3 phase stacking GSLC's

Dual Input, 120/240, stainless steel form factor design for unsurpassed surge capability
Post Reply
JasonPAtkins
Forum Member
Posts: 9
Joined: Wed Sep 11, 2019 12:50 pm

3 phase stacking GSLC's

Post by JasonPAtkins » Wed Sep 11, 2019 1:33 pm

Hi all, this is the third 10kw+ off-grid system I've been involved with, but the first using Radian inverters. I'll be using 3 x GS7048e's to generate 3ph 400v 50hz power for a non-profit in West Africa.

I'm sure I'll have more questions, but I'm currently spec'ing and ordering. I'm ordering the GSLC's for the three inverters, despite realizing the installation manual advises against stacking more than 2 inverters in GSLC's because of wiring complexity, because I'm going to take care of all of the AC distribution external to the GSLC's, so there should be plenty of room in the GSLC's for the DC stuff.

Actually, my question would be the same even if I were just stacking two units - so ignore the above, haha...

My question is, what's the official way to tie the DC buses of separate GSLC's together? In other words, I don't want to run battery connections to/from each GSLC, so the DC+ and DC- busses need to be tied together. I'll mount the units close with a big horizontal conduit connection between the GSLC's, so would you just run a 2' length of 4/0 from one to the next? I want to install the flexnet dc monitor that uses the shunt, so I assume that would have to happen in only one spot - cause if current is passing through one of the other negative busses and over that shunt, it won't get counted.

So I'm envisioning the battery connections coming into one GSLC (let's call it the main), and it'll be the main where the dc monitor is installed. Then, on the other (non-battery) side of the shunt of the "main" GSLC's dc- bus, I can make connections to the other GSLC's dc-'s, and similarly tie all three DC+ cable plates together with fat cables. That will give me solar controllers plus inverters tying into the now-common dc busses, but all of the current going to or from the batteries all going through the single shunt, letting the dc monitor correctly meter.

Does this seem like the best way to do it, or is there a better/easier way?

kayak1
Forum Whiz
Posts: 25
Joined: Sat Aug 02, 2014 10:27 am
My RE system: 77xRec 315 Watt Panels
GS8048A Inverter, 2xFM100 Charge Controller, Mate3S, Flexnet DC, 12xEnergyCell PLR 600 AH AGM. (Attached with Tigo Otimizers to 38 panels)
Solar Edge 10KW inverter Attached with Solar Edge Optimizers to 41 panels.

Re: 3 phase stacking GSLC's

Post by kayak1 » Thu Sep 12, 2019 7:03 am

Have you looked at the midnight solar Battery Combiner, it has a version with the correct shunts built into it, to allow you to use the flexnet dc.
-Jason

EMCF
Forum Virtuoso
Posts: 95
Joined: Thu Jul 21, 2016 10:22 am
My RE system: FX3048, 4 200a batteries, FM60. 3 kw array

Re: 3 phase stacking GSLC's

Post by EMCF » Thu Sep 12, 2019 4:52 pm

The safe and official way to connect the DC busses is, from battery bank, run one separate positive cable in series with a breaker to each GSLC. Yes, your negative can be common, and without a breaker. Run one cable to "phase A" GSLC, then make jumper cable to "phase B" GSLC, and another jumper cable to "phase C" GSLC.
This configuration will enable you to power down one or two of the inverters, and can still have a 230VAC output. If you power down one, you can still have a 400VAC output.
Outback Fanboy at your service

JasonPAtkins
Forum Member
Posts: 9
Joined: Wed Sep 11, 2019 12:50 pm

Re: 3 phase stacking GSLC's

Post by JasonPAtkins » Fri Nov 08, 2019 12:11 pm

Thanks for the answers.

EMCF, your illustration touches on one of my biggest irritations with Outback hardware (only beating the MATE's UI and menuing system). The ability to do what you just described, supplying two phases of a three phase stack, is incredibly dangerous to connected 3ph equipment, and in my opinion, should almost never be done - but should DEFINITELY not be the only option, or even the default option, for handling the condition where one inverter of a 3ph stack goes down. Xantrex XW's handle the situation correctly - if one of the inverters has an error that causes its output to drop out while part of a three phase stack, both of the other inverters shut down as well. You could fry an entire workshop's worth of 3 phase motors in about 15 seconds if you didn't have external phase protection installed. I have installed a phase protector that opens a contactor feeding my entire shop if it detects such a condition, but shouldn't need to do that - the inverters should (at the very least be able to be configured to) all cut output together if one goes down. This is obviously a different story where you have two or three inverters operating in parallel, where if one goes down you want the other two to take up the slack and continue powering the load.

