I have 2x 5kVA Multi II in parallel. I have installed a hobby site 100A CT (SCT013) at the grid entry into the DB and configured the Master to use the External CT. ESS was configured to use the Inverter/Internal grid metering. I have seen some values which does not make sense.
I get a reading on the GX for AC-In and out L1 with voltage 228v, current at ~15A and power at ~2400W. P=IV does not compute here. There is about 1000W missing. ([edit] I think I found the missing 1000W in the apparent power reading) I then clamped it with a Fluke next to the CT and found the current to be around the 7.5A mark. When I disconnect the CT the current drops to 0 , but there is still ~800W measured on L1.
Does anyone have any idea why this could be? Is this a config issue, CT wiring issue? I don’t find much info on the subject.
I may be way off base here, so take this with sufficient salt. This is based on a discussion I had with one of the engineers, and what I know of the internals of the thing. You should probably look in the manual to be certain. If you find it in the manual, let me know as well. I could not find it.
When you say that you configured the master to use the external CT, do you mean you removed the loop on the connector D?
What that does is really quite similar to what an audio jack does when you plug in headphones: It disconnects the internal speaker, and instead connects the headphones. Here it disconnects the internal current sensor and substitutes the external one. The Multiplus doesn’t know that it’s getting the reading from an external source now.
So right now the master is using an external CT, and the slave is using the internal CT, and the sum of the two is used in calculations. So I absolutely will expect some weirdness.
So you will have to get another SCT013 for the parallel unit. Either that, or you can look at using an external transfer switch. This is a fairly new feature. Then the current sensor is connected only to the master. The parallel unit’s input is left disconnected and only the AC-output is paralleled with the master.External Transfer switch is only certified for generators, my bad!
Yes, loop removed on D and external CT connected to COM-EXT. I configured the Master with the external CT option in VEconfig, so it should know that it is using an external? I will try and connect my second CT to the slave, config in VEConfig and see how it goes. It seems that the Multis’s are adding the currents together.
There was apparently a quote somewhere, sometime to plug a dummy jack in the slave unit (when they still used the jacks) to disconnect the AC-in CT of the slave. I did try to just remove plug D from the slave, but there was no change in current.
Step 1
Master - Ext CT configured/connected
Slave - Int CT Configured/External Connected
Current still double what the Fluke measures, Power now doubled and calculated correctly using the CT Current (although current incorrectly measured).
Step 2
Master - Ext CT Configured/Connected
Slave - Ext CT Configured/Connected
Current still double what the Fluke measures, power still as in Step 1
Step 3
Master - Ext CT Configured/Connected
Slave - Ext CT Configured/Disconnected
Current still double what the Fluke measures, power back to 1/2 of Step 1 and 2.
Step 4
Master - Ext CT Configured/Disconnected
Slave - Ext CT Configured/Connected
Current 0A, power still as in Step 3
Step 5
Master/Slave - Int CT Configured/Loop connected
Current and power seems to be correct.
No ESS feeding of non-critical loads on AC-In.
Now I am wondering… I have the YHDC branded SCT013-000 100A:50mA. I had a look at a few datasheets and it seems to note that it has a 50mA current output. I see there is a SCT013-000V with a 1V voltage output. Which output type is the Multi expecting?
If I am reading what Mr. P says then the Multi doesn’t know if the physical sensor is internal or the one extended by the cable on the outside (not calling it External). Set the setting back to ‘Internal’ but use the ‘extended’ one on the outside. For the Slave, set it to Internal also but remove the loop. I am guessing that ‘External’ means it’s an ET112 perhaps.
I tried the above suggestion and had the same results.
Master - Int CT Configured/ Ext Connected
Slave - Int CT Configured/Connected
Current double that of the Fluke, Power correct according to Fluke current reading.
I opened one of the CTs and it looks like a TVS Diode. Still need confirmation of what the Multi is expecting from the CT.
I converted the 100A:50mA CT from current output to voltage output by removing the TVS diode and populating the existing space for the burden resistor with a 20Ohm resistor. This was calculated using the 100A rating, 2000 turns on secondary (assumption from the Datasheet) and a 1V output voltage.
I have the master and slave both set to External CT, but have unplugged the slave connector and the master has the converted CT connected.
This seem to have solved the issue of the incorrect current being measured. Now it seems the lowest current that can be read is 2.2A. Again, is this a CT or Multi issue. my ESS Grid point cannot be lower than 100W at this stage. This is either the meter or the current measuring not working optimally yet.