Branch Circuit Monitoring

[jordi29]

Hello all,

I am evaluating a Branch Circuit Monitoring (BCM) for my company. Only for a 42-pole panel. We have hundreds of these panels, and no breaker configuration is the same. We have 10A, 20A, 30A, 50A and 100A breakers... 1-pole, 2-pole, and 3-pole... you get the idea.

I need to develop a solution that read data output(Amps per pole) and input (Amps per phase for main ABC feeders). My concern is how to gather this input feeder data. One solution is to use Enercepts for the feeders... easy enough.
Another alternative I have found is that some products have no CTs in the feeders, and rather "add up" the Amps per pole to provide feeder data.

Ok, bear with me, I am a Mechanical Engineer.... so keep your answers simple. But I recall college classes on phasor diagrams, inductive loads, etc... that leads me to believe that you can't simply "add up" Amps per pole. When I asked the manufacturer (PDI) for their "algorithms", they simply gave me the formula for RMS readings.. duh!

I am afraid with our load diversity (although almost everything is pF=0.98... close to resistive), this "adding up" can get very inaccurate...

Can someone please explain this to the Mechanical Engineer...? ;-)

Thanks!
Jordi

 

[wareagle]

You are correct. You can not add the amps on each phase together but you can add the KW. What is the data going to be used for?

 

[jraef]

Looking at each and every 1, 2 and 3 pole feeder breaker current in hundreds of 42 circuit panels is going to be hugely expensive. Why do you need this kind of granularity?


"If I had eight hours to chop down a tree, I'd spend six sharpening my axe." -- Abraham Lincoln
For the best use of Eng-Tips, please click here -> faq731-376

 

[racobb]

You say you want to measure input and output, why not measure the input (main) and look at the branch circuits on an as needed basis.

Agree with jraef that this is going to be very expensive for what you may glean from the info.

Are you having balance problems and need to switch loads around....just a guess. If so, you can always do spot checks with a clip on meter for the ones that might be giving you problems. Balance problems are usually a ont time fix and done with it until you add more single phase loads.

Not trying to tell you how to do your job, but do you really need to do this?

What are we missing here?

Alan

 

[itsmoked]

Me thinks the bean counters have gotten loose again.

Keith Cress
kcress -

http://www.flaminsystems.com

 

[rbulsara]

Regardless of "why", if you implement a 42 pole BCM, monitoring the main is just one more data point and a set of split core CTs. It usually is an option.

As a consultant I have incorporated them in the design upon customer request but was never convinced of the justification for such a system. If this is a data center, installing plug-strips with ammeters is a much better option.

Rafiq Bulsara

http://www.srengineersct.com

 

[waross]

If you really want to know the current on each circuit, look at applying a Thermon 1818a panel.

http://www.thermon.com/catalog/aus_pdf_files/TEP0027u.pdf

This panel is intended for heat tracing control and monitoring, but it has a lot of features that would be useful for monitoring circuits.
It will monitor:
Current
Ground leakage current.
It will alarm
Over current
Under current
Ground leakage current
These parameters are all programmable.
A communication option is available.
For you application you may not need power modules, but if you want to add control, there are both Electro-mechanical relays and Solid State relay output modules available.
Soft start is available with the solid state modules.
The Thermon cabinet is mounted near the breaker panel and the circuits to be monitored are run to the Thermon cabinet and then out to the field.
For circuits of more than 30 amps and feeders, a second CT may be fitted to scale the current down.
You may remotely control by using one ambient sensing RTD and remotely setting your set point for each circuit above or below ambient.
If this were my project I would also check out what is current for monitoring and control in the refrigeration industry.



Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[jordi29]

Wow... lots of replies, but no answer to my question (except for Wareagle... thanks!). Lots of questioning as to "why" I need this. Ok, I can answer that... we are a datacenter, with over 1000 customers.... we bill them per power consumption per individual ckts... got it?

Now, my question is... again, how accurate is it to add all the 42 poles to get feeder data on a panel? I am looking for an answer that talks about phasor diagrams and gets into the mathematics behind it.... and why you just can't simply add Amps on a panel....

We will use BCM data for customer billing... we will use feeder data to balance our panels (and hence our PDUs upstream). Hope that helps!

 

[davidbeach]

If your measuring equipment can come up with Watts per circuit, you can accurately add those up to your heart's content. The total of the branch circuit Watts will equal the panel input Watts less minor losses (I²R). Metering errors will probably account for more discrepancy than losses.

Adding current magnitudes will definitely get you into trouble as you will have a collection of apples, oranges, kumquats, pears, nectarines, and apricots; adding them up won't provide much information.

