diy solar

diy solar

Sol-Ark 15k will only invert 13k of DC PV to AC

This is a site issue, and the cause could be a momentary rise in the grid voltage as seen by the inverter as it is the source, not load triggering as UL1741-SA power reduction as I linked in the previous post.

Your graph above are DC values, I'm discussing AC values as the trigger. We don't need to prove the mppts are the cause, they are just how the inverter throttle the harvest. There maybe a separate issue of how the unit handles edge of cloud events, ( the zeroing and resuming of power delivery ) but one issue at a time

While turning on loads some where else could help, it may not if the issue is rise in voltage as the inverter sees it with wiring from the inverter to the main panel.

Get yourself a DVM that has peak hold function, this will see detect and hold the reading. Here is one on amazon the might do the job, but you can search using digital multimeter hold as the key words.

Adding a large load, at the inverter on the grid port itself should be the next test and or having a peak hold to measure the AC voltage. The data logging you have just points to the direction, 5 minute logging can't do a smoking gun on issues like this.
 
I’m not sure this has anything to do with it but in some of your attachments I see your load curves all “spikey” and not relatively smooth. This is not how mine appears at all. Not sure why but definitely different than mine. I have load spikes but they are relatively smooth. Probably nothing to do with it but it is definitely different
 

Attachments

  • 9AAAB2E4-DC62-4DA5-A852-90249ADC1D13.png
    9AAAB2E4-DC62-4DA5-A852-90249ADC1D13.png
    263.6 KB · Views: 23
  • 19566CEA-7657-4795-90B8-26FCCE5EBA28.png
    19566CEA-7657-4795-90B8-26FCCE5EBA28.png
    422.1 KB · Views: 19
  • 7FC4BCD3-7448-4181-AD42-5F3153849311.png
    7FC4BCD3-7448-4181-AD42-5F3153849311.png
    982.9 KB · Views: 19
Clint - do you have batteries connected? In closed loop or open loop? When my batteries batteries are in closed loop is when I get the drops and spikes. In open loop all is well.
Solar Guppy - In my case I asked Carlos at Sol-Ark about monitoring the AC power. That I could hook up a scope that records. His response was that the AC side is not where the problem is. It was after this he determined that the battery closed loop communication had something to do with it.
jfpetesn - Wow, it's good to see how these graphs are supposed to look.Do you have batteries?

Perhaps these drops and spikes can have multiple causes? Either a short duration event on the AC or the Sol-Ark responding incorecctly to a battery event. Or a battery event that should not happen and the sol-ark is responding correctly?

Clint took some power off the DC side by using microinverters. For a couple months I took one 400w panel out of each of the 2 PV strings. During this time the drops and spikes were significantly reduced. While the batteries were in closed loop mode.
 
Clint - do you have batteries connected? In closed loop or open loop? When my batteries batteries are in closed loop is when I get the drops and spikes. In open loop all is well.
Solar Guppy - In my case I asked Carlos at Sol-Ark about monitoring the AC power. That I could hook up a scope that records. His response was that the AC side is not where the problem is. It was after this he determined that the battery closed loop communication had something to do with it.
jfpetesn - Wow, it's good to see how these graphs are supposed to look.Do you have batteries?

Perhaps these drops and spikes can have multiple causes? Either a short duration event on the AC or the Sol-Ark responding incorecctly to a battery event. Or a battery event that should not happen and the sol-ark is responding correctly?

Clint took some power off the DC side by using microinverters. For a couple months I took one 400w panel out of each of the 2 PV strings. During this time the drops and spikes were significantly reduced. While the batteries were in closed loop mode.
Yes I am running 928 aH of Trophy batteries open loop and 13.4 kW of DC coupled array.
 
I’m now running EG4-LL stack of 6 of them and a total of 609ah. Interesting about the closed loop. We did some testing around it with solark a while back and couldn’t find an issue. We set them up as lead acid etc.

