What's new
What's new

Another Explanation of RPCs, and Their Capabilities

peterh5322

Diamond
Joined
Dec 15, 2002
Location
Monterey Bay, California
Referring to the attached diagram ...

The C-N-A phasor remains largely the same irrespective of load.

The values of the Cab and Ccb capacitors affect the magnitude of the N-N'-B phasor. The load affects this phasor, too.

There is a point at which, for a given load, N-N'-B is 208 volts, if C-N-A is 240 volts. This is the balance point.

If the load is lower, then there are too many VARs (volt-amperes, reactive) and N-N'-B will be greater than 208 volts.

If the load is higher, then there are too few VARs and N-N'-B will be lower than 208 volts.

Although it is usual to measure C-B, A-B and B-C, it is also possible, in some cases, to measure N-N'-B.

The single-phase line is supplying the real power as C-N-A.

The RPC's idler and its capacitors are supplying reactive power as N-N'-B.

Combining the real and reactive powers, at the load motor's terminals, results in the apparent three-phase power.

An RPC can "source" three-phase power, but it cannot "sink" three-phase power.

Therefore, an RPC is not a useful source of power for a load which regenerates three-phase power to the line during "regenerative braking" as only the A and C phases can sink such power.

A utility three-phase source can both source and sink three-phase power.

A Phase Perfect device can also source and sink three-phase power.
 

Attachments

  • Small_Medrez_RPC_Phasors.jpg
    Small_Medrez_RPC_Phasors.jpg
    86.5 KB · Views: 479








 
Back
Top