Charging One Off-Grid System from Another

[hogback]

My current system has been working more or less flawlessly for four years. One Schneider xw+, 2x mppt 60150, 5.7kw panels, 14kw kohler genny that almost never is needed, scp, gateway, and 4x simpliphi 3.5s, voltage controlled for 56V bulk/absorb for 6min and 54V float. Also ags for the kohler.

So I inherited 2x xwpros, with unknown history, but they might be nib. Also, 8x sunpower 6.5kwh units, which is a separate discussion, but long story short, I'm not going to use their bms, and have bought 4x of the newest jk units to trial, and then maybe get four more to complete the utility of these batteries. So, the 2x xwpros were originally meant for replacements for my xw+ if it ever busted, but I chucked both of them in the wall in my garage which is pretty far from the existing electronics room, but does have power from the existing setup. The eventual plan is to make sure I can make the sunpower units work with the aftermarket jk bms, and then add all 8 to the current system, and move the four simpliphis to the garage, with one or both xwpros connected.

Right now I have one xwpro connected to one hacked sunpower battery with a jk bms - no closed loop going at the moment. No mppts or solar at all here - just messing with it with no loads and no charging.

My question is how to, at least temporarily, charge the sunpower/hacked with jk units from my existing power in the garage? My thought is to use the AC2/generator charger and just manually flip a breaker when I know my four simpliphis are charged midday. Or, I could use AC1 and pretend the xw+/simpliphis are grid, and get a little fancier with charging. Or, I could see if I could get the jks to work in closed loop with the schneider pros, and charge off of soc.

I should mention too that I have an insight facility on the isolated xwpro system right now.

Again, I'm imagining getting all the sunpower units working with the isolated xwpro system with no charging capability, and then moving them to the xw+ system, and maybe moving the 4x simpliphis to the garage to work with the xwpro(s). There is the possibility that I could get another mppt60150 and add some panels over in the garage area, as I do have a bunch of panels laying around. Maybe just 9 for one mppt60150, which would be half my current solar supply.

My main question still is how I might connect my current system to charge the isolated system - use AC1grid or AC2 generator? Probably limit it to 10A or so.

 

[Crowz]

Since your just playing with it I'd use the gen breaker switch when needed for now if it was me.

 

[hogback]

Thanks. Sounds right to me - seems like the ac2/gen input is more limited, in a good way, in terms of the system considering it an input only (no selling back....

 

[hogback]

I think that when my main/house system is charged in the morning, 2kW would be reasonable to charge this aux system. If I set the 'breaker size' to 10A, it would limit the charging to 80% of this, so about 8A, which is roughly 2kW at 240V coming in to AC2 on the aux system. Anything else I need to switch in settings before trying?

Oh, and I'm assuming that I should NOT tie the earth ground between these two systems, as it is bonded to N in the main system, and this aux system also has earth bonded.

 

[hogback]

Can we assume that an xwpro would qualify to charge on AC2 from the load output of an xw+?

 

[BentleyJ]

Can we assume that an xwpro would qualify to charge on AC2 from the load output of an xw+?

Theoretically that should work. No harm in trying, worst case the Pro doesn't qualify the AC Input and nothing happens.

Be sure to set Charge Current (or use Gen Support feature) on the Pro such that charging Watts AND all the other loads on the XW+ don't exceed the XW+ max output.

 

[hogback]

Was only going to try to charge at 8A or so. To do this, I just set the 'generator breaker size' setting to 10A. Any other settings other than this 80% derating breaker size thing?

 

[BentleyJ]

Honestly, I'm not sure how the Breaker Size setting effects other settings. That may work, but I would either set the Pro Charge Current to 40A (which is the equivalent of less than 10A at 240V) or set Gen Support to 10A.

 

[hogback]

Cool thanks. Haven't delved into the manual in awhile....

