Kirkoff's Law Violated

[Martin4321]

The battery current in does not equal the battery current out. Can someone explain to me what is going on with my battery? Using a DC Hall Clamp current meter I measured the current going into and out of my battery bank. These wires go directly from my battery to the input of a packaged MPPT controller and 24 VDC to 240 VAC converter. I don't know what else is in the box. Also attached are the solar panel wires. I measured multiple times with the load plugged in and unplugged (the load is off). There is no path to ground (that I know of).

As shown in the attached photo, the Current on the negative wire is +0.985 amps and the current on the positive wire is -0.590 Amps (note the negative and positive values are a function of the direction I put the meter on; if I flip the meter over the signs change. This was taken at 11:00 pm so no solar power input. I took the reading because I wanted to see how many wats my Controller Box was using at night.

How can the current in not equal the current out? I took about ten readings and got very similar results.

[solarix]

clamp style DC amp meters are definitely not so accurate.
Batteries consume energy - especially as they age
Malfunction behind the eyepiece...

[bob-n]

Martin: You're right that it doesn't make sense.

You mention 11PM so no solar generation. Is it possible that the solar panels or something connected to them are consuming power? We're only looking for 0.2A * 24V = 4.8 watts. That's not a lot of power. A few LEDs or optimizers might use some power. Electronics is often designed to go to ultra-low power mode when the input is below a voltage threshold. Your batteries are keeping the input high, so it might be (really wild guess) that the electronics doesn't go to sleep.

Can you sketch a diagram of the whole network showing where the panels, etc. connect? There has to be some current sneaking in around your meter.

solarix suggests that clamp meters are not great. That's true. But this is a huge error. They shouldn't be that bad unless defective.

The charging circuit is probably putting out pulses due to PWM control. If these pulses are not well filtered, it could be that the high frequency signal is interfering with the meter measurement. The meter is marked "True RMS". That means that the meter has either analog or digital processing to compute the RMS value of the current waveform. This could be confused by a high frequency signal. In this context, high frequency is >1kHz. The meter is only expecting 60Hz variations.

Another possibility is that the meter has an offset. But your "flip" experiment means that there is no offset.

I assume that the meter is set to DC amps, not AC amps, for all readings.

Do you have anything that you can use to test the meter, such as an old-fashioned flashlight or a small battery and low-value resistor?

The only other thing I can think of is stray magnetic fields in the area. It looks like the meter has moved 2" or so. If there is a huge magnetic field from a transformer or other high-current device, that could interfere wtih accurate measurements.

But it's really hard to think of any of this causing that much error.

Do you have any sense whether 0.59A is correct or 0.985A is correct, or whether it is somewhere in between?

[PVAndy]

Has to be a measurement error. Without another current path the current in a series circuit has to be equal. As others have said, lets see the rest of the circuit.

If you have a 1 ohm resister put it in circuit and measure voltage drop.

Andy

[Martin4321]

I just went out to the garden to measure again. I have two of the exact same meters so I was able to make some measurements and compare them. It turns out the meters give different readings on the 2 amp scale when I place them 1 inch apart on the same wire! When I put it on the 20 amp scale I get 1.35 amps on both lines. This matches what the manufacture says it should be when idle (about 35 watts which is about 24 x 1.35 amps. When it is on the 2 amp scale it is also unstable giving me a different reading by up to 25% each time. It is very stable on the 20 amp scale.

I think the best explanation for what was happening was by solarix above......"A malfunction behind the eyepiece." LOL, We used to call it "a loose nut behind the wheel" back in the days when I used to work on my own car.