Sol-Ark with EG4 PowerPro - coms working but now pullis from the grid when it shouldn't

[Kylepharmd]

I just recently setup a Sol-Ark 12k with a EG4 powerpro (with the help of an electrical engineer). The system is grid tied and permitted for net metering and I have TOU setup so that once the sun is down it uses the battery instead of the grid. Overall goal for the solar system is to run load, charge battery and sell excess back to the grid and then run load off of battery most of the time (with the grid as back up for the occasional scenarios where we don't have enough battery to get us through the night or some cloudy days). Today I was able to get the closed coms to work between the two.

However, I still have one puzzle. For some reason during the evening the sol-ark is pulling 10-20w from the grid while running the majority of the load from the battery. I have the load limit under TOU set well above the current load and yet for some reason it pulls very low wattage from the grid (see below).



It seems to do something similar when PV is running during the day. Today I was pulling 4.5kW of solar into the inverter which was being used to charge the battery, but then the sol-ark was still taking anywhere from 50-250w from the grid to run the load.

r
That purple line that is flat until about noon is the grid and it's showing a draw of 250w (and then later 50w) from the grid during the morning while the battery was charging. You can see once the battery hits 100 SOC it starts sending electricity back to the grid until the sun goes down. Then it turns flat again, this time pulling about 10-20w from the grid (even though the battery is charged and capable of meeting the relatively low load of <2kW.

I talked to someone at sol-ark today briefly who told me the EG4 BMS messes the sol-ark up so it's best to remove the closed communications between them, but I thought others had reported using EG4 successfully with sol-ark before so I wanted to see if anyone can point me in the right direction here. Are there disadvantages to disconnecting the closed communciations between the two of them? I like that the percentage shown on the sol-ark matches what the battery bms is actually saying, but are there other advantages beyond that? Is there any way to do what I want while not disconnecting the coms between the two?

Thanks in advance for your help!

 

[slypig]

Have you taken a look at the 'Settings' in Powerview for the inverter?

I think it might be doing this...... 'Zero Export Power'

Go to Inverter
find the three dots on the right
click to get menu
Choose Parameters Setting


(can't see it above.. yeah it's small) close up


Choose 'System Work Mode'


I'm using 'Limited to Home'
What is your Zero export power' set to?


Is this setting set to something between 10-30?

If so, I think you've got it, if not, then it's something else.

 

[Quattrohead]

It's a known and good thing that inverter is doing this, it stops the inverter accidentally pushing any power back to grid when it shouldn't.
Also it keeps it synced up to the grid for instant switch over on overload or ups mode.
Do you have the CT clamps installed and configured correctly ?

 

[Nobodybusiness]

I just recently setup a Sol-Ark 12k with a EG4 powerpro (with the help of an electrical engineer). The system is grid tied and permitted for net metering and I have TOU setup so that once the sun is down it uses the battery instead of the grid. Overall goal for the solar system is to run load, charge battery and sell excess back to the grid and then run load off of battery most of the time (with the grid as back up for the occasional scenarios where we don't have enough battery to get us through the night or some cloudy days). Today I was able to get the closed coms to work between the two.

However, I still have one puzzle. For some reason during the evening the sol-ark is pulling 10-20w from the grid while running the majority of the load from the battery. I have the load limit under TOU set well above the current load and yet for some reason it pulls very low wattage from the grid (see below).

View attachment 213395

It seems to do something similar when PV is running during the day. Today I was pulling 4.5kW of solar into the inverter which was being used to charge the battery, but then the sol-ark was still taking anywhere from 50-250w from the grid to run the load.

View attachment 213397r
That purple line that is flat until about noon is the grid and it's showing a draw of 250w (and then later 50w) from the grid during the morning while the battery was charging. You can see once the battery hits 100 SOC it starts sending electricity back to the grid until the sun goes down. Then it turns flat again, this time pulling about 10-20w from the grid (even though the battery is charged and capable of meeting the relatively low load of <2kW.

I talked to someone at sol-ark today briefly who told me the EG4 BMS messes the sol-ark up so it's best to remove the closed communications between them, but I thought others had reported using EG4 successfully with sol-ark before so I wanted to see if anyone can point me in the right direction here. Are there disadvantages to disconnecting the closed communciations between the two of them? I like that the percentage shown on the sol-ark matches what the battery bms is actually saying, but are there other advantages beyond that? Is there any way to do what I want while not disconnecting the coms between the two?

Thanks in advance for your help!

Mine does the same thing and I have never asked why.
All the settings seem correct.
Turn the trickle to 10watts.

It’s a small amount being drawn but constant.

Be interested to hear from other 15k owners on this.

I just turn off the grid to the inverters if I know I won’t need any.

 

[Kylepharmd]

I think it might be doing this...... 'Zero Export Power'

Bingo!

It's a known and good thing that inverter is doing this, it stops the inverter accidentally pushing any power back to grid when it shouldn't.
Also it keeps it synced up to the grid for instant switch over on overload or ups mode.
Do you have the CT clamps installed and configured correctly ?

So should I NOT set this zero export to 0? Keeping a small draw under 20W is advantageous for these reasons?

I believe the CT clamps are installed/configured correctly (that was done by the electrical engineer), but how might I check?

So that solves the nighttime component of my question, but I still can't figure out why it's pulling from the grid when the sun is out (and pulling from grid at wattage exceeding the zero export setting mentioned above):

This is right now in the morning making ~1300W PV with a load of only 267W. For some reason the sol-ark is pulling 63W from the grid even though under system work mode the energy pattern is set to "Load First" and the PV production should be more than enough to cover the load. Any thoughts what's going on here?

