Why does my main house consumer unit trip when the inverter switched to AC backup?

[Tonyboy]

Hello All

Can I say to start with I am not an electrician but used to be a heating engineer (yes I can hear the groans). I have wired the installation myself using the correct wiring and isolators, racking and trunking and it is not grid tied it is purely off grid with a separate input for a backup supply from a generator or the grid. The system does not feed back to the grid and when the mains is disconnected the system isolates the ac input internally in the cabinet with relays

In my solar setup I have a purpose built cabinet containing all protective devices, with isolators and an Axpert MKS V4 5.5kw inverter. The kit is housed in a fireproof cupboard outside the house. The unit is supposed to be a plug and play with 2 x us5000 pylontech batteries beneath the cabinet connected in parallel into the main input on the cabinet as well as AC input and AC outputs on the side of the cabinet. The PV goes into the cabinet and then into the appropriate connections fitted to the cabinet I have separate isolators external to the cabinet for both PV and AC and the supply for the AC backup in, runs off of a 32 amp line from the consumer unit.

The issue at present is that the main breaker in the house trips when the inverter switches over to utility to charge the batteries over night. The 32 amp m.c.b does not trip but the main isolator to the board does, I think it is an RDC isolator main switch ( I have attached a pic) The cabinet has a soft start relay for a generator but as the supply is from the mains the relay becomes active as soon as an AC supply is applied to the inlet socket.

Is there a device I could add to the line in before the solar cabinet and on the supply from the consumer unit in my house that would stop what seems to be a surge when the inverter calls for AC power.

My setup is that I have a separate solar supply to my house hold appliances via a new consumer unit next to the cabinet and then separate cabling to all outlets in the house.

I am trying to have a system where I power all of my regularly used stuff from the batteries and solar panels in the day and then just use the AC into the house as a way to supplement additional power to keep the batteries fully charged with cheap rate overnight electricity. At this time of year the solar gain will be fairly low, I am in the U.K.

Any suggestions would be welcome. I am new to solar and am finding my way albeit slowly.

Regards and thanks Tonyboy

 

[pc1966]

A couple of possible causes:

• You have some sort of N-E fault in your installation and it is only on-inverter that enough difference is present to cause enough current imbalance (N not matching L) to trip things.
• You have something else fundamentally wrong with the inverter or its wiring (N not earthed when running, or L & N swapped wihch is very bad)
• It is simply the act of bringing the supply on quickly that trips the RCD. This is quite likely if it is a 30mA device for the whole installation. If you switch the mains AC on and off via the isolator switch (RCD on and ready to go, not enabling inverter at all) does it also trip? If so its down to the arrangement you have.

 

[pc1966]

Just to add, a diagram would make it a lot clearer as to how things are wired up!

 

[Tonyboy]

A couple of possible causes:

• You have some sort of N-E fault in your installation and it is only on-inverter that enough difference is present to cause enough current imbalance (N not matching L) to trip things.
• You have something else fundamentally wrong with the inverter or its wiring (N not earthed when running, or L & N swapped wihch is very bad)
• It is simply the act of bringing the supply on quickly that trips the RCD. This is quite likely if it is a 30mA device for the whole installation. If you switch the mains AC on and off via the isolator switch (RCD on and ready to go, not enabling inverter at all) does it also trip? If so its down to the arrangement you have.

Hi pc1966
I am currently working through a number of tests to see what particular combination is causing the problem. This afternoon I set the inverter to u.s.b and set the time for utility to come on while I was standing next to it. I had turned off the p.v. isolator and when the inverter switched over it started charging the batteries with the 10 amp limit I have set. Tomorrow it is supposed to be sunny so I am going to seen if the P.V. on its own charges the system with no issue. I will leave the AC isolated completely.
I have turned the AC supply on and off a few times with the inverter not enabled and it did not trip, I will try it again tomorrow with the PV both on and off as well as with the inverter enabled and the pv on and off to see if there is something that sets it off in particular.

