Random shutoff of Charge controllers

ChuckKautz

I have a grid tied system that allows net metering.  I have an 8100 W array with 2 Schneider Electric XW6048 inverters, 4 charge controllers,and battery for backup.  The battery is flooded 875 Ah and 48V.  The charge controllers seem to shut down randomly.  Sometimes all of them are working, sometimes none, and everything in between.  Is this a problem with the communications board or could it be something else?

Dave Angelini

Any faults or warnings?  Pretty old system, who installed it please?

BB.

Did the system work correctly in times past? Any changes recently? How old of system? What is the battery bank bus voltage during these issues?

Dave is the one to help you here--Just my basic debug questions.

-Bill

ChuckKautz

Dave,

There are no faults.  It was installed by Hill Country EcoPower that has now been renamed Native. 

ChuckKautz

Bill,

The system worked well for 8 years and then started having problems with the batteries as the perceived problem.  I hired Joe Kam who came out and sold me a new battery.  Since the new battery was connected 7 months ago, I have had all kinds of problems.  Joe abandoned me shortly after he convinced me to pay him.  (I know that was a poor decision).  I have gone to someone else who says that he thinks it is one or two bad communication boards in my two inverters.  I do not know what a battery bank bus voltage is (no luck on a web search either, was that a typo?)  Thanks for your and Daves help.  I have put in hundreds of hours of research and have the system running.  It seems the only issue that is current is the charge controllers shutting off when they should be sending electricity to my system.

BB.

The communications issue, while a pain, should be a secondary issue. Primarily, each charging source should be looking at the battery bus voltage and making "decisions" on how to charge.

And it is OK if the charging sources are not in sync if each is programmed correctly. I.e., you may have one charging source in float, while another is still in absorb. As long as the battery bank is "happy" (i.e., correct charging voltage at terminals, and absorb time is roughly 2-6 hours--shorter if shallow discharge, longer if 50% of more discharged), then that is a good start.

Get a good quality voltmeter and measure the DC terminal voltages of each device (when current is flowing--Charging, discharging, and resting). You should not have much more than a few tenths of volts difference between the battery bus voltage and the terminal voltages of each of the major DC devices (XWs, charge controllers, etc.). We are looking for "differences" here... You have a bunch of "matched" components and they should be all running about the same voltage/conditions. Any "differences" need to be further researched.

And look for any corrosion (can be from battery electrolyte, or even just rain water following a cable to a terminal block), loose DC wiring connections, and signs of overheating at the terminals (browning).

Also, I suggest getting an AC+DC Current Clamp Digital Multimeter... The current clamp is very easy and safe to use when measuring current flow--Just zero (DC clamp meters need zeroing) and clamp on a single wire to measure current flow. Great for debugging and understanding what your system is doing.

https://www.amazon.com/UNI-T-Digital-Handheld-Resistance-Capacitance/dp/B0188WD1NE (cheap, good enough)
https://www.amazon.com/gp/product/B019CY4FB4 (nice mid-range cost meter)

I am not the person to talk about detailed debugging of the XW system and communications issues. I have seen here that a good place to start is to turn off the AC/DC power to each of the units (inverter, charge controller, etc.) for at least 5 minutes, and then restart in the proper order (DC bus on first, then connect solar array next, etc.).

Another thing that has sometimes helped... Resetting to factory defaults, then programming again (make sure you have all the setting written down). On occasion, the memory can be corrupted...

Other than that, check the comm bus wiring (inspect the connectors, make sure wires have not been pulled while replacing the battery bank, no corrosion, that wires and bus terminators are in place, etc.).

That is about my limit on the XW/solar chargers... 

Perhaps Dave or others can help with the details.

-Bill

BB.

Also, I should ask... What do you mean by "shutting off" randomly... Do the displays go blank, or do they say something else (0 Amps, or whatnot)?

And what is stopping charging... The MPPT solar charge controllers, the XWs, a combination of both? 

With networked gear--You have the behaviour of both the individual components, and the behaviour of everything networked together. You might find that one MPPT controller is acting "squirrely", and when connected to the network, the others start misbehaving too.

-Bill

PS: Any other events (like lightning in the area 7 months ago) that may have happened?

ChuckKautz

The MPPT solar chargers shut down, sometimes individually and sometimes together.  They register 0 W PV when it is nice and sunny out.

The event was the new battery placement, on an otherwise nice, uneventful day.

ChuckKautz

Thank you for all of the information.  I am learning a lot.

BB.

Probably the simplest place to start is with a voltmeter and measure the Vpanel input voltage and Vbatt battery bus voltage on each solar charge controller.

You want the Vbatt voltage at the controller and the battery bus to be less than 1/2 volt difference between the two.

The solar panel voltage, you would want it to be at least 1.3x the Vbatt voltage--Especially if you are reading 0 Watts (1.3x 59 volts = 76.7 Volts)... If you know the Vmp of the panels, then you should see at least Vmp-array*0.80 (hot panels, under MPPT loading).

If you have no current/power flowing, something closer to Voc-array (X number of panels in series * Voc -- Voltage open circuit).

-Bill

BB.

And one thing to emphasize--When connecting solar chargers--Connect the battery power first to the controller--And then connect Vpanel power... If you do it backwards (panels first, then DC battery bus), it can confuse the controller and in (usually rare) some cases even damage the charge controller (lots of variables here).

And when shutting down solar charge controllers, turn off solar array power first (or do the work at night), then turn off DC battery power.

