48. Test No. 6: Effect of Variation of Power Factor-Both Elements in Circuit.-The effect of variation of power factor upon the registration of the meter carrying constant load shall not exceed that specified below. Each meter shall be tested on a balanced polyphase circuit.

49. Test No. 7: Effect of Variation of Frequency.-The effect of variation of frequency upon the registration of a meter carrying constant load shall not exceed that specified below. shall be tested with a balanced polyphase load.

Each meter

50. Test No. 8: Effect of External Magnetic Fields.-The change produced in the registration of a meter with a balanced polyphase load at 10 per cent of rated current by the application of an external magnetic field, in the manner stated in section 38, shall not exceed 2.5 per cent.

One meter shall be subjected to an alternating magnetic field, as is specified in section 38, first with the current in the external conductor in phase with the voltage applied to one element of the meter, and then with this current in phase with the voltage applied to the other element of the meter.

51. Test No. 9: Effect of Variation of Temperature. This test shall be applied to not less than three meters. The temperature coefficient shall not exceed 0.15 per cent per degree centigrade at either 10 per cent or 100 per cent of rated current. The per cent difference in the rate of registration of the meter at the higher temperature from that at the lower temperature, divided by the difference in temperature, is the temperature coefficient.

The meter shall be placed in a space having a temperature of approximately 20° C or room temperature, as specified in section 17 (a), and allowed to stand for not less than two hours with the voltage circuit of the meter energized. A load of 10 per cent of the rated current shall then be applied, and after running for one hour the meter shall be tested at this load. A load of 100 per cent of rated current shall then be applied for one hour and the meter again tested at this load. These operations shall be repeated in the same order at a temperature approximately 20° C higher. 52. Test No. 10: Effect of Temporary Overloads.-The maximum percentage deviation in registration of a meter due to temporary overload shall not exceed 1 per cent at 10 per cent of rated load (item 1) and 100 per cent of rated load (item 2).

A temporary load of ten times the rated current shall be applied three times to each meter under test, the duration of each overload being approximately two seconds. The overload current shall be approximately in phase with the voltage applied to the voltage circuit of the meter.

Each meter shall be tested with 10 per cent of rated current and 100 per cent of rated current before the application of the overload, and again afterwards.

53. Test No. 11: Voltage Drop or Watts Loss in Current Coils.— The voltage drop or watts loss in the current coils of each meter shall be measured at 100 per cent of rated current.

In meters of 25 amperes or less capacity, the loss in the current coils of each element shall not exceed 5 watts.

In meters of more than 25 amperes capacity, the drop in the current coils of each element shall not exceed 0.15 volt.

4. AMPEREHOUR METERS

At the present time wide differences of opinion exist with reference to the accuracy and suitability for general service of amperehour meters.

The meter committee of the National Electric Light Association in its report for 191237 calls attention to the desirability of having a cheap meter.

It would seem that a meter cheaper than the ones now in use, especially for continuous currents, is an absolute necessity. Apparently the hope of such a continuous current meter must be based on an amperehour meter rather than a watthour meter. The objection to the use of an amperehour meter rests entirely on the fact that it does not take into account the voltage variations of the circuit and consequently the difference between the voltage for which the meter is calibrated and the service voltage might cause an error additional to any error arising in the meter itself. It is, consequently, necessary to know not only the accuracy of the meter but the average voltage of the circuit on which it is used, to determine the accuracy of the registration.

While the voltage regulation in some companies is doubtless so good as to permit of the use of an amperehour meter without its accuracy of registration falling outside the limit now prescribed in States where this matter is regulated, in many companies there are likely to be parts of the system where the voltage at times varies above or below the usual limits prescribed for meter accuracy.

On the other hand it has been urged that lack of familiarity with amperehour meters is the cause of much objection. Large numbers of amperehour meters are used in England,38 France, and Germany, and give satisfactory service. Again, it is said that direct-current circuits are very well regulated as a rule so that the accuracy of amperehour meters on such circuits may be as good as that of watthour meters.

The New York Commission, second district, issued an order on July 3, 1913, as follows: 39

Whereas, commutator type meters, when used on alternating-current circuits, are inaccurate and subject to change in degree of accuracy, and amperehour meters do not record energy on commercial electric circuits, except by uncertain deviation, owing to variations in voltage; it is therefore ordered:

1. That on and after July 1, 1915, no corporation, person or municipality within the jurisdiction of this commission shall furnish, set or keep in use any commutator type, watthour meter for the purpose of measuring electric energy on alternatingcurrent circuits, as a basis for computing bills.