I should mention I had an F12 fault occur last week and it is a dc.dc overcurrent fault. The suppliers wants it back at my cost and says it is an mppt issue. I want to see if this is the case before I spend money getting a brand new inverter checked and probably charged for the privilege.
Initially I was concerned it was something to do with the pv. array being earthed to the battery earth and the ground rod so I disconnected the cable from the arrays to see if this solved the problem but it returned this morning when both ac and pv were connected to the inverter. Albeit the AC was the charging priority. Funnily enough though when the pv arrays were earthed to the ground rod the inverter never tripped the house at all. The only issue was when the earth was connected, the pv voltage measured at the connection to the cabinet went from 250v dc to380v dc and it was sufficiently close to the max pv input of 500 v to make me remove the earth cable, I wondered if the mppt was pulling voltage as a result of the earth connection in some sort of loop?
I will add more details when I have done some more tests.
This is from the suppliers site re the cabinet. I am not sure if this will conflict with my home system.
The unit is ready for direct connection to a back up source such as a diesel generator or the grid. The AC loads are connected a fully protected AC output plug , there is one 30mA 40A, RCD and MCB with TN-S earthing arrangement (common Neutral to Ground).
Thanks for taking the time to reply to my initial post I do appreciate it.
Regards

Tonyboy

 

[Tonyboy]

Just to add, a diagram would make it a lot clearer as to how things are wired up!

Just got this pc1966 I will sort out a drawing tomorrow hopefully, regards

Tonyboy

 

[pc1966]

What does the inverter manufacturer say about supply RCD protection?

You would be well advised to clamp the tails supplying the board that trips with an ammeter that can resolve mA to see what leakage is present before and after the inverter is running. If it is above 10mA then tripping is quite likely (as "30mA" are specified to trip below 30mA but above 15mA)

This bit:

Funnily enough though when the pv arrays were earthed to the ground rod the inverter never tripped the house at all. The only issue was when the earth was connected, the pv voltage measured at the connection to the cabinet went from 250v dc to380v dc and it was sufficiently close to the max pv input of 500 v to make me remove the earth cable

Really does not look good! For a start, if you have an independent supply you must have an independent means of earthing! You cannot rely on the DNO still being connected during any outage. Also is it not the inverter that gets earthed, and the array of PV pannels is floating so it can check the insulation resistance? (at least a friend who has an ABB inverter works that way) Again, what does the manufacturer say?

Also 380V is nowhere near a 500V limit! I would expect an over-voltage error from the inverter if it were the case.

 

[Tonyboy]

What does the inverter manufacturer say about supply RCD protection?

You would be well advised to clamp the tails supplying the board that trips with an ammeter that can resolve mA to see what leakage is present before and after the inverter is running. If it is above 10mA then tripping is quite likely (as "30mA" are specified to trip below 30mA but above 15mA)

This bit:

Funnily enough though when the pv arrays were earthed to the ground rod the inverter never tripped the house at all. The only issue was when the earth was connected, the pv voltage measured at the connection to the cabinet went from 250v dc to380v dc and it was sufficiently close to the max pv input of 500 v to make me remove the earth cable

Really does not look good! For a start, if you have an independent supply you must have an independent means of earthing! You cannot rely on the DNO still being connected during any outage. Again, what does the manufacturer say?

Also 380V is nowhere near a 500V limit! I would expect an over-voltage error from the inverter if it were the case.

Hi pc1966
I will have to look into an ammeter to clamp the tails, I don't have one of them at present.
Re the earth, the batteries are connected to the earth rod next to the cupboard with 25 mm cables and I ran a separate 6mm from the roof to the earth rod to protect the arrays from lightning etc. The cabinet has its own surge protection on both the in and output of the ac and the pv but I have no information from the supplier regarding what earthing arrangement to apply to the cabinet. I initially ran 6 mm earth from both the arrays and from the metal cabinet and they all terminated on the 6ft earth rod installed adjacent the cupboard. This was tested and was 0.003 ohms if I recall correctly.
My concern was that the cabinet being wired as as TN-S adding an additional earth path may in some way cause an issue with the inverter and its wiring. This may well be completely unfounded so I am more than happy to be proved wrong on that. The voltage did spike a couple of time near to 480v on the pv supply input and I did notice when the earth from the array was disconnected it went back down to 258V or there about.
It was suggested that the whole of the system was best to be connected to earth via the Battery earth and therefore I had assumed this would also be accounted for in the way the whole cabinet was wired.
Does connecting additional earth cable to the chassis of the cabinet and to the earth input of the solar pv cause any kind of issue with the separation of the electrical side of things and the mechanical side of earth bonding the arrays and frames?
Many thanks and regards

Tonyboy

 

[Tonyboy]

What does the inverter manufacturer say about supply RCD protection?

You would be well advised to clamp the tails supplying the board that trips with an ammeter that can resolve mA to see what leakage is present before and after the inverter is running. If it is above 10mA then tripping is quite likely (as "30mA" are specified to trip below 30mA but above 15mA)

This bit:

Funnily enough though when the pv arrays were earthed to the ground rod the inverter never tripped the house at all. The only issue was when the earth was connected, the pv voltage measured at the connection to the cabinet went from 250v dc to380v dc and it was sufficiently close to the max pv input of 500 v to make me remove the earth cable

Really does not look good! For a start, if you have an independent supply you must have an independent means of earthing! You cannot rely on the DNO still being connected during any outage. Also is it not the inverter that gets earthed, and the array of PV pannels is floating so it can check the insulation resistance? (at least a friend who has an ABB inverter works that way) Again, what does the manufacturer say?

