Conext ComBox aparently showing incorrect Load Watts

Ganglefang

I am off on another project now.  But the Combox has lost one MPPT and the BattMon (neither appears in system devices.)  The BattMon refuses to synchronize after I hold both < > for more than 3 seconds.  None of the values reported by the Combox appear to be correct except perhaps the solar input which is consistent with our solar radiation sensors.

As an EE my instinct is to trust my meters.  What I think is happening is that the data being sent via the Zanbus is either incorrect or being interpreted incorrectly.  I will do a Gatesboot when I get back and see if that helps.  I never got a case number --the second email i sent back on April 22 is shown below.  (The email included a screenshot which doesn't seem to copy.  Recently downgraded to Win10 and a lot of stuff doesn't work right anymore.)
-------------------------
To: re.techsupport@schneider-electric.com
From: xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx
Subject: Current reporting.
Cc:
Bcc: ƒ\Personal\POWER
Attached:
(I sent this about three weeks ago, but have not gotten an answer.) I have noticed that the battery current
reported in the System Performance report from Combox is not consistent with that reported by the XW.
See attached screenshot. The XW is reporting approximately ten times what the Performance screen is
reporting. As you can see, the energy equations don't balance.
Configuration:
· On grid (off grid test, battery only)
Conext XW+ 6848 Inverter SN M70012350
2x Xantrex XW 80 600 MPPT Battery Charge Regulators
Conext COMBOX connected to the local network SN B13067792
Conext Battery Monitor
Xantrex AGS Automatic Generator Start Control
Generac 20KVA generator
20x Batteries 8A4DLTP-DEKA 210Ah batteries
24x Roof Mounted REC 255W panels

new2PV

Not to be rude but do you have the xanbus terminator on the combox empty port if your only using 1 port with a cable in it? Try and re-flash the firmare on the battery monitor also. I do know for a fact the combox will report wrong, with grid support enabled, but not thoussands of watts wrong.

Ganglefang

Not rude.  Yes, Xanbus is terminated.  It all worked at one time, except for the incorrect values.  I will flash everything when next I am in-house.  I hate to do that as then I need to offload all the data. 

An observation--most people on this site who were kind enough to reply don't seem to have Comboxes.  Even when I was talking to someone on the phone at Schneider, I asked them what their Combox said, and the reply was that their box wasn't working.

In my opinion, using a Combox to monitor without a BattMon is sort of useless.  I need the BattMon to tell the generator when to start when the batteries reach a certain level of discharge.  But if the BattMon keeps failing, ....  Anyhow, I just looked at the Combox/PerfMonitor and it is reporting 60A out of the Batteries. while the BattMon display is directly reporting 0.5A which agrees with my meter.  (Not too unusual since the Combox isn't seeing the BattMon, and so I think the Combox is estimating based on voltages--not a good idea.)  My installer tells me most installs do not include a Combox.

And another observation--in one place in the manuals it says the Xanex cables must be greater than 6 feet, and another it says less than 6 feet (2 meters).  This leads me to believe, beyond a likely typo, their Xanbus protocols are electrically or logically fragile.  I deal with Cat5/5e/6 all the time and never have any problems even at LAN speeds.

If the BattMon is gonzo, I will have to go through the warranty process again.  But that doesn't explain why one of hte 2 MPPTs is not being reported by the Combox.  And if I have to replace a BattMon every few months, I wonder if someone else makes something that will work.  I'd build my own, but don't want to invest my time in deciphering the Xanbus protocols, and if I did, I would have no recourse.

new2PV

All the manuals I have read say min 2 m in length. You need to get that 2 nd mppt working as one is the primary other secondary and they will sync charging stages. Reflash the combox , reboot all devices and make sure the cables are long. Even try to make sure the devices are in the same xanbus loop. The device number on each mppt must be different. I set that up before plugging in my mppt 60s to the network.This might be your issue two devices with the same ID.

Dave Angelini

I can't spend any more time on this but must say that if you or your installer did not get a case number,
 Schneider will not follow up on this.
 I would recommend reading the posts here again.
You need to eliminate the battery monitor and shunt for testing, make sure your devices have different addresses.
You may want to just get the system running as if it is offgrid first for testing. It should look something like this.
There are DC loads running below, some error/loss, but the numbers should add up roughly after 5 seconds if the load is constant.

mike95490

new2PV said:

Not to be rude but do you have the xanbus terminator on the combox empty port if .......

