Growatt Data Missing

[Tsa]

Has anyone had the scenario before where there is no data available via the App or Web? Have unplugged and reseated the wifi dongle, but no luck. 

Doesn't give any data and gives these values

Connect Status：Lost Update Time：2020-03-25 12:52:01  (Last time I had data)

[charlvnk]

Hi,

Yes, I'm experiencing the same kind of problem at the moment. I have a SPF 5000TL HVM-P inverter with Dyness Powerbox F-10.0 battery. These were purchased just a few days before the lockdown, so they could probably be from the latest batch.

When the inverter is set up correctly to communicate with the Lithium battery, the Wi-Fi data logger malfunctions and no data gets sent to the server.

When I change the battery type setting on the inverter to USER or AGM or pretty much any setting that does not communicate with a BMS, the Wi-Fi logger works perfectly and data is sent to the server, no problem.

Change back to Lithium (Li01 in my case), and the Wi-Fi logger stops sending data immediately.

When logged onto the server, I can see that the Wi-Fi logger's Connect Status is reported as "Connection". The inverter (under "Storage") is reported as "LOST".

I have tried everything. Even sent videos to Growatt technical support (engineer by the name of "Mahone"), and he sent me the latest firmware + BMS comms software. I updated the unit with this firmware, but still no luck.

I'm now waiting for Mahone's response to my latest mail giving him this news...

Perhaps if you're willing to try and see if your unit sends data when you change the battery seting to SLA or something, then at least we could get to a point where it can be proved that this problem isn't just an isolated case ?

[Tsa]

@charlvnk Thanks for the update. My side also SPF 5000TL HVM-P and Dyness B4850 (x3), basically the same batteries as you have in the Powerbox. Powerbox just has 4x B4850s.

In my case this has happened a few times in the last +- 2 weeks and also since moving to Li(program 05) and L01(program 36). It's not permanent, but happens for anything from 1 hour to 9hours once. Doing any kind of reset on the datalogger, unplugging datalogger, deleting the datalogger has no effect. Restart router, CPE, whatever, no effect.  Just do nothing and then hour\s later the data returns and functions as normal again. Like now I haven't had the issue for 2 days again. All good.  

38 minutes ago, charlvnk said:

When logged onto the server, I can see that the Wi-Fi logger's Connect Status is reported as "Connection". The inverter (under "Storage") is reported as "LOST".

Exactly the same here. 

Have also been chatting to Mahone about this and he also has many screenshots and photos to confirm the issue from my side. His last email indicated that he has sent this to the R&D team and then also suggested the usual restart datalogger\reset datalogger\delete and reconfig datalogger and and and which up to now hasn't worked anyways. 

I have not tried setting it to USE again once this occurs, but will do so and confirm. Also want to do a traceroute to the Growatt server when this happens. tracert 47.91.67.66

BUT the fact that you say it works when on USE mode it clearly isn't a comms issue then. 

What is the BMS comms software you are referring to?

Please also keep me posted when you do hear anything back from Mahone. I now 1st have to shutdown my inverter and make it lose all charge for a while before I will reply back to him. Just waiting for the time to switch the whole house off and over to Eskom first

 

[Tsa]

@charlvnkToday again I lost data on my app/web. When I saw it, data stopped reporting about 2hours prior. I changed program 05 to USE (only thing I changed) and 5mins later I had data again. I then changed it back to Li and continued having data as normal. 

The trend I am seeing happening is that when my batteries get to 99/100% charge via PV and then automatically take the load and start discharging, it is sometime during this phase that the data stream is lost. Might be too early to say, but seems to happening more and more during that phase, but will have to monitor more to confirm. 

I will update Mahone with these findings and hopefully R&D can sort this out asap. On his suggestion I have now switched off the inverter, no power sources are connected and it is totally off and will leave it now for the whole night so that it can properly discharge.

My dreaded power tripping on setting Li(05) has also re-appeared and hoping they can also resolve this in a firmware update. I just need to catch it during the day so that I can confirm the warning codes and if it is again warning 04 and 20. This morning it tripped at 00:35! Power to house off for a few seconds, inverter beeping,  and then everything came back on. 

