Induction kWh meter theory

Hi all,
I'm wondering if I can draw on the knowledge of some of the electrical engineering experts here. This coming week I'm going to be talking to a
group of high school students about engineering. They will mostly be 17 year olds who study physics and mathematics and have a pretty good understanding of science. People who might become future scientists or engineers, or who might be drawn into IT, management consultancy or whatever. My job is to offer them a brief insight into engineering and to take them through a few questions of the kind that might be asked by university interviewers.
Usually I take along a few small machines for us examine and discuss. Last time I took a box of small electric motors in various states (some complete and working, some disassembled) and this worked well. This time I was thinking of taking a few domestic induction meters. I know that energy measurement is on their syllabus, but that the exact theory behind the meters is a little above it.
I have hunted for the oldest meter I can find, which offers a good view of the magnets and coils. As I understand it the conductive disc, which is connected to the counter via a gear train, is acted upon by three magnetic fields, all of which act perpendicular to the disc. One field is produced by a coil wound with a few turns of thick wire. This coil carries the line current and shows little inductance, so the magnetic field it produces is in phase with the line current. Another field is produced by a coil wound with many turns of thin wire. This is connected between the live and neutral of the supply, and is highly inductive, so that the magnetic field it produces is almost in quadrature with the supply voltage. I think I'm right in saying that for the meter to work, this field must be at least as strong as the field produced by the maximum allowable line current through the first coil - perhaps someone can confirm this for me? A third field is produced by strong permanent magnets. This produces a retarding torque which is proportional to the rotational speed of the disc. When the load is purely resistive, the two alternating magnetic fields are in quadrature, and the disc experiences a moving magnetic field which drags it around. When the load is purely inductive, the two fields are in phase, and the disc experiences a pulsating magnetic field which does not drag it around. Of course the usual situation is for the load to be partially resistive and partially inductive, and in this case the meter only registers the consumption of real power.
So that's my understanding of how an induction meter works. Do correct me if I've got anything wrong. Now (at last) on to my questions. The oldest meter I have has an arrangement of coils like this (the more modern meters have encapsulated coils which are small and hard to see):
|----------------| -----| |----- | | High L | | | -| |- | | | |----------------| | | | | | | | -------------------- | | Core | | ------- ------- | | | | | | | | | ---------- | | | | | | | | | | | Low | | | | | | L | | | | | | | | | | | ---------- | | | | | | | | --- ---- --- ---------------------------------- <- Disc ---------------------------- | Core | ----------------------------
Here are some pictures of the meter:
http://www.mythic-beasts.com/~cdt22/elec_meter1.jpg
http://www.mythic-beasts.com/~cdt22/elec_meter2.jpg
http://www.mythic-beasts.com/~cdt22/elec_meter3.jpg
So here are a few questions:
1. Why does this meter use the arrangement of coils and core shown above? Someone is bound to ask me.
2. Does anyone know of any eloquent, succinct explanations of induction meter theory online which I can read over?
3. Does anyone know who invented the induction meter? I was under the impression that it was Elihu Thomson, but I'm not sure of this.
I'm not firmly decided on using the induction meter as an example for discussion, but these are bright students and the meters can be found in almost every home in England, so it seems like a good topic.
I'd be interested to hear your thoughts.
Best wishes,
Chris
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
Christopher Tidy wrote:

Dunno, but I suspect that the section of core between the 'high L' and the rest is to shunt the magnetic field for even _higher_ L, and to help match the high L field strength at the disk with the low L field strength. The pole pieces would be separated to help give a traveling magnetic field.

Dunno, but if you find one before someone posts it, could you post what you find?

You're missing out on an important point: The torque developed by the high-L, low-L windings will be proportional to real power used -- so it'll be slow when real power consumption is low, high when real power consumption is high, etc.
--

Tim Wescott
Wescott Design Services
  Click to see the full signature.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
Tim Wescott wrote:

Of course. Thanks Tim. It slipped my mind while I was typing that bit about the magnetic field produced by the high-L coil being at least as strong as the maximum magnetic field produced by the low-L coil.
Best wishes,
Chris
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
Hey Chris,
Can't help with the on-line explanation, but in my former occupation it was necessary to have some means of determining/guaranteeing that three phase power was being supplied in the correct phase relationship. If it was "backwards", the lift would go up when it should go down, and vice versa. Bit more to it than that of course, but prior to the contemporary solid state devices used now, the device Otis Elevator used world-wide for many years from sometime in the 1920's until well into the 1970's (and still functioning on many installations today) is referred to as a "Reverse Phase Relay", and commonly called "The J switch". It actually is just a very similar device to what you have described the meter, being a quadrature of four coils perpendicular to a 1/16 thick vertical copper disc about 5" in diameter (from memory) mounted on a horizontal needle axle . Application of power would cause the disc to rotate in one direction or the other, and the axle had a switch (the J switch) mounted to it which would shunt a contact pair allowing motor start if the phases were correct, and the disc would stall and remain stalled by the contacts until the power was removed. Phase reversal would cause it to rotate away form the contacts ands stall against a stop provided. Never made more than maybe a 20 degree rotation motion in use.
Wouldn't surprise me if you could contact any Otis branch office and ask if they have an old one kicking around that you might have or borrow fore the week.
We also used "stalling torque motors" to lift cams, but they were very ordinary in construction.
Take care.
Brian Lawson, Bothwell, Ontario. XXXXXXXXXXXXXXXXXXXXXXXXX
On Sun, 09 Jul 2006 22:43:08 +0000, Christopher Tidy

Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
On Sun, 09 Jul 2006 22:43:08 +0000, Christopher Tidy