 

[hogback]

So, I had mentioned in another thread that I thought I might have fried one of my xw pros. All settings in insight home greyed out - not changeable. I had thought this had correlated with my messing with some bms from sunpower, and possibly connected some over voltage wire/adapters between the insight facility and the bms and/or xanbus. Anyways, I think what was happenings is as follows.

Some background to start. I have an operating single xw+ (off grid) with mppt 60150s, ags, gateway, what was 4x simpliphi 3.5s (now 8x sunpowers that I hacked with jk inverter bms (not using soc control). So I got a great deal on 2x xw pros and an insight facility. I originally thought that the pros would just be spares for the xw+ when it might hop the twig, but to get them off the floor and taking up space, I mounted them on the wall of a small room I built inside my garage. One thing led to another, and I thought I'd try messing around with them and possible have a separate system in the garage. The garage is 100' or so from the house, and it has a subpanel from the main xw+ system. The xw pros are mounted on plywood and backerboard, and this is mounted on a wall that has metal lath for, hopefully, a lime plaster experiment (I hate sheetrock). So, the xw pros might not be insulated from the lath as the screws likely contact the lath.

I originally powered up the xw pros individually with one of the sunpower hacks. Managed to upgrade firmware on the facility and both pros, plus the grid codes. I think the problems started when I screwed another subpanel directly to the lath, and connected load wires and N/G from one xw pro to this subpanel. By the way, I was trying to be safe with the firmware upgrades, so had the insight facility powered with a wall wart on the main xw+ system. This might have been the issue. I never actually drew any intentional loads. I had an earth rod outside the garage that wasn't being used, so connected this to the neutral bar in the subpanel, thinking that this would be my bonding point for this auxilliary system. The subpanel enclosure itself hadn't yet been earthed, but it would have been to an extent from the main system, as the lath touches steel outlet boxes which are grounded to the main system, which happens to have its bond in the generator. So I don't see the subpanel itself as the issue (might have been too if I earthed the enclosure), but because I connected the xwpro chassis to it's neutral, and this would be a double bond if the xwpro chassis mount screws touched some lath. Sometimes when powered on, the pro would make bad noises and seem to draw a couple hundred watts (but sometimes just the 20 its supposed to) and give me EEPROM errors.

Does it make sense that this wall wart power supply could be the issue for the greyed out insight local options? Last night I let it be powered by just xanbus, but also removed the NG bond. No problems so far. Menu options came back, and I paralleled both xw pros with the same battery and xanbus network, plus the ac sync cat5. Verified that all relevant settings are identical between both xw pros. When trying to get them to invert, the seem to want to do AC pass through (I see ---- in the display). That was this morning and I haven't gotten to the point in the manual where it says how to 'enable' invert if this is in fact what I need to do. Might go try that now.

Lastly, I had originally thought of this system being totally isolated from the xw+ system with the exception of charging with the two hot legs from the xw+. But I now see that the generator input is both hots and the neutral. So, do you think I should just keep the neutral and ground the same metal as the xw+ system, with the only bond being at the generator? Is it ok that the two pairs of hot legs are then totally independent (with the exception of when I'm charging off the main xw+ system?). I think this makes sense, just want to check.

I have the xwpro generator breaker size set to 10A, so derated to 8A, this is 1900W which I think is reasonable to draw from the xw+ system considering my other possible loads on that system. Now I think I should change the settings to 5A. It's 5A for EACH xw pro, right? I think I might save some headaches by investing in a pdp (have a mini on the xw+ system.), especially if I'll be adding mppts later on....

I have entertained the idea of temporarily charging the batteries for this aux system with the output of my main xw+ system. Eventually, I'd have an scc and a few panels for it.

 

[hogback]

So, by pressing both power and equalize buttons, but only momentarily, will 'enable' invert. This only worked on one of them - the primary - which is not the one that ever had a problem. It hums and gives appropriate V on the load terminals. The other does not. I can get it to say 0kW after a quick EN, but only a few stray mV on the load terminals. Do the load terminals actually need to be tied together between two xwpros for the secondary to invert? Mine are totally floating right now. Maybe I have indeed fried this inverter. I might go back to standby and reselect the grid codes. Possibly I missed this after the firmware update. It said it was right, but I heard from one of the guides that you should re-select the grid codes/region settings after a firmware update, even if it says it is right.