Just a another question, when you set program 15 to BOF, does your button presses still make the beep sound? 

[charlvnk]

Hi Tsa,

On Friday evening @ 21:00, it started sending data again.  Nothing changed around that time (Line mode, battery 100% charged), notmal load etc...

BUT - it only sent 3 measurements (15 minutes' data) and then stopped again.  Until now, nothing at all.

Switch from Li to USE, and all is well.  This is so damn frustrating - I don't want to run it in USE mode just so I can see telemetry data, because I'm worried this might perhaps have a negative impact on the Lithium battery.

Regarding your tripping problem - do you have the latest firmware loaded on your Dyness' BMS ?  I found out about some bug in that firmware, but in order to fix it they have to come out to your premises to update the firmware as well as some PCB in there (I read this in another thread).  Apparently on the Powerbox (and I assume the others as well), they still have to go out and do this after the lockdown.  I checked with the local importer (Andy), and he asked me to take a pic of the inside of the box.  Then he let me know mine is already updated in terms of hardware and firmware (he can see some negative cable installed, which the older ones don't have yet).

Regarding the "BMS software" - I assume it's some part of the firmware that handles communication with the BMS.  It was also just a .bin file he sent me (like the firmware update), loaded using the same utility app.

Yes, my BOF function doesn't work anymore after the update.  It beeps when you press the buttons, regardless that setting.

I haven't had any reply from Mahone again, after the last time I sent him more information regarding the Wi-Fi problem on Thursday.

[willietes]

Hi All 

 

I am not the only one it seems. Mine logs perfectly yesterday but today it started again. It stops at about 9am and then starts at 4:30pm

I thought is was PV so I shutdown the inverter restarted it with PV switched of and still... no dice. I also went as far as to install PVkeeper and check if it is a USB issue 

and it works perfect. Plug the wifi unit in, and no dice. Tonight its running again. 

 

I do have a few suspicions

Network service not running on their servers Day for us night for them

Internet service provider issues Problems related to international connectivity. 

 

Next to me running a network analyser and capturing network packets (of which I have none of the hardware available atm)

[willietes]

4 hours ago, charlvnk said:

Hi Tsa,

On Friday evening @ 21:00, it started sending data again.  Nothing changed around that time (Line mode, battery 100% charged), notmal load etc...

BUT - it only sent 3 measurements (15 minutes' data) and then stopped again.  Until now, nothing at all.

Switch from Li to USE, and all is well.  This is so damn frustrating - I don't want to run it in USE mode just so I can see telemetry data, because I'm worried this might perhaps have a negative impact on the Lithium battery.

Regarding your tripping problem - do you have the latest firmware loaded on your Dyness' BMS ?  I found out about some bug in that firmware, but in order to fix it they have to come out to your premises to update the firmware as well as some PCB in there (I read this in another thread).  Apparently on the Powerbox (and I assume the others as well), they still have to go out and do this after the lockdown.  I checked with the local importer (Andy), and he asked me to take a pic of the inside of the box.  Then he let me know mine is already updated in terms of hardware and firmware (he can see some negative cable installed, which the older ones don't have yet).

Regarding the "BMS software" - I assume it's some part of the firmware that handles communication with the BMS.  It was also just a .bin file he sent me (like the firmware update), loaded using the same utility app.

Yes, my BOF function doesn't work anymore after the update.  It beeps when you press the buttons, regardless that setting.

I haven't had any reply from Mahone again, after the last time I sent him more information regarding the Wi-Fi problem on Thursday.

We have both already done that upgrade. Andy was VERY helpful and did the upgrade free of charge. 

[charlvnk]

Hi Willietes,

Yes indeed, I've also found Andy to be VERY helpful !

As for the Wi-Fi problem - mine just started sending data again around 17:30.  This time only ONE set of readings (5 minutes' worth...).

