Calibrating GEM against meter

[turbo]

Hi all,

I'm looking at my power usage recorded on SEG and seeing the results be quite different than what my meter is recording.

For example, SEG is showing the combined power usage yesterday from my two mains CT's is around 20kwh. The meter reading for the house for yesterday's use was 31kwh.

I would think 'yesterday' on SEG would be midnight to midnight, but not sure of that. I'm reading a 24 hour period from 7 AM to 7 AM on my meter.

Also, I just ran a washer all morning with a Kill-a-Watt on the outlet, 120V appliance. SEG reports .240kwh on that circuit for today, the Kill-a-Watt reports .350kwh used.

I'm consistently getting this result. This is not the first time I've done this test.

The ratio of difference is approximately the same for the whole house reading as it is for the single appliance.

What could be going awry here? Why is my GEM reporting around 33% less usage than the meter is and how can I calibrate this?

[Teken]

1. Verify that the main CT's are set correctly and fined tuned using low, mid, high loads.

2. Measure the line voltage where the PT is plugged into and fine tune it to a true RMS DMM.

3.


Ideals are peaceful - History is violent

[turbo]

1. Verify that the main CT's are set correctly and fined tuned using low, mid, high loads.

2. Measure the line voltage where the PT is plugged into and fine tune it to a true RMS DMM.

3.

Okay, thanks.

By 'set correctly' do you mean the orientation of the CT's on the physical circuit or the configuration in the GEM's setup? I'm fairly certain the GEM setup is correct.

I'm not clear what tuning would involve. I can run a simple load with all else off, then run a medium pull load and large load. I could do this for a circuit on each main breaker using a Kill-A-Watt or an amp meter on the load, then compare it against the house meter to see if it's reading the same or not. Perhaps I would find one mains CT is reading correctly and the other is not.

[Teken]

First thing is to verify the primary 100/200 amp CT's are facing the correct direction and not canceling each other out.

My reference for a small, medium load is a 60, 100 incandescent light bulb.

Place it on one side of the electrical leg and confirm those two loads are seen. If not write down the variance and report them here keeping in mind there will be a slight variance which we can compensate via CT value changes.

A high load is anything like a hair dryer / heater which consume 1500-1800 watts. Place this load again in each leg of the electrical system and report back the measured wattage values.

As indicated you need to measure and validate the line voltage and compare it to a true RMS DMM. If the GEM indicates 126 VAC but your (reliable) meter says 120 VAC then you will change the PT setting to reflect this.

Note this measurement must be taken at the same outlet as the PT transformer .

Please post up some of the settings for your GEM's CT values along with what the PT setting is set for now.

Do a search under my name and you will see a thread which details how I calibrated both above.


Ideals are peaceful - History is violent

[turbo]

I've got a 20 year old Radio Shack Micronata that says it has true RMS. It's saying 117V at the PT outlet. My GEM setup is saying 119.4V, PT type is 238, PT range is 4.

[Teken]

I've got a 20 year old Radio Shack Micronata that says it has true RMS. It's saying 117V at the PT outlet. My GEM setup is saying 119.4V, PT type is 238, PT range is 4.

If you're confident in the accuracy of the DMM you can adjust the PT Type up / down to reflect what you recorded on the meter.

Once you select save you can navigate to the live tab in the GEM and see the voltage reading. If its correct and reflects what you saw in the DMM you're good to go.

The next step is to verify low, mid, high loads on each electrical leg and report back their values here. If for example the 60 watt bulb shows up as 45 watts.

In the CT tab for that specific channel you can adjust the *Type* to closely match the load. Please keep in mind this assumes the load is actually (verified) as being what ever in this case 60 watt incandescent bulb.

I call this out because you would be surprised even a common light bulb may not actually consume what it states such as a 60 watt bulb.

[turbo]

In the CT tab for that specific channel you can adjust the *Type* to closely match the load. Please keep in mind this assumes the load is actually (verified) as being what ever in this case 60 watt incandescent bulb.

I'm not particularly confident my DMM is giving good readings. I went ahead and adjusted PT voltage to match.

All my non-main CT's are CT-50's. I'm not clear how I would adjust the CT type to match the load. I can see the menu and choose CT sizes.

If I test the load with a light or a blow dryer, to verify the reading I see in the GEM I can use a Kill-a-Watt to get the watts used.

So, right main, CT6

Breaker 22: 60watt led bulb
kill-a-watt reads 17W, GEM reads 18watt
hair dryer
kill a watt = 363 gem = 350

[Teken]

In the CT tab for that specific channel you can adjust the *Type* to closely match the load. Please keep in mind this assumes the load is actually (verified) as being what ever in this case 60 watt incandescent bulb.

I'm not particularly confident my DMM is giving good readings. I went ahead and adjusted PT voltage to match.

All my non-main CT's are CT-50's. I'm not clear how I would adjust the CT type to match the load. I can see the menu and choose CT sizes.

If I test the load with a light or a blow dryer, to verify the reading I see in the GEM I can use a Kill-a-Watt to get the watts used.

So, right main, CT6

Breaker 22: 60watt led bulb
kill-a-watt reads 17W, GEM reads 18watt
hair dryer
kill a watt = 363 gem = 350

Hello Turbo,

For now the first step is to ensure the line voltage from the POCO is accurately being displayed and configured in the GEM. Changing the line voltage by guessing will only make things worse not better.

If you can borrow a True RMS DMM that would be great. If you can't just go to the store and buy one and once you're done with it just return it!

With respect to your lighting load it must be a incandescent load not a LED. Because a standard incandescent bulb is resistive in nature where as a LED bulb has a whole lot of electronics which are reactive, resistive, and inductive.

As you noted LED bulbs can emit the same amount of light (lumens) as a normal bulb but at a fraction of the energy consumption. Your high load in this case (hair dryer) ideally should be anywhere from 1000 - 1800 watts.

The reason for using (known verified loads) is to determine and have reference to what the CT is showing in terms of low, mid, high, wattage values. Some things to keep in mind adjusting the PT value is a global change to the GEM and how it portrays the energy readings.

When you change the CT value it only applies to that specific CT and nothing else. In the tab where you enrolled each micro 50 CT you will see there are values like the PT where you can change the (TYPE) also.

Again, you will adjust this value until it matches the value obtained by the True RMS DMM. I would not be using the Kill a Watt as a means of absolute reference. Its a good tool for ball part values but it is by no means as accurate as a DMM.

[ben]

Are you posting directly to SEG from the GEM?

[turbo]

Are you posting directly to SEG from the GEM?

Yes, it's direct posting.