Mystery current in GT inverters

bill von novak
bill von novak Solar Expert Posts: 891 ✭✭✭✭
I recently installed some monitoring hardware to monitor power consumption and production within my house. I am looking at solar generation, pool pump usage, air conditioning power usage, electric vehicle charging usage etc in an attempt to better understand overall loads. To that end I built a data logger and placed snap-on current sense transformers around a dozen circuits on my system.

I have two older Sunpower transformer based GT inverters, one southwest facing 3500 watt and one southeast facing 5200 watt inverter. The array itself is 9.88kW. They seem to work well. Per their displays they come close to maxing out on sunny cool days, and this has been reflected in our Zigbee based smart meter energy monitor. When they are generating 8000 watts total we see about 7500 watts being fed back, with the 500 watts traceable to refrigerators, fans etc.

The inverters are wired to a 240VAC subpanel (no neutral) and the subpanel is then fed back to the main panel. Thus I initially installed current sensors on both inverter feeds to the subpanel. It didn't work.

Today I delved into it more deeply, using both the current sense transformers and calibrated AC/DC current probes connected to an oscilloscope. Here's what I saw:

East inverter display: 4728 watts (thus 19.7 amps at 240 volts)
West inverter display: 2766 watts (thus 11.5 amps at 240 volts)

Thus total current out based on the inverter displays should be 31.2 amps.

Total current measured leaving subpanel: 30.7 amps
Current measured from east inverter feed: 20.3 amps
Current measured from west inverter feed: 0 amps (no AC, no DC on either leg, no neutral in box or to inverter)

This, of course, was puzzling. I tried both legs on both inverters. East inverter 19-20 amps consistently. West inverter 0 amps. Tried it with both the current sense transformers and the scope probe; they agreed.

Then I tried shutting down the west inverter via the AC breaker. The west inverter current remained at zero. Total current out of the subpanel dropped to 20.1 amps. East inverter current remained the same.

I can't figure this out. The power is coming from somewhere but it does not look like it's coming in from the west inverter. Kirchoff can't be wrong. If I can't figure it out I will start measuring DC in and perhaps even break the AC line to insert a resistive sensor, but I can't imagine that there is any sort of current that can "fool" an inductive current sensor; in this case, Ampere can't be wrong.

Anyone have any ideas?

Comments

  • BB.
    BB. Super Moderators, Administrators Posts: 33,431 admin
    Re: Mystery current in GT inverters

    Broken sensor (or wiring), or both legs going through the CT would give false/bad readings.

    The only other I can think of--You mention AC/DC sensors. Is something set to DC measurement (would be net zero current with many DC meter).

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • bill von novak
    bill von novak Solar Expert Posts: 891 ✭✭✭✭
    Re: Mystery current in GT inverters
    BB. wrote: »
    Broken sensor (or wiring), or both legs going through the CT would give false/bad readings.

    Same sensors give accurate readings on east inverter. All attached to one wire at a time.
    The only other I can think of--You mention AC/DC sensors. Is something set to DC measurement (would be net zero current with many DC meter).
    Looked at that, but the probe is good for DC and AC up to 1 MHz. Again it sees the current through the east inverter accurately.

    Some pictures:
    Subpanel showing three(!) current sensors - the simple ones (black) the original monitoring transformer (donut) and the probe.
    Subpanel output ~30 amps
    West inverter output ~20 amps
    East inverter output 0 amps

    Attachment not found.Attachment not found.Attachment not found.Attachment not found.
  • BB.
    BB. Super Moderators, Administrators Posts: 33,431 admin
    Re: Mystery current in GT inverters

    Why a pair of blue wires from the left and a pair of "grey" wires from the bottom of the box going to the same breaker pair?

    I assume this is simple single channel measurements--No A-B with B being the wrong polarity (all looks single channel "1" although you are trigging on channel 2?).

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • bill von novak
    bill von novak Solar Expert Posts: 891 ✭✭✭✭
    Mystery solved
    BB. wrote: »
    Why a pair of blue wires from the left and a pair of "grey" wires from the bottom of the box going to the same breaker pair?

    I was about to say "the thinner gauge black wires go to the monitor circuit from the previous owner and the heavier gauge goes to the inverter." Then I realized that was an assumption I made, so I checked it and . . . the thinner wires DO to go the inverter, which explains the mystery. But does lead to the question - why are the 14ga wires carrying the inverter power and the 10ga wires carrying a few milliamps for the monitor? Fortunately max inverter current is about 14 amps.

    Thanks for the question that made me check my assumptions.
  • SkiDoo55
    SkiDoo55 Solar Expert Posts: 414 ✭✭✭
    Re: Mystery current in GT inverters

    BB good catch on double stacked wires on L/H CB's. Where do they feed too?
    Question why No neutral wire?
    Could the donut current transformer be interfering with the clamp probe? May be generating enough magnetic field to cancel reading, just something to consider.

    looked up info on sun-power older inverter the one I found pictures and spec sure read like a old Zantrex.

