22 Charge controllers & 1 battery bank

mike760

I have my two inverter systems connected to a common battery bank.  The charge controllers are split between the two systems as far as xanbus goes but connected to a common bus as far as output goes.  I am having trouble with the two separate banks of charge controllers staying in matching modes (bulk/absorb/float) even through the parameters match.  I was thinking they would be much better if they were all controlled by one inverter system but then I exceed the maximum number of devices on the xanbus network.  Theoretically the bus should support than number of devices but SE does not recommend that many for traffic reasons.  Has anyone ever exceeded the 18 device max recommendation?  Is there a better design opinion?  Any suggestions or ideas?  

System A:
6 @ XW6848 Pro

8 @ MPPT 60/150

6 @ MPPT 100/600

1 @ SCP

1 @ Gateway

System B:
4 @ XW6848 Pro

8 @ MPPT 100/600

1 @ SCP

1 @ Gateway

2 @ Battmon

1 @ AGS

48V Battery Bank:
48 @ Rolls Surrette 2YS 31P (2VDC 2527AH)

Generator:
1 @ Volvo 100KVA Diesel

PV Panels:

488 @ 250W

BB.

In general, I thought that there is not really a "need" to have paralleled charge controllers always being "in sync" (bulk, absorb, float, etc.)... If one controller is "ahead" of another controller--It really does not matter much.

The controller that tripped "first" (if programmed correctly) things that battery bank is "full". The second controller is going to add a bit more charge before it also goes into the next mode.

As long as the battery bank is "hitting" the appropriate "marks", it does not matter if it happened with two controller going to "float" or if one goes to float first, then the second.

If you have "issues" with battery voltages/current/state of charge--You may have some programming issues with one or the other controller (i.e., if battery bank is over charging--One controller is over charging while the other is in "float").

There was one report (years ago here) where charge controllers behaved "differently" during charging (different hardware and/or software levels). When the bank was running with a "battery desulftator" (which sends "impulses to break up hard sulfate deposits on "problem" lead acid batteries). When the desulfator was on, one of the paralleled charge controllers shutdown long before the second. With the D.S. off, the chargers tracked each other fairly closely.

-Bill

mike760

I guess I forgot to mention that this is an off-grid system if that makes any difference in your analysis. 

There are several scenarios that I believe are related to a lack of common control.  One example being that the inverters on system B might start to draw down on the batteries around 3pm and the charge controllers on that system appear to be putting out maximum wattage given the amount of sunlight (let's say 10KW).  When you look at system A, the charge controllers on that system  will not be putting out much if any wattage (<1KW).  So this creates a scenario where the batteries end up supplying the difference instead of the other charge controllers using the available PV to make up the difference to maintain the AC load.

It is also prolonging the absorption stage.  Actually, I don't believe I even have enough time to properly complete the absorption stage when half of my charge controllers are actually "sleeping."  I'm almost at 4pm and my bank is still taking in >150A when the absorption timer is expended.

On another note, I noticed that the 60/150 charge controllers have 4 charge modes available which are stand alone, primary, secondary and echo.  Anyone know what echo is?  I looked for documentation and came up empty.  I also noticed that this charge mode selection is not even available in the configuration of the 100/600 chargers.

mike760

This is an earlier picture during installation so it does not show all of the equipment listed.  Both walls are packed so it's hard to get everything in one picture.

BB.

Impressive system.

I would suggest using a higher resolution DC volt meter (I.e., XX.XX to XX.XXX whereas a XX.X meter may not have enough digits) and measure the voltage on the output of each charge controller (I.e., solar when charging with sun, and hybrid AC systems when/if charging from gensets), and measure the charger(s) terminal voltages.

My first guess is that under heavy loads/current, your terminal voltages are significantly different between the chargers--The chargers with the lowest DC output terminal voltages may be doing most of your charging (due to voltage drop of the various cable lengths and possibly unmatched battery bus voltages.

Also, double check the voltages across each battery (or cell) and see if they voltages match (under load, under charging, etc.). If there are any "unbalanced" voltages, check cell Specific Gravities, look for voltage drop across terminals (I.e., between lead battery terminal and battery wire terminal/copper, even end to end drop on cable to bus bar connections, etc.). You can even check temperature of connections/cables under heavy load.

Looking for anything that could indicate poor electrical connections (loose/dirty/corroded terminals/etc.). Basically--Looking that everything "matches" voltages under heavy current (during loads and currents).

You could also use a DC Current Clamp meter and insure that current in "identical cables" is being shared equally (I.e., suggest that within 10% current sharing is probably "ideal". Where as >50% differences in sharing is not good).

Looking for "higher" voltage drop and lower current flow in specific locations and see if you can find a reason for the differences (I.e., bad connections, different cable lengths--I.e., longer cable runs have higher voltage drop and lower "share of current").

With physically large systems, it is going to be difficult to have "equal" cable lengths between everything for matched resistances.

Ideally this is the choice between Star and Daisy Chain wiring for battery banks (generic wiring--AKA sub battery for encoder):

The problem with Daisy Chain is the battery(ies) closest to the source will tend to absorb/source the most current. While the ones at the "end of the ladder" will tend to participate less (due to increased wiring resistance). (in the above star configuration, ideally looking for equal wiring lengths for source to each battery--equal wiring resistance.

In the above STAR wiring, the source could/would be a common bus bar connections--All batteries connect to common point, all chargers connect to same common point, all loads connect to same common point --This way all devices "see same DC Bus voltages".

There are lots of variations to "balance" battery wiring/voltage drops with multiple series batteries. Example of matching total battery bus length--Splitting + and - wiring destinations to end up with identical total path lengths:

I am sorry, I don't have the knowledge to discuss the different controller setups (alone, primary, secondary, echo). One of those modes may offer better overall charging performance (I.e., Master makes the "decisions" and the slaves simply follow).

I am no expert in these systems--Just my overall experiences in other power systems over the years and how I would start some debugging/thinking of system issues.

Best wishes,
-Bill

Dave Angelini

Which model Gateway?     Insight home, Insight Facility, or the original gateway?

mike760

Insight Facility information

Insight Facility name
Model designation	865-0335
Application version	v1.18
Application build number	41
Application build date	2023-07-25_20-09-43

mike760

And BB, thanks for all of the information.  I'm trying to make sure I have the Schneider equipment connected in the best way possible given the higher number of devices and then move on to troubleshooting if necessary.  And there are some things that you touched on that I am struggling with but I don't want to open that can of worms in this thread.      

Not sure If I'm going in the right order but I want to make sure any communication/synchronization, or whatever that is necessary for the charge controllers to operate optimally, has been setup properly.  My first concern is the proper xanbus setup for two separate systems of inverters, gateways and charge controllers operating independently but sharing a common DC bus / battery bank.  

BB.

No problem. Just take one step at a time

Good luck,

Bill