Jump to content

Growatt -5000TL Earth Leakage not tripping.


Harryc
 Share

Recommended Posts

Hi Jaco

I realized yesterday that when the inverter feeds from Eskom the earth-leakage works.

In other words power to the house is fed from eskom through the inverter to the house

When the house runs from batteries only it does not work.

Thanks

 

Link to comment
Share on other sites

On 2020/07/25 at 9:53 AM, Harryc said:

The earth leakage does not trip.

How did you test? I assume with one of these?

Selection_614.png.1075bb8233033e42cf7d3c4f3d90edc8.png

It is rather unlikely that a working earth leakage would fail to trip, usually the reverse happens. I can think of only one way this could happen... you have no earth.

The way these devices work is they take a small amount of current from the live connector and conduct it to the earth conductor. The earth conductor is supposed to be tied to neutral upstream, so this causes the current to bypass the RCD and this difference in current trips it.

If your earth is not properly connected, it has no way to conduct current past the RCD, and hence the RC won't trip.

Not all RCD testers have the features of the one above.  With one like this, one would think you'd see the fault on the lights display.

If the earth is properly connected (as in you can measure 230V from live to earth, an 0V from neutral to earth), and the RCD won't trip, it means the RCD has failed. But you already replaced it and it works on the other side... which brings me back to the conclusion that you must have a bad earth. The tester has no way to "leak" current to earth...

Link to comment
Share on other sites

38 minutes ago, Harryc said:

When it is fed from batteries it does not trip.

The only other thing then is bonding. If neutral and earth is not bonded, then again the earth leakage tester has no way to complete the path and create an imbalance.

Bonding must be done before the RCD, and must be done by the inverter (otherwise you're not SANS compliant). But I kinda sorta remember @jykenmynie said something about how some of these inverters simply don't do that...

Here is a test you can do. Turn the inverter off, and isolate the mains input to the inverter. Install a copper wire from the neutral wire at the top of the RCD to earth (ie bond it before the RCD). Turn on the inverter (leave the AC input disconnected). Test the RCD. Chances are it will work now.

Remove the bonding wire (SANS says you're not allowed to have two bonds) before restoring AC input.

Edited by plonkster
Link to comment
Share on other sites

If this turns out to be the case, it is a perfect practical example of how floating supply can't trip on a single earth fault. When you hit the button on the RCD tester, it deliberately creates an earth fault... but only one. A second one is needed before the circuit is complete. Bonding T and N is done so that a single earth fault will trip the RCD.

You said earlier that bonding it causes the RCD to trip all the time. I can only guess that you had it bonded on the wrong side.

Link to comment
Share on other sites

4 hours ago, plonkster said:

But I kinda sorta remember @jykenmynie said something about how some of these inverters simply don't do that...

Indeed (and I did again, a few posts up). My installation might have been faulty, who knows. For one, my one (of three) neutral bar in the main DB literally floats (not fixed to the board). It is the one connecting the neutrals coming from the inverter’s output. I only found this out recently.

Not sure if that was be a problem (that they aren’t connected to the other neutral bars, of which the one has the incoming neutral which is bonded to earth). But I do know that with my MP II, I do not have a floating neutral, even though I have the literal floating neutral bar.

As such, I know I had a problem with the Growatt (same model), I know I don’t have it with the MP II, but I don’t know if the Growatt would have had the problem had the neutral bar been fixed to the board.

Link to comment
Share on other sites

12 hours ago, Harryc said:

Is it safe to leave it like that?

In the past some sparkies have argued that it is safe. It is certainly safer than the alternative. It is not SANS compliant though. Bonding earth and neutral a second time like this allows working current to run on an earth wire, and that is incorrect. Earth conductors should only carry fault current.

If possible, I would try to get my installer/seller to swap the inverter for one that does properly bond earth and neutral.

Also, I want to compliment you on actually testing your RCD. Many people never do that...

Link to comment
Share on other sites

OK WOW WOW WOW...so you saying with me running this Growatt SPF5000TL and it running on batteries/solar and I touch a live wire, EL wont trip?

Easy method to confirm this? 

19 hours ago, plonkster said:

If the earth is properly connected (as in you can measure 230V from live to earth, an 0V from neutral to earth),

Would this be a valid test when running on batteries? How about when running on grid via inverter (changeover switch still on Inverter)?

@HarrycHow were you testing if it trips or not? 

Edited by Tsa
Link to comment
Share on other sites

1 minute ago, Tsa said:

OK WOW WOW WOW...so you saying with me running this Growatt SPF5000TL and it running on batteries/solar and I touch a live wire, EL wont trip?

It may, or it may not. The RCD trips on the residual current, the difference in current between the live and neutral conductors. Now this is where it gets complicated, because today we have so many different appliances with switch mode power supplies and filters and surge arrestors, and all of these cause tiny bits of current to be conducted between the three inductors, to the point that some people have reported that low power LED lamps continue to glow even after they have been switched off (there is enough leakage across the conductors to make this happen!). So an RCD might well trip even without the bonding because there is enough additional leakage already... but that is not really good enough.

The way to test it is to unplug everything from your backup circuits, so that you have no additional leakage currents. Then run on battery only (grid disconnected), and use an earth leakage tester like the picture above. The earth leakage must trip at 30mA, but a good one will usually trip between 15mA and 20mA. I've never had an RCD that didn't trip at 20mA.

The second test, again while running from batteries: Use a volt meter (such as a DMM, digital multimeter) and measure from neutral to earth at any plug on the backup side. In a properly bonded setup this measures zero (or very close... some multimeters will show a bit of noise at most), in an onbonded setup this usually sits around 80V-100V.

Link to comment
Share on other sites

8 hours ago, plonkster said:

The second test, again while running from batteries: Use a volt meter (such as a DMM, digital multimeter) and measure from neutral to earth at any plug on the backup side. In a properly bonded setup this measures zero (or very close... some multimeters will show a bit of noise at most), in an onbonded setup this usually sits around 80V-100V.

OK so initial test with everything connected as it would be every day and inverter supplying load from PV gives 0V AC between N and E on a plug. I did not unplug anything else. In this case Grid is still connected to the inverter,along with PV and batteries. Disconnecting grid from the inverter via the breaker into inverter also gives 0V AC between N and E. This is still with load being supplied by PV. 

Disconnecting the grid and PV from inverter, so purely batteries still gives me 0V between N and E.

I take it this is a  good sign, but the test to confirm all is 100% would be with the EL tester as indicated above.

Link to comment
Share on other sites

2 hours ago, Tsa said:

Disconnecting the grid and PV from inverter, so purely batteries still gives me 0V between N and E.

I know with the Axperts, some models had the bonding relay and some didn't. Maybe it is the same with the work-alikes. Or someone already bonded that one on the output side 🙂

Link to comment
Share on other sites

I know they have a few 5000TL models. There’s the HVM-P I had (I think, because the manual I got was for the HVM without the P and didn’t match the settings I saw on mine), and there’s a HVM-WPV or something as well. Probably a whole lot more. Who knows whether there’s difference or not between them. All I know: I had a floating neutral, with the MP II and nothing else changing, no floating neutral.

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

 Share

×
×
  • Create New...