Preventing inverter operation under low light conditions

[cyd]

I have an issue with a 5kva AIO inverter waking up when the morning sky gets bright. It flicks between trying to run on hopelessly low PV energy and draining the battery bank.

Then in the evening as the sun goes down the same problem happens.

So I wrote a program that senses the PV voltage using a PIC microcontroller.

It controls the PV panels via a 2.5KW DC solid state relay on the positive PV line (I use an east-west array and the inverter is deliberately oversized)

The PV panels are connected when PV OC voltage is > 117 VDC and disconnects when load voltage is < 85 VDC.

A time delay of ten seconds allows for brief low PV voltage excursions. If the inverter cannot pull itself from draining the battery after ten seconds the PV arrays are disconnected for 20 minutes. The inverter goes to standby and switches to bypass mode. In the event of a power failure it will of course have full battery capacity available.

 

[LakeHouse]

What am I missing here? I don't think I understand what problem you're tying to solve.
Yours might be different, but my particular AIO operates as though the inverter part and the charge controller part are separate components that just happen to be in the same box. In fact, turning off the AIO 'power' switch only turns off the inverter, not the charge controller.
Running the inverter under no load burns about 50-60 Watts, but running the SCC with no solar power (and the inverter off) uses something like 2 Watts. And importantly, disconnecting the solar panels doesn't reduce the SCC power requirement at all, so it seems like you're probably using more power monitoring the voltage and operating a disconnect than you're saving by disconnecting the PV panels.

 

[OM617YOTA]

The MPPTs in my LV2424s will burn nothing when there's no light at all, but if there's enough light to generate a signal and wake up the MPPTs, then the MPPTs will burn a few watts each until enough power is harvested to overcome that. Dawn and dusk will do this, and so will very overcast days - like right now. This is my MPPTs alone drawing power, inverters are off.

I like what the OP has done, but the same phenomenon hasn't drawn enough power to be of concern to me.

 

[Crowz]

What am I missing here? I don't think I understand what problem you're tying to solve.
Yours might be different, but my particular AIO operates as though the inverter part and the charge controller part are separate components that just happen to be in the same box. In fact, turning off the AIO 'power' switch only turns off the inverter, not the charge controller.
Running the inverter under no load burns about 50-60 Watts, but running the SCC with no solar power (and the inverter off) uses something like 2 Watts. And importantly, disconnecting the solar panels doesn't reduce the SCC power requirement at all, so it seems like you're probably using more power monitoring the voltage and operating a disconnect than you're saving by disconnecting the PV panels.

I remember there being a thread or two on the forum here where certain aio's had the issue the op is talking about and I think it effected more than one particular line of aio's. One line of them was fixed with a firmware update and the other I never saw a fix for.

I can't remember which ones it was though.

 

[pollenface]

In the past I have used a 10w solar panel connected to a 12v relay.

You could also use a photo sensing relay adjusted to trigger when the sunlight is strong enough.

 

[cyd]

My AIO is dumb and assumes that a high PV OC voltage means that there are plenty of amps available.

For example it tries to run a 450 va load from PV at 6am when the sun hasn't even risen above the horizon. All because at 6am the PV OC voltage is 118v.

So I created this solution to assist the AIO make smarter choices, i.e PV voltage-under-load must not sag excessively under certain low light conditions.

 

[pollenface]

So I created this solution to assist the AIO make smarter choices, i.e PV voltage-under-load must not sag excessively under certain low light conditions.

Ah, so problem solved?

 

[CoolWill]

Have you actually tried this? I ask because the OC voltage can be quite high but the available amps be low. Seems like relying solely on voltage would result in the same issue.

 

[cyd]

Have you actually tried this? I ask because the OC voltage can be quite high but the available amps be low. Seems like relying solely on voltage would result in the same issue.

The circuit I designed and programmed is working well.

The inverter remains in bypass mode until the circuit detects a PV OC voltage rise above 117v.

Then the SSR is activated and the inverter swaps to powering the load in PV mode

As long as PV voltage-under-load remains above 85v the circuit holds the SSR on