Anyway, that complaint aside, I don't need the ability to do what you're describing, so I would like to have a common positive bus (among the GSLC's) as well, but want to do it in a way that will run all of the current through one GSLC's shunt so the flexnet can still tie in there. The only problem I foresee is the current limitation of the shunt - does anyone know what it is? In other words, what I'd like to do is connect the battery positive to one side of one of the GSLC's shunts, and then come off of the other side of that same shunt to tie into the DC+ of the other two GSLC's.

Kayak1 - are you saying the Midnite product has a shunt, or has a shunt that can handle 3 inverter's worth of DC current? These GSLC's have the shunt, I believe, I'm just not sure it's appropriate for the case I'm describing. I know the manual advocates a different solution if stacking more than two inverters, but in my case, if this shunt issue isn't an issue, there will be enough space to make all of the connections, because I'm not going to handle any of the AC stuff inside these boxes, those three legs will get pulled out to a disconnect and then off to AC distr panels.

Second question - does anyone know how much space is between Radian inverters when mounted per the instructions? I ask because I plan to use the 2" knockouts in the GSLC's to run cables between them - but if the GSLC's are touching, I'll just need a plastic knockout protector to keep cables from getting cut on the metal of the case. If there's some gap, I assume I'll need a conduit nipple to bridge the gap.

JasonPAtkins
Forum Member
Posts: 9
Joined: Wed Sep 11, 2019 12:50 pm

Re: 3 phase stacking GSLC's

Post by JasonPAtkins » Fri Nov 08, 2019 2:05 pm

Hmmm... it appears that Outback had already anticipated my need? The docs for the FN-DC say it can monitor up to three different shunts at once. So if that's the case, I can tie the three DC+ buses together, run separate from BATT- into each of the three shunts, and in this way still monitor them all (assuming the FN-DC can sum the three shunts to get total system draw?), while not running all of the current through just one shunt!

https://d1819pwkf4ncw.cloudfront.net/fi ... -44031.pdf

EMCF
Forum Virtuoso
Posts: 95
Joined: Thu Jul 21, 2016 10:22 am
My RE system: FX3048, 4 200a batteries, FM60. 3 kw array

Re: 3 phase stacking GSLC's

Post by EMCF » Fri Nov 08, 2019 3:35 pm

Jason, this is a reply to your 1:11 pm post.
I may have not described my DC connections very clear. The negative and positive bus are actually both have common connections, with the difference that the battery breakers are in the positive cables. The basic GSLC usually comes with 2 shunts, one for CC, and one for inverter. These shunts can carry DC amps up to 500ADC, so a single shunt can carry the DC current of all 3 inverters combined. Attached is my handwritten diagram on how I can wire your system using a single shunt.



On another note, there is actually a use case for a three phase power supplying two inverters configured as three phase. Here in North America, in some industrial buildings there are cases where the service connection is 120/120/208. This means these are L1-N-L2 of a three-phase distribution system. For 2 Outback FXR 120v output inverters configured for 3-phase, those are exactly the input that is needed.
Attachments
GSLC.JPG
Outback Fanboy at your service

kayak1
Forum Whiz
Posts: 25
Joined: Sat Aug 02, 2014 10:27 am
My RE system: 77xRec 315 Watt Panels
GS8048A Inverter, 2xFM100 Charge Controller, Mate3S, Flexnet DC, 12xEnergyCell PLR 600 AH AGM. (Attached with Tigo Otimizers to 38 panels)
Solar Edge 10KW inverter Attached with Solar Edge Optimizers to 41 panels.

Re: 3 phase stacking GSLC's

Post by kayak1 » Sat Nov 09, 2019 1:52 am

As I can't post links in replies and have them not be deleted please google:
"MIDNITE SOLAR 1000ADC MAX 1000A/100MV SHUNT BATTERY COMBINER"

Go to the midnight solar link not one of the amazon ones and you can see the base box.

They have an option positive and optional breakers you can install in it.

With outback you can install larger shunts but the ADC MV ratio needs to be maintained.

