Monitoring during power failure

General questions and discussions focused around the EKM Metering System. Discuss your meters, installations, and meter communications here.
Post Reply
Zaaphod
Posts: 20
Joined: Fri Oct 23, 2015 2:04 pm

Monitoring during power failure

Post by Zaaphod »

I'm using an Omnimeter pulse V.4 to monitor the mains coming into my shop, and yesterday I had a power failure. I the way the power failed was terrible, I think someone crashed into a power pole or something, because it was rapidly off then back on, then low voltage.. etc.. I RAN to kill my main breaker as fast as I possibly could to avoid damage to my electronics and motors none of which like that kind of thing at all.

So, I found that I wanted to monitor the incoming voltage to make sure it was stable and within tolerance before I turned my main back on. This omnimeter is hooked to to the mains before the main breaker, so it would monitor the power coming in and allow me to monitor the incoming while the main breaker is off... however since I was having a power failure, all I would get is RS485 errors of course, until the power returned, and then I was able to monitor the power conditions... but this brought up an interesting idea.

I have a gigantic battery backup system which can run all my computers and electronics through a complete blackout for 20 hours, it is a constant duty double conversion uninterruptible power supply that outputs true sine wave regulated power from the battery bank which is charged all the time when power is available. I am wondering, what would be the best way to power this one omnimeter with my battery backup system so that during a power failure, I would get actual voltage readings of 0 volts on L1 and L2, instead of communication errors.

I wanted to run this by the experts here to see if I'm overlooking anything...

I don't really want this omnimeter to monitor anything other than the incoming power, which is 240V single phase with neutral, so I'm only using L1, L2, and neutral, I'm thinking I could send power to it using L3 and neutral, but not hook up any CT for that circuit, that way the L3 line would read the voltage of my battery backup power supply, but would have a current reading of 0, with a current of 0, I would also have zero power on the L3 line, and therefore, (I'm hoping) my total kW/h readings for the mains based on L1 and L2 with their CTs would still be very accurate. There are a few things I need to consider...

am I correct that with zero current on L3 my L1 & L2 and total readings would still be accurate? or would the voltage being present on L3 put the omnimeter into some kind of 3 phase mode where the calculations of L1 and L2 might be different than they would be for single phase?

L3 may end up being completely out of phase with L1 or L2, and might even be a slightly different frequency, because it's a double conversion system, it produces it's own output frequency regardless of the input frequency, and I don't believe it even attempts to sync with the input frequency. It outputs 60.00Hz regardless of what comes in. would this phasing difference cause any problems with the omnimeter?

what should I do with the CT input for L3? Should I short the input to make sure it always reads, 0 amps, or should I leave it open?

What would be the maximum current draw the Omnimeter would use when powered only by the L3 supply during a power failure? My plan is to run fairly small voltage supply wires for this purpose, and I would fuse them as low as practical, I'm thinking like 50mA or so (or whatever the max power would be)

Should I run a second neutral wire back to the omnimeter from the battery backup source, or just let it be connected through the building's neutral?


or Is there another way of powering this omnimeter during a power failure that I'm missing? or is this the best way to do it?
I have several power supplies available which are powered by the battery backup, so I have available 5VDC, 12VDC 24VDC 48VDC, 48VAC, 75VAC, and of course the 120VAC from the system itself, but I don't see any way of doing it besides to use 120V on L3, unless I'm missing something.
Jameson
Posts: 860
Joined: Fri Nov 04, 2011 7:42 pm
Location: Santa Cruz, CA
Contact:

Re: Monitoring during power failure

Post by Jameson »

Yes, I think that powering the meter during a power outage on line 3 would work. Can you set it up so the meter only gets a 120 VAC connection on Line 3 from the inverter and a neutral connection from the inverter when the grid power is out? Im thinking of a double pole switch that would only power the Omnimeter and connect the neutral, when the grid power is out.

If your inverter neutral is not connected to the grid neutral, Im not sure what you would get if you connected the inverter neutral and the grid neuttral together at the inverter. Hopefully they are already connected together somewhere else in your system.

You are correct that if you only have volts on Line 3 and no current on line 3, your metering of L1 and L2 will still be accurate. You can verify this on a v.4 meter as it will give you kWh on Line 3, this will remain at 0 kWh. I don't think you should have to shunt the CT3 connections. Id not think the frequency on L3 being off from L1 and L2 should matter.

The Omnimeter draws less than a watt of power. We use 1 amp fuses to protect the wire that goes from the voltage reference to the meter.

Hope this helps
Jameson
EKM METERING
http://www.ekmmetering.com
831.425.7371
Zaaphod
Posts: 20
Joined: Fri Oct 23, 2015 2:04 pm

Re: Monitoring during power failure

Post by Zaaphod »

Thanks for the advice. I just checked and the neutral on the output of my inverter IS connected to the building neutral.

If the neutrals weren't connected, I suppose I could also use an isolation transformer to prevent any kind of neutral loop problem.

It would be easy to wire up a tiny relay to apply the L3 signal when the battery backup system is reporting it is no longer charging, the only issue with the definition of grid power being out.. it normally doesn't just go 'Out' as in a nice sharp 240v to 0v in an instant. it's normally low voltage and up and down and all over the place before it fails. My battery backup system considers even the frequency being off by more than 2Hz to be an alarm condition, and the voltage dropping to less than 108v to be considered out of tolerance.

I could just get 2 small relays and wire their normally closed contacts in parallel and the coil of one to L1 and neutral and the coil of the other to L2 and neutral.. that way it would only apply L3 when neither L1 or L2 were able to hold the coils any longer. I could also wire the normally open contacts to disconnect L1 and L2 at the same time, as I really don't need them connected until the incoming voltage is high enough to pull in the relays.

but I'm thinking since my neutrals are all connected, I shouldn't have a problem. I am thinking that running a neutral from the inverter back to the omnimeter and connecting it with the other neutral is probably a bad idea.. just stick with one neutral and that eliminates any possibility that some current would flow on a second neutral wire that ends up being in parallel with the building neutral wiring.
Post Reply