question

mike-lynch avatar image

Generator in ESS system on ACIN1 fails to connect when Grid available on ACIN2

Existing system:

Grid supply is 2 phase (120 degree). 2 x 48/8000/110 - 100/100. One on each of the 2 phases.

9.6kWpk on 2 x Delta Solivia 5.0 AC inverters (one per phase - old Grid Tie system) on AC Out of Quattros.

6.6kWpk Solar on 2 x 150/85 MPPTs DC Coupled.

Battery is 9 x BYD B-box LV 23kWhr

Latest firmware in everything. Quattro's are running the UK grid code and are feeding excess solar to the Grid.


I'm attempting to connect a 15kVA 3 phase backup generator to the ESS system and have connected two of the three phases to the ACIN1 of both Quattros. (The grid is connected to ACIN2). To balance the load on the generator I've connected the third leg to a Titan 48/50 battery charger. I had to disable LOM on the ACIN1 to get the generator to stay connected. It all works perfectly in regards to connecting and charging. I'm not concerned about the interaction between Gen/Quattros/AC Inverters as the generator won't be running when the sun is out anyway.

The problem I have is that the Quattros will not connect to the generator when the Grid is available on the ACIN2. It hooks up every time if I isolate the Grid.

I figured I could possibly rectify it with an Aux input/Assistants to "Ignore ACIN2" but it didn't work either.

As soon as the Generator is available on ACIN1 it diconnects the Grid on ACIN2 but won't connect to the generator. If I then isolate the Grid it hooks up and starts to charge. I can then reconnect the grid and the generator stays connected as it should. When I shut the generator down it disconnects and the grid reconnects on ACIN2 as it should.

I've read all I can find on the subject but can't find a solution.

Any help would be appreciated.

Mike L.



MultiPlus Quattro Inverter ChargerESSGeneratorAC PV CouplingBYD
10 |3000 characters needed characters left characters exceeded

Up to 8 attachments (including images) can be used with a maximum of 190.8 MiB each and 286.6 MiB total.

5 Answers
mike-lynch avatar image
mike-lynch answered ·

Hi @Daniël Boekel (Victron Energy Staff)


Spent a bit of time on this today and learnt the following:

Changing the L2 to floating made the problem worse. I couldn't get the generator to hook up at all (with the grid isolated) after that. Would still connect to the grid OK but not the generator.


Back-dated the firmware first to 459, then to 433. Same issue with 465 so re-installed 465. Tried making the other inverter L1 (master) and set the other to 120 degrees - no change. Hooks up to generator but not with grid present on ACIN2 - same as before. Put them back as they were (Dual phase 240 with L2 fixed).


Changed the grid code from "UK: G98/1 March 2019, G99/1 May 2018" to "Germany: VDE-AR-N4105:2018-11, external NS Protection". Disabled LOM detection on ACIN1 and used default Type B for ACIN2. Everything works as expected. While connected to the grid, you can start the generator and within 6 seconds the grid is disconnected from the ACIN2 and within 25 seconds its connected to the generator. After shutting down the generator it re-connects to the grid within 15 seconds.


So while I'm happy that I've isolated my issue to something in the firmware associated with the UK grid code, it raises a few questions:

1) I'm in New Zealand and using the UK grid code because it works well with the ESS and because VE Configure wont let me select "New Zealand: AS/NZS 4777.2.2015" I noticed that when I selected the Germany code above it has warnings about using the Quattro with an external Anti-Islanding devise. There is an additional waring in the ESS setup about same. I thought we were able to use the Quattro and make it compliant here in NZ with the use of external Anti-Islanding equipment as well. If that is true how do we select the NZ grid code. I don't see how the Quattro would recognise it was connected to an Anti-Islanding device automatically? A pop up with "This device is not equipped with internal interface protection! So you cannot choose option: New Zealand...…….", and you can't use it. Could this be changed so it could be selected for use with external Anti-Islanding or does this error relate to something else other than Anti-Islanding?


