Online assistance for electrical trade people in New Zealand and Australia 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: generator SDoC

Apr 11 2018 11:37

are 230v generators under the medium or high risk list? therefore requiring an SDoC.

the I\'ve been supplied with is an inverter type 2600 w. it doesn\'t have a RCD fitted to it and so it must (?) be an isolated type.

Its a Kipor IG2600

I cant see them there.

Apr 11 2018 14:57

Had a quick look and couldn\'t see it on the medium risk or high risk database. ESR80 requires new fitting or appliance to comply with AS/NZS3820 or Schedule 4 or a CCA. Evidence of compliance with standards (ESR81) is in the form of a test report or certificate of compliance. I would ask the supplier to provide a test report or COC if its not covered by a CCA.

Apr 11 2018 14:58

gensets are not on either list.
they don\'t even have to comply with relevant Standard: AS 2790.

That is getting very dated, and is likely to be withdrawn; but the relevant bits are now in AS/NZS 3010: 2017.
Which you can get via EWRB link

Apr 11 2018 21:10

Many inverter generating sets are a isolated output.

Check the handbook and the set nameplate of the generting set to see if genset maker has made call.

Otherwise you have to meansure the output voltage when running on load. I wouls suggest at least a 1kW load and then measure the output voltage active and neutral with respect to the frame, to see if there is an internal earth to neutral strap.

It the genset has RCD protection built-in it will NOT be an isolated supply as you must have an earth bond to ensure that the RCD will operate correctly.

AS/NZS 3010:2017 Appendices A and B has all the thoery you will need on the use of isolated supply and use of RCDs with generating sets. The EWRB website has copies of AS/NZS 3010:2017.

Apr 11 2018 23:54

ESR 80 does not \"require\" compliance with Standards; It simply requires \"electrically safe - and compliance with those listed Standards gets \"deemed to be electrically safe.
Many small gensets do not comply with relevant standards.

Apr 16 2018 06:56

This gen is being used on a transportable structure, and has no RCD on it. so i\'m going by 3001 - page 19.

Its a \'inverter type generator\', would this be the same as an isolated type?

what volatages should I expect across Ph, N and E?


Apr 16 2018 09:00

In order to comply with \"3001\", the genset must comply with clause; ie be either RCD-protected output or isolated output i.a.w \"3010\".

Whether the genset has an inverter is completely unrelated to whether it is isolated output, RCD-protected, centre-tapped, or any other output configuration.

As Pluto said, if it is RCD-protected (ex factory, to ensure compatibility of RCD with output waveform); you would expect 230 V A-N, and A-E, and 0V N-E. Basically one side of the output - regardless of whether simple winding or inverter output stage - will be tied to frame, and becomes \"neutral\".

For isolated output, you should get 230 V A-N, and 0 V A-E & N-E. Again this is unrelated to whether or not there\'s an inverter stage; and in this case NO PART of the winding / output is connected in any way to frame - allocation of one side as \"neutral\" is just arbitrary; and as a source it\'s electrically equivalent to an isolating transformer.

An on-load voltage test is a good indicator, but you should also do an IR test between output A+N to E and to frame. A clear test (both ways, \'cos your IR tester puts out d.c. and there may be semiconductors) to confirm isolated output if that\'s what the voltage test indicates.


Apr 16 2018 09:04

Just a follow up question.

Is it permitted to have an RCD protected generator with a switch for the MEN link?

I\'ve heard that some rental generators might have a switch?

So that it can be plugged into an installation, and also used to power standalone equipment.

I\'ve had a look through the new 3010, but I can\'t seem to find anything about switching the MEN link.


Apr 16 2018 10:56

With no load I get
L-E 240
L-E 83
N-E 83

with approx 1600w load I get

L-N 240
L-E 100
N-E 3

So this is neither an rcd type or isolated type. Back they go!
N-E 3

Apr 16 2018 11:04

Try measuring the voltage across L-E with a load connected from L-E

Inverter generators have a three terminal capacitor across the output and you\'re probably measuring that voltage.

Try and google inverter wiring diagram. All those small models are virtually identical wiring and inverter block.


Apr 16 2018 11:21

I had a look for the diagram. And that Kipor model came up but doesn\'t show the capacitor.
If you check the diagram in the link below, you\'ll see \"noise filter\" in the inverter block.

I\'m not sure if these truly qualify as \"isolated\" for NZ standards, but a similar Inverter is the type that I use to supply my MEN installation and they work fine.

Apr 16 2018 11:26

If there is voltage to earth then it can\'t be an isolated generator type then can it?