Not picking on your controllers--Just good practice for most. The controllers usually use DC Battery voltage to power their internal computer. Powering from solar first may give you a "bad computer boot" and/or read the wrong battery bus voltage (most controller auto-configure to the bus voltage 12/24/48 volts when they boot. Connecting solar power first can really confuse them).

-Bill

Dave Angelini

How does OP know please that there are no faults or warnings?

ChuckKautz

When there is a fault, it shows up on my Conext Combox.

mcgivor

Check to make sure that each controller is on its own separate array, multiple controllers cannot share a single array. Should each have a separate array, disconnect all except one controller  to assess the individual controller's performance in a divide and conquer approach, repeat the process with each controller.

BB.

The system did work fine for 8 years... And during a battery bank replacement (only work done) is when things started going south (I guess mostly the several MPPT charge controllers are producing 0 Watts at times/randomly).

-Bill

mcgivor

BB. said:

The system did work fine for 8 years... And during a battery bank replacement (only work done) is when things started going south (I guess mostly the several MPPT charge controllers are producing 0 Watts at times/randomly).

-Bill

Agreed if that if that's all that was done, however there have been 3 cooks in the kitchen, a verification wouldn't hurt.

BB.

True... When debugging a confused product with more than one set of hands in the mix, it is usually a good idea to start at the beginning, and assume nothing.

Charge controllers usually make their "decisions" based on voltage and sometimes time (2-6 hours absorb timer, then fall back to float voltage). Starting with a voltmeter and seeing what voltages are present tells us if the basic wiring is correct.

-Bill

Dave Angelini

In a network the decision is based on the set-point and this is globally set to all devices. The chargers will drop off and output zero if the set-point is reached. 

The reason I asked about faults and warnings is this is all logged. There are historical logs as well as current logs. These are logged in the network system and each device logs it's own faults and warnings. They have dates also.

Chuck is not answering the question even though he thinks he is.   Are there Faults and Warnings ? What are they?

In the end this most likely is a configuration problem if there are no faults or warnings.

mcgivor

Dave Angelini said:

In a network the decision is based on the set-point and this is globally set to all devices. The chargers will drop off and output zero if the set-point is reached. 

Wouldn't that be the case only if the setting are manually copied to other devices from a single device?  If this is the case, then couldn't  the settings of individual controllers be different if adjusted by someone attempting to troubleshoot after being copied, for example?

BB.

There could even be the issue where the XW is charging the batteries instead of the solar chargers (their setpoints/other configuration parameters are higher than the solar chargers--The solar chargers will stop while the XW uses utility power to charge the batteries).

-Bill

ChuckKautz

Dave,

I was checking historical faults in the two inverters only.  I went through all 4 charge controllers and 2 of them had historical faults that were not erased.  I think that the others were erased while troubleshooting.   The shutoff of the charge controllers happens daily, not showing up on the fault log.  Maybe it is a fail safe type setting that I am not aware.   I will copy and paste below.  Hopefully this sheds some light on the issue.  Also, the guy who put in the battery also replaced a charge controller that he thought would fix the issues. 

Thank you all for putting your heads together, I appreciate it.

Chuck

Bill,

I have the inverter battery charger disabled.  It did not make sense to have the grid charge the batteries. 

Conext ComBox

System Performance   System Devices   XW (1)   XW (2)   MPPT (3)   Settings   Events   MPPT (4)   Settings   Events   MPPT (5)   Settings   Events   MPPT (6)   SCP (7)   AGS (0)   ComBox Configuration

Today's Events Event Type:

Time ID Name 2020/10/16 07:27:19 69 Configuration Fault 2020/10/16 07:16:48 69 Configuration Fault 2020/10/11 18:24:29 69 Configuration Fault 2020/10/11 18:24:14 69 Configuration Fault 2020/10/08 08:52:14 69 Configuration Fault Timestamp unavailable 69 Configuration Fault Timestamp unavailable 69 Configuration Fault Timestamp unavailable 70 DC Over-voltage shutdown

2020/11/09 03:03

Conext ComBox

System Performance   System Devices   XW (1)   XW (2)   MPPT (3)   MPPT (4)   MPPT (5)   MPPT (6)   Settings   Events   SCP (7)   AGS (0)   ComBox Configuration

Today's Events Event Type:

Time ID Name 2020/10/16 11:21:29 69 Configuration Fault 2020/10/16 07:27:18 69 Configuration Fault 2020/10/16 07:16:47 69 Configuration Fault 2020/10/08 08:52:13 69 Configuration Fault Timestamp unavailable 69 Configuration Fault Timestamp unavailable 70 DC Over-voltage shutdown

2020/11/09 03:08

Dave Angelini

I will say this again, you need to look at not just the faults. There are historical and current for each type.

There is the second thing I asked, and that was warnings.
You might want to read how XW system treats warnings, how they can escalate, and how they become faults.
Read the warning and fault numbers and you should post a 24 hour screenshot of the battery summary screen. A picture is worth alot and it removes reading errors.

A config error can happen from the grid. A surge can change a default or programmed setting. In offgrid we surge protect everything to avoid the loss of power.

mcgivor

The over-voltage is an escalating fault  for the DC input, or battery voltage if you like. If the voltage exceeds the maximum allowable which I believe is 66V  3 times in 30 seconds it regesters a fault.

The configuration fault could be related to various things such as the sync cables, more than one controller being set as master, errors occurring during a firmware update, interference being induced into the Xanbus communication cables to name a few.

Since there have been multiple people working on the system which worked prior to the batteries being changed  I would be inclined to reset to factory defaults then reconfigure the whole system.