Online assistance for electrical trade people Login  |  Register  |   Forgot Password
Assistance for electrical trade people
 

 

 

 


Click here to send Ron a pdf document for publication on this Topic

Documents must be less than 200k in pdf format

Posted By Topic: Connectable installation two on board inverters

Rhino
Dec 11 2017 16:20

Any problems with having two “plug in” type inverters onboard connectable installation- both connected directly to battery’s . There will be one plug inlet connection near the msb . This is where the supply lead will plug into msb from inverter therefore only one inverter can be plugged in to msb at a time - the other is used as back up .. there will be 10a rcbo as mains switch ..
   

DougP
Dec 11 2017 16:54

You're saying it's "onboard", but then saying the supply lead will plug into the inverter.

You can't have it both ways. Either it's onboard, and connected as per 3001 section 3.4.4.3 including a changeover switch, or it's connected to the "portable" inverter by the supply lead under section 2.4

Personally, I would go for the changeover switch. As you can see in Fig. 3.1, the inverter outlet connector is shown, which can be the plug to connect to the inverter, or a permanent connection.

In my motorhome, I have a built in inverter which has a remote control panel and I have the switching as per Fig. 3.1. I also wired an additional pole on the changeover switch to turn off the 230V battery charger when being supplied from the inverter.

I can dig up the model of my inverter if you're interested. It's just a 1500W modified sine wave. But it's fine to run a (basic dial type) microwave and a 1kW coffee machine.
   

pluto
Dec 11 2017 22:29

I'm currently working on an amendment to as/nzs 3001 for using an inverters as another energy source.

A few years when the output transformer was the norm the use of an inverter as described by Doug P was OK as being electrically safe.

Modern UPSs don't have an an output transformer which do not give the old level of isolating from ELV input to the LV output was causes a number of high risk situations which require careful consideration and installation.

My suggestion as an interim arrangement use the UPS as a plugin device on the input and use the output sockets of the UPS to directly supply the load. And make sure the inverter complies with AS 62040 and in particlar, has been tested for correct operation of the backfeed protection.
   

Rhino
Dec 11 2017 22:34

Your comment;
You're saying it's "onboard", but then saying the supply lead will plug into the inverter.

My comment;
Yes thats correct two onboard plug-in isolated inverters, one is 24V 3000W and one is 24V 500W(side by side). One inlet connecter below the MSB feeding the MSB directly- therefore only one inverter can be plugged into the MSB at a time via short supply lead. You have to physically remove the lead from the inverter and plug it into the other inverter to change between each inverter.

Your comment;
You can't have it both ways. Either it's onboard, and connected as per 3001 section 3.4.4.3 including a changeover switch, or it's connected to the "portable" inverter by the supply lead
under section 2.4

My Comment;
Yes section 2.4 looks most applicable FIG 2.1. HOWEVER this allows for no MEN connection and still has RCD which notes also rcbo won't work. This contradicts FIG 3.1 when a isolated inverter is used a MEN link is provided?

Your comment;
Personally, I would go for the changeover switch. As you can see in Fig. 3.1, the inverter outlet connector is shown, which can be the plug to connect to the inverter, or a permanent connection.

My comment;
Yes agree however, the ‘intention’ is not to plug into mains, supply only from inverters( or generator as back up,) therefore FIG 2.1 looks most relevant. However is contradicted by FIG 3.1, so a little confused in which would be best practice. There is no need for a changeover switch as there is only ONE inlet connection with no future dual supply planned. Would it be a good idea to link the earth and neutral at the inlet connection behind the plug to enable the RCD to be operation ? Although i see its not actually required following FIG 2.1..?
   

pluto
Dec 11 2017 22:58

You need to seriously consider and check that the inverter has an output transformer to obtain full isolation from the d.c. input to the a.c. output.

If there is no ouput transformer is a transformerless design and these can be very lethal in some configurations and have caused fatal electrical deaths in the past.