I don't think there's any problem connecting to their server.  Remember, mine works 100% without any problems or missing data for days on end, the moment I change the battery setting from Li to USE.  Don't know if perhaps this could be because I don't have my PV's connected yet...

Also, I can see that there is communication between the Wi-Fi logger and the server, by looking at the "Connect Status" (says "Connection" green), as well as "Update Time" (which is never more than 10 minutes behind).  So, the problem must be between the Wi-Fi logger and the inverter, or between the server (via Wi-Fi logger) and the inverter.

In my mind it can only be firmware.  If it was hardware related (that communication PCB) causing some interference on the USB when there's BMS comms, you would have experienced problems while using the PVkeeper app over USB as well.

Still no reply from Mahone...

[Tsa]

19 hours ago, charlvnk said:

Switch from Li to USE, and all is well.  This is so damn frustrating - I don't want to run it in USE mode just so I can see telemetry data, because I'm worried this might perhaps have a negative impact on the Lithium battery.

@charlvnk Agreed 100%! 

 

19 hours ago, charlvnk said:

Regarding your tripping problem - do you have the latest firmware loaded on your Dyness' BMS ? 

Yip Andy came to do my battery FW upgrade and PC board replacement just before lockdown started. So now I am wondering if anyone else still has the tripping issue post these fixes. Mine was running perfectly after the FW and board fix, probably close to 2 weeks, and then all of a sudden it reappeared. Twice in 24hours. 

19 hours ago, charlvnk said:

Regarding the "BMS software" - I assume it's some part of the firmware that handles communication with the BMS.  It was also just a .bin file he sent me (like the firmware update), loaded using the same utility app.

Ah yes, OK I also updated this. This is the FW specific to the BMS. Sorry I though you were referring to some software to be able to interrogate the Dyness batteries.

19 hours ago, charlvnk said:

Yes, my BOF function doesn't work anymore after the update.  It beeps when you press the buttons, regardless that setting.

Yip painful, I think now after weeks of raising this with Mahone, and now that you guys have also started noting this, he eventually "understands" what I am referring to. .

R&D team need to sort out these bugs asap and release new FW update. Well at least we can update it ourselves vs having to have on site visit, which will not happen now with lockdown.

15 hours ago, willietes said:

I do have a few suspicions

Network service not running on their servers Day for us night for them

Internet service provider issues Problems related to international connectivity. 

Dont think it is related to comms as I can see my wifi data logger connected to 47.91.67.66 on my router and using the same ports the app/web indicate the datalogger using at that time. A traceroute to 47.91.67.66 also end up at server.growatt.com so all fine there when testing this in the time that I have no data. 

 

[willietes]

Ok Sooo….

 

I left mine on and at about 11am I got an error 500 pop up and then the Logger was offline.

Bottom line - even though their server is contactable the service taking the logs are not handling it. 

 

So I recon its a server service issue

[SARED]

Ok... you all seem much more informed than me on this topic. 

I have 2x Growatt SPF 5000TL HVM-P with 4x Dyness B4850's  

what i have noticed: 

1. I had to fire up the 2.4kHz bunny ear router just to get connected.. that unit tends to drop the signal constantly (i can hear my teen's frustration ...as its directly proportionate to his gaming lag time) 

2. the flippen buggers used all the wifi data the other day... 25Gb in 5days! No wonder i had no info  

3. settings are USE on both and sometimes at random times the data just stop. 
 

i can not believe that a these new Growatt units do not support 5khz wi-fi. 
 

is that not the problem: the networks are using 5khz and we're still trying to crank up 2.4khz + old tech router, dropping connectivity = huge lag time: system loose connectivity: no data update: lost data. 
 

Do anyone know how much data these data loggers use per 24hr period? 

[Luminous]

I doubt their data would be much. I think rather it is your wifi router. On the cheap one's it does not like multiple divices using it. CPU pin 100% it overheats and then start dropping packets.

Hardwire where you can and switch to lite commercial use, something like Mikrotik or Unify.

You can also check if you don't have multiple wifi transmitters on the same channel. I have a mikrotik and I do around 15.-2.5TB at home per month and never have to restart routers or switches.

