Goodwe 5Kva - Panels will not charge the batteries if Eskom is not present

[LLTHB]

Hi to all.

My heading explains my problem.

The Goodwe is 5 years old and it used to charge the batteries during load shedding.

About 2 years ago they did a firmware upgrade and now the unit no longer chargers the batteries if Eskom is off.

The inverter supplies power to the appliances etc, but lets say I am only using 1000 watts on a bright sunny day and it could make 5000 watts and charge the batteries, it no longer does this.
The agents are about as effective as the ashtray on my motorbike and just speak goobledegook rubbish.

Anyone, anyone anyone?

Thanks in advance.
Cheers.

[Scorp007]

7 hours ago, LLTHB said:

Hi to all.

My heading explains my problem.

The Goodwe is 5 years old and it used to charge the batteries during load shedding.

About 2 years ago they did a firmware upgrade and now the unit no longer chargers the batteries if Eskom is off.

The inverter supplies power to the appliances etc, but lets say I am only using 1000 watts on a bright sunny day and it could make 5000 watts and charge the batteries, it no longer does this.
The agents are about as effective as the ashtray on my motorbike and just speak goobledegook rubbish.

Anyone, anyone anyone?

Thanks in advance.
Cheers.

Let's hear what @Bobster. can tell us as he knows the Goodwe inverters. 

[Bobster.]

14 hours ago, Scorp007 said:

Let's hear what @Bobster. can tell us as he knows the Goodwe inverters. 

I have nothing to offer but sympathy.

@LLTHB has described the situation very well. The only difference is that I get this intermittently (or it looks that way). It is happening this morning. On a lovely sunny morning I was getting 400W of PV (or the inverter was demanding 400W) at 8:30. The two previous days it was delivering > 2kW with the same loads at the same time of day. Load shedding started at 8:00.

So typically when I have the problem described, the inverter is pulling roughly load + 50W from the panels (irrespective of conditions) with the 50 odd going to the battery. If you push you loads up, the system will draw more, but still only 45 to 50 W going to the battery. So it seems to be a question of how much the inverter demands from the panels.

We have load shedding this morning. Until this morning I thought the problem was caused by low grid voltage which has been a problem where I live. But that has been resolved, the system was behaving beautifully since the voltage was restored, and until this morning. The reality is, I now think, that when the voltage was low, the inverter was disconnecting from the grid and so it may as well have been load shedding.

I need to start keeping a record of how the system behaves when we have morning load sheds (on a good day the system is fully charged by 11:00), but I suspect my experience will tally with that of LLTHB. I think I was led down the wrong path by the reoccuring low grid voltage issues where I live.

My system is about 4.5 years old, another match with what  @LLTHB describes. 

The frustrating thing is that I KNOW from other experiences early on that Goodwe can get a very good look at our machines remotely and can do configuration and check logs. They can even do firmware upgrades - but don't do so without an instruction to do so. But at some point the support channels changed and also Sunsynk has taken the market and these days when you say "goodwe" to somebody in this business you will, on a good day, get "I've heard of them" and mostly you will get a "huh?". Which is no judgement on these people, but a reflection on how few Goodwes there are in the field and that resources are line with their market penetration.

Edited February 8 by Bobster.
Include time load shedding started

[Bobster.]

21 hours ago, LLTHB said:

Hi to all.

My heading explains my problem.

The Goodwe is 5 years old and it used to charge the batteries during load shedding.

About 2 years ago they did a firmware upgrade and now the unit no longer chargers the batteries if Eskom is off.

The inverter supplies power to the appliances etc, but lets say I am only using 1000 watts on a bright sunny day and it could make 5000 watts and charge the batteries, it no longer does this.
The agents are about as effective as the ashtray on my motorbike and just speak goobledegook rubbish.

Anyone, anyone anyone?

Thanks in advance.
Cheers.

What battery do you have?

[Bobster.]

There is a Goodwe user group on Facebook. I've posted a question there. I am also trying some other channels. @LLTHB I will let you know if I get any bites.

[Bobster.]