Normal MEN thinking when using RCDs with transformerless inverters does NOT produce an electrically safe installation, you need to cover a number of potential risks to make it electrically safe and including when there is NO supply lead connection to a TNC-S (MEN) supply is being made.

NOTHING IN AS/NZS 3001 comes close to what is actually required.
   

AlecK
Dec 12 2017 08:45

For your intended set-up, just follow Fig 2.1.

Don't worry about the not re the RCD, it's something of a red herring. The RCD simply has to exist, just in case the CI is ever plugged into "shore power".
If your inverters are RCD-protected types, the RCD in the switchboard becomes redundant. And if your inverters are isolated output, then you're achieving "fault protection" by separated supply, and not by automatic disconnection; so again the RCD in switchboard is (almost) redundant.
And if your inverters are neitherof these two types, then your set-up is non-compliant and potentially unsafe. It comes down to complying with 2.4.1.

Ignore Pluto's remarks about UPSs, you're not using UPSs so simply not relevant. His other remarks relate to using the kinds of inverters you're not allowed to use under 2.4.1.

And don't fret about the fact that for a change-over configuration, Fig 3.1 sets up a N-E connection. As above, the switchboard RCD is only there because the CI has to be suitable for "shore supply", even if it never actually gets connected to "shore supply". So ALL CIs have to have RCD protection.

The background to that, and to those notes in the Figs, is that some (very few) kinds of RCDs require a functional earth connection to operate. But most don't, and work perfectly well; because they don't rely on an intact PEC back to an N-E connection for operation; they simply compare outgoing & return current. And the RCD's test button works perfectly well, again because it doesn't rely on using the PEC to create the test current imbalance.
However RCD testers pretty much universally work by applying a test current onto the PEC - and if the PEC isn't connected to N somewhere upstream, the test won't work.


   

pluto
Dec 13 2017 07:25

RE-read my warnings and rfecomendations re transformer-less inverters very carefully again.

Much of the comments of Alec K are NOT correct and incorrectly describes the actual arrangement being used.
   

AlecK
Dec 13 2017 09:03

The fact is that currently only two sorts of inverter are permitted. If output not protected by RCD, must be isolated output. Transformerless inverters without built-in RCD are not permitted. So while those dire warnings of yours are technically valid, such a configuration is banned.
There's been NO suggestion that the inverters in this scenario have RCD protection built in. Therefore must assume they have isolated output, and that being the case they can be used as per Fig 2.1.
Result will be both safe and compliant.

There is a problem, in that while the Standard limits the use of unsuitable inverters, there's no practical enforcement. In this scenario the inverter is simply a source, and is not part of the CI's "installation" (unless hard-wired).
It's also not part of a fixed installation, subject to certification by a licenced person. So the "requirement" to only use safe types can't be enforced but can be easily ignored; leading to an unsafe situation.

It's also a fact that (with exception of those having a "functional earth" connection) RCDs do NOT need a PEC to operate. They simply sense current imbalance between A & N (or As & N for 3-phase), and react accordingly.

Where the PEC comes in is to satisfy requirements set in clause 1.5.5.3 of "3000", and also in 2.4.2; that fault protection by automatic disconnection must occur "on the occurrence of a fault" rather than later; and that an RCD by itself - with no PEC - does NOT comply.
Such an RCD will still work; the problem being simply that it won't operate when the fault occurs but when a person touches the now-live conductive parts and so establishes a fault current path. This exact scenario used to be suggested by Energy Safety as a means of improving safety of old installations with no PEC on the lighting.

There also the problem of output waveform, which may adversely affect operation of the RCD. But if the inverter has an on-board RCD ex factory, can assume no compatibility issues - and the caravan switchboard RCD is effectively redundant.

And assuming these inverters have an isolated output, then fault protection is not by automatic disconnection, and again the switchboard RCD is redundant. In the unlikely event of a dual-fault situation causing a current imbalance, the RCD will operate