Apr 16 2018 12:12


Diagram I have just shows supplies to outlets coming straight from Inverter unit, and earth not connected to anything. fairly basic line diagram though.
Looking on the net for Gens, there are lots of them that are of this same type, inverter type. Ryobi, Kipor, Honda etc. surely they cant all be non compliant?


Apr 16 2018 12:12

The problem with having a \"switched link\" type connection is that it is certain to be switched to the wrong position at some stage; which could lead to it being electrically unsafe.

You really can\'t have it both ways; needs to be either one or the other; ie either set up for use as portable supply or set up for use to supply installations.

Centre-tapped units are only any use as portable supplies direct to equipment.

RCD-protected are OK for direct supply too, and can be used for some (modern) types of connectable installation; but no use for anything with a N-E link.

The only type that\'s OK for all applications is isolated output, and no that does not include having noise filters or anything else between output & frame.

Apr 16 2018 13:22

But even on an isolated type, the frame would still be connected to the installation earth though? So having this noise filter connected to the earthed frame and connected to the MEN installation shouldn\'t be unsafe right?

It\'s no different to an isolated generator which may have a slight amount of leakage to the earthed frame due to moisture or whatever which we might be able to measure a slight voltage, but it would still pass an insulation test.

Really, the only reason we even know these voltage are there, is because we\'re measuring with a modern, very high impedance digital meter. 30 years ago, if we measured with our Duspols, we wouldn\'t see any voltage.

Apr 16 2018 14:42

IR test on gen are >500M ohm on L-E and N-E, same to frame.
So this show isolated type, so maybe voltages I\'m seeing are only because of high impedance meter?

Apr 16 2018 14:51

I get different voltage readings from ph and e when I use different meters, meters are fluke tm600 and fluke 1653.

No duspols here either.

Apr 16 2018 16:06

Yes, with isolated output the frame and the earth contacts of sockets are connected to exposed conductive parts of equipment. In this set-up it is no longer earthing for fault current, but simply equipotential bonding (to deal with double-fault scenarios).
For supplying a house, it\'s back to being earthing, but sort of in reverse; the genset frame is earthed to the \"MEN\" switchboard which in turn is connected to an electrode.

Apr 17 2018 16:06

So what is the correct sort of Generator to use and to look out for, just wondering incase a customer wants one one hooked up to a changeover switch, i.e. is the little Honda EU20/22 ok

Apr 18 2018 08:15

look for one with isolated output.

Apr 18 2018 10:43

Action - You\'ve checked the output of the generator with your IR tester and found that the A & N are not connected to the frame. That\'s all that NZS3010 specifies for isolated output in D3.1.9 and other places.

I know Alec or someone may that there are specific standards that the generator manufacturers may have to comply with. And it\'s quite possible that the generator manufacturers aren\'t specific in that compliance. I\'ve checked my inverter\'s nameplate, brochure, instruction manual and manufacturers website and I can\'t see anywhere that it states \"complies with NZS????\". And without looking up regulations or standards, I can\'t really comment on the requirements.

All I can say, is that the \"voltage\" that you measured to the frame, doesn\'t indicate that it isn\'t an isolated generator. If you measured that voltage with a 2k ohm resistor in parallel, like a touch voltage test, I think the results would be different. I may try that later.

I had exactly the same concern when I first tested mine, and posted the same question on here before I carried out further investigation and tests. But I can tell you that they work fine for supply to a MEN installation. And once they are connected, of course there will be no voltage between N & ht frame as the frame will be bonded to E.

As for Alec\'s comment \"not include having noise filters or anything else between output & frame.\", in my opinion, that is not the required \"test\" for being an isolated type. In practical terms, for an inverter model that the market share is touted as \"electronics\", as soon as anyone plugs in a laptop supply or a phone charger, there\'s a very good chance that there will be a filter in those as well, and the measured voltage to the frame will reappear, even in a pure isolated output.

Apr 18 2018 11:20

The problem with filters & similar is that if / when they break down, the isolation is no longer effective. Even before they break down, they can cause problems. Filters in VFDs are what causes a lot of tripping problems with RCDs. If there\'s a current path, some of the current will take it.

To be properly isolated - which after all is a safety function - there has to be NO connection. When we rely on an isolated supply as our means of fault protection, it\'s instead of the more common protection-by-automatic-disconnection.

Inverters are more complicated, they need isolation not just between output & \"earth\" (actually the equipotential bonding system of whatever is being supplied, which may or may not be connected to actual mass of earth) but also between output & input.

Exactly the same as for an \"isolating\" transformer - and therefore the same sort of test is required: min 1 Megohm.

It would be nice if gensets were declared articles, and had to comply with Standards. But right now they aren\'t & they don\'t; it\'s buyer beware. And when certifying connection to an installation; installer beware.

