diy solar

diy solar

My SolArk 12k install with Enphase Micros

The issue I am dealing with now is that I have Solar Assistant running but cannot get it to communicate with the SolArk. I get the message, "No response". I am using the cable I purhased from Watts247. I am wondering if I need to restart th inverter for it to see the USB port. I saw that Ian did that in one of his videos.
EDIT: It was a simple fix. I found the answer on the solar-assistant.io help page. Instead ot the RS485 port the cable goes into the CAN port on the SolArk 12K.
 
Last edited:
Now that I have Solar Assistant running I am able to tweak my TOU settings and see the results immediately. I changed to Voltage settings instead of Battery % settings because voltage is more consistent for my purposes. For peak times when I want to use the battery I set the voltage during that time to 52 Volts (3.25V per cell). During the current season I also want to charge the batteries from the grid starting at 3AM so I set the voltage to 54 V at that time so I would have a full battery in the morning. It seems to be working and I will give it a few days and then look at the daily charts to see what has been happening over time. It has been cloudy and rainy for the past few days so I am not getting much solar production.
 
Does this actually do anything..? I tried both options awhile ago and it seemed to make no difference (have 12k).
When you're producing less than your loads are using it determines whether the solar power generated will recharge your battery or offset loads. This is also dependent on time-of-use settings, so it may not be instantly apparent.
 
When you're producing less than your loads are using it determines whether the solar power generated will recharge your battery or offset loads. This is also dependent on time-of-use settings, so it may not be instantly apparent.
That is helpful if I understand it correctly. In my case, especially in winter, I want to prioritize battery charging to make sure my pack is full from my least expensive source (solar) to have enough in my pack to get me through the peak period each day.
 
I have been trying to get my Orion BMS to communicate with the SolArk and some of that process is contained in a thread started by another member.

My situation is slightly different because I have a very old Orion Jr. which does not have all the capabilities of the Jr 2 BMS. Orion support modified my profile and I am working on setting up the CAN communication cable. To be continued.
 
My new Orion Jr2 arrived and I spent a rainy day rewiring the BMS wires. Fortunately when I assembled my 3P16S pack using copper busbars I provided a separate tapped connection for the BMS wires so I could keep the pack working while I disconnected the old and connected the new BMS wires. The Orion BMS Jr2 also uses a Hall effect current sensor which means one less set of connections. The 4/0 cable just feeds through the sendor, which resembles a CT but without the removable section to allow it to slip over the cable. I got it all working late in the day and need to give it a couple of cycles to begin giving me a reliable SOC.
I may work on trying CAN communications today, which was my whole purpose in upgrading to the Orion Jr2.
 
I got CAN communications working and the great thing about that is I have much more confidence in my BMS SOC percentage numbers than what I was seeing on the SolArk. Also Voltage is almost exactly what i measure with my VOM and which my active balancer reports.
The value in using SOC% in programming the SolArk is that it is much more intuitive for me to look at that table and control where I want to the battery to discharge to at any time of day. My goal is to peak shave during the 3PM to midnight high rate periods. In winter I never have enough solar to consistently provide enough power because my heat source is a heat pump and I am using between 2-3kWh per hour. That amounts to forty to fifty percent of my battery. With higher chance of power outages I want to charge my batteries up to about eighty percent after 3AM.
In Spring through Summer and early Fall I use less power overnight and have a greater chance of covering that with solar the next day. Also the high rates and longer solar days in Summer allow me to build dollar credits so my October True Up zeros out.
 
Last edited:
I posted the procedure of getting my Orion JR2 connected to the SolArk but I thought I would also post it here for continuity.