Also, when I first got the original 15k I had an old lead acid battery bank and was seeing this issue as well.

Today, I disabled grid charge, which I like to have because I do take advantage of TOU during peak utility grid charge times to supplement my load and then charge back with cheap power at night. The only reason is, so if I had a power outage early in the morning I don’t want to have dead batteries. That would really suck after dumping so much cash. lol ?

We are having a cloud/storm system here in Oklahoma for the next few days so I won’t be able to get much useful data until this moves out. But I’m curious about maybe letting my batteries pull a healthy load like was suggested earlier in thread. That would be great, even though just selling back to the grid would be better!

One last thing. My grid voltage usually stays at the following:

At rest - no PV production
L1- 122.0
L2- 122.5

Moderate PV production
L1- 123.8
L2- 124.0

A few times a day spiking during full PC production and system throttling out at 13.2~kw
L1- 125.4
L2- 125.3
 
Mine is either at 118-120V per leg When under normal load or may dip slightly under a heavy load.

It would be interesting to hear what other 15k owners are seeing.
 
One last thing. My grid voltage usually stays at the following:

At rest - no PV production
L1- 122.0
L2- 122.5

Moderate PV production
L1- 123.8
L2- 124.0

A few times a day spiking during full PC production and system throttling out at 13.2~kw
L1- 125.4
L2- 125.3

That's an issue, those values indicate a resistance of .103 ohms ( rise of 6.2V Volts @ 60 amps ). That's really high.

You need to find where the line rise is located from the inverter to the utility meter ( 251 is very close to the typical 254 trigger point in some UL1741 parameters ). Since this is a UL1741-SA Inverter there is automatic derating also based of active and reactive AC power.

Start at the main breaker and work towards the inverter with the voltage measurements, you will likely find a point that shows the most voltage increase.

Can you describe the wiring from the utility connection to the inverter?
 
Last edited:
My grid power has always been high at ~249-253V (even before my solar install).

With 2x15K's on 28 Jan:

At rest - no PV production
L1- 125.0
L2- 124.7

Highest sample during PV production (13 kW production w/ 10.6 kW export)
L1- 126.6
L2- 126.7

No alarms or even dark shaded measurements on the Master's details screen. Running the default grid setup so UL1741-SA and high voltage faults start at 263.8V.
 
My grid power has always been high at ~249-253V (even before my solar install).

With 2x15K's on 28 Jan:

At rest - no PV production
L1- 125.0
L2- 124.7

Highest sample during PV production (13 kW production w/ 10.6 kW export)
L1- 126.6
L2- 126.7

No alarms or even dark shaded measurements on the Master's details screen. Running the default grid setup so UL1741-SA and high voltage faults start at 263.8V.

UL1741 SA has auto derating of maximum power to grid ( on active and reactive power each with different trip points ) before the high voltage trip values, that's what I am thinking about here.

We have two sites, both running high line side AC voltage that both are throttling Maximum PV harvest to grid.

One test would be if the unit allows different grid profiles such as just UL1741 ( not SA ) and see if it resolves the issue.
 
We have two sites, both running high line side AC voltage that both are throttling Maximum PV harvest to grid.
Oh no no no...I had home loads and charging going on that ate up the missing 2.4 kW. I don't believe it was throttling judging by the PV curve and the intermittent clouds that were rolling by that day. The earlier peak was 17.3 kW of production but of course that's two 15K's combined and a 20.4 kW array (total). Output/Grid Voltage during that time was L1-125.8 L2-126.0.

I just posted to add data to the high(er) grid voltage behavior/rise.
 

Attachments

  • Snap.gif
    Snap.gif
    34.3 KB · Views: 10
Last edited:
Question - if it was related to grid voltage would that mean that the upper PV production limit would be limited based on what that voltage was? I’m not seeing any correlation on that. Here’s a graph with PV and L1/L2 voltage. Thoughts?
 