From: www.radianresearch.com/PDF/Bul_102.PDF
BULLETIN 102 Copyright © UTEC – 1999. All rights reserved. 1 043099 Document Revision 1.0 INTRODUCTION TO WATTHOUR METER TESTING The information presented in this bulletin has been compiled from several sources by Utility Test Equipment Company (UTEC), in an effort to provide a general description of the functioning of watthour meters and test and calibration techniques. Of necessity, the material is very general in its nature as it applies to all makes and types of meters and testing equipment. Care should be taken in the application of this general information to specific types of meters and testing equipment and the information given in this bulletin should be carefully checked for correctness with the manufacturer’s information and instructions for the particular make and type of meter or testing equipment used. THEORY OF OPERATION OF WATTHOUR METERS Basically, the watthour meter consists of a motor whose torque is proportional to the power flowing through it, a magnetic brake to retard the speed of the motor in such a way that it is proportional to power, and a register to count the numbers of revolutions the motor makes. There are three principle torques involved in the operation of a watthour meter; first, the propelling torque of the motor element; second, the retarding torque of the magnetic brake; and third, the retarding torque due to friction. The motor is made up of a stator with electrical connections as shown in Fig. 1, and a disk. The stator has two windings. One of them, the Current Coil, is connected in series with the load and the other, the Potential Coil, is connected across the line and carries a current proportional to the voltage of the circuit. The split phase effect causing rotation is developed by winding the current coil with few turns and by winding the potential coil with many turns of fine wire making its magnetic circuit of low reluctance and high reactance. As a result, the current in the potential coil is made to lag almost 90º behind the line voltage. The potential coil with its core is commonly referred to as the Voltage Electromagnet and the current coil with its core as the Current Electromagnet. The magnetic flux set up by the voltage electromagnet extends across the air gap over to the iron core of the current electromagnet. Similarly, the magnetic flux set up by the current electromagnet extends across the air gap over to the iron core of the voltage electromagnet. The resultant flux of the voltage and current electromagnets then passes through the disk of the meter, and since there is a difference in phase between the two separate fluxes, the resultant flux undergoes a continual shift or “sweep” from one side to the other, always in the same direction. The eddy currents set up in the disk as a result of the magnetic flux penetration, react with this shifting flux pattern and cause the disk to rotate. BULLETIN 102 Copyright © UTEC – 1999. All rights reserved. 2 043099 Document Revision 1.0 Basic Single Stator Electromagnet Figure 1 The torque on the disk caused by the interaction of fluxes tends to cause constant acceleration. Without a brake the speed of rotation would be limited by the supply frequency, by friction, and by certain counter torques at higher speeds but the speed of rotation would be very high. Therefore, some method of making the speed proportional to power and also of reducing it to a usable value is needed. A permanent magnet performs these functions. When the disk is rotated in the field of the permanent magnets the eddy currents set up in the disk react with the magnetic flux from the permanent magnets in such a manner that there is a retarding torque or “drag” applied to the disk which is always directly proportional to the speed. For this reason, the permanent magnets are referred to as “Drag Magnets”. The retarding torque due to friction does not vary with the speed, and increases only as the bearings and register become worn. Minor amounts of friction can be compensated for, as long as they remain constant, by means of the ‘Light Load’ adjustment. To register the amount of energy measured by the meter mechanism, a register is geared to the meter disk shaft. The reduction gearing in the register is designed to make the register read directly in units of kilowatt hours. It is therefore necessary to determine not only that the meter element has the correct speed when a known load is applied, but also that the gear ratio and register constant bear the proper relation to each other to correctly register the energy passing through the meter. Multi-stator watthour meters, usually referred to as “Polyphase” watthour meters are essentially a combination of single-stator meters on a common disk. Therefore, we can rely on the basic meter theory of the single-stator meter for an understanding. The differences are mainly in a few special features and in the various applications to polyphase power circuits. The theory of polyphase metering was set forth on a scientific basis in 1893 by Andre E. Blondel, engineer BULLETIN 102 Copyright © UTEC – 1999. All rights reserved. 3 043099 Document Revision 1.0 and mathematician. His theorem, known as “Blondel’ s Theorem”, applies to the measurement of power in a polyphase system of any number of wires. The theorem is as follows: If energy be supplied to any system of conductors through ‘N’ wires, the total power in the system is given by the algebraic sum of the readings of ‘N’ wattmeters, so arranged that each of the ‘N’ wires contains one current coil, the corresponding potential coil being connected between that wire and some common point. If this common point is one of ‘N’ wires, the measurement may be made by the use of N- 1 wattmeter. From this theorem it follows that basically a meter containing two stators is necessary for a three-wire, three-phase circuit and a meter with three stators for a four-wire, three-phase circuit. HISTORY OF WATTHOUR METERS AND TESTING EQUIPMENT Since the first “Thomson Recording Wattmeter” manufactured by the Thomson-Houston Electric Company in 1899, manufacturers have made many improvements in the accuracy and reliability of the watthour meter. With these improvements has developed the necessity for faster, more reliable, and more accurate testing and calibration equipment. In the early days, calibration of the watthour meter was a major problem because suitable standards of comparison were not available. At first, only indicating instruments were used for calibrating purposes. In order to make a complete calibration it was necessary to measure time along with voltage and current or power. Prior to 1900, voltage was measured with a Cardew hotwire voltmeter. Current was measured by means of a Siemens dynamometer or by a Kelvin balance and power was measured by the Siemens watt-dynamometer or by a Kelvin wattbalance. The portable rotating standard watthour meter was introduced by Westinghouse in 1899. M. Mowbray designed and built the first multiple-range portable standard watthour meter. In 1904, Westinghouse developed a “Precision Wattmeter” which was an improvement of the dynamometer type Kelvin bridge. The development of these reference standards made possible the testing method commonly used today where the standard meter and the meter under test are connected in series with a suitable load. This method of testing greatly speeded up the calibration procedure since any fluctuations in load affected the meter and standard alike and therefore did not alter the result. In the late 1920’s it was realized that the tremendous increase in the number of meters in service necessitated more efficient methods of testing to maintain the high standards set for metering electric energy. Testing of meters on the customer’s premises was a slow and costly process. In many sections of the United States, it became increasingly apparent that more satisfactory results could be obtained by testing large quantities of meters in centralized shops where automatic test equipment could be used. By 1925, development in electronic devices had progressed to the point where their use as auxiliaries in the testing process gave promise of both greater speed and accuracy. The first electronic development was in 1925 by A.R. Rutter of Westinghouse. This development used a photoelectric device cut by holes in a meter disk to produce marks on a printed tape that were compared to master clock marks to determine the speed of the meter. In 1927, H. P. Sparks, also of Westinghouse developed the use of the stroboscopic principle which allowed the meter BULLETIN 102 Copyright © UTEC – 1999. All rights reserved. 4 043099 Document Revision 1.0 to be adjusted visually, without the necessity of counting revolutions. By 1940, some utilities had developed watthour meter test boards that were essentially fully automatic. In 1960, Weston Instruments developed the inductronic wattmeter which was the first electronic wattmeter. Also in the ‘60’s the single revolution method of calibrating watthour meters and the application of using digital counters was. introduced. In 1968 a method for computer-controlled meter calibration was patented by Duncan Electric (Landis & Gyr). By 1969 photoelectric test equipment utilizing programming for sequence of tests, computers to calculate meter accuracy, digital readout and printout of meter accuracies, and solid-state circuitry was being widely used in test equipment designs. In the early 1980’s a method of generating the precision voltage and current using solid state amplifiers was introduced by test equipment manufacturers. This method eliminated the testing problems associated with watthour meter burden affecting the accuracy of the test system and allowed precise control of the phase angle between the current and voltage. Further improvements in the reference standard were made by Radian Research by their introduction of a fully auto-ranging ‘summing’ reference standard with three current circuits making possible the testing of watthour meters with the potential clips closed. All of these improvements have made testing equipment faster, more accurate, more dependable, and more versatile. METHODS OF TESTING There are basically two methods of testing watthour meters. One is where the load during the test is controlled and the disk is timed and the other is where the meter being tested is compared with a known precision reference standard. There are times when a simple quick method of checking watthour meters for accuracy is needed. The method outlined here is used for various purposes by many companies, both large and small, for making an approximate check with a fair degree of accuracy. The accuracy of this method of checking should not be expected to be better than ±2% consequently it should not be used for calibrating watthour meters. This method is most commonly used in field testing, when a load box is not available, for determining approximate service load, suspected cases of meter tampering etc. The method consists merely of connecting a known load to the watthour meter in the conventional manner and timing the disk for a desired number of revolutions. One of the most consistent and readily available loads is a standard incandescent lamp. The accuracy of the field check can be substantially improved by measuring the service voltage in each case and adjusting the “known” wattage accordingly. For voltages within ±10 volts of lamp rating, the watt load of the lamp will increase or decrease 1.5% for each 1% of voltage above or below their rating. The disk of the meter should be timed for a convenient number of revolutions depending on the rating of the meter and the load used. It is usually desirable to run meters for about one minute or more to minimize errors in reading time. It is preferable to use a stop watch or a synchronous timer, however, any digital watch may be used with good accuracy. BULLETIN 102 Copyright © UTEC – 1999. All rights reserved. 