 

 

 

 

 

[Tsa]

2 hours ago, willietes said:

I left mine on and at about 11am I got an error 500 pop up and then the Logger was offline.

Bottom line - even though their server is contactable the service taking the logs are not handling it. 

 

So I recon its a server service issue

Thats good to know. I actually also saw 500 error on my web login earlier, but had it open so not sure when it happened. 

I hope you are also sending these updates to Mahone. The quicker they get all the feedback from as many sources as possible, the better chance we have of getting it fixed.

[Tsa]

So it seems around 17:35 yesterday my data started returning, after a long absence, and has been running fine since.

Just wondering, if this was a Growatt server issue, should we all not be seeing the same issue at the same time? That's IF we all connect to the same server on their end. 

[SARED]

On 2020/04/06 at 1:51 PM, Luminous said:

I doubt their data would be much. I think rather it is your wifi router. On the cheap one's it does not like multiple divices using it. CPU pin 100% it overheats and then start dropping packets.

Hardwire where you can and switch to lite commercial use, something like Mikrotik or Unify.

You can also check if you don't have multiple wifi transmitters on the same channel. I have a mikrotik and I do around 15.-2.5TB at home per month and never have to restart routers or switches.

 

 

 

 

 

Thanks for the advice L, i will change our router. 

[charlvnk]

Yesterday morning I went to the inverter and simply changed the battery setting from Li to USE again.

Data started coming in on their server immediately, and until now it has been working perfectly.  Blue LED flashing at a rate of about 1 second on, one second off, constantly.

The moment I change back to Li battery setting, it will stop sending data again.  Blue LED flashing 3 - 5 times, then nothing for at least 30 seconds (sometimes up to a minite or so), then another 3 - 5 blue flashes, and so on...

I think it's safe to say that in my case it can't have anything to do with my router or internet connection whatsoever ?

[Tsa]

I keep mine on Li, inverter and BMS comms is still 100% so it does what it needs to do. 

Mine still going string since 17:35 yesterday. So yes I agree with you that it would be a per inverter (ie firmware) issue as if it was a server issue we would all be having the same data drop at the same time.

What will be interesting to see if mine again, as before, works fine until the batteries reach 99/100% and then start to auto discharge. Usually during this  discharge cycle to 95% my data drop happens. 

[Tsa]

Well it was fun while it lasted... nearly 48hours later and my data is gone again at 15:35.

Graph for 09/04 is totally screwed, but good thing I built that extra out building to house all my batteries......LOL 

SOC graph is correct though. 

 

 

[willietes]

@tsa

Sadly mine has also started the tripping thing come back very quickly.

 

Something I dont like about how the inverter works

Its supplements from storage to meet power demand when solar not strong enough - I personally think that is quite daft as it cost more. Anyone cares to chime in ?

per kwh to run from battery than from utility. I would really like if it only used storage when there is no power solar.

otherwise use solar only when it is capable to meet the full demand. 

 

Wish this inverter was programmable like an arduino controller 

[Tsa]

On 2020/04/10 at 2:14 PM, willietes said:

Sadly mine has also started the tripping thing come back very quickly.

@willietes Oh no, that's not good news. I am however happy that I'm not the only one, so points to an issue somewhere and still wondering if this is not a FW issue. The whole thing of us losing connection to the server, some settings not taking effect (beep program 15) and this battery tripping issue.

I have changed my host\master to a different battery in my stack and monitoring. So far so good. 

On 2020/04/10 at 2:14 PM, willietes said:

Something I dont like about how the inverter works

So you basically need SUB setting for 01 like the Mecer's have it seems. Only when SOL and UTI isn't available, only then run BAT. Might be worth raising this with Mahone as a feature request. Wondering if setting your Back to battery (13) very high would assist with minimising batter usage? Not sure. 

 

 

 

[willietes]

good point. 

27 minutes ago, Tsa said:

@willietes Oh no, that's not good news. I am however happy that I'm not the only one, so points to an issue somewhere and still wondering if this is not a FW issue. The whole thing of us losing connection to the server, some settings not taking effect (beep program 15) and this battery tripping issue.

