Temperature limits in large machines

The prevent Institute standards allow a maximum limit of 105 (leg. for fibrous insulation and 125 deg. for mica insulation with an allowance of 150 deg. subject to special guarantees of the manufacturer. In high-voltage machines of large size, the effect of actual copper temperatures higher than 105 deg. may cause softening and disappearance of binding materials, bulging of insulation, and consequent powdering of insulation under periodic pounding of copper on softened material, and, ionization at voids so created. Operating experience of four large size machines demonstrated the above effects to repeatedly take place in machines operating at copper temperatures of 150 deg. cent. None of these effects were noted in machines operating at copper temperatures of 105 deg. cent. or under. Two machines have safely operated for over three years at maximum copper temperatures of 130 deg. cent. Incorrect conclusions may be made as regards the safe temperature limits by judging the performance of machines unless actual copper temperatures are known. Lower ambient temperatures and fractional loads may reduce the operating temperatures 25 to 35 deg. below the assumed limits. Machines designed for high temperatures are less efficient than machines designed for cool temperatures, in one instance the difference being as great as several hundred kilowatts at all loads. The calculation of ventilation of large machines is relatively uncertain; it is of importance to aim at a conservative limit rather than set it too near the danger point. From the standpoint of economy as well as greater safety, it appears that large machines should not be operated at higher copper temperatures than 105 deg. cent. This means that with outside air ventilation seldom exceeding 20 deg., the maximum standard limit with the standard reference of 40 deg. ambient should be 125 deg. equivalent to 85 deg. maximum rise at the copper. In all cases where the room air is close to 40 deg. the maximum copper rise should be limited to 65 deg. Other correlated and important features discussed incidentally in the paper are the typical proportionality of life at different temperatures for fibrous insulation, and the new tentative conventional allowance for reducing to maximum copper temperature, readings taken outside the insulation.