5 043099 Document Revision 1.0 The required number of seconds with a known watt load for a given number of revolutions of the disk in an accurately calibrated meter is given by the equation: = t W Kh x 3600 x R where: Kh = Watthour (or Disk) Constant (Wh per revolution) 3600 = 60min. x60 sec. = 1 hour (needed to convert Wh to wattseconds) R = Revolutions of meter disk for time of test W = Watt load on meter (E x I x Cosq) t = Time of run in seconds The watthour or disk constant, Kh, will be found on the nameplate of all modem meters. On some older types it was marked on the disk. They may also be found in the Handbook for Electricity Metering or from the manufacturer’s data sheets. EXAMPLE: Assume a 15 ampere, 240 volt, 3 wire meter, Kh = 2, connected to a nominal 240 volt service with voltage actually 240 volts and a check is made using lamps having a total rating of 600 watts at 120 volts; the meter is run for 5 revolutions, the required time for the run in seconds is as follows: If the observed time is 62 seconds, it would show the meter to be slow approximately 3.33% or if 59 seconds, about 1.67% fast. Since the accuracy of this method should not be expected to be better than ±2%, in either case the meter is in all probability within commercial accuracy and a big part of the small apparent error is in the method of testing. ( ) Theoretical Time Theoretical Time Actual Time x 100 Percent Error - ( ) 3.33% 60 60 62 x 100 Percent Error = - - ( ) 1.67% 60 60 59 x 100 Percent Error = + - When testing 3 wire meters, the load should be applied to both current coils. This may be done by dividing the load between the two line conductors and the neutral. When using this method, be sure all loads, other than lamps, to be considered are turned off before making the check. Look out for hidden load such as lamps in closets, basements, etc., that may be on or go on with switches for other lamps. It is desirable in all cases to see that the meter is not running before applying the test load. 60 seconds 600 2 x 3600 x 5 time = BULLETIN 102 Copyright © UTEC – 1999. All rights reserved. 6 043099 Document Revision 1.0 Time-Load Testing - Polyphase The time load method may also be used for testing polyphase meters although it may be less convenient. The important thing to remember is that watt load (W) on the meter is the sum of the watts in all elements of the meter. a a a b b b c c c W = E I Cosq + E I Cosq + E I Cosq The formula therefore becomes: a a a b b b c c c E I Cos E I Cos E I Cos Kh x 3600 x R q + q + q = t where: Kh = Watthour (or disk) Constant (watthours per revolution) 3600 = 60 minutes x 60 seconds = 1 hour (needed to convert watthours to wattseconds) R = Revolutions of meter disk during test W = Watt load on the meter (sums of EI Cosq) t = Time of test in seconds EXAMPLES: Assume we desire to test a form 16 meter (3 stator, 3 phase, 4 wire, wye) that is rated at 120 volts, 30 amperes and has a Kh of 21.6. We examine the connected load and find it to be resistive in nature which would make the power factor unity (1). We next measure the current and voltage of each phase and compute the watts. A Phase: E = 120 Volts I = 5 Amperes Load = Resistive (PF=1) B Phase: E = 119 Volts I = 2.5 Amperes Load = Resistive (PF=1) C Phase: E = 121 Volts I = 10 Amperes Load = Resistive (PF=1) How many seconds should it take for the meter to make two revolutions under this load if it is 100% accurate? W Kh x 3600 x R = t ( ) ( ) ( ) t 120 x 5 x 1 119 x 2.5 x 1 121 x 10 x 1 21.6 x 3600 x 2 + + BULLETIN 102 Copyright © UTEC – 1999. All rights reserved. 7 043099 Document Revision 1.0 t 2107.5 155520 73.8 = t (seconds) Remember, when using this method for polyphase, you must calculate the watts in each phase. Some meters such as a 2 stator, 3 phase, 4 wire, wye are very tricky. This meter, because of its design, looks like it has four elements. Therefore, in calculating watts you must consider all four current coils. Comparison Testing - Single Phase Probably the best way to test watthour meters is the comparison method. In this method, the meter under test is compared to a highly accurate meter, commonly called a reference standard. This method applies, the same power, or watts, is to the test meter and the reference standard for the same length of time, and the rotating time of the test meter is compared to that of the reference standard. When older style ‘tap’ standards are used, this comparison is based on revolutions of both the meter under test and the standard. The newer style ‘summing’ standards display in watthours which simplifies testing procedures by not having to compute the revolutions ratio between the meter under test and the standard. ‘Summing’ standards always have a Kh value of 1.0 for all voltages, currents and power factors. If the same number of current inputs are used for the meter under test and the reference standard, the ‘summing’ type reference standard will display the meter under test Kh for every revolution tested. In order to apply the same power, regardless of whether you are testing a simple single phase meter or one of the more complex polyphase meters, there are two things that must be done. First, the potential coils of the test meter and the reference standard must be connected in parallel with the same voltage source. Secondly, the current coils of the test meter must be connected in series with the current coils of the reference standard and with a source of known current. Because the same voltage and current are applied to the test meter and reference standard, both have the same power (watts = voltage x amps) applied; and therefore, any variations in voltage and/or current during the test will have an equal effect on both the test meter and the reference standard and will not effect the accuracy of the test. Consequently, it is not necessary to apply precise values of voltage and current, nor is it necessary to maintain the voltage and current at exact values. It is however, important to use test sources that are free of noise, distortion and harmonics. It is usually very easy to obtain the desired test voltage, since it can readily be obtained from the power line. However, a more complex arrangement is necessary to obtain the desired test current because it must be adjusted to different values depending on the type of meter being tested. When using the comparison method, the accuracy, or more properly, the percent registration of the test meter is given by the following formula: BULLETIN 102 Copyright © UTEC – 1999. All rights reserved. 8 043099 Document Revision 1.0 Percent Registraton R x KH r x kh 100 x where: r = revolutions of test meter kh = watthour constant of test meter R = revolutions of reference standard KH = watthour constant of reference standard Watthour Constant Before continuing, some discussion of Basic and Test or Nameplate Kh is necessary. The Basic Kh is the Watthour Constant defined at 5 amperes and 120 volts. ‘Tap’ reference standards, such as the Scientific Columbus SC-10 and SC-10V, are specified in Basic Kh since they can be configured to many combinations of voltage and current. Before a comparison test can be made the Kh value for the ‘tap’ standard in its test configuration must be computed for standards that readout in revolutions. Manufacturers of ‘tap’ standards usually provide a chart in the lid of the standard with the correction factors and the Test Kh values listed. For ‘summing’ standards that readout in watthours, such as the Radian RM-10, 11, 15 and the Scientific Columbus SC-30, this calculation is not necessary because they are designed to have a Kh of 1.0 for all values of voltage, current and power factor. 5 Test Current * x 120 Test Voltage Test Kh = Basic Kh x * This is the ‘tap’ value not the test current value. For example: if the test current was 30A and the 50A tap on the standard was used for testing; Test Current = 50A. The Basic Kh of a watthour meter can be computed by working backwards from the Nameplate Kh. For example, a common residential 2-3 wire, form 2 meter rated at 240V and 30 Amps has a nameplate Kh of 7.2. If we compute the factor for voltage and current (last two fractions of above formula) we get 2 for voltage and 6 for current making a multiplication of 12. If we now divide the nameplate Kh by 12 we obtain the basic Kh of 0.6. Note that the basic Kh for the form 2 meter is the same as the normal ‘tap’ standards basic Kh. The following example will show how the testing formula is applied. Assume you are required to test a meter rated at 240 volts, 30 amperes, and having a Kh of 7.2. A standard with a basic Kh of 0.6 will be used. Because 30 amperes will be used to test the meter, the 50 ampere coil of the standard must be used. The value of the Test Kh for the standard using the 50 ampere coil is 0.6 x 2 x 10 [twice the basic voltage (120 volts) and 10 times the basic current (5 amperes)] 12.0. Substituting these values in the formula we have: R x 12 r x 7.2 Percent Registration = 100 x It can be seen that the ratio of r to R is as 0.6 to 1. When using ‘tap’ standards, it has been determined that the revolutions of the standard should always be 10 or more, and since r must always be a whole number of revolutions the nearest value of r that will make R 10 or more is BULLETIN 102 Copyright © UTEC – 1999. All rights reserved. 9 043099 Document Revision 1.0 20, making R= 12. Using these values then: R x KH r x kh Percent Registration = 100 x 100% 12 x 12 20 x 7.2 Percent Registration = 100 x If the meter runs fast or slow, then the value of R in the formula will be less or greater than 12. Suppose that for 20 revolutions of the meter, R of the reference standard is 12.12 then: 99% 12.12 x 12 20 x 7.2 Percent Registration = 100 x or the meter is 1% slow. Now suppose that R = 11.76 then: 102% 11.76 x 12 20 x 7.2 Percent Registration = 100 x or the meter is 2% fast. A quick way to find the ratio of meter revolutions to standard revolutions is to find the ratio of the product of the current rating of the standard and its basic Kh (120V at 5A) and the product of the current rating of the meter and its basic Kh. m m s s s m Kh x C Kh x C R R where: Rm = Revolutions of meter Rs = Revolutions of standard Khm = Basic Kh of meter Khs = Basic Kh of standard Cm = Current rating of meter Cs = Current rating of standard Using the same meter example as above: 12 20 3 5 0.6 x 30 0.6 x 50 R R s m = = If you are using a standard that reads out in watthours, such as the Radian RM10, 11 and 15, or Scientific Columbus SC-30 the calculation of percent registration is much simpler. The formula for this type of reference standard is as follows: BULLETIN 102 Copyright © UTEC – 1999. All rights reserved. 