JasonPAtkins
Forum Member
Posts: 9
Joined: Wed Sep 11, 2019 12:50 pm

Re: 3 phase stacking GSLC's

Post by JasonPAtkins » Thu Nov 14, 2019 11:36 am

EMCF wrote:
Fri Nov 08, 2019 3:35 pm
Jason, this is a reply to your 1:11 pm post.
I may have not described my DC connections very clear. The negative and positive bus are actually both have common connections, with the difference that the battery breakers are in the positive cables. The basic GSLC usually comes with 2 shunts, one for CC, and one for inverter. These shunts can carry DC amps up to 500ADC, so a single shunt can carry the DC current of all 3 inverters combined. Attached is my handwritten diagram on how I can wire your system using a single shunt.



On another note, there is actually a use case for a three phase power supplying two inverters configured as three phase. Here in North America, in some industrial buildings there are cases where the service connection is 120/120/208. This means these are L1-N-L2 of a three-phase distribution system. For 2 Outback FXR 120v output inverters configured for 3-phase, those are exactly the input that is needed.
Thanks for the diagram - that's pretty similar to what I was imagining.

Last I heard, the 110v versions of the vfx (I think) and Radian (I'm pretty sure) can't be configured in a three phase stack, so if that's true then your case isn't supported anyway. Either way, "keep supplying partial 3 phase power" should DEFINITELY not be the only way, and shouldn't even be the default. If 110v inverters can 3ph stack, and your use case is a supported one, then there should be a "Keep supplying if other phases drop out" setting, and it should default to off, so that your use case can turn it on.

EMCF
Forum Virtuoso
Posts: 95
Joined: Thu Jul 21, 2016 10:22 am
My RE system: FX3048, 4 200a batteries, FM60. 3 kw array

Re: 3 phase stacking GSLC's

Post by EMCF » Thu Nov 14, 2019 12:20 pm

JasonPAtkins wrote:
Thu Nov 14, 2019 11:36 am
Last I heard, the 110v versions of the vfx (I think) and Radian (I'm pretty sure) can't be configured in a three phase stack, so if that's true then your case isn't supported anyway. Either way, "keep supplying partial 3 phase power" should DEFINITELY not be the only way, and shouldn't even be the default. If 110v inverters can 3ph stack, and your use case is a supported one, then there should be a "Keep supplying if other phases drop out" setting, and it should default to off, so that your use case can turn it on.
1. Outback's designed for North America inverters have rated output of 120VAC not 110VAC
2. The successor to VFX types, the FXR types inverters can be configured for three phase, that is a fact, not my opinion.
3. Radian E types can be configured for 3 phase, as what you are planning. Of course, the Radian A, being split phase, 120/120/240, can't be configured for 3 phase
4. If you read the VFXR manual, there is such a thing called "Multi-Phase Coordination". Here, if the master or sub-phase master inverter do not sense an acceptable AC source, the entire system disconnects from the source. None of the inverters will reconnect until the source is acceptable. You can disable it too. Does that take care of your "keep supplying..." concern?
Outback Fanboy at your service

JasonPAtkins
Forum Member
Posts: 9
Joined: Wed Sep 11, 2019 12:50 pm

Re: 3 phase stacking GSLC's

Post by JasonPAtkins » Wed Nov 20, 2019 11:20 am

I'm less familiar with the north american lineup, so I'm sure you're right about those first three points.

I don't think so on the 4th though. Multi Phase Coordination, at least in the quick setup guide I found a reference to it in, refers to the inverters working together to reject a flawed external 3 ph source. This capability indicates that Outback realizes it's a problem to pass thru an incomplete 3 ph signal. However, their inverters are still happy to supply it. Maybe the functionality is described better in a more detailed manual? Maybe the feature also precludes an incomplete 3ph signal from being output from the inverter, no just AC pass thru? It should, but there wasn't mention of that on the quick setup guide.

EMCF
Forum Virtuoso
Posts: 95
Joined: Thu Jul 21, 2016 10:22 am
My RE system: FX3048, 4 200a batteries, FM60. 3 kw array

Re: 3 phase stacking GSLC's

Post by EMCF » Wed Nov 20, 2019 4:28 pm

JasonPAtkins wrote:
Wed Nov 20, 2019 11:20 am

I don't think so on the 4th though. Multi Phase Coordination, at least in the quick setup guide I found a reference to it in, refers to the inverters working together to reject a flawed external 3 ph source. This capability indicates that Outback realizes it's a problem to pass thru an incomplete 3 ph signal. However, their inverters are still happy to supply it. Maybe the functionality is described better in a more detailed manual? Maybe the feature also precludes an incomplete 3ph signal from being output from the inverter, no just AC pass thru? It should, but there wasn't mention of that on the quick setup guide.
I am sure you are misunderstanding/misreading the concept of enabled multi phase coordination. In multiple inverter configuration such as 3-phase (3 OB inverters) with a 3-phase input source, if one of the 3-phase input line is missing or unacceptable, the whole 3-phase input is rejected and isolated from the system. But that does not mean that the 3-phase configured inverters will stop functioning as a 3-phase system supplying a 3-phase load. With the battery bank powering the inverters, the 3 phase input source is now totally out of the picture, but is being continuously monitored and possibly re-accepted upon return to normal or acceptable values.
By the way, can you quote your source word for word from the the quick start guide as you claim?
Outback Fanboy at your service

