Adventures in Schneider/Xantrex Diagnostics

Since troubleshooting the Schneider/Xantrex equipment is turning out to be an extended and significant effort, I'm collecting information here for ready access. Open issues are shown first.

Open Issues

XW6848 Inverter F63 Faults (January 2015 - present)

Schneider Case Numbers 16450612, 16875454

With the addition of a diesel generator and (finally) autostart control, I decided to add some additional loads to the XW+6848 inverter. That was done in the form of a "Desirable Loads subPanel" (DLP), a sub-panel of the original critical loads panel (CLP) with a contactor upstream of the main breaker. An automatic load management scheme was also designed and implemented, so that only one of the three biggest loads (well pump, ground sourced heat pump, hot water heat pump) could be active at the same time.

But, we starting experiencing F63 faults (AC Overload), something that had never occurred before with the original XW6048. The F63 fault is nasty because the XW+6848 cuts all power to the CLP (and of course DLP) with no or minimal warning (the start of the beeping and the loss of power are simultaneous). We had grown rather spoiled by the way the original XW6048 had seamlessly tolerated utility glitches. I was too busy to worry about the first F63 incidents, and didn't have any instrumentation to quantify the loads and initally assumed that, in spite of careful calculations and measurements, the loads were in fact excessive. The F63 fault incidents would typically occur in conjunction with the many frequent short utility outages experienced here, one or more outages of a second or two from utility automatic recloser trips.

So first I implemented the automatic load management to prevent the well pump (which had been on the critical loads panel all along) from running at the same time as the biggest loads on the DLP, the upstairs heat pump and the hot water heat pump. We still continued to experience F63 faults, even though a logging data meter (Fluke 287) never recorded loads in excess of the 28A nominal capacity of the XW+6848. That's excepting the inrush current of the well pump, of course, which is on the order of 40A for less than 400ms.

Next was wiring a failsafe cutout for the entire DLP; so that when neither grid nor generator power are present the DLP contactor opens and the entire DLP is dead. Still we experienced F63 faults; clearly something is wrong as the CLP proper now has lower loads than for the nearly two years of successful operation with the original XW6048 (note several CLP circuits were moved to the new DLP).

But, there was still the nagging question of what the loads were on a cycle-by-cycle basis, as the Fluke 287 logs transitions between different voltage levels and not waveforms. So, we acquired a power quality analyzer (Dranetz 658) capable of logging the load in detail (four channels with waveform capture on all four channels).

The gory details of the F63 diagnosis ordeal from this point on can be found here.

AGS (August 2014 - present)

Schneider Case Numbers 13718734, 13787391, 13428287, 13969583, 15570921

In 2014 I spent months -- and mandays -- trying to figure out why the AGS (Automatic Generator Start) module would randomly (and inappropriately) start the diesel generator. Eventually Schneider tech support determined the AGS firmware was defective. Last I heard the firmware upgrade is due in mid-2015. The gory details of that ordeal can be found here

Note that during the extended process of diagnosing the AGS issues, Schneider supplied a new XW+6848 hybrid inverter under warranty to replace the original XW6048. That turned out not to be the cause of the AGS issues. Later (early 2015) it appears that replacement may have introduced new problems (see next section).

I gave up on the Xantrex AGS and now use a Pentrametric State of Charge (SoC) meter for generator auto-start; no more false starts. If and when new firmware is available I will probably give it a try, but if it's not rock solid reliable I'll stick with the Pentametric (the AGS false starts were abusing my expensive diesel generator). On paper the AGS supports load based autostarts, whereas the Pentametric only starts based on low battery SoC. That is a desirable feature that never worked in practice so we'll see how the new firmware works.

Troubleshooting Notes

In fairness to Schneider/Xantrex the XW6048/XW+6848 inverter and AGS module are complex devices and they have been generally responsive in talking to me on the phone and via E-mail. Some of our troubleshooting sessions have taken the better part of an hour and they're still talking to me. In the hope that we may yet succeed in figuring out and correcting the various problems I offer the following thoughts:

• I will cheerfully configure any logging the Combox is capable of.

• I have a logging multimeter (Fluke 287) which can do detailed logging of any one parameter of interest (e.g. voltage, current) for periods of up to several days. Since the failure events (AGS false starts, F63 faults, etc.) don't happen predictably or every day it may take awhile to capture the data of interest.

• I have a PQ analyzer (Dranetz 658) and will supply graphs and waveforms as relevant or as requested. This device can monitor up to four channels of voltage or current and capture the waveforms on all four channels when programmable out-of-bounds events occur on any channel. It can log for months at a time, but consumes 100W of power when in operation, and generally requires that the subpanel covers be removed.

• I can configure direct access to the Combox by Schneider/Xantrex personnel. I'll need the source IP address to forward inbound connections through the firewalls.

• The XW+6848/battery/generator installation supports an office and business that generates the revenue needed to pay for that expensive equipment. This office also supports servers accessed 7x24x365 from around the world. Reliable operation of this office and servers was the reason for the purchase and installation of the Schneider/Xantrex equipment in the first place. We're happy to do whatever is truly necessary to debug these very disruptive failures, but please keep in mind that any such experimentation can have a very adverse impact on the operation of this business.

What I don't want to do is reflexively "reset to factory defaults". I've been asked to do that multiple times now. The very detailed configuration must be painstakingly typed in manually, a process that takes about twenty minutes during which time the critical loads are without power. I can accept an occasional mysterious glitch in heavily computerized hardware, but such a "hard reboot" is a last resort intervention that is not acceptable on a routine basis.

Closed Issues

TBD, no closed issues yet.