10 043099 Document Revision 1.0 Display Number x ME 100 x Kh x R x SE Percent Registration where: Kh = Disk constant of meter under test (MUT) R = Number of revolutions of MUT SE = Number of standard elements included in test Display Number = Display reading of standard in watthours ME = Number of MUT elements included in test Comparison Testing – Polyphase When testing polyphase meters with two or three current coils, they must be connected in series aiding; thereby driving the disk in the same direction and with the same force that the coils would produce under normal operating conditions. Because of the series connection of the current coils when testing polyphase meters, the accuracy formula must be modified to account for the number of current coils of the meter under test through which current is passed. This is necessary since the test current passes through a multi-tap standard only once, but goes through the meter being tested as many times as there are current coils. The following formula takes the current circuits into consideration: R x KH x C r x kh Percent Registration = 100 x where: r = Revolutions of meter under test (MUT) kh = Watthour constant of MUT R = Revolutions of reference standard Kh = Watthour constant of reference standard C = Number of current coils energized in MUT as given in the list below SINGLE STATOR METERS 2-wire All tests C = 1 3-wire Testing all current windings series C = 1 Testing individual current windings C = ½ TWO STATOR METERS 3-wire, 3-phase Testing individual stators C = 1 Testing stators in series C = 2 BULLETIN 102 Copyright © UTEC – 1999. All rights reserved. 11 043099 Document Revision 1.0 4-wire Y, 3-phase Testing individual circuits, single coil C = 1 Testing double coil (Z-coil) or all circuits in series with only one potential coil energized C = 2 Testing all circuits in series C = 4 4-wire delta, 3-phase Testing individual circuits, 2-wire coil C = 1 3-wire coil, windings separately C = ½ 3-wire coil, windings in series C = 1 Testing all circuits in series C = 2 THREE STATOR METERS 4-wire Y, 3 phase Testing individual stators C=1 Testing two stators in series C=2 Testing three stators in series C=3 The following example will show how the testing formula is applied. Assume we desire to test a form 16 meter (3 stator, 3 phase, 4 wire, wye) that is rated at 120 volts, 15 amperes and has a Kh of 5.4. A standard having a basic Kh of 0.6 will be used. This particular standard does not have a 15 ampere range but it does have a 12.5 ampere range. The value of the Kh of the standard is 1.5. We must first determine the number of revolutions the standard rotates to each revolution of the test meter using the formula: m m s s s m Kh x C Kh x C x ME R R This is the same formula used in the single phase section that has been modified to account for the number of current coils of the meter under test ME through which current is passed. In our example: R x KH x C r x kh Percent Registration = 100 x 100% 12 x 1.5 x 3 10 x 5.4 Percent Registration = 100 x BULLETIN 102 Copyright © UTEC – 1999. All rights reserved. 12 043099 Document Revision 1.0 As in previous examples given for single phase meter testing, if the standard rotates less than the expected revolutions for 100%, the test meter is greater than 100%: conversely, if the standard rotates more than the expected revolutions for 100%, the test meter is less than 100%. Suppose in the example above the standard rotated 11.8 times for 10 times of the test meter, then: 101.7% 11.8 x 1.5 x 3 10 x 5.4 Percent Registration = 100 x or the meter is 1.7% fast. Now, suppose that R = 12.2, then: 98.3% 12.2 x 1.5 x 3 10 x 5.4 Percent Registration = 100 x or the meter is 1.7% slow. TYPES OF TEST EQUIPMENT There are three basic methods for developing a calibrated current for meter testing. They are Resistance Loading, Phantom Loading, and Solid State Loading. Testing equipment is most commonly called by the name of its loading method...such as a Phantom Load Box etc. Resistive Load In the resistance loading method the current coils of the test meter and the standard are connected in series with the loading resistance across a voltage source. Because of this connection, the current which is permitted to flow by the selected resistance passes through both the test meter and the standard. As illustrated in the figure below, a Resistance Load usually consists of several fixed resistances of various values which can be selected to obtain different current values needed for testing. The resistors are calibrated for specific voltages and the switches are generally marked to indicate the current each allows to flow. The major disadvantage to Resistance Loads is the problem of dissipating the energy consumed by the I²R loss at high currents since the current being supplied to the test meter is also taken from the line. Resistance Loads are primarily used when non-inductive loading is required such as in protective relay testing. BULLETIN 102 Copyright © UTEC – 1999. All rights reserved. 13 043099 Document Revision 1.0 Phantom Load Through years of use, it has been found that the safest and the best method of producing the test current is by using a Phantom Load. As illustrated in the figure below, a Phantom Load basically consists of a special loading transformer that reduces the line voltage to a lower voltage which is applied through loading resistors to the meter under test; thereby, producing current. Phantom loading reduces the power dissipation in the current circuit because of the reduced voltage across which the load is connected and therefore, requires less line current by the ratio of the transformer. Let us assume that it requires 5 volts to cause 50 amperes to flow through a selected resistor. If the phantom load is powered by 120V only 2.08 amperes is required from the line source to produce 50 amperes in the loading circuit... thus the name ‘phantom load’. Even though there appears to be magic in this method in its capability of being able to supply 50 amperes to the meter while taking only 2.08 amperes from the service, the laws of physics prevail. The VA (Volt Amp) of the primary is equal to the VA of the secondary. Primary = 120V x 2.08A = 250VA Secondary = 5V x 50A = 250VA BULLETIN 102 Copyright © UTEC – 1999. All rights reserved. 14 043099 Document Revision 1.0 Solid State Loading New technology coupling solid state amplifiers and the phantom loading method is now being used to generate test voltages and currents. This new method eliminates several problems that exist with the standard phantom loading method. Because many transformers were needed to generate matching currents and the required voltages; test circuits were sensitive to the different burdens presented them by different types of meters. These burden errors caused changes in power factor and many times changes in the values of voltage and current from one test to another generating errors in the tests results. While these errors were small (usually less than .2%) they have been a concern to Utility Companies, Meter Manufacturers, and Test Equipment Manufacturers alike. The new technology basically consists of an amplifier, much like your stereo amplifier, that generates a line frequency signal to drive an output transformer like the one in the phantom load. The difference is that a computer monitors the amplitude of the output signal and its phase relationship and makes adjustments to the input signal of the amplifier to compensate for any changes due to changing load etc. It is a closed loop system which corrects itself so as to be exact at all loading points. This new technology lends itself to easy and exact control of the power factor for the test. Conventional methods of attaining 0.5 power factor were to select two phases of a three phase delta service or use a calibrated gaped core inductor in the current circuit to shift phase. These methods limit the selection of power factors available (namely 0.5) and result in very approximate phase shifts. The new technology can provide power factors of from 0 to 1 lead or lag easily and with great precision making possible the testing of VAR and QHour meters as well as Watthour meters on the same test system. Meter Shop Test Tables All modern meter shop test tables use solid state loading for producing test voltage, current and phase angle. The primary difference between a meter shop test table and the phantom load box is that it generates the test voltage, current and phase angle from digital synthesizers and not the service line; which eliminates the noise and harmonics from the service. It is, of course, more sophisticated which increases the efficiency and accuracy of testing meters. BULLETIN 102 Copyright © UTEC – 1999. All rights reserved. 15 043099 Document Revision 1.0 Some manual test tables can, with a series of selector switches, offer the tester the flexibility of easily and quickly matching the meter current coils to test hookups without tearing down each set up. This is especially convenient when a large number and variety of meters must be tested. Some manual test table manufacturers use what is called a cookbook approach to meter testing. By simply looking up the form number or type of meter in their instruction manual, the hookup and control switch settings are clearly shown. The meter tester has little problem hooking up even the more complex meters. Other more sophisticated test tables are controlled by computers. In this case every function is precisely timed and monitored so that each second of test time is used to its fullest extent; thereby performing tests in the fastest possible times. The operator initializes the test by placing the meter into the test jack, aligns the pick-up, enters the meter information and test sequences to be tested and gives the command to test. The test system performs the test automatically, printing the results after each test or sending the test results to the main frame computers history file. Many of these systems use bar code to automatically set up the meter test and program the test system. CONTRIBUTING ERRORS TO METER TESTING There are several things that can contribute error when testing watthour meters. Some of the sources of error apply only to the conventional phantom loading method and some apply to both the conventional phantom loading and solid state loading method. The first contributing error to meter testing is resistance in the potential circuit. While a small amount of resistance does not appreciably effect full load unity power factor test, it is of major concern when testing 0.5 power factor contributing as much as .2% for 0.2 ohms of resistance. Resistance is introduced most commonly from dirty contacts on connector jacks, loose or corroded connections and either too long or too small a wire gauge for potential leads. The Phantom Loading method is subject to all of the possible problem areas while the Solid State Loading is subject only to dirty contacts on the connector. Because this method senses the voltage and phase angle at the meter socket, dirty meter blades etc. are outside the control loop and therefore will cause errors in testing. The second contributing error to meter testing is wave form distortion. Improperly designed transformers, oscillating amplifiers, and voltage regulators coupled with poor transformer designs are the major reasons for wave form distortion. This type of error applies to both phantom and solid state loading methods. This error can be eliminated by choosing a solid state test system that monitors the wave form for distortion and stop the test should distortion appear. The third contributing error to meter testing is timing errors from the photoelectric counter. The relay used to control the standard potential becomes worn and dirty with use which causes unreliable and unpredictable drop-out and pull-in times. Also changing lighting conditions and power supply voltages change the threshold point at which the relay is instructed close and open. Errors produced from the photoelectric counter can contribute as much as ±0.2%. This error applies only to testing systems using switched voltage standards such as the Scientific Columbus SC-10. This error can be eliminated by using only gated display solid state reference standards, such as the Radian RM-10, 11, 15 or Scientific Columbus SC-10V, 20, 30 and UTEC 711 or 712 electronic revolutions counters. BULLETIN 102 Copyright © UTEC – 1999. All rights reserved. 