I have changed my host\master to a different battery in my stack and monitoring. So far so good. 

So you basically need SUB setting for 01 like the Mecer's have it seems. Only when SOL and UTI isn't available, only then run BAT. Might be worth raising this with Mahone as a feature request. Wondering if setting your Back to battery (13) very high would assist with minimising batter usage? Not sure. 

 

 

 

Good point on the sub setting will try it out. Will chat to Mahone as well

[Tsa]

14 hours ago, willietes said:

Good point on the sub setting will try it out.

Unfortunately there is no SUB setting on Growatt, but I know the Mecer\Axpert? ones have it from what I have seen on the forum. 

[willietes]

Hi All

So I wrote a whole letter to Mahone from growatt

 

Lets see how it goes.

 

--------------------------------------------------------------------------------------------------------------------------

Out of pure frustration I have gone and done some troubleshooting.

It seems that the connection to the server is lost when PV provides power to both the batteries AND supplying power to the home grid. Once it stops giving power to the household via PV the data-logging continues.

 

This makes me believe the following

a) Its not an internet connection. (Tried different service providers)

b) Its a possible firmware issue.

 

Also I have studied how the Invertor works and there is one function it needs

It needs to have a function where it will only switch to PV once PV watts are high enough to meet household demand. At the moment as soon as the inverter can it will use power from storage and PV to provide power and that is very expensive power as Battery storage is not cheap.

 

The Mecer inverter has a SUB function which allows you to provide power in the following order.

 

1) PV

2) Utility

3) Battery ( if no PV or utility is available.)

 

This would be a great function in your firmware.

 

Below are some diagnostics

Traceroute to server

C:\WINDOWS\system32>tracert server.growatt.com

 

Tracing route to server.growatt.com [47.91.67.66]
over a maximum of 30 hops:

 

1 1 ms <1 ms 1 ms homerouter.cpe [192.168.8.1]
2 * * * Request timed out.
3 20 ms 18 ms 17 ms 105-187-234-162.telkomsa.net [105.187.234.162]
4 23 ms 21 ms 29 ms 105-187-253-74.telkomsa.net [105.187.253.74]
5 44 ms 23 ms 22 ms ndn-ip-hsll-1.telkom-ipnet.co.za [165.165.235.173]
6 207 ms 199 ms 200 ms 10.189.30.10
7 220 ms 218 ms 209 ms alicloud.interxionfra11.nl-ix.net [193.239.118.179]
8 * * * Request timed out.
9 * * * Request timed out.
10 * * * Request timed out.
11 218 ms 217 ms 219 ms 47.91.67.66

 

I could telnet the port.

 

I have also included a video of what the device is doing. Also a screenshot of the Wifi device config.

 

It starts recording data if you change from BMS to USE. So the problem appears to be related to the BMS system on the unit. As soon as the invertor supplies power to the house with PV on, it does not log anymore data.

 

The latest firmware does something funny. It does not hold the bulk charge voltage setting and defaults to 54v manufacturer requires 53.2v also option 2 defaults back to 10A even though I set is to 20A

 

This latest release seems to be really buggy.

 

Please assist.

[charlvnk]

Hi,

Yes indeed, the bulk charge voltage setting going back to 54V all by itself and the "beep" that can't be disabled, are two clear problems that I noticed right after updating the firmware.  I haven't even mentioned these other problems to Mahone yet, because I want him to "focus" on my main problem which is NO data logging at all while battery type is set to Li.

Mahone contacted me on Saturday via WhatsApp, and requested me to do a few things and then send him video clips to show what happens.

Eventually he realised that something is definitely wrong and so he gave up, then told me he was going to refer the matter to the R&D team.

I then sent him an e-mail (copied the SA importer as well as two other persons from Growatt), asking how long this process was going to take.

Waiting for a reply...

[willietes]

Awesome news.

At least they know there is a problem and will respond with something better than "take a picture of your setup and a video"