) at 60 Hz Sample At 150 C after At 25 C after No. At 25 C 3 days at 150 C 3 days at 150 C B1 25.6 30.2 1.4 B2 3.5 23.1 1.2 B3 4.5 25.8 1.8 B4 32.8 23.9 1.8 Samples B2 and B3, which had been allowed to cure in the N2 stream, showed much lower power factor values than the ones which had been left in air. This was not too surprising since the latter two samples were still very tacky, even after two weeks. The others were firm and tack-free after a few hours in N2.

To illustrate this lack of cure in Samples B1 and B4, all four specimens were postcured 3 days at 150 C and the power factors remeasured at 25 C and also at 150 C. These data are also shown in the above table, At room temperature all four samples gave similar values (between 1.2% and 1.8%) and also at 150 C (23.1 to 30.2%).

Thus, the indication is that the samples exposed to the N2 stream were more fully cured than the ones left in air.

For a typical Class F motor, an additional heat treatment of the materials would not appear to be necessary since full cure could be obtained after the coils have been wound into a stator because the operating temperature of the motor would be high enough to bring about a full cure.

Also, the winding of partially cured coils (assuming resin is non-tacky) might be advantageous since the mica insulation would still retain an appreciable degree of flexibility. Conventionally made coils are sometimes stiff and difficult to wind without cracking the insulation.

Claims (22)

WHAT WE CLAIM IS:

1. A method of forming a cured, resinous coating on an article comprising applying an anaerobic resin to the article in the presence of oxygen, and placing said article in a gaseous atmosphere which does not contain oxygen.

2. A method according to claim 2, wherein the gaseous atmosphere is of nitrogen, carbon dioxide or a mixture thereof.

3. A method according to claim 1 or 2, wherein the anaerobic resin is an acrylic resin.

4. A method according to claim 1, 2 or 3, wherein the article is a conductor covered with insulation.

5. A method according to claim 4, wherein the insulation is mica and the conductor is copper.

6. A method according to claim 4 or 5, wherein the insulated conductor is placed in a vacuum and then the insulation saturated with the anaerobic resin under pressure.

7. A method according to claim 6, wherein the pressure is from 45 to 100 psi.

8. A method according to claim 1, 2 or 3, wherein the article is a wire which is continuously passed through the anaerobic resin and thence through the gaseous atmosphere which does not contain oxygen.

9. A method according to any of claims 1 to 7, wherein the article is a motor or generator coil.

10. A method according to any of claims 1 to 9, wherein the anaerobic resin is solventless.

11. A method according to any of claims 1 to 10, wherein the anaerobic resin contains an accelerator.

12. A method according to claim 11, wherein the accelerator is benzoid sulfimide.

13. An article comprising a conductor covered with insulation which is impregnated with a cured anaerobic resin.

14. An article according to claim 13, wherein the insulation is at least one of mica, glass, asbestos and organic resins.

15. An article according to claim 14, wherein micacontaining insulation includes from 3 to 30% (by weight based on said insulation weight) of an organic resin co-reactive with the anaerobic resin.

16. An article according to claim 13, 14 or 15, wherein the anaerobic resin is an acrylic resin.

17. An article according to claim 16, wherein the acrylic resin contains copolymerized styrene.

18. An article according to any of claims 13 to 17, wherein the amount of anaerobic resin is from 5 to 35% (by weight based on the weight of said insulation).

19. An article according to any of claims 13 to 18, wherein the conductor is copper.

20. A method of forming a cured, resinous coating on an article as claimed in claim 1 and substantially as described herein with particular reference to the foregoing Examples.

21. Coated articles when made by a method according to any of claims 1 to 12 and 20.

22. Articles as claimed in claim 13 and substantially as described herein with particular reference to the foregoing Examples.