Apr 18 2018 11:43

My issue is that this gen is being used as the supply to a transportable structure, via the power lead, and as such the installation doesn\'t gave a MEN link at the switchboard, only a RCBO.
There is no change over switch as it will be fed either power lead or by gen.

Apr 18 2018 13:00

Alec. I\'m not sure if you can say they aren\'t \"properly isolated\" just because they have a filter which could break down, but it tests as being isolated. Even an isolating transformer could break down.

I agree they should be covered by additional regulations. There\'s so much confusion about their connection and the different types - not just from the public, but also electricians that don\'t understand the standards and how to apply them.

Action. Section 2.4 of 3001 covers the connection of transportable structures to a generator with the supply lead. It is allowed to be an isolated type. The test button on the RCD will work, but your plug in RCD tester will not because there is no MEN link.

Because the supply is isolated, a single fault to earth should not be dangerous. The RCD will still provide protection in the case of a double fault - like if the filter has broken down as Alec suggested, and there is another fault to earth.

Apr 18 2018 17:28

Thanks Pluto, I tested RCD and It tested as you said.

Apr 18 2018 17:50

not Pluto... he\'s the good looking one..

Apr 18 2018 17:56

Ha, sorry Doug. Help much appreciated

Apr 18 2018 18:02

That\'s ok...
So I dropped into the local hire shop to check what they had. Various ones from around 2kVA to 8kVA. It looked like they were all RCD protected. I asked the guy if they had any that could be connected to a house for a temporary supply, but of course he didn\'t really know (he wasn\'t one of the more experienced staff..)

He directed me to a larger generator hire company. But in any case, I didn\'t see any evidence of MEN link switches - which wasn\'t really surprising.

Anyway, I found one interesting one with an earth stake. Maybe because it has rubber wheels?

Apr 18 2018 18:21

I use something like this to make the flukes act more like an old Analog meter. Helps prevents inductive voltages.

Apr 20 2018 16:41

ppaw1965 Apr 18 2018 18:21

Your comment
I use something like this to make the flukes act more like an old Analog meter. Helps prevents inductive voltages.

My comment

I recommend the use of the 3 k ohm module instead of the 20 k ohm module, this would be a closer match to the 2 k ohm used in AS/NZS 3017 and many IEC equipment testing standards.

Apr 20 2018 17:01

Filters in inverter gensets and other high power electronic devices.

The noise filter is to reduce EMC issues and there are serval configurations in use some are better for isolated supplies and others are suitable to TN-C-S (MEN) supplies.

Many sets available in NZ will be made to IEC standards (mainly manufactured in Europe) the balance will be made in Asia or North America and many will have been built using a 110 Volt based standards. Many cheap gensets up to a 2 to 3 kW have a centre alternator output winding tapped earth and connecting these as an alternative supply to an electrical installation will in many cases result in the genset alternator bweing burnt out due the MEN link connecting neutral and earth togther.

There is a very simple genset connection in AS/NZS 3010 Appendix C, this system will allow any genset to be used. Take a look and you will see how simple a a domestic house connection for an alternative supply can be, needs a bit a work to connect the load to the genset and to return the plugin connections when the normal supply is available again.

Apr 20 2018 18:42

Pluto, I think you might have missed the fact that this is an inverter. There\'s lots of block diagrams or wiring diagrams on Google for inverters.

Apr 21 2018 08:25

DougP Apr 20 2018 18:42
Your comment
Pluto, I think you might have missed the fact that this is an inverter. There\'s lots of block diagrams or wiring diagrams on Google for inverters.

My comment
In my statement about filters it should be remembered that all inverter type generating sets will always have a filter for EMC reasons due to the high switching frquencies used.

Some EMC filters just consist of shunt connected capacitors and the more complex ones typically have series chokes.

It is the combination type of filters, Chokes and capacitors, that the connection of capacitors is arranged differently for isolated supplies when compared to gensets which are not classified \"isolated output\", of course thay also cost more.

Sorry I failed to make clear in simple rotating magntic field alternatorss, the only EMC filter is typically only shunt capacitors.

Apr 21 2018 13:08

Pluto.. No apology necessary. You were talking about both types anyway (now that I read it more clearly).

Regarding the 3k ohm vs. 20k ohm. Yes 3k might be fine for quick measurements, but keep in mind you\'re looking at less than 3W for 20k, and over 19W for 3k, so I wouldn\'t leave it connected for too long if there\'s actual full voltage present.

I checked my inverter generator output. With no shunt on the meter P-N ~235V. P-E ~115V. N-E ~115V. So this is obviously measuring across the split noise filter.
With a 3k3 resistor across the meter, P-E and N-E were around 10V and 3mA. Not sure how this fits with being isolated or not..