Also 380V is nowhere near a 500V limit! I would expect an over-voltage error from the inverter if it were the case.

Hi pc1966
I have done a rudimentary drawing showing my system overview, I hope it is sort of self explanatory.
I did have the earth connected from the bus bar to the earth rod but with it connected the voltage at the pv input in the cabinet was up to 450 + volts. since disconnecting it the reading is 252 v with the inverter off, and 350 to 400 with it on. I considered adding an earth from the consumer unit serving the house to the ground rod as this would then protect all of the house appliances and I was considering adding an earth to the cabinet as well as adding back the pv earth all to the ground rod.
I wondered if it was better to earth the pv rails and panels to a separate rod in a different part of the garden a fair bit away from the house so that this was not connected to either the battery earth or the ac earth to the consumer unit.
Currently I am trying to get information from the supplier but it is quite difficult, I had expected to receive a wiring diagram or at least a sheet showing the test results before the item was shipped but at present that has not been forthcoming either. If you have any suggestions please feel free to let me know. I want the installation to be safe and I am not using it to power the house at present so it is only running to test the p.v. charging and the a.c. charging.
Adding more ground rods and cable are relatively easy and would be better if that is what is needed.
Hopefully you can view the picture, many thanks and regards Tony

 

[marconi]

Please could you clarify the bit in bold in:

. The AC loads are connected a fully protected AC output plug , there is one 30mA 40A, RCD and MCB with TN-S earthing arrangement (common Neutral to Ground).

On the inverter load side including what you may have connected to it have you connected the Neutral and Ground/Earth together? This would be wrong to do if you are going to connect the inverter to the mains supply which already makes the N-E connection at or before your home's main fuse.

When the inverter output is supplied by the mains input, the presence of an inverter output N-E link after the house RCD, creates an alternative path for the neutral current to flow which is detected by the house RCD as a leakage of current to earth - so it trips.

 

[Tonyboy]

Please could you clarify the bit in bold in:

. The AC loads are connected a fully protected AC output plug , there is one 30mA 40A, RCD and MCB with TN-S earthing arrangement (common Neutral to Ground).

On the inverter load side including what you may have connected to it have you connected the Neutral and Ground/Earth together? This would be wrong to do if you are going to connect the inverter to the mains supply which already makes the N-E connection at or before your home's main fuse.

When the inverter output is supplied by the mains input, the presence of an inverter output N-E link after the house RCD, creates an alternative path for the neutral current to flow which is detected by the house RCD as a leakage of current to earth - so it trips.

Hello Marconi and thanks for the reply

The whole kit is in a purpose built cabinet, and the input and output are on the side of the cabinet with waterproof sockets. The main ac supply is via an isolator switch I installed outside of the cabinet at the end of swa cable and before the input socket, this is from a 32 amp mcb line from the consumer unit in the house.

The output of the cabinet goes to my separate consumer unit for the solar supplied sockets in my house. I have the line neutral and earth connected to the consumer unit and into MCB's and none of them are joined together. I was also going to be adding an earth from the consumer unit (solar) to the ground rod I have installed adjacent to the cupboard the unit is installed in, from the solar consumer unit so that in the event the dno power is cut off (which switches off and isolates the ac to the cabinet) I have a separate earth path as the dno's earth would effectively be removed also.
I am therefore assuming the output of the inverters AC arrangement in the cabinet is where the TN-S arrangement is carried out leaving the supply to be connected straight to the consumer unit (solar) in the manner I have described.
I have read a re read the description of the earthing arrangement (in bold) in the cabinet and can not see any other explanation for the assumption I have made.
That said I am at bit of a loss as to how I can get over this without having a separate small consumer unit fitted next to my current home mains unit with an mbc of 32 amp and then taking the supply to the cabinet from this.
Is that the only solution or is there a way to negate the issue by connecting the supply in a different way?
many thanks for taking the time to reply.

Regards Tonyboy

 

[marconi]

Have you an Ohmmeter/multimeter? Do know how to use it to measure resistance?

 

[davesparks]

What disconnects the installation from the DNO supply when the inverter is connected?

 

[marconi]

Is this the boxed system you have or something similar?

 

[DPG]

What disconnects the installation from the DNO supply when the inverter is connected?