Attachments

  • A8C31F1D-AECE-4C89-91A2-EBA5C9B3C7C7.jpeg
    A8C31F1D-AECE-4C89-91A2-EBA5C9B3C7C7.jpeg
    109 KB · Views: 18
It is clear, when you AC grid voltage rises, the clipping starts, I put pointers and a box on your photo

You have an issue on your site voltage, you are having a 6-7 volt rise on the grid port, when it should be more like 1 to 2 volts at most.

I've written this numerous times, you need to find out from the inverter to the main breaker WHY you have this much voltage rise, it is NOT normal

Can you please detail the path from the inverter grid port to the main breaker panel, wire gauges and lengths.A8C31F1D-AECE-4C89-91A2-EBA5C9B3C7C7.jpeg
 
It is clear, when you AC grid voltage rises, the clipping starts, I put pointers and a box on your photo

You have an issue on your site voltage, you are having a 6-7 volt rise on the grid port, when it should be more like 1 to 2 volts at most.

I've written this numerous times, you need to find out from the inverter to the main breaker WHY you have this much voltage rise, it is NOT normal

Can you please detail the path from the inverter grid port to the main breaker panel, wire gauges and lengths.View attachment 132518
I think your on to something.
 
It is clear, when you AC grid voltage rises, the clipping starts, I put pointers and a box on your photo

You have an issue on your site voltage, you are having a 6-7 volt rise on the grid port, when it should be more like 1 to 2 volts at most.

I've written this numerous times, you need to find out from the inverter to the main breaker WHY you have this much voltage rise, it is NOT normal

Can you please detail the path from the inverter grid port to the main breaker panel, wire gauges and lengths.View attachment 132518
I often think that it would be helpful if everyone posted a one-line diagram of their setup. I'm curious if that voltage fluctuation is being caused by the utility company or the inverter.
 
Without the solark on I have a perfect 121.3 on both L1 and L2.

The grid side of the solark connects directly to the service shutoff switch for my property. That box connects to meter. All 3/0 wire.

My meter service is a 325a (400 they call it).

It’s dark out so I’ll try to see what the transformer is soon. It’s about two years old and doesn’t feed anything but my shop and house.

It sound like several folks feel like we are on a good path for more testing now in this. Maybe I can get my power company to dial down the voltage on my transformer if that’s possible. Of course even it it is possible they may not do it.

My solark is set in general standard by the way. We’ve a had it in UL1741sa as well with same outcome. I’ve never tried ul1741/IEEE whatsoever it is though.

Another consideration is that this does happen whether or not I have microinverters enabled and flowing power in to the microinverter (smart load) input of the Solark 15k. During these higher voltage times we are seeing, the micros have no issue. They produce perfectly. I’m producing about 2kw max power input with my 4 micros online. If I pull them out, the high line voltage still persists and it was all happening (the PV flatline clipping at 13.2kw) before enabling the micros.
 
Last edited:
Pulling a 13kw load (grid charge) would cause a voltage drop, if it's a wiring/connection or supply issue.

Just thinking of a test that didn't require full sun.

And as @Solar Guppy says, check voltage at the inverter to mains or as close as you can to the utility. If applying that load causes a voltage drop at the main, a Blinking Light call to the utility should be made.

Where I work, 350kcmil Al is run for a 320A meter socket and appropriately sized transformer, a 400A should be all of that. We sometimes contract out work and find that some of those guys think a piece of solid #6 cu will power the world, so the utility may have something going on.
 
Without the solark on I have a perfect 121.3 on both L1 and L2.

The grid side of the solark connects directly to the service shutoff switch for my property. That box connects to meter. All 3/0 wire.

My meter service is a 325a (400 they call it)..

What is the distance from the shutoff to service transformer?

What is the wire gauge on the inverter to shutoff?

There is no way that service could support 325a based on the Voltage swings your seeing.
 
Last edited:
Back
Top