2) Do my findings indicate there is an issue with the UK grid code that can be rectified or is the way it is handling the generator the way it has to be for some other component of the grid compliance?

Thanks again. Mike L.

3 comments Share
10 |3000 characters needed characters left characters exceeded

Up to 8 attachments (including images) can be used with a maximum of 190.8 MiB each and 286.6 MiB total.

Hi @Mike Lynch,

Glad to hear that you have found some resolution to the immediate problem.

When using an external grid protection device, that external device is the one that is compliant with the grid code and making sure that the AC supply is complying with all the required regional grid settings, and not the MultiPlus/Quattro.

There is no interface to communicate to the Quattro when this device is connected, so it is the responsibility of the installer to ensure compliant equipment is installed and configured correctly (and the reason for the grid code password).

This includes selecting the grid code that is compliant with the other requirements of the local utility (eg power factor correction).

Hi @Mike Lynch

Thanks for finding this out and the extensive report, I'll forward the ACin 1/2 gridcode behavior to the right person.

Hi @Mike Lynch

Just to let you know: this is on the list now, will report back when it is fixed (might take a while)

Also: please make sure you use an anti-islanding device when required due to local rules.

Daniël Boekel (Victron Energy Staff) avatar image
Daniël Boekel (Victron Energy Staff) answered ·

Hi @Mike Lynch

Did you try with 'L2 floating' in the settings?


4 comments Share
10 |3000 characters needed characters left characters exceeded

Up to 8 attachments (including images) can be used with a maximum of 190.8 MiB each and 286.6 MiB total.

Hi @Daniël Boekel (Victron Energy Staff),

No I haven't tried that. I can though and see if it makes a difference.

Can you explain exactly how this works? When does it actually "float"? Obviously it will stay locked to the grid, and the generator at 240 degrees while connected to either ACIN, so does it just float while in Inverter mode? If that is true, then is the graphic on the screenshot a true representation of the floating phase shift? It looks like it will only float between 120 - 240 degrees from the master?

Thanks for taking another look.

Mike L.

for example if you have a multi-phase and single phase source, you can use this. (of course the loads should all be single phase, and not shared between inverters)

I don't know if it will help you here, but worth a try, maybe your grid source is 120 degrees and generator 240 degrees apart?

You do mention you're using a grid-code, this can / will be of influence of what is accepted and not accepted as input.

If the problem persists, it might be best to use a relay to disconnect grid when running the generator.

mike-lynch avatar image mike-lynch Daniël Boekel (Victron Energy Staff) ♦♦ ·

Thanks @Daniël Boekel (Victron Energy Staff).

I will give it a try and report back.

I've proved beyond any doubt that both the grid and generator inputs are at 240 degrees apart. I also know from testing that with the Quattros set up as Dual Phase 240 with L2 fixed, there is no way they will connect to a 120 degree input.

The Quattros accept both of these inputs and will swap back to the available grid when the generator shuts down perfectly so I don't believe the grid-code itself is the issue here.

I'm not attempting to do something outside of what Victron documentation states can be done, so would sooner get to the bottom of why this is not working as it should be, instead of introducing "work-arounds".

Cheers..............Mike L.

Please let me know the results:)


(I was confused by your opening post where you stated that grid was 120 degrees apart, but that's of course just a phase swap away from 240 degrees)

Guy Stewart (Victron Community Manager) avatar image
Guy Stewart (Victron Community Manager) answered ·

Hi @Mike Lynch,

There is a new feature in VEConfigure that should give some indication WHY the grid or generator is rejected:

https://www.victronenergy.com/blog/2019/07/17/ve-bus-firmware-460-veconfigure-and-venus-os-v2-33/

Can you please have a look and see if that provides any more information?

3 comments Share
10 |3000 characters needed characters left characters exceeded

Up to 8 attachments (including images) can be used with a maximum of 190.8 MiB each and 286.6 MiB total.

