watt-hour meter operation in 6 phases

Suppose the following scenario is set up (and yes, I know it would be a mess):
ONE of the following is chosen to step distribution voltage down to utilization
voltage:
1. 3 separate transformers with 2 phase "Edison split style" 120/240 volt secondary. Each transformer is fed by a different distribution phase.
2. 2 separate transformers with 3 phase 208Y/120 volt secondary. These two transformers are fed at 180 degrees relative to the other, such as by reversing the L-L connections on the 2 bushing primaries, OR by reversing the 120 volt secondary windings.
The neutrals are all solidly interconnected. The end result is a 6 phase system with 7 total conductors. I'll label the phases A,B,C,D,E,F around in a circle. So phases A and C would be 120 degrees apart, and phases B and E would be 180 degrees apart.
This 6 phase system feeds 5 panels. 2 of the panels are 3 phase panels for 208Y/120 systems. One of these panels is fed with phases A,C,E. The other is fed with phases B,D,F. 3 of the panels are 2 phase panels for 120/240 "Edison split style" systems. The first of these is fed with phases A,D. The second is fed with phases B,E. The third is fed with phases C,F. Most loads are 1 phase loads (120 volt L-N). Some loads are 240 volt L-L and some are 208 volt delta.
What I am interested in finding out is how watt-hour meters would behave in this setup. Two possible setups could be used:
1. 2 separate 3 phase watt-hour meters, one wired for phases A,C,E and the other wired for phases B,D,F.
2. 3 separate 2 phase ("Edison split style" again) watt-hour meters, the first wired for phases A,D, the second wired for phases B,E, and the third wired for phases C,F.
The neutrals would be solidly interconnected as needed, and the meters could be in the same metallic enclosure.
I'm not expecting any particular strangeness from any type of power utilization based on which transformer scheme is chosen. What I would be concerned with is how accurate the meters would be in either of the meter configurations.
I suspect the first meter choice (2x 3 phase) would be the most accurate as I believe these meters are designed as 3 separate voltage-current product meters internally. I worry the second meter choice (3x 2 phase) *MAY* be inaccurate in certain cases, because they are designed as a true single phase meter with a single current transformer where the 2 phases of the 180 degree split system pass through that one CT in opposite directions, and the voltage is measured only between those phases, not involving the neutral.
Perhaps as long as the voltages are reasonably balanced across the phases, either metering scheme could work.
If you were designing a watt-hour meter for such a 6 phase system, would it be possible to reduce the costs of the meter by using 3 CTs instead of 6, by having the 180 degree paired phases running through in opposing directions, and maintain accuracy within the range suitable for use as electric service meters, within the service voltage remains reasoably in balance?
Some of you may be wanting to ask why have a system like that in the first place. THIS THREAD is not about WHY to have such a system. THIS THREAD is about how to meter such a system. If you want to discuss WHY to have such a system, please start a new thread for it, with a subject "why a 6 phase system".
--
|WARNING: Due to extreme spam, googlegroups.com is blocked. Due to ignorance |
| by the abuse department, bellsouth.net is blocked. If you post to |
  Click to see the full signature.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
snipped-for-privacy@ipal.net wrote:

(described mess)
Phil Read up on Blondel's theorem. For accurate power measurement, the number of power measuring elements needs to be one less than the total number of conductors (including the neutral). A Blondel solution should be accurate under any condition. I haven't looked at the specific case you have presented, and I am not inclined to at the moment, but you should be able to prove mathematically that the sum of the power readings from all of the elements will be the accurate total power if you have N-1 elements.
Many metering schemes are not Blondel compliant, and will only be accurate under balanced voltage and load conditions. They are used for some applications where imbalance is unlikely (i.e. to meter only a three-phase pump motor, you can get away with a single element and multiply by three). I beleieve that your "180 degree pair" proposal falls under this category, as it would assume that both wires are at the same potential with respect to the neutral.
Ben Miller
--
Benjamin D Miller, PE
www.bmillerengineering.com
  Click to see the full signature.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

I suppose some may call this 'cheating', but for such an unusual setup, maybe it would be easier to just use PT/CT and instrument the primary side of your transformers with traditional 3-phase metering??
daestrom
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.