16 043099 Document Revision 1.0 The fourth contributing error to meter testing is a rotating standard. Rotating standards suffer from a variety of possible problems; among them are dirt, friction, worn bearing, tilt errors, temperature drift, and coasting; all of which contribute errors to meter testing. These errors can be eliminated by using a solid state standard. The fifth contributing error to meter testing is the resolution to which the reference standard is capable of reading. Most rotating standards have resolving abilities of only 1%. In order to get increased resolving ability, multiple revolutions of the standard are necessary. This error is eliminated when the test system uses a solid state standard. The sixth contributing error to meter testing is magnetic offset both in the meter under test and the testing equipment’s standard meter and transformers. Magnetic offset is most commonly caused from switching the load on and off at points other than zero on the sine wave. This error is eliminated with test systems that use zero crossing switching or use a ramp function to start and stop the test. The seventh contributing error to meter testing is related to solid state meters. Most solid state meter designs require that the load be applied a few seconds before a measurement of accuracy is taken. This time delay ranges from about 3 to 7 seconds. To eliminate this source of error, energize the MUT with potential and current for at least 10 seconds before beginning a test. The eighth contributing error to meter testing is low service voltage. The problem primarily affects induction (disk type) meters since they typically do not have linear or flat voltage response curves. This condition of low voltage is usually created by the additional load drawn by the phantom load when connected to the PT secondary and can cause test error as much as 15%. This error does not exist when using solid state test devices or when testing selfcontained meter installations or transformer-rated installations that do not use PT The last and probably the largest contributing error to meter testing is the human error factor. Improper load adjustments, improper test sequences, improper application of correction factors, improper connections, improper recording of test data, and improper selection of testing parameters are among the most common human errors. These errors apply to any test method and are the most difficult to control. Fully automatic solid state test systems minimize these errors. TESTING SAFETY Safety should be on the mind of every meter tester. When performing field tests, the voltage levels of the service and the fault current capabilities are very dangerous and should not be dealt with lightly. Safety glasses, hard hats, low voltage insulated gloves, and long sleeve fire retardant protective clothing should be worn at all times when a service connecting device such as a meter socket is exposed. The dangers of testing are increased when the tests involve a transformer rated service. These services contain Current Transformers which reduce the high primary currents to lower currents (usually 5 amperes) so that a electricity meter may be used. The Current Transformer is a device that has a low voltage secondary as long as the secondary connection is a continuos connection. If however, the secondary connection is opened and there is current flowing in the primary, the current transformer becomes a step up voltage transformer and the secondary voltage can rise to many thousands of volts. The high voltage that is present on the open secondary of an energized current transformer generates two great hazards. The first hazard is ELECTRICAL SHOCK TO THE TESTING BULLETIN 102 Copyright © UTEC – 1999. All rights reserved. 17 043099 Document Revision 1.0 PERSONNEL. The second hazard is THE BREAKDOWN OF THE CURRENT TRANSFORMER INSULATION resulting in the destruction of the current transformer. Both hazards can be avoided provided that the secondary of the current transformer is never opened. The safest current transformer installations for testing are those that have as part of the secondary loop a Test Switch. A Test Switch is a device that will facilitate shunting of the current transformer secondary loop without the danger of opening the circuit. This device provides a make-before-break connection to prevent accidental opening of the current transformer secondary when isolating it from the metering circuit. In addition, the test switch provides for the safe insertion of other instruments in the CT secondary loop such as ammeters using a test switch safety plug sometimes referred to as a duck bill plug. The test switch safety plug, like the knife switches, provides a make-before-brake connection so as to guarantee that the CT secondary is never opened. On installations that do not have a Test Switch included in the current transformer secondary loop, THE SECONDARY TERMINALS OF THE CURRENT TRANSFORMER MUST BE SHORTED BEFORE THE LOOP IS OPENED! The shunt or short connected across the CT secondary should be a BOLT-ON or CAPTURED type of connection. Test clips, or any spring type connection, should never be used for shorting a CT secondary. When the CT secondary has been shorted with a bolt-on connection, the electricity meter may be isolated from the secondary circuit for testing. The CT secondary shunt must remain in place until the electricity meter is again wired back into the CT secondary loop so as to complete the circuit. Do not forget to remove the CT secondary shunt before leaving the service site. Leaving the CT secondary shunt ON will, of course, cause the electricity meter not to register energy for that CT and cause a decrease in the customer billing. PROCEDURE FOR TESTING WATTHOUR METERS In the case of single stator meters there are two adjustments to be made in calibrating a watthour meter; one, the “Full Load” adjustment, which involves changing the drag torque developed by the permanent magnets (drag Magnets), the other, the “Light Load” adjustment which compensates for friction. In single stator meters, the power factor adjustment is made at the factory and can not be easily changed in the field. Multi-stator meters (polyphase), however, have power factor adjustments. Typically there is one FL adjustment for the meter, and one LL and PF adjustment for each electromagnet assembly in a multi-stator meter. To properly adjust a watthour meter, its present accuracy must first be determined. This is known as the “As Found” test. After the percentage registration of the meter has been determined by the ‘as found’ test, the necessary adjustments to the FL, PF and LL adjustments are made to bring the meter within the desired accuracy. The final test made on the meter after adjustments are made is known as the “As Left” test, because it is that test made on the accuracy of the meter in the condition in which it was left by the tester. The procedures for “As Found” and “As Left” tests are identically the same as far as meter connections and readings are concerned. Because of the different procedures for testing, depending on what type and manufacturer of equipment you are using, only the method for using a field load box will be discussed in this paper. However, the basic testing principles can be applied to any piece of testing equipment. Before using any piece of testing equipment, be sure to always consult the manufacturer’s operation manual for complete and correct operation. BULLETIN 102 Copyright © UTEC – 1999. All rights reserved. 18 043099 Document Revision 1.0 Connect the meter under test to the portable load box and to the reference standard in such a way that the potential coil of the meter under test is in parallel with the potential coil of the reference standard. The current coil or coils of the meter under test should be connected in series with the current coil or coils of the reference standard and in series with the current circuit of the portable load. The particular current coil of the reference standard should in every case be one which will easily carry as a minimum at least 50% of the testing current but one where the testing current will never be greater than 150% of the reference standard current tap values. In general, two different reference standard current coils must be used for the FL and LL test when using a ‘tap’ type standard. Apply voltage and current to see that the meter under test (MUT) and the reference standard rotate in the proper direction. In those cases where the testing facility consist of a load box and a reference standard which must be connected to the meter by means of removable leads, it is easily possible to accidentally reverse the polarity on either meter or reference standard. In case of reverse rotation, the following should be done: 1. If both meter and standard rotate backward, interchange either the current potential connections or the current connections of the load box. Note, solid state standards will not run backwards, they simply will not run at all if either the voltage and current is wired backwards. In this case, the meter will rotate backwards and the standard will not operate. 2. If the meter only rotates backward, interchange current feed connections to the meter. 3. If the reference standard only rotates backward or does not operate at all, interchange either the potential connections or the current connections of the reference standard. 