That's what I was going to say, but there also needs to be a 2nd terminator plug on the other end of the Xanbus line:

term--[CC]--------[XW] -------[Monitor]------[Combox]--term

new2PV

When I forgot the 2nd terminator everything worked fine, except the battery monitor was not updating  the charge cyce data, number of charges, etc.  It seemed to be missing bits of data randomly. Rebooted with the 2nd terminator and it works again!

new2PV

You can;t be going through so many batter monitors, this has be an issue with your install. Please update us if you managed to get the mppts online.

Ganglefang

Both ends of the bus are terminated.  I did get the MPPTT back online.  (Rebooted again.)  And I re-flashed the Combox.  The BattMon is still missing in action.  I started another email chain with Schneider about the BattMon and possible another warranty replacement.  But each round of email takes about a week, and a replacement takes a month+.    When I get that resolved I will try again about the issue of incorrect values being reported by the Combox.  Having multiple issues seems to confuse Schneider.

When I said I thought the Xanbus was fragile, I meant that electrically or protocol-wise, it is prone to intermittent failure.  Certainly incorrect termination can cause reflections that confuse the logic or protocol.  But I am religious about terminating both ends of buses.  Except perhaps on computer motherboards, or a Cray computer, bus length should not make a difference in open-air buses.  I routinely run gigabit cat 5e/6 from 1 foot to 300.  And I also note that the parts you can order from Schneider for this equipment come in 3ft and 25ft lenghts.  Nothing listed in 2m/6.5ft which is what came with my equipment. 

I will build an adapter and use it with my scope to look at the signals--easy enough to see reflections if they are there.  It is possible one of the devices has a faulty buss pass-through which is causing problems.  However I am also changing out all the cables, and adding some longer ones and changing the position of the devices on the bus to see if that makes any difference.  All this takes time and I work for a living.

I think the solution to the Combox incorrect values issue lies in reverse-engineering their code, something I do not propose to do.  I can live with the incorrect values, and now use the Combox primarily to observe and change settings.

Dave Angelini

Xanbus and it's cousin Canbus are not robust as you might think. The Xanbus has a max length of 130 feet not what you said about bus length. I suggested long back that if you want to fix this, do what a technician would do not someone designing a Nuke as you said.
Get the system working as an offgrid system. An inverter, charge control, and the combox. Start adding devices until you find the problem.

Ganglefang

I am back.  I replaced the Combox and BattMon, and added an SCP.  I got both MPPTs working.  I was working fine with new Cat5e cables, but put the original cables back in just to to see what would happen.   The system continued to work correctly   I let the system "cook" for several months to be sure.  

I see there are other people with energy equations that don't balance, but mine now balance within reason.  The numbers reported by the Combox from the BattMon agree with my own meters.  I am seamlessly selling and buying energy to the grid.  I survived several lightning storms without losing any equipment (phones, modems, computers, microwave, satellite, etc.)      I don't know for sure if it was firmware updates or new boxes or divine intervention.  But I am happy.   Thanks especially to Jimmy Fernandez at Schneider for suffering through a long trial and error process to replace some equipment.    Note that the XW and MPPTs were always working without a problem.  My impression is that the Xanbus is very fragile, and the additional boxes might need some better engineering.

Thanks to everyone for all the help.  

Ganglefang

I forgot to mention, the energy equations balance but only if the energy in/out of the battery is near zero.  If not, then the numbers reported by the Combox (system performance screen) do not balance.  (You may recall this was the original question.)  The battery energy is off by factors of 10 to 20.  For now, everything is working again so I will let sleeping dogs lie.  I will reopen that issue when I have time.  And patience.

Ganglefang

One more thing.  People without a Combox and who are off-grid do not understand the issues us on-gridder, anal retentive Combox owners are having.   Look at a few of the other threads to see what I mean.

mike95490

The Combox only reports.   All the math and calculations are done in the XW inverter, which is how it controls itself.   My SCP and Combox agree all the time. As power goes into the inverter, less comes out, it uses a bit for it's internals.  I've heard it uses a lot when grid connected and trying to load shave or backfeed.  
 So I can't explain it, except the calcs are all done in the Inverter, and then passed onto the bus.

Ganglefang

I thought I'd update on my problems.  After replacing all those components back through December, everything is running smoothly even after several power outages.  Values being reported by the Combox for the battery continue to be off, but I just multiply them by a fudge factor of 11 and I'm good to go.  Thanks to all for your help.