49 minutes ago, Bobster. said:

So typically when I have the problem described, the inverter is pulling roughly load + 50W from the panels (irrespective of conditions) with the 50 odd going to the battery. If you push you loads up, the system will draw more, but still only 45 to 50 W going to the battery. So it seems to be a question of how much the inverter demands from the panels.

And load shedding ends, grid comes back, loads are low, but now the PV side is pumping. So it is exactly as @LLTHB describes.

Edited February 8 by Bobster.

[Bobster.]

An interesting aside: If this is a bug in the firmware, or even a design feature, Goodwe users in most other places won't be noticing it, because they have nice, stable grid supply. It is possible that Goodwe have not seen it or that it's not a condition they test for. 

Moral of the story: If you want to be sure the hybrid inverter you make is bullet proof, get some test sites in South Africa.

[Bobster.]

I have interesting feedback from another Goodwe ES user.

1) He thinks it never makes best use of PV when in general mode and there is grid
2) He thinks it makes best use of PV in off-grid mode (which could be tested by switching between the two modes)
3) But he doesn't see this problem. In fact he sees the opposite - when grid goes, his PV production INCREASES.

I did speak to the guys who installed my system. They just called back to say that Goodwe would be logging into my inverter to check settings. I have load shedding schedules for 10:00 tomorrow, so another chance to watch it misbehave. I will post any feedback I get/have.

[Bobster.]

So this morning the situation at Chez Bobster is as follows
1) Loadshedding at 2:00
2) Something went pop and power never came back on
3) Another loadshedding slot starting at 10:00 
4) So no grid
5) Overcast weather (actually starting to clear a bit now, but we have rain forecast from mid day). 

I am running in off-grid mode, and it definitely is not doing the load + 50W thing. Currently I am getting about 1.6kW from the panels, and I've turned off all the big loads. The battery is getting most of that 1.6 and is charging (probably won't get all the way). 

Now the scientific and helpful thing to do would to be switch back to general mode and see what happens. But 
a) I am not going to mess with the current situation since I would like to have a cold beer tonight whilst watching Midsummer Murders reruns.
b) I think I know what it is going to do, and I don't want it to do that.

[Bobster.]

So this morning I notice that the firmware version is changed. Now 2121E. 

The installer now has a help desk, and they told me they were going to take this issue up with Goodwe. So I am guessing that this is how the update happened. 

I now have to wait for the right combination of conditions. The City are not helping by restoring our power two hours early. 

I am now wondering if this condition hasn't been present all along, but was only exposed by the increased load shedding. 

The other user I mentioned is on 2121H

[Bobster.]

OK, so here is PV (blue), load (yellow) and SOC (green) for my system today

PV was very variable and downright poor early in the day. Power was out until 9:25 and then from 12:00. So two sections of this graph are of interest to us

1) 8:35 with the grid out, we see PV considerably greater than load. As this brief sunny period was happening, I was busy switching the system into economical mode. PV tails off slowly over the next 20 minutes, but it was gloomy and we don't see an immediate shift to just tracking the load.

2) 12:00 the grid goes down for the 2nd load shed of the day. But look what happens next. PV is still quite variable (or demand from PV is), but it carries on charging the battery.

What I don't have time to do is look out of my window to see how the sun is shining, but definitely the system, in general mode now, carried on charging after the grid went down. 

So two possibilities
1) 2121E has made a difference
and/or 2) The problem is in eco mode 

The latter is not clear to me yet. There's a voice in my head (shut up! You all have them  ) saying that the way I use eco mode to charge between 15:00 and 16:00, the system doesn't know for the rest of the day if it should be charging or not.

I need some more data, but I'm skittish. Twice in the last week the power has stayed off after load shedding. Once for 1 hour, on the other for 10. So if the weather is not so lekker, and even then, I like to get some charge into the batteries because you don't know when the power is going to come back.

[Bobster.]

So this morning we have no grid since 7:00 and lots of sunshine. 

I changed from general mode to economical mode at 8:35 or so. Then the grid came back at 8:40. Meh. This is not in the interests of scientific research.

But clearly it was charging from PV whilst the grid was down and the system was in general mode.

 

 

Edited February 13 by Bobster.
Finish the post

[Bobster.]