That was what I was asking about yesterday when I asked about relays and isolation. But I posted it in the wrong thread! Doh

 

[Tonyboy]

Is this the boxed system you have or something similar?

Hi Marconi and davesparks

Yes the cabinet is similar to the one in the picture, it is from the same people. It is connected to the mains via a side input and in the event the power goes of the soft start relay cuts off the power to the inverter. I disconnect it from the mains manually if I don't want utility power to provide any input but it can be left on and it comes in when the timer is set to use utility.
The battery will still supply power to the solar consumer unit via the inverter when it is set to in the program but it cannot go back to the main supply as this either disconnects when the power goes off, and as the output from the inverter only goes to the solar output consumer unit
In terms of it disconnecting from the main supply in normal use I think the supply side stays connected but not used and it only powers the inverter when the timer calls for ac from the ac input.
Yes I have a multi meter I used it to check the resistance of the earth cable to the rod when I installed that.
Tell me more please!
regards

Tonyboy

 

[marconi]

The switching inside the inverter box is normally done by solid state switches to ensure no-break of supply when moving between mains to output or inverter to output.

Tonyboy - where have you happened upon the term TN-S? I suspect what you have bought or set up is an off grid systemwith genie back up to the inverter mains input. This would require the inverter output to be made tn-s in order for automatic disconnection of the inverter supply to the output socket in the event of an earth fault. This would be done on the inverter output. For uk mains input in lieu of genie one cannot link the output N with Cpc/earth/pe for the reason I mentioned earlier.

 

[Tonyboy]

The switching inside the inverter box is normally done by solid state switches to ensure no-break of supply when moving between mains to output or inverter to output.

Tonyboy - where have you happened upon the term TN-S? I suspect what you have bought or set up is an off grid input with genie back up to the inverter mains input. This would require the inverter output to be made tn-s in order for automatic disconnection of the inverter supply to the output socket in the event of an earth fault. This would be done on the inverter output. For uk mains input in lieu of genie one cannot link the output N with Cpc/earth/pe for the reason I mentioned earlier.

Hi Marconi
The description of the inverter outputs says it has a TN-S arrangement and when I purchased it I was told it was suitable to connect the mains as a standby/ additional source of charging power for overnight cheap rate electric.
It is indeed an off grid solar supplied unit with either standby genie or mains power to charge the batteries overnight.
Is there a solution to the issue of the cabinet being TN-S and does it involve changing or adding anything else to the system to rectify the issue. Currently I am having an issue with the inverter showing an F12 fault so I think that it will be going back to the supplier next week, however given they are selling this unit with the specific option of direct connection to a mains supply I had really expected it to work once it was connected.
I am content to change things if it will sort the problem but I don't want to be having to alter stuff in the cabinet as it is all pre wired and supposedly plug and play.
Many thanks again

tonyboy

 

[marconi]

Please see attached electrical art. Please study and note my remarks beside the three balloons. The ohmmeter tests at A, B,C and D must be done with all sources of electrical energy ie mains, battery and pv panels isolated. With ohmmeter set to low range eg 0 to 200 measure between N and E/cpc and tell me results.

 

[Tonyboy]

Please see attached electrical art. Please study and note my remarks beside the three balloons. The ohmmeter tests at A, B,C and D must be done with all sources of electrical energy ie mains, battery and pv panels isolated. With ohmmeter set to low range eg 0 to 200 measure between N and E/cpc and tell me results.

Hi Marconi
I think this will take me some time to do, as I will need to get access to the mains in my house and I am assuming the insides of the cabinet trunking and connections. To be honest it may be more than difficult to start dismantling the internal components of the cabinet as its all concealed, I will get in the cabinet and see what I can access and hopefully get back to you soon, I am working over the next three days on odd shifts so it may not be until the weekend or Monday.
I am assuming I is the inverter and the consumer unit on the right is my solar out unit.
Thanks for taking the time to do the drawing, I am very grateful.
My fears at present are that I am going to have to do some serious alterations to the set up and that was why I got a self contained cabinet to avoid this.

Quick edit I measured the point (b and c) before and after the rcd and got 00.2 for both and the consumer unit (D) with feed unplugged and all both on and off and got no reading. The main consumer unit is in a built in cupboard over the door and will take some getting to, so it may take me a while to get reading A.

We will see.
Regards

Tonyboy

 

[marconi]

What literature do you have that you can send me or a link to? Do you have a circuit diagram for the cabinet?

If your cabinet has on one side as in the earlier video clip two c forms for mains in and inverter out then I can spell out a way to do the measurements at the socket or plug N and E pins without dismantling anything. But that will be tomorrow because I am starting to cook for visitors tonight.