JasonPAtkins
Forum Member
Posts: 9
Joined: Wed Sep 11, 2019 12:50 pm

Re: 3 phase stacking GSLC's

Post by JasonPAtkins » Thu Nov 21, 2019 7:39 am

What you are describing is exactly how I think the Multi Phase Coordination feature does work. What you're describing is also not the problem I'm describing.

With inverters in a 3 phase stack, when one of them experiences an error that causes its inverting output power to stop (for example, it is overloaded and errors out), the other two inverters should STOP inverting as well. The system should NEVER (or at least have the option to be configured to never) send power on only two legs of a three phase system. This is a sure way to burn up every three phase electric motor connected to the system in a matter of seconds.

Your description of the Multi Phase Coordination feature admits that Outback knows the system should reject incomplete (2 of 3 phases) external AC input, because it would be dangerous to connected equipment to pass that electricity on. In the same way, the inverters shouldn't CREATE incomplete three phase and send it on to the powered equipment, which is exactly what they do if one inverter has an error.

JasonPAtkins
Forum Member
Posts: 9
Joined: Wed Sep 11, 2019 12:50 pm

Re: 3 phase stacking GSLC's

Post by JasonPAtkins » Thu Nov 21, 2019 7:45 am

As a counterpoint, Xantrex XW's, when configured in a 3 phase stack, when one inverter experiences an error that causes it to stop inverting, the other two also suspend their output, waiting for the faulted inverter to come back online, and then they all come back online at the same time. This should be the default, or at the very very least, an option.

I understand your assertion that there is a use case for wanting 2 phases of a 3 phase stack in N.A. Fine, but that setting should have to be turned on. The default should NOT be the setting that burns up three phase equipment.

EMCF
Forum Virtuoso
Posts: 95
Joined: Thu Jul 21, 2016 10:22 am
My RE system: FX3048, 4 200a batteries, FM60. 3 kw array

Re: 3 phase stacking GSLC's

Post by EMCF » Thu Nov 21, 2019 4:04 pm

JasonPAtkins wrote:
Thu Nov 21, 2019 7:39 am
Your description of the Multi Phase Coordination feature admits that Outback knows the system should reject incomplete (2 of 3 phases) external AC input, because it would be dangerous to connected equipment to pass that electricity on. In the same way, the inverters shouldn't CREATE incomplete three phase and send it on to the powered equipment, which is exactly what they do if one inverter has an error.

It is not dangerous. On the contrary it is perfectly safe. You can have a single phase input (120 or 230 for you) to the master inverter, and the system will accept it (Multi phase Coordination disabled). The 3-phase output will still be a normal 3-phase output that will feed the load..
JasonPAtkins wrote:
Thu Nov 21, 2019 7:39 am
What you are describing is exactly how I think the Multi Phase Coordination feature does work. What you're describing is also not the problem I'm describing.
Ok, let us go to a pure off-grid 3-phase configured 3 outback inverters. The use case you are saying is about having exclusive 3-phase loads. In that case I agree with you that it presents some issues when an inverter's output malfunctions or completely lose its output for a phase loss especially when a 3-phase load like a motor is on the line. But not all use cases are like that. Some just like the possibility of having a single phase 230VAC, or a single phase 400VAC, or in our case here 120 or 208. We can afford to put to rest one or two of the inverters once in a while if there are no use for the higher voltage.
Outback Fanboy at your service

User avatar
Kurt Lundquist
Alpha Energy Engineering
Posts: 1447
Joined: Tue May 31, 2011 7:23 am
My RE system: VFX3524, Mate3, 200Ah 24V AGM Battery Bank
Location: Arlington, WA

Re: 3 phase stacking GSLC's

Post by Kurt Lundquist » Fri Nov 22, 2019 8:58 am

For any application with sensitive/expensive 3-phase loads, it is a good idea to install a phase monitoring relay as close as possible to the load. Even the grid can lose a phase so I consider it cheap insurance.