I hope @LLTHB has been following this.

After Tuesday's test I left the system in economic mode.

This morning we had load shedding starting at 6:00. The inverter reports utility loss at 6:01 and at time of writing power is not restored. It's a nice sunny morning.

Here's the graph

As usual, the PV starts shooting up around 8:15. But as early as 7:45, PV is double load and the battery is charging. Then at 8:15 we see the load increase a little, but PV shoots up a lot. PV is no longer tied to load when the grid is down.

This is under the following conditions

Grid: no grid
Inverter mode: economical
Safety code: 230V 50Hz default.
Firmware: 2121E

At 8:31 I changed the safety code back to the South African code.

At 8:33 SolarGo tells me that I have 2.08 kW coming from the roof, and 1.53 going to the battery. The battery is still charging.

At 8:35 the grid comes back.

According to current load shed schedules, the next useful test slot is 6:00 to 8:30 on Saturday.

Edited February 15 by Bobster.

[Bobster.]

On 2024/02/10 at 7:35 AM, Bobster. said:

So this morning I notice that the firmware version is changed. Now 2121E. 

The installer now has a help desk, and they told me they were going to take this issue up with Goodwe. So I am guessing that this is how the update happened. 

The helpdesk have confirmed that Goodwe pushed that update through. So despite not selling Goodwe inverters any more, they still have an open channel to them.

[Nelis Volschenk]

Consider switching to a different brand if this problem persists.

[LLTHB]

Had such poor service from the agents, useless actually.
Removed a perfectly good working inverter and replaced with a Sunsynk 8.
I believe there is a patch available or still coming out for the Goodwe, but I could not wait.
Really a lesson in how to piss off your clients and kill your brand.
If I was in this industry, I would monitor a site like this and send clients with issues a message discussing contact and solution problems.
Poor show.
We had no Eskom in the Karoo area I live for 12 days.
You think you are prepared for it, I can assure you, a 12 day blackout makes you rethink your entire strategy.
And no, I cant just go off grid - 9 geysers, 7 ovens, 14 air cons plus loads of pool pumps etc etc.

 

[Scorp007]

1 hour ago, LLTHB said:

Had such poor service from the agents, useless actually.
Removed a perfectly good working inverter and replaced with a Sunsynk 8.
I believe there is a patch available or still coming out for the Goodwe, but I could not wait.
Really a lesson in how to piss off your clients and kill your brand.
If I was in this industry, I would monitor a site like this and send clients with issues a message discussing contact and solution problems.
Poor show.
We had no Eskom in the Karoo area I live for 12 days.
You think you are prepared for it, I can assure you, a 12 day blackout makes you rethink your entire strategy.
And no, I cant just go off grid - 9 geysers, 7 ovens, 14 air cons plus loads of pool pumps etc etc.

 

Sad to hear about the service. It seems solar systems have few with REAL great support. Just my uninformed view. 

Your geyser problem is perhaps the easiest to solve. If they are in close proximity a well insulated storage tank and using PV to heat via heat pump. The other option perhaps better is gas geysers as you only use on demand. No point in keeping all hot if not needed every day. 

12 day must be a nightmare. Our longest break has been 36 hrs in 34yrs when a mini sub burnt out. 

 

[Bobster.]

On 2024/02/10 at 7:35 AM, Bobster. said:

So this morning I notice that the firmware version is changed. Now 2121E. 

The installer now has a help desk, and they told me they were going to take this issue up with Goodwe. So I am guessing that this is how the update happened. 

I now have to wait for the right combination of conditions. The City are not helping by restoring our power two hours early. 

I am now wondering if this condition hasn't been present all along, but was only exposed by the increased load shedding. 

The other user I mentioned is on 2121H

Well smack the cow and paint the barn red! This morning I am on 2525J.

Now I have to start all my checks all over again. Under current conditions with no load shedding from 5:00 to 16:00 there is no way to do any useful testing. I was just getting relaxed about 2121E. 

[LLTHB]

On 2024/02/18 at 12:23 PM, Scorp007 said:

Sad to hear about the service. It seems solar systems have few with REAL great support. Just my uninformed view. 