Hi @Guy Stewart (Victron Community Manager)

Yes I've seen that feature and would be great but what it was saying makes no sense.

The blue line moves from left to right and when it reaches the end it fails to connect, goes back to the start and momentarily the text "Out of Phase" pops up at the top of both phase monitor screens. It disappears and has another attempt.

You can let it do that 10 times with the same result, but as soon as you isolate the grid from ACIN2 it connects on the very next attempt. It's like its trying to synchronise with ACIN1 but looking at ACIN2

Thanks for taking a look. Mike L.

There are several possible phases configurations to swap.

Grid can be physically swapped, generator can be physically swapped, and Multi's can be swapped in programming.

If you swap the phases around on each of the possible variations, grid, generator and multi programming, do you get the same error message in VEConfigure and behaviour, or is it different?

From there on in, usual AC work caveat applies. I am not an electrician, and the following work is potentially fatal without the correct equipment, training and safety procedures. If you are not qualified to undertake this info in your area, please employ a suitably qualified technician to troubleshoot any further.

Have you put a multimeter between the actives of the grid phases, and the respective (should be in same phase) actives of the generator and made sure that they are in phase relative to each other while connected to the multi?

Does the generator have a neutral? Is there a common Neutral between multi's, the grid and generator?

mike-lynch avatar image mike-lynch Guy Stewart (Victron Community Manager) ♦♦ ·

With the setup being configured to "Dual Phase 240" with "L2 Fixed" the options are 1 & 3, 2 & 1, and 3 & 2 with the same rotation, all of which are 240 degrees apart. No choice with the grid - it's only a two phase supply. Same result in each case with the VE.Configure Monitor.

I had to use two multimeters, one on each of the two Quattros. I had to reduce the engine speed of the unloaded generator to try and get the frequency similar to the grid. I managed to get it close enough to prove that the voltage between grid and gen actives were the same on both phases at the same time, albeit continuously changing.

The generator does have a neutral.

The Grid, Generator, and the inverter chargers AC inputs all have their neutrals connected in the same neutral bar.

Thanks again. Mike L.

Daniël Boekel (Victron Energy Staff) avatar image
Daniël Boekel (Victron Energy Staff) answered ·

Hi @Mike Lynch

What firmware are you running on Quattro and GX?

1 comment Share
10 |3000 characters needed characters left characters exceeded

Up to 8 attachments (including images) can be used with a maximum of 190.8 MiB each and 286.6 MiB total.

Hi @Daniël Boekel (Victron Energy Staff)

The firmware in the Quattros is 465, and the CCGX has 2.40.

Thanks.

Matthias Lange - DE avatar image
Matthias Lange - DE answered ·

Maybe there is a twist at the phases!? So you don't have the same order with grid and generator.

4 comments Share
10 |3000 characters needed characters left characters exceeded

Up to 8 attachments (including images) can be used with a maximum of 190.8 MiB each and 286.6 MiB total.

Thank you for your suggestion. The Quattros are configured to "Dual Phase 240" with "L2 fixed". I tried changing the generator phase rotation but it wouldn't connect at all when it was wrong.

"As soon as the Generator is available on ACIN1 it diconnects the Grid on ACIN2 but won't connect to the generator. (...)"

How long did you wait? If the generator voltage and/or frequency fluctuate to much, it's hard for the Multi to synchronize and it can take up to 2 minutes.

mike-lynch avatar image mike-lynch Matthias Lange - DE ·

I wasn't timing it but was waiting for easy 3, maybe 4 minutes. The generator runs quite stable unloaded at 236V L-N, and I did lower the frequency to about 51.3Hz. This is what I don't understand: With the grid isolated from the ACIN2 the generator connects every time within 20 seconds. Thanks again.

Hm ok, I'm out off ideas at the moment...
I never used such a setup at my own so I can't imagine what can cause this behavior.