The first step on the SolArk is making up a CAN communication cable. Any RJ 45 cable will do. Pin 4 is CAN Hi and Pin 5 is CAN Lo. This cable is connected to the CAN cable coming out of the Orion Jr2. ( Red is CAN Hi and Black is CAN Lo) No resister is needed. In the case of the SolArk 12k this is plugged into the RJ45 port labeled Battery CAN Bus. Details for other models can be found in the SolArk document titled "Battery Integration Guide". The battery communication needs to be set up on the SolArk by checking the Box "BMS Lithium Battery under Battery Setup. Make sure the address is 00. Once you are done you should see a new selection on the SolArk main screen. It won't be populated yet until the next step.
The next step is setting up CAN communications on the Orion. This is done on the Battery Profile page. Click on the CANBUS Settings tab and at the bottom of the page is a window in which you need to scroll down to "AC Inverter". Clip to drop down choices and click on Victron Inverter. After doing this be sure to upload the profile to the BMS from the programming interface,
Once you are done you should be able to see the BMS values in the new tab you added above in the SolArk. I just set mine up and have not reviewed all the values on the attached.
PXL_20230111_215131544.MOTION-01.COVER.jpg

For example, I am not sure why Charge current limit: 0A and Nominal_Cap: OAh. Battery charge voltage: 56.0 V , comes from a BMS setting and I will be watching how this relates to lower settings I have set on the SolArk.
EDIT: I checked the BMS and one of my cells had tripped the BMS so the BMS must have lowered the Charge limit. Now, later the battery is discharging, that cell is back under the limit and the Charge current limit is back to normal_
I see no correlation to the Ahr Capacity set in the Orion (800 Ah) and the above Nomical_Cap:0Ah. The Alarm digits do not correlate to anything I see on the Orion.

Last edited: Yesterday at 3:59 PM
SolArk 12k hybrid inverter with a battery pack of 42 kWh of LF 280 cells monitored by an Orion BMS. 7kW Mission Solar panels with Enphase IQ7+ microinverters AC coupled to the Skybox. An additional 2kW Sunpower panels DC coupled to the SolArk inverter.
Currently own two EVs and been driving EVs since 2012.
 
This is a screenshot of my last 24 hours.

2023-01-13sshot.png

Some explanation of my programming would help explain the chart. They legend is not visible so here it is:
Yellow is the solar on the left from about 9AM tapering to about 3:30PM
Blue is the load which is pretty consistent throughout the day. Most of the load is a heat pump and you can see it varies between 400 Watts and 1500 Watts.
Red is the grid. You can see the export from solar until about 3:30PM. The additional export after the solar shuts down is apparently the way I initially had the sell settings configured until 9PM. Apparently inverter will export from the battery based on the Watt settings when the xell box is checked. That box is not checked after 9PM. The spike at midnight is the battery charging up to 75% from the grid and again to 80% at 3AM. Then again in the morning to 100%, hence the taper in power until the battery is full. Normally Solar would start charging the battery beginning around 9AM but that day it was cloudy.
This particular chart does not show Battery SOC, Voltage or current going in but you can get a pretty good idea of SOC from the about. The SolArk allows seasonal settings and those were my winter settings. In Spring, Summer and early Fall my heating load will not be so great and the Probability of cloudy weather and power outages will be less so I won't need those early morning grid charging settings because the battery will probably only go to 70-75 percent and will easily charge to 100 percent before Noon.
 
For graph analysis use you would be much better off using Powerview on a PC.
It graphs so many more items together and gives you a much better sense of exactly what is happening at any point in time.
I find Solar Assistant to be great for realtime data but not analysis and programming the Inverter.
 
I just looked at Powerview on my laptop and it is informative. This is all new compared to what I have had in the past. I love the data.

I am not going to invest much time in it until the new app comes out. I will have some useful data by then as well.
 