 

[AshleyL]

*DELETE* My mistake, I didn't fully read.

 

[Quattrohead]

Yeah try tweaking that zero export power both ways and see what it does with it.

 

[slypig]

My “grid” readings on powerview do not always display what is really going through them, either import or export. I know this because I have two devices that have CTs between the main breaker panel and the grid meter that agree on what is going where and by how much, within 1-4%.

This might be because my inverter still needs some updates, but my reason for having the inverter is only for :

1. Store as much excess solar as I have storage space for in my batteries.

2. Use that stored energy for the home until the next day’s solar production starts again.

 

[Kylepharmd]

Well as an update, I tried making a few changes and am now at the following.

I think it might be doing this...... 'Zero Export Power'

This was in fact the case. When I set the value to zero, the import when running on battery goes down to zero.

Why the sol-ark imports low wattage during the day when the sun is out still perplexes me.
As an example:


PV is making more than enough to run the load and charge the battery, but I usually see import from grid between 30-70W all day long.

Interestingly, on the sol-ark screen itself, it will look something like this (sorry for glare):


The glare is a little confusing, but you can see 1670W PV generated,, 420W to the load, and 1130W to the battery.

It is also showing 40W at the grid, but interestingly there is no arrow indicating the direction (although since it's a positive number I assume that means it's pulling from the grid and that would match the PV pro app).

This seems to ONLY occur when the PV is running and the battery is charging. As soon as the battery is charged up, the PV runs the load and exports to the grid (via net metering) with no problems. Once the sun is gone, load runs completely off the battery with no grid import (as long as 'zero export power' is set to 0). It's just that period where sun is up and battery is charging that I'm getting this low wattage import from grid, even when PV is more than enough to cover load.

Any idea what's going on here?

 

[Quattrohead]

That reading is within a percentage of error, remember that current sensor reads up to 60A.

 

[Kylepharmd]

That reading is within a percentage of error, remember that current sensor reads up to 60A.

Interesting. So if it's showing 40W (presumably at 240V?), that's <0.2 amps....

It is odd that it doesn't show a direction on the sol-ark screen itself though. I can't find anything in the documentation to suggest what that is supposed to be indicative of...

 

[slypig]

Each time I’ve reached out to sol-ark, I get the feeling that information and documentation is a constantly fluid area. One tech asked me if I wanted a document outlining more details on a certain area.

felt like saying “no, send me everything on all topics related to this unit, not one”

 

[Kylepharmd]

Each time I’ve reached out to sol-ark, I get the feeling that information and documentation is a constantly fluid area.

I know the feeling. After not getting anywhere on the phone twice (other than them telling me to disconnect BMS) I tried emailing support. Will see what they have to say...

 

[Kylepharmd]

Update: after emailing support, they gave me a call and suggested it may be because I have CT sensors installed unnecessarily. I don't have a separate critical loads panel (sol-ark load output just feeds the entire panel). In my case, he suggested the CT sensors aren't needed since I don't have separate loads being run from the grid that need measured. I'm not able to shut everything down to disconnect the sensors now to see if this will resolve it, but I guess it makes sense. In my situation he suggested the CT sensors are feeding the sol-ark info it doesn't really know what to do with and since the currents are so low, it could just be noise or just related to the margin of error on the sensors as @Quattrohead suggested.

Will try to disconnect the CTs sometime this week and see if it resolves the issue.

 

[Quattrohead]

You definitely need those current clamps connected, that's how it knows how to fine-tune it's output compared to the load. They should never have told you that.

 

[Kylepharmd]

You definitely need those current clamps connected, that's how it knows how to fine-tune it's output compared to the load. They should never have told you that.

Wow, if that's the case that is pretty frustrating it came from Sol-Ark's "Technical Support II Specialist".

He indicated that the sol-ark has it's own sensors built into the load output on the inverter and the clamps are just for when you have a critical load separate from the full panel where the sol-ark feeds the critical load (which it can measure output directly) and uses the clamps to measure the additional load from the rest of the panel since it can't measure it directly.

The 12k manual itself only mentions using the CT sensors for Meter Zero and Grid Peak Shaving.

 

[Quattrohead]

Someone is confused, they are for your incoming grid connection if you are using it to supplement your batteries and loads, but don't want to ever feed back to the grid. They have nothing to do with a critical loads panel.

 

[Kylepharmd]

Someone is confused,

It's probably me

they are for your incoming grid connection if you are using it to supplement your batteries and loads, but don't want to ever feed back to the grid. They have nothing to do with a critical loads panel.

To you point, though, I have net metering in place and am not worried about back feeding the grid. The only reason I was initially concerned was that it appears the opposite was happening when charging with PV (i.e. PV running enough to cover load and charge, but sol-ark pulls small amount from the grid anyway).

So since I'm not worried about backfeeding when using the grid to supplement loads, you still think the CT sensors are necessary?

 

[LydMekk]

The CT should always be installed.
On the Live AC wire coming from the grid into the inverter.

 

[Kylepharmd]

The CT should always be installed.
On the Live AC wire coming from the grid into the inverter.

I'm not disagreeing with you, but is there any documentation of this somewhere? I don't see it in the manual as being required outside of meter zero and peak shaving situations. I can email support back but it would be nice to point to some of their own documentation when writing to them ( and I can't find anything specific in the manual).