Here are 3 other reasons to consider installing phase monitoring relays for each of your 3 phase loads:

1. With stacked single-phase inverters there will inevitably be some amount of latency before the other two inverters realize the 3rd inverter is offline.

2. If the site has inductive or capacitive single-phase loads the phase angle might be unbalanced which can also damage a 3-phase load.

3. Inverters can only see voltage and current at the inverter output. The load can see something much different depending on how the AC distribution is set up.

Cheers,

-Kurt
Kurt Lundquist
Renewable Energy Project Engineer
http://alphaenergy.us/

JasonPAtkins
Forum Member
Posts: 9
Joined: Wed Sep 11, 2019 12:50 pm

Re: 3 phase stacking GSLC's

Post by JasonPAtkins » Tue Dec 31, 2019 1:43 am

EMCF wrote:
Thu Nov 21, 2019 4:04 pm
JasonPAtkins wrote:
Thu Nov 21, 2019 7:39 am
Your description of the Multi Phase Coordination feature admits that Outback knows the system should reject incomplete (2 of 3 phases) external AC input, because it would be dangerous to connected equipment to pass that electricity on. In the same way, the inverters shouldn't CREATE incomplete three phase and send it on to the powered equipment, which is exactly what they do if one inverter has an error.

It is not dangerous. On the contrary it is perfectly safe. You can have a single phase input (120 or 230 for you) to the master inverter, and the system will accept it (Multi phase Coordination disabled). The 3-phase output will still be a normal 3-phase output that will feed the load..
JasonPAtkins wrote:
Thu Nov 21, 2019 7:39 am
What you are describing is exactly how I think the Multi Phase Coordination feature does work. What you're describing is also not the problem I'm describing.
Ok, let us go to a pure off-grid 3-phase configured 3 outback inverters. The use case you are saying is about having exclusive 3-phase loads. In that case I agree with you that it presents some issues when an inverter's output malfunctions or completely lose its output for a phase loss especially when a 3-phase load like a motor is on the line. But not all use cases are like that. Some just like the possibility of having a single phase 230VAC, or a single phase 400VAC, or in our case here 120 or 208. We can afford to put to rest one or two of the inverters once in a while if there are no use for the higher voltage.
Of course they aren't all that way - but some *are*, so why would the default behavior be to damage equipment in that (easily anticipatable) use case? We can argue about which configuration is more common (your case probably is) and that can influence which the default behavior should be (I'd still argue for the default being the one that doesn't damage equipment - because if mine is the default, it doesn't damage anything in your proposed case). I'll chalk the "which way should be default" up to a difference of opinion.

However, that still doesn't excuse Outback, because it's not that they chose the default behavior your way over mine. You *CAN'T* change it to work the way I'm saying. And that is ridiculous, given that other vendors have been defaulting to it for almost a decade.

JasonPAtkins
Forum Member
Posts: 9
Joined: Wed Sep 11, 2019 12:50 pm

Re: 3 phase stacking GSLC's

Post by JasonPAtkins » Tue Dec 31, 2019 1:49 am

Kurt Lundquist wrote:
Fri Nov 22, 2019 8:58 am
For any application with sensitive/expensive 3-phase loads, it is a good idea to install a phase monitoring relay as close as possible to the load. Even the grid can lose a phase so I consider it cheap insurance.

Here are 3 other reasons to consider installing phase monitoring relays for each of your 3 phase loads:

1. With stacked single-phase inverters there will inevitably be some amount of latency before the other two inverters realize the 3rd inverter is offline.

2. If the site has inductive or capacitive single-phase loads the phase angle might be unbalanced which can also damage a 3-phase load.

3. Inverters can only see voltage and current at the inverter output. The load can see something much different depending on how the AC distribution is set up.

Cheers,

-Kurt
Kurt, you are right of course. My argument with EMCF is only because I don't understand why he's so blindly defending Outback, who could've provided an additional layer of security for 3ph loads easily, but didn't.

Yes, you are correct, there are reasons that it's better to have phase protectors on each load device rather than only relying on a system-wide protection like the group-inverter-fault configuration would provide. I'm not arguing otherwise. All I'm saying is that Outback could easily have given an *additional* layer of security here that would have brought it's three phase output back on par with the safety of a generator's output (where it's very difficult for the device itself to supply only two of three phases), but Outback chose not to despite the competition working that way for a long time.

I'm not saying the devices faulting together is a comprehensive 3 phase protection solution. I'm just surprised that company that makes stuff as good as the rest of Outback's equipment didn't choose to include this protection, at the very least as an option you could turn on.

Post Reply