I ran into an interesting problem not related to the SolArk but related to my Enphase Encore. About three days ago it stopped reporting production and lost communication with the micro inverters. The micros continued producing power according to Solar Assistant, The SolArk and my Emporia Energy Monitor. I thougtht maybe it was something in the Gen Port that was preventing the PL:C communications. I jumpered the AC coupled micros over to my critical loads subpanel but that did not change anything. I hint from an Enphase led me to suspect EMIs based on that suggestion about faulty flourescents. I had just eliminated some flourescents and replaced them with some four foot tube LEDs. I turned off the new LEDs and eventually communications was restored. Fortunately I can return the LEDs but it is a real disappointment because they put out a lot more light than the four foot flourescent LED replacements that required rewiring the old fixtures. I am not sure what the lesson to be learned is because other LED shop lights did not create the same issue.
 
Powerline Modulated Communication can be problematic with "noisy" switching power supplies or other devices that transmit on or near the same kHz range. If you wish to filter a device or even the entire Enphase circuit I've had good luck with TDK line filters. Here is a link to a 20A version and they have 50 and 60A filters also that could be used on the input or output of an Inverter.

 
I just requested a firmware update. I am on 6.2.1.6. The latest is 6.2.2.2 for the 12K
Here is a link for how to request an update.

As posted on another thread there are no changelogs posted so we are on our own as to what to expect.
 
Last edited:
I will try to remember to do that. I am hoping Solar Assistant will allow me to do that. Also @robby just posted about a new interface so that may also help. It is good I made the purchase now because the winter is the challenging time because loads are high and production is often down. It just started rainingba few hours ago and you will probably see it in the morning.
So what grid profile have you used to get the AC coupled Micros to work? Mine work fine on grid, but refuse to produce and power when AC coupled off grid. Have you tested this with grid down or you just use it to load shift you usage to off peak?
 
So what grid profile have you used to get the AC coupled Micros to work? Mine work fine on grid, but refuse to produce and power when AC coupled off grid. Have you tested this with grid down or you just use it to load shift you usage to off peak?
I have not yet been able to get the micros to work when grid is down, but I have not focused on that yet. It may have been as simple as having enough loads for the micros to serve. Some of that process is discussed in another thread here:
 
I recently connected my Orion JR2 BMS in closed loop to my SolArk and experienced an unexpected change in parameters. My preferred charging voltage is 3.45 per cell which translates to 55.2 volts. That was the setting for charging on my SolArk.
My Orion BMS has some settings which I had assumed were alarm settings and I set the voltage for max charging at 56 volts because I wanted it to be outside my inverter setting. After I connected the BMS to the inverter I was looking at some charts and saw that my battery voltage was peaking at 56 volts. I went in and changed the inverter settings back to 55.2 volts. After a few minutes I checked again and it was back to 56 volts. I realized that the BMS setting was overbidding the inverter setting. I finally changed that setting in the BMS to 55.2 volts and the SolArk setting went to 55.2.
The point is that when using closed loop communications, check your parameters to make sure there are not unintended consequences such as a change in a parameter. Also, in the event that communications between the BMS and the inverter fails, make sure you have a correct setting on the inverter as a backup.
 
I finally wired up a 16 Amp 240 volt receptacle to my SolArk via the critical loads panel so I could test the off grid AC coupling. My battery was fully charge\d so I did not have that load and the EVSE was only a 16 Amp device so the most my EV could pull was 3800 Watts. The micros went offline but the good news was the frequency was stable at 61 Hz. I am encouraged that I am going to get this working. I will try again in the morning when the battery charging is pulling 5kW and see what happens. The good news is that I was able to get the EVSE to be powered by the SolArk, which is something I never did with the Outback Skybox. That is not a slam on the Skybox because I was using another EVSE then. I may try to plug in a higher Amperage EVSE to test this further. I also have my micros on three strings of 4 Amps, 10 Amps and 14 Amps. That gives me some latitude to test various combinations as well. Ninety nine percent of the time the grid is up but we did have a three day power outage three years ago and I want to make sure I don't have any surprises. This is not a major issue but, I am retired and this gives me something to do until my significant other gets back from a business trip and the Honey do list gets longer.
:)
 
Last edited:
Back
Top