4. If a solid state standard is used and either the potential or current polarity is reversed, the standard will not run at all. Reversing either the potential or current connections to make the standard run. In certain phantom load designs such as the UTEC 402, 403, 404, 406, 407, 408, 440, 441, 443, 452, 453 and 454 meter test kits, permanent connections between the load box and the reference standard eliminate all guesswork because the proper combinations of current and reference standard current coil selection are made automatically. Apply full load to the meter and standard. FL is taken to mean the rating of the meter in amperes, as noted on the nameplate of the meter (TA). Proceed with the full load test, allowing the meter under test to rotate a sufficient number of revolutions so as to give at least 10* revolutions for the reference standard. This is in accordance with the general practice to obtain a sufficient number of revolutions on the rotating reference standard to be able to detect meter inaccuracies of a fraction of a percent. For instance, the large dial of a rotating reference standard is always subdivided into 100 divisions. When the sweep hand pointer of the reference standard has completed 10 revolutions, it has passed over 10 x 100 = 1000 readable divisions, so that the meter accuracy can be read to 1 part in 1000, or 1/10 of 1 percent (0.1%). The check on meter accuracy is accomplished by allowing the reference standard to rotate only during that interval of time which is needed for the meter under test to complete a given number BULLETIN 102 Copyright © UTEC – 1999. All rights reserved. 19 043099 Document Revision 1.0 of revolutions. For example, assume that 20 revolutions of the meter under test are required to make a check. With load turned on and the MUT rotating, set the reference standard to zero, with the click switch off. Then snap the click switch on at the instant the black mark on the edge of the test meter disk passes a conveniently visible stationary reference point, which is usually taken as the space between the front edges of the drag magnets or the mark usually provided above the disk opening of the meter nameplate. Then count test meter disk revolutions until 20 revolutions have been completed and snap the click switch off when the black mark has passed the reference point for the 20th time after the start. In cases where photoelectric counting equipment is used, the photoelectric counter is used in place of the snap switch. The accuracy, or more properly the percent registration of the meter, is given by the following formula: R x KH x C r x kh Percent Registration = 100 x where: r = Revolutions of meter under test (MUT) kh = Watthour constant of MUT R = Revolutions of reference standard Kh = Watthour constant of reference standard C = Number of current coils energized in MUT as given in the list below After the full load test, apply light load to the meter. The usual light load value is 10% of the TA value. The procedure for obtaining a light load check is exactly the same as the procedure outlined above for a full load check, except that the meter under test now runs at a speed which is only 10% of its speed at full load. So as not to make the time required for testing too long, 2* revolutions of the meter under test will usually suffice. The procedure for obtaining a power factor check is exactly the same as the procedure outlined for a full load check, except the power factor must be set to 0.5 which will cause the meter to run at 50% of its speed at full load. It is important that an “As Found” test be made on the meter without any cleaning or adjustment, since in case of a dispute with a customer it is essential for the power company to know exactly what condition the meter was received in before adjustments were made. For this reason, an “As Found” test is made without touching the meter more than is absolutely necessary in order to make the electrical connections required for test. When the “As Found” test has been completed, worn and defected parts of the meter should be replaced. The procedure for the “As Left” test, both full load, power factor, and light load, is the same as for the “As Found” test, as far as connections and calculations are concerned. BULLETIN 102 Copyright © UTEC – 1999. All rights reserved. 20 043099 Document Revision 1.0 The meter is run at full load current (TA), and the necessary adjustment made to the drag magnets to bring the percent registration of the meter at full load within the prescribed limits. After the full load check is completed, the meter is tested on light load and power factor and the necessary adjustments are made to bring the loads within the prescribed limits. The “As Left” test is made with the register in place on the meter. Detailed procedures for meter testing can be found in the Handbook for Electricity Metering, 8th edition, chapter 15 or the 9th edition, chapter 14. ( * Less revolutions are possible when using solid state reference standards.) INTERPRETATION OF “AS FOUND” TEST RESULTS Of course, once the “As Found” tests are made they must be interpreted as to how to proceed. If the results are within acceptable limits the test may be complete. If, however, the results are outside the acceptable limits further action is necessary. Below are listed some common conditions that are found in ferrous meters and their possible causes. FOR FERROUS (DISK TYPE) METERS ONLY CONDITIONS FOUND POSSIBLE CAUSES 1. Meter slow principally at light load. 2. Meter slow full load and light load. 3. Meter fast principally at light load. 4. Meter fast full load and light load. 5. Meter creeps but is correct on full and light loads. 6. Meter creeps either forward or backward and is either fast or slow on light load. 7. Meter slow on full and light loads, much faster on loads of low power factor. 8. Disk revolves but meter does not register. 1. Inaccurate previous adjustment: friction or dirt in register, or on magnet, worm, worm wheel, or upper or lower bearing. 2. Inaccurate previous adjustments, iron filings in magnet gaps; ground or short circuit in current electromagnet. 3. Inaccurate previous adjustment; disappearance of friction which has formerly been compensated for by light load adjustment. 4. Inaccurate previous adjustment; weakened permanent magnet. 5. Presence of excessive friction which has been compensated for by changing the light load adjustment instead of removing the friction. 6. Short circuit in voltage electro-magnet; inaccurate previous adjustment. 7. Short circuit in current coil. 8. Worm or worm wheel out of mesh; dogs at rear of register out of mesh, defective register. Bulletin 102 Copyright © UTEC – 1999. All rights reserved. 21 043099 Document Revision 1.0 METER ADJUSTMENTS IN SOLID STATE METERS Solid state meters are typically not adjustable for calibration by the utility. If a solid state meter test falls outside of acceptable limits, first make sure the meter was put into the test mode. Next the meters programming should be checked for accuracy, particularly the test pulse value. If the programming is ok, the testing equipment should be checked for proper programming and accuracy. If the test equipment is found to be programmed correctly and in calibration; and the meter continues to test out of acceptable limits, it should be returned to the manufacture for repair or replacement. Solid state meters will some day be self adjusting. The meter will control the test via a solid state test bench and will make the necessary calibration adjustments in its own software. METER ADJUSTMENTS IN FERROUS METERS Full Load Adjustment... The full load adjustment is made by varying the amount of damping flux passing through the disk. In modern meters this is done by a steel screw mounted between the pole faces of the damping magnet which, depending on its position, shunts more or less flux resulting in the speeding or the slowing of the disk. In older meter designs, the magnet position was changed to accomplish the same result. Light Load Adjustment... The light load adjustment is made by varying the amount of light load compensating torque. These adjustments are most commonly screws or wheels which, when turned, shifts a coil so that its position with respect to the element potential coil pole is changed. When this coil is shifted, torque is produced in the meter disk which will turn the disk in the direction of the shift. Over adjustment of the light load may result in “creep” which is a condition where the meter disk rotates with applied voltage and no applied current. Lag or Power Factor Adjustment... The lag adjustments are normally made only in the meter shop. These adjustments establish the flux produced by the potential coil to lag the flux produced by the current coil by exactly 90°. Some older meter designs used a coil with exposed pigtail ends that were soldered so as to lengthen or shorten the overall length of the coil, thereby changing its resistance. Other designs used a lag plate which was adjusted by a screw. In most modern single phase meters the lag adjustment is made by punching a lag plate during the manufacturer’s testing. This type cannot be easily adjusted in the field. Element Balance... A polyphase meter is actually two or three single phase meters sharing the same disk. The result shown by the register of a polyphase meter is, of course, the polyphase watthours or the sum of the individual phase energies. It is therefore necessary to make certain that all phase elements are calibrated correctly. This is accomplished by testing and calibrating each element of a polyphase meter so that the individual elements are as equal as possible for each load point. After the elements are balanced, a series element test is made to verify the summing of the watthour meter. A more complete description of these adjustments can be found in the Handbook for Electricity Metering, 8th edition or 9th edition, chapter 7. Bulletin 102 Copyright © UTEC – 1999. All rights reserved. 22 043099 Document Revision 1.0 MECHANICAL REGISTER INSPECTION The register is the counting or totalizing device which ultimately translates the revolutions of the meter into kilowatt-hour readings. It is important for the register to function properly as for the meter to rotate at the proper speed. The following points concerning the register should be noted: Friction... The projecting dog or gear which meshes with the meter disk drive should be spun around rapidly with one finger. It should move smoothly and easily. Meshing... With the register in place on the meter, there should be a slight amount of “wiggle” or backlash between the first register gear and the worm or pinion on the meter disk shaft, so that there will be no friction due to binding. Register Ratio... All registers are marked with a number which is known as the “register ratio”. This ratio is the speed reduction between the first or meshing gear of the register and the units dial of the register. To actually check the ratio of the gearing within a register, special laboratory facilities are required. The tester in the field must necessarily assume that the register ratio has the value which is marked on the register, and it is the function of the tester to see that the value agrees with the rated capacity of the meter. RECORDING TEST RESULTS After the meter has been cleaned, calibrated, and the register inspected, an “as left” test should be run. The result should be recorded together with the serial number, date tested, tester, test equipment number, and in many cases the “as found” data for the meter history file. In the more sophisticated testing equipment this may be done automatically. In other cases the procedure outlined by your meter shop foreman should be used. The meter cover should now be replaced and sealed so that the meter is ready for installation on a customer’s service.
--
Posted via a free Usenet account from http://www.teranews.com


Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
clare wrote:

<snip>
Thanks for the link. That's a very useful document.
Best wishes,
Chris
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
On Mon, 10 Jul 2006 02:05:47 +0000, Christopher Tidy

The magic part of the rotating watthour meter is the fact that force is related to current multiplied by magnetic field. The current coil induces a current in the disk circulating around the central pole piece and this current interacts with the magnetic field produced by the voltage coil. All of a sudden you have a force/torque that is proportional to the instantaneous product of the voltage and the current... This is of course, the instantaneous power. The brake magnet produces a constant field and a current in the disk proportional to the speed of the disk, therefore a braking force/torque proportional to the speed of the disk. Add it all together with a bit of inertia (ignoring friction, saturation, resistivity of the disk etc) and you get a disk rotating at a speed that is proportional to the average of the power represented by the voltage and current circuits.
Mark Rand (who spent 7 years tending. calibrating and carrying kWh meters around the world for power station performance tests in a previous life) RTFM
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
Mark Rand wrote:

Thanks very much. That's a nice explanation. Do you know why the designer might have chosen the interesting shape of core shown in my earlier post? I know there are several variations of this, but they all see to have one three-legged core and one two-legged core.
Best wishes,
Chris
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
On Mon, 10 Jul 2006 21:55:45 +0000, Christopher Tidy