    Looks like a big drop in RMS from output @ 307 to inverter showing 20 amp of @ 198
    need some voltage measurements. L1 - L2 and this is where a neutral come in for checking L1 - N, L2 - N of each inverter. Input and outputs of each inverter with the opposite one breaker off.

    sounds like it is actually working but something is messing with your amp clamp readings.
    GT3.8 w/4600W Trina 230W, TX5000 w/5000W ET-250W, XW4024 w/1500W ET-250W, 4 L16, 5500W Gen. (never had to use) Yet!!
  • BB.
    BB. Super Moderators, Administrators Posts: 33,431 admin
    Re: Mystery current in GT inverters

    Those 14 awg wires are too small. 0.80 * 15 amps = 12 amps continuous loads maximum. And 20 amp breaker is not "legal" for 14 AWG cable (for some reason, in NEC 14 AWG is rated to 20 amps, but they then derate it to 15 amps--Never found the justification for doing this. They do this same derating for 12 and 10 AWG cables too).

    The reason for no Neutral connections. The older GT inverters do not have a neutral connection (just L1-L2 for 240 VAC). The newer ones do have a neutral connection--But that is just used by the Inverter to qualify that the neutral is "good" (checks for ~120 hot to neutral). I had a newer GT inverter installed for a little bit, and had no neutral. The installer just connected safety ground to neutral input on the inverter (neutral to ground are "close enough" for otherwise working system).

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • bill von novak
    bill von novak Solar Expert Posts: 891 ✭✭✭✭
    Re: Mystery current in GT inverters
    BB. wrote: »
    Those 14 awg wires are too small. 0.80 * 15 amps = 12 amps continuous loads maximum. And 20 amp breaker is not "legal" for 14 AWG cable (for some reason, in NEC 14 AWG is rated to 20 amps, but they then derate it to 15 amps--Never found the justification for doing this. They do this same derating for 12 and 10 AWG cables too).
    Yep. Looks like I have some work to do this weekend. Serves me right for making assumptions.

    I assume that the 20 amp rating you mentioned above comes from NEC table 310-16. Where are the deratings called out?
  • BB.
    BB. Super Moderators, Administrators Posts: 33,431 admin
    Re: Mystery current in GT inverters

    I have not cracked open an NEC book in decades to know what is there "Today"--But from one of the online excerpts:

    Wire
    Current Ampacities NEC Table 310-16

    1 The load current rating and the overcurrent protection for conductor types shall not exceed 15 amperes for 14 AWG, 20 amperes for 12 AWG, and 30 amperes for 10 AWG aluminum and copper-clad aluminum after any correction factors for ambient temperature and number of conductors have been applied

    Fixing your setup... Need to rewire to 12 AWG (minimum) for the 2nd AC GT inverter.

    And--The 12 AWG blue cable sharing the 20 amp breaker + ~14 amps from the GT inverter--That gives the blue wires a source capable of ~34 amps total if there is a short or excessive load... Way too much for 12 AWG cable. That blue cable should be a) disconnected, b) box upgraded with a second pair of 240 VAC breakers, or something else (hack in some fuse holders for the blue cable--But that is not something you would really like to do).

    Some breaker brands/types are available in "thin line" versions... You can pull the two ganged single pole breakers and replace them with a pair of ganged two pole breakers (assuming the box is rated for this, and the up stream breaker and wiring in the main panel was done correctly).

    Oh the joys of working with N+1 power/redundant power supply systems.

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • SkiDoo55
    SkiDoo55 Solar Expert Posts: 414 ✭✭✭
    Re: Mystery current in GT inverters

    Bill. If this is an inherited system, my best advice would to do a complete review of your system from the panels to your main panel. Checking for correct wire and gauge for amp capacities and being in conduit. If there are string combiners, with correct fuse protection if required. Check the Inverter DC inputs and AC outputs for correct and complete wiring and again correct wire for in conduit and amp capacities. Normally GT inverters are wired based on output of there max continuous rating, so you will need to refer to there manuals or hopefully it's still on a information tag on the inverter.
    A new 6 space 100 amp capacity load center is pretty cheap at Home Depot/Lowes etc. One that I used was a HOM612L100RB. If you don't need outdoor Rain-Tight then I think they have an indoor rated version. It will take 3 standard 2 pole 240 VAC breakers.

    if you have pictures of the inverters and there model numbers would be curious about them.

    Be Safe.😊
    GT3.8 w/4600W Trina 230W, TX5000 w/5000W ET-250W, XW4024 w/1500W ET-250W, 4 L16, 5500W Gen. (never had to use) Yet!!