[sci.electronics] How a watthour meter works

steve@ingr.UUCP (Steve Conklin) (04/13/89)

From article <3312@ihlpm.ATT.COM>, by wrv@ihlpm.ATT.COM (Vogel):
> Thanks to everyone who responded to my pump question.  I've
> learned alot just from all the responses.  I tried to send email to
> everyone who responded but about half of it bounced back.
> 
> Here's a quick summary:

[ text deleted ]

> So how does the power company measure current?  If its just on the
> ground wire, then it would miss all current on Path #3.  If its the
> sum of 110v and 110v', then current on Path #3 would be metered twice
> (wouldn't it?), once goin' out and once commin' back!  And since the
> voltage there is 220, wouldn't the power be 4X what it really was?

Here is a description of how a watthour meter works. I grabbed this from the
net two years ago.

		Steve Conklin		uunet!ingr!tesla!steve
		Intergraph Corp.	tesla!steve@ingr.com
		Huntsville, AL 35807	(205) 772-4013

===========================================================================

You can find descriptions of watthour meters in electrical engineering
texts.  Check at your local university's library.

First, you referred to "two phases".  Well, yes, you could think of
a single-phase service as two phases 180-degrees apart, but we'd better
stick to power company nomenclature and think of it as a single-phase
supply, with two hot conductors carrying current in opposite directions.
(Otherwise we'd have somewhere between 9 and 12 phases running around
by the time we'd dealt with a 3-phase meter.)

I will Keep It Simple and Stupid by discussing only an ideal meter,
neglecting the dozen or so minor ways they deviate from perfection.

A (single-phase) watthour meter is a two-phase electric motor:
 - The spinning disk is the rotor.
 - A forward force proportional to the power consumed is applied to it
   by a pair of electromagnets driven by currents 90 degrees out-of-phase.
 - A retarding force proportional to the speed of the rotor is applied
   by a permanent magnet.

Thus the rotor rapidly reaches an equilibrium speed proportional to the
power consumed, and its rotation (which drives a gear-train to an analog
counter) integrates the power.  It will also run backward if you feed
power to the grid.

(A polyphase watthour meter has two or three separate disks on a common
 shaft, one to receive a forward force proportional to the power delivered
 by each phase of the service.  Under some conditions you can get away
 with two disks in a 3-phase service.)

The two electromagnets that apply the force are a "voltage" magnet and a
"current" magnet.  Assuming they are driven by sine waves, the force they
apply is proportional to the product of their field strengths times the
sine of the phase angle between them.  Thus if the "voltage" magnet has
a field strength proportional to the voltage, but delayed by 90 degrees,
and the "current" magnet has a field strength proportional to, and in
phase with, the current, the total force will be proportional to the power
through the meter.

The "voltage" magnet is wound with many turns of thin wire, placing a large
resistance in series with its inductance.  This causes it to carry a small
current, retarded about 90 degrees from the voltage's actual phase.  It is
connected between the two "hot" wires on the input side of the meter (thus
making the assumption that the "neutral" wire is at a voltage exactly
midway between the two "hot" wires.)  The field strength is proportional
to the current times the number of turns, which is what we wanted.  (The
current is small, but there are a >lot< of turns, so the field is usable.)

The "current" magnet has two windings, one for each of the "hot" wires.
There are very few (3-ish) turns of very heavy wire, so there is negligible
voltage drop through the meter.  The windings are connected so that if you
pull your current from hot wire to hot wire, the fields will add.  Thus if
you pull one ampere from hot to hot (220 volts), you'll cause twice as much
of a field as if you pull one ampere from one hot wire to neutral (110 volts).

Does it all make sense now?

===========================================================================
  "I've got code in my node."	| UUCP:  ...!ihnp4!itivax!node!michael
				| AUDIO: (313) 973-8787
	Michael McClary		| SNAIL: 2091 Chalmers, Ann Arbor MI 48104
---------------------------------------------------------------------------
Above opinions are the official position of McClary Associates.  Customers
may have opinions of their own, which are given all the attention paid for.
===========================================================================

gdelong@cvman.UUCP (Gary Delong) (04/21/89)

In article <4958@ingr.com>, steve@ingr.UUCP (Steve Conklin) writes:
> Here is a description of how a watthour meter works. I grabbed this from the
> net two years ago.
> 
> 		Steve Conklin		uunet!ingr!tesla!steve
[description deleted]

That sounds like it would make the accuracy of the meter very dependent
upon the type of load.  ie: wouldn't an inductive load really screw
things up?