 

[hogback]

By floating, I mean they are connected by the same batteries, xanbus, and sync cable, but nothing else.

 

[hogback]

Condensed questions:

Should the N and G be shared between two 'independent' off grid systems?

Do two xw pros need to have their load lines tied together (when connected with xanbus and the sync cable) for both to invert?

Thanks!

 

[hogback]

So I did discover one interesting thing just now...

So I restored factory defaults on both pros, reset grid codes, and filled in all values to match on each. So, I hadn't discovered the "identify device" function so I wasn't sure which inverter I was working on, and I set up the primary and secondary backwards from last time. Now, it's the other inverter that doesn't 'hum' and present 120/140 on the loads terminals. So it's seems like both inverters are perhaps ok, so some settings or wiring is wrong. I'm thinking of bridging L1 L2 and N between the units to see if that gets the second one going. As a side note, I did try to set the association with the second inverter to 'Loads 2' or whatever it is called, but that didn't help get it inverting, which I suppose it should have if each L1 N L2 on each inverter are not connected. Maybe just the N's need to be tied? Will try that now....

 

[hogback]

No change really....

I am noticing that the ACV of the inverter that is now working is about 121V L to N. On the other inverter, it was about 118V or a little less. I know there is voltage calibration available in this latest firmware, but I don't see instructions anywhere on how to use it.

Another point to note is that neither inverter is showing 0.02kW when 'on,' and I know my xw+ shows this when all loads are disconnected.

 

[hogback]

Perhaps this thread should be in one of the inverter sections?

 

[hogback]

I calibrated the voltages. They are each within 0.5V of 240, L1 to L2, and to each other. Still, to actually get V to show up on the terminals, that unit must be set to 'primary.' With one secondary and the other primary, I see nothing on the secondary terminals. I can switch primary/secondary, and the 0V on the terminals always stays with the secondary. The manual seems to imply that I should be seeing V on both inverters, even when they do not have the L1s and L2s bridged. I could try bridging them and drawing a load, and then see how the power draw looks on each. The multi unit commissioning document has a procedure for fine tuning the secondary inverter V to get the power draw under load within 10%. What I'm interested in here is seeing if I get ANY power from the secondary.

Thoughts?

 

[hogback]

Got a response from se that with firmware 2.4, the primary and secondary do indeed need to be "physically common" due to a new balancing feature. This is at odds with all the current literature online, so a heads up for those looking to parallel more than one XW pro.

So what do you all think - should I connect with the neutral from the XW+ system or bond NG separately in this dual pro system? They will necessarily have to share G, and I think should be fine with their own NG bonds, but the question comes back to how this will play out if I try to charge the pro system from the + system.

 

[hogback]

I just confirmed with these two inverters that the load lines do indeed need to be bridged for both to go into invert mode.
Strangely, independently on, I had calibrated the invert V of each to within 0.5V. Now, bridged, they are about 1V out of 240 off. I should re-confirm the independent deltaV, and perhaps recalibrate the V. Bridged, I tried a measly load of 200W, and one was inverting about twice the power. Schneider has a secondary fine tuning V calibration for ensuring the inverted power is within 10% for each inverter. I'm not sure how much load should be present to aim for 10%, but I'm guessing more than a couple of hundred W.

To rephrase one of my questions above, in regards to an off-grid system sharing an N with another, does the xwpro need to have a charging source with a neutral? What would go awry with AC1 or AC2 just seeing L1 and L2, and no N? If it does need to see a N, then if I was going to charge this dual pro system from the xw+, then I would need for them to share a N all the time, and hence not bond the NG is the dual pro that I'm playing around with. Thoughts?