Your geyser problem is perhaps the easiest to solve. If they are in close proximity a well insulated storage tank and using PV to heat via heat pump. The other option perhaps better is gas geysers as you only use on demand. No point in keeping all hot if not needed every day. 

12 day must be a nightmare. Our longest break has been 36 hrs in 34yrs when a mini sub burnt out. 

 

Hi. At one guest house we preheat the electric geysers from the inverter between 10h00 and 15h00, then the power to the geysers goes off.
65Deg hot water then flows through a Paloma Hybrid gas geyser. First person showers, the gas does not even come on. When the 3rd person showers and the temp drops below 55 Deg, the gas geyser slowly comes to life and we use gas until the next day when the solar heats up the geysers again.
This system works so well for us.

[LLTHB]

On 2024/02/18 at 3:31 PM, Bobster. said:

Well smack the cow and paint the barn red! This morning I am on 2525J.

Now I have to start all my checks all over again. Under current conditions with no load shedding from 5:00 to 16:00 there is no way to do any useful testing. I was just getting relaxed about 2121E. 

Morning Bobster, please let me know if this has solved my problem - Goodwe not charging batteries unless Eskom power is present.
Fingers X this patch will work, otherwise I sit with 2 perfectly functioning inverters I cannot use.
Thanks. LLTHB

 

[Bobster.]

5 hours ago, LLTHB said:

Morning Bobster, please let me know if this has solved my problem - Goodwe not charging batteries unless Eskom power is present.
Fingers X this patch will work, otherwise I sit with 2 perfectly functioning inverters I cannot use.
Thanks. LLTHB

 

Hi @LLTHB

Let's try to recap:

1) I was experiencing the same problem as you. Battery would not charge when there was solar but no grid.
2) this problem could be overcome at a pinch by switching the inverter into off-grid mode
3) I was on firmware 2121D
4) Goodwe upgraded me to 2121E.
5) Test results above after this upgrade showed that when the grid was down, solar was abundant, the battery charged - in general and in economical mode
6) So it does seem to be down to firmware version
7) Since then Goodwe have upgraded me a second time. Now I have to repeat my tests again. With sunny days and no day time load shedding at present I can't do meaningful tests.

Note that I didn't contact Goodwe. I contacted my installer who themselves contacted Goodwe. They called me to confirm the serial number because that is what Goodwe use to find and connect to your inverter.

Edited February 21 by Bobster.

[Bobster.]

19 hours ago, Bobster. said:

7) Since then Goodwe have upgraded me a second time. Now I have to repeat my tests again. With sunny days and no day time load shedding at present I can't do meaningful tests.

So serendipitously we have load shedding this morning AND we have some PV but it's not great.

But look at this

Loadshedding starts at 10:00 and you can see the load drop as the pool pump is no longer on. But the battery is charging. Not at a great rate (it's overcast) but it is charging. This is with the inverter in economical mode.

So @LLTHB I think you need to check the firmware version on your Goodwes checked and then upgraded. I say to check first because if something goes badly wrong you will have the previous version number recorded and can ask a roll back to a known version (this is my attitude towards all firmware upgrades on any type of device)

To get upgraded you will have to have the inverters powered up and connected to the internet. Goodwe identify them using the serial number. They can push the update onto the inverter, and no restart by you is necessary.

Your challenge now is to get through to somebody at Goodwe SA who can make this happen, or get whoever sold you those inverters to intervene on your behalf.

It can be done! What I record above shows that. If Goodwe SA can't help you, then try going to Goodwe EU or Goodwe themselves.

[Bobster.]

Just to drive the point home, with inverterter still in economical mode and loadshedding ongoing

[Bobster.]

OK... my last word on this. For about 25 minutes today I had a situation where there was load shedding, not a lot of PV, battery not completely charged, and (because of the shedding) light loads. For that time span the Goodwe drew about 1.2kW from the panels whilst the load was about 450 - 520 W. So, again, the battery charged and the draw from PV only dropped off when SOC hit 100%.

I'm still on 2525J.

So now I'm as sure as I can be that this problem can be addressed by firmware.