In all cases it will be so that the currents induced in the disk by one winding are circulating in the field produced by the other winding. in the one shown in your photos the currents induced by the centrally placed current winding circulate under the poles of the voltage winding.
The forces produced follow the " Fleming's left hand rule". hold our left hand so that the thumb, first finger and second finger are at right angles to each other. then you have the directions given by:_
thuMb=motion First finger=field seCond finger=current.
So you can see that with the voltage coil generating its field through the two sets of poles (down through one and up through the other, and the current induced by the field from the current coil (circulating the pole piece) going into the screen/paper on one side and out of the screen/paper on the other side, that the force generated in the disk will be in the same direction at each of the outer poles.
Also you may have noticed that the currents induced by the field from the voltage coil can interact with the field from the current coil (why not, the same rules apply to all the fields and currents involved). If I haven't completely lost the plot (quite possible), the forces produced by this combination of fields and induced currents, add up in the same direction as the other lot.
Does your head hurt yet? mine does:-)
Mark Rand RTFM
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
Christopher Tidy wrote:

Actually, I think I might be figuring it out now. Still got some figuring to do, but the shape is starting to make sense...
Thanks.
Best wishes,
Chris
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
On Mon, 10 Jul 2006 21:55:45 +0000, Christopher Tidy
snip

For the watthour meter to respond to real power and accurately reject reactive power the current and voltage magnetic fields must be pretty precisely 90 deg apart when feeding a resistive load. (0 deg and hence no torque when feeding a purely reactive load)
The field produced by the current coil is is inherently 0 deg in phase but the field produced by a similarly wound voltage coil would lag by a lot less than 90 deg dependent on the L/R ratio of the coil.
The iron cross leg is a magnetic shunt which produces a large increase in the voltage coil inductance and this permits the current lag in the voltage coil to closely approach the ideal 90 deg lag. It is equivalent to a large external inductance in series with a conventionally configured voltage coil
Jim.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
The mechanical SAGMO units were rated at 120 years minimum. The useful life would be that of the house or more. It was like pulling teeth and kicking but when they changed over to electronic readouts... That dropped the life due to semiconductors. But the Electric companies wanted data logging, time of day..... Data logging is over the power lines - no meter readers - a computer. Time of day - multiple (2 or more) rates during a 24 hour time. Martin
Martin H. Eastburn @ home at Lions' Lair with our computer lionslair at consolidated dot net NRA LOH & Endowment Member NRA Second Amendment Task Force Charter Founder IHMSA and NRA Metallic Silhouette maker & member http://lufkinced.com /
snipped-for-privacy@yahoo.com wrote:

-
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
Martin sez: "But the Electric companies wanted data logging, time of day.....

Reminiscent of our Motorola days, what?
Bob Swinney
wrote:

Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
Yep Bob I was cleaning up the shelves for more books and ran across : 'my copy' "Load Management Study of Irrigation Consumers - Farmers Electric Cooperative Corp. Newport, Arkansas Dated Jan 1979." An Engineering Report by Allen & Hoshall of Memphis.
Remember this one - I think we were running 173MHz and had to 'swap or beg' frequency tones from another company in the area as the pine trees - the needles absorbed... So 403 ? we came. Seem to want to use those numbers and reason. I think you determined the issue and worked the transmitter problems.
Jean Seweat was the Manager at Farmers Electric at the time - The blue print copies of graphs are still good but the edges are turning...
We did a lot of good for people and enjoyed a lot of visits across the south and out to Sackoftomatoes as well!
Actually the meters I was talking about - the new time of day by Sagmo has some of my code inside - binary to bcd conversion. It was a Schlumberger / Motorola site (Moto after a while) - and I was SLB Sr. Scientist in Test Equipment (1 $M and up type). I had machine language background (not assembly) real machine - and was contacted off my background list of abilities. So I worked some nights to develop code and guidelines so they could re-write and own the code.
Best regards, Martin
Martin H. Eastburn @ home at Lions' Lair with our computer lionslair at consolidated dot net NRA LOH & Endowment Member NRA Second Amendment Task Force Charter Founder IHMSA and NRA Metallic Silhouette maker & member http://lufkinced.com /
Robert Swinney wrote:

-
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
Question for Jim. . .
Is the iron cross leg an adjustable (calibrateable) feature of watthour meters? It would seem that the perm magnetic shunt would be subject to some degradation over time and thus would be "proximity" adjustable.
Bob Swinney
wrote:

Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
On Wed, 12 Jul 2006 11:34:02 -0500, "Robert Swinney"

I don't know but I don't think so - I would expect the primary calibration adjustment would be the strength of the permanent magnet "drag" field. Most soft magnetic materials have "u" which varies with flux density but is pretty stable against minor temperature change and time.
Jim
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
Thanks Jim.
Bob Swinney
wrote:

Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

Hmm. Brake magnet.
What happens if the brake magnet happened to be, say, too strong for some reason?
Jim
--
==================================================
please reply to:
  Click to see the full signature.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

If they catch you it can get expensive :-). A simpler solution is a 1/64" or smaller hole in the case with a bristle sticking through it.
Mark Rand RTFM
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
On Wed, 12 Jul 2006 22:54:52 +0100, Mark Rand

Hell, if they catch you, expensive won't be the end of it. Lawyers fees, bail bondsmen, etc. That's called Theft Of Energy, and they almost always prosecute.
You might end up spending some quality time - 6 months to a few years - at The Graybar Hotel. You check out when THEY say you can.
--<< Bruce >>--
--
Bruce L. Bergman, Woodland Hills (Los Angeles) CA - Desktop
Electrician for Westend Electric - CA726700
  Click to see the full signature.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

Polytechforum.com is a website by engineers for engineers. It is not affiliated with any of manufacturers or vendors discussed here. All logos and trade names are the property of their respective owners.