Vector addition of currents won't tell you a whole lot more. The vector sum of the phase currents will tell you the vector sum of the neutral and ground currents. If the three phases are all balanced, the vector sum of the three phase currents will be zero. If you want to meaningfully sum up the currents on a single phase, the vector sum is what you need.

But in all cases, Watts are your friend, there is no phase angle to Watts, they can be added without error, and that is ultimately what you need to be charging your clients for anyway.

 

[waross]

Sometimes you can add directly. That is when the loads are connected line to neutral and the power factors of the loads are the same.
If the phase angles of the loads are not the same (not the same power factor) you can not use simple addition.
Line to line loads are at a different angle than line to neutral loads.
The load on "B" phase is the vector sum of the line to line load on "A" to "B" and "B" to "C" plus the line to neutral load on phase "B".
You should have a good EE working with a good lawyer on this. Billing based on current alone (KVA vs kW) may leave you open to an expensive and hard to defend class action suit.
It's all about those pesky phase angles.
And, what davidbeach said.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[jghrist]

If you are trying to add peak loads, then you have more to worry about than power factor and phase angles. The peak loads for each branch circuit will occur at different times, so the sum of the branch circuit peaks will not equal the main feeder peak. This is also a problem with David's suggestion of using watts instead of amps.

 

[rbulsara]

I am looking for an answer that talks about phasor diagrams and gets into the mathematics behind it....

Here is a preview and you can take it from there. There no reason why you cannot learn it.

For the starters:

Assuming a perfect sine wave ( no harmonics), I=(I*cos#)+j(I*sin#).

Where # is the phase angle. Cos#=Power factor. j is the 90 degree phasor operator and square of j is -1. Use j for lagging pf, -j for leading pf.

For line to neutral loads, you do this for current in each circuit on the same phase and mathematically add them to get the total line to neutral load in a phase.

If you do a graphical addition, Icos# will be at the right angle to Isin# component. You can add all cos# components together and add all sin# components together and come up with a final vector addition for a particular phase current.

For two pole (208V) circuits on a three phase system, it gets more complicated as there is a 120 degree phase difference between each pair of phases. For example, for 10A of resistive single phase load (pf=1)between phase A and B and , 10A of load on A and C would add up as 17.32A on phase A and not 20A.

For more refer to books that explains AC poly-phase circuit analysis and sit down with an electrical engineer friend or take some formal classes. You will need to spend some time and effort to master it but it can be done.

Explaining impact of harmonics and their addition is even more complicated, for me at least, so I would not attempt it here.

Rafiq Bulsara

http://www.srengineersct.com

 

[jordi29]

Great answers guys! That's exactly what I was looking for. Perfectly understtod, even for an ME.

I only have one question regarding rbulsara's post. You wrote:
"For example, for 10A of resistive single phase load (pf=1)between phase A and B and , 10A of load on A and C would add up as 17.32A on phase A and not 20A."

I think you meant to write:
"For example, for 10A of resistive single phase load (pf=1)between phase A and B and , 10A of load on A and B would add up as 17.32A on phase A and not 20A."

Correct?
In the same example, would you also see another 17.32A on phase B?

Lastly,
Why the comment on the "lawyers"? What is risky about it? We are in the colocation business... we rent space/power and cooling. That's our business model. are there concerns with the accuracy?

Thanks again to everyone!!
jordi29

 

[rbulsara]

jordi29:

No, the statement, as I had written, is correct. Also read the fourth line in waross's post. The BCM will read 10A in each phase of a two pole circuit.

The lawyer' concern is valid and you want to be sure as to what you are charging them has the correct technical basis, if you use the calculation method. That is why the direct mesuarement is much more convenient and less complicated.

Much worse would be if you are charging less than what you should have if you are not accounting for the losses between your measuring point and your utility meter.

It all depends on what the agreement says between you and your clients, I am sure that has been reviewed by your company's attorneys.

Imho, there are much simpler and less expensive ways create basis to charge tenants for energy consumption, but I do not run your business.






Rafiq Bulsara

http://www.srengineersct.com

 

[davidbeach]

You may also have a legal problem in the local utility finds that you are reselling watts (kWH) to your customers. Generally only utilities are authorized by law to sell kWH. Check your utility regulations and state laws.

 

[stevenal]

davidbeach finally brought up the unit you want, kWh. Your customers consume energy (kWh) at varying rates (kW). The way to measure this consumption is with a watthour meter built to ANSI or IEC standards. A BCM probably does not make the grade.