National Electric Light Association, thirty-fifth convention, vol. 3, p. 25.

28 For house service the watthour meter is not popular in England, the amperehour meter being in much greater demand. (Brooks, Electrical Instruments and Meters in Europe, p. 8.)

29 1913 report, p. 86, vol. 1, New York Public Service Commission, second district.

2. That on and after July 1, 1915, no corporation, person or municipality within the jurisdiction of this commission shall furnish, set, or keep in use any type of electric meter, the readings of which are used as a basis for computing bills, which shall be of such design and construction that its action and registration depend wholly upon the current in the circuit and not upon the energy (amperehour meters).

3. That every person, corporation, or municipality having in service on July 15, 1913, any meters of the types herein prohibited, shall file statements with the commission as follows: Each statement shall show:

(a) Number of commutator type meters owned by respondent which are in use on alternating-current circuits for the purpose of computing bills;

(b) Number of "amperehour" meters owned by respondent which are in use for the of computing bills.

purpose

Provided that if any statement shall show that respondent has no meters of the types herein prohibited in service on the date for which such statement is given, such respondent shall not be required to file any subsequent statements.

On March 25, 1914, the commission issued an order as follows:

By order dated July 3, 1913, this commission prohibited the use of certain types of electric meters in the second public-service district of the State of New York on and after July 1, 1915. Some question has been raised as to the exact interpretation which should be made of such order.

Ordered: 1. That the prohibition of meters, the registration of which depends upon the current in the circuit and not upon the energy, shall apply to integrating (amperehour) meters, and not apply to meters used to determine the maximum current or power used on the circuit.

Ordered: 2. That the disapproval of amperehour meters shall not be understood as indicating that this commission would not approve the use of a thoroughly satisfactory amperehour meter, under conditions such that it could be relied upon to give a reasonably accurate record of the energy consumed.

The Wisconsin Railroad Commission has made special interpretation of its rules in the case of electrolytic amperehour meters, 40 and has held in part as follows in its opinion dated November 23, 1911:

In order to comply with rules 14 and 15 it is necessary to place the oil film over the surface of the electrolyte and keep the cover and stopper on the meter whenever the meter is in service.

In order that the error in registration due to voltage variations may be determined, and also that there may be full compliance with rule 23, voltage records shall be carefully taken in each locality at least once every three months, or whenever changes are made, and proper correction shall be applied to all bills rendered.

A station log shall be kept giving the readings of all indicating instruments on the station's switchboard at least once every hour, and oftener during the hours when the lighting load is changing rapidly.

The correction constant for voltage shall appear upon the consumer's bill together with the meter readings.

40 Eighth report, W. R. C., p. 270.

It seems obvious that meter practice in so far as amperehour meters are concerned, is in a formative state, tending, however, toward the increased use of cheap amperehour meters, particularly for small consumers.

5. DEMAND METERS

Devices for indicating and recording the "maximum demand” of consumers' installations are still in an experimental stage of development, and there is a wide variation in practice and opinion with reference to them. Such devices are classified by the meter committee of the National Electric Light Association as follows:

A maximum-demand device is one which indicates or leaves a record of the maximum load which has existed in the circuit, limits the supply, or differentiates the consumption with reference to a fixed value of demand or period of time.

Maximum-demand devices are not yet in wide general use, though there is a widespread interest in such devices. The Board of Gas and Electric Light Commissioners of Massachusetts has investigated the operation of certain types of demand meters to "determine whether or not it is expedient to regulate or prohibit the use of such meters." 11

The use of demand meters, as pointed out in that report, is directly and intimately associated with differential rates for electric service. The question of rates is not discussed in this Circular, and only the accuracy and reliability of the devices used are of interest in this connection, as is also the case in the report of the Massachusetts commission above cited. It is found by that commission that if "properly constructed and calibrated, and properly handled and inspected by the companies' employees while in use upon customers' premises, they may be expected to operate correctly within certain reasonable limits." The board, however, further says:

The board is equally convinced that it is not practicable to test and seal these appliances in the manner long pursued with respect to gas meters, or to apply the methods employed in the periodic adjustment, examination, testing, and sealing of balances, weights, and measures.

The recommendation of the commission, that the provisions of sections 36, 37, and 38 of chapter 141 of the revised laws of Massa

41 Twenty-eighth annual report, Board of Gas and Electric Light Commissioners of Massachusetts, 1912, p. 322a and following.