-- 
  _____ 
 /  \    /   Gary A. Delong, N1BIP   "I am the NRA."  gdelong@cvman.prime.com
 |   \  /    COMPUTERVISION Division                  {sun|linus}!cvbnet!gdelong
 \____\/     Prime Computer, Inc.                     (603) 622-1260 x 261

larry@kitty.UUCP (Larry Lippman) (04/23/89)

In article <559@cvman.UUCP>, gdelong@cvman.UUCP (Gary Delong) writes:
> > Here is a description of how a watthour meter works. I grabbed this from the
> > net two years ago.
> 
> That sounds like it would make the accuracy of the meter very dependent
> upon the type of load.  ie: wouldn't an inductive load really screw
> things up?

	You are correct.  That is why virtually all commercial customers
of electric power utility companies with an electric service of greater
than 200 amps are also metered for reactive kva (RKVA).  Newer meters have
dual indicators for both measurements within the same meter housing, and
older power installations or those for large services use separate meter
housings.

<>  Larry Lippman @ Recognition Research Corp. - Uniquex Corp. - Viatran Corp.
<>  UUCP:  {allegra|ames|boulder|decvax|rutgers|watmath}!sunybcs!kitty!larry
<>  VOICE: 716/688-1231, 716/773-1700        {att|hplabs|utzoo}!/
<>  FAX:   716/741-9635, 716/773-2488      "Have you hugged your cat today?" 

henryb@hpspdra.HP.COM (Henry Black) (04/25/89)

In article <3110@kitty.UUCP>, larry@kitty.UUCP (Larry Lippman) writes:
>> That sounds like it would make the accuracy of the meter very dependent
>> upon the type of load.  ie: wouldn't an inductive load really screw
>> things up?
>	          ......  That is why virtually all commercial customers
>of electric power utility companies with an electric service of greater
>than 200 amps are also metered for reactive kva (RKVA).  Newer meters have
>dual indicators for both measurements within the same meter housing, and
>older power installations or those for large services use separate meter
>housings.

Not exactly, commercial Utility energy meters tend to be accurate at all
power factors, inductive or capacitive.  This is defined in standard
ANSI C-12, and can be loosely expressed as accurate to plus or minus one
percent of energy used at all rates of consumption down to one percent
of full power rating of the meter, and at all power factors (provided
the energy component is at least one percent of full power meter
rating).   KVARs (kiloVolt-Amperes-reactive or 90degrees from watts) are
metered on heavy duty services because commercial tariffs often charge
for KVAR consumption, (in addition to energy consumption), not because a low
power factor causes the energy to be metered inaccurately.  KQs (60degrees
from watts) are often measured as an alternative to KVARs on old meters.  

Incidentally many meters rated OVER 200 amps do have a severe source of
inaccuracy related to their using current-transformers rather than
measuring the current directly. You see if the load includes any D.C.
component (half-wave) the current-transformer core may saturate and
become non-linear. 
---------------------------
Henry Black  (415) 857 6655    henryb@hpspdrs.HP.COM
These are my views, endorsement by my employer is not intended.

gdelong@cvman.UUCP (Gary Delong) (04/29/89)

In article <3110@kitty.UUCP>, larry@kitty.UUCP (Larry Lippman) writes:
| In article <559@cvman.UUCP>, gdelong@cvman.UUCP (Gary Delong) writes:
| > That sounds like it would make the accuracy of the meter very dependent
| > upon the type of load.  ie: wouldn't an inductive load really screw
| > things up?
| 
| 	You are correct.  That is why virtually all commercial customers
| of electric power utility companies with an electric service of greater
| than 200 amps are also metered for reactive kva (RKVA).  Newer meters have
| dual indicators for both measurements within the same meter housing, and
| older power installations or those for large services use separate meter
| housings.
| 
| <>  Larry Lippman @ Recognition Research Corp. - Uniquex Corp. - Viatran Corp.

Hum.. Sounds like I might have a use for all those 5KVA Topaz transformers
I picked up.

-- 
  _____ 
 /  \    /   Gary A. Delong, N1BIP   "I am the NRA."  gdelong@cvman.prime.com
 |   \  /    COMPUTERVISION Division                  {sun|linus}!cvbnet!gdelong
 \____\/     Prime Computer, Inc.                     (603) 622-1260 x 261