Growatt SPF 3000TL LVM-24P - Bypass Error Issue

[Off Gridin' It]

The GW is working just as I described it will under low light or low yield conditions. There won't be a fix, the problem really isn't the GW, the problem is PV.

Flat mounted panels seem more prevalent to this issue, any shading with the flat mount and the PV can not supply the load the SCC puts on PV.

Are you trying to say there was never a problem? Both Signature Solar and Growatt agree that there is an issue here. The PV input voltage range is 30 volts to 115 volts. The incoming voltage from the array was between 36 volts to 49 volts. The voltage from the array was within the specification of the Growatt SPF 3000. Both signature solar and Growatt agree that the 3P configuration should have worked without issue. They believe it may be a bug in the firmware and are investigating it further. For now as long as you a series configuration for Series/parallel configuration to get the voltage up then it will work.

So you say there won't be a fix but Growatt says they are now working on a fix. So.. there is that fact.

 

[Off Gridin' It]

Good job of figuring out what was causing the situation. Since the voltage was not lower than spec it should not cycle. That is something for the Growatt programmers to fix. Probably won't since few people will run into it. Those that do and complain they will likely just be told to increase panel array voltage.

I was able to bring this to the attention of both Signature Solar and Growatt. And Since Signature Solar works directly with Growatt hopefully they will be able to fix this sooner than later.
But it was a tough solve but I do feel accomplished for figuring it out. lol

 

[Zwy]

Are you trying to say there was never a problem? Both Signature Solar and Growatt agree that there is an issue here. The PV input voltage range is 30 volts to 115 volts. The incoming voltage from the array was between 36 volts to 49 volts. The voltage from the array was within the specification of the Growatt SPF 3000. Both signature solar and Growatt agree that the 3P configuration should have worked without issue. They believe it may be a bug in the firmware and are investigating it further. For now as long as you a series configuration for Series/parallel configuration to get the voltage up then it will work.

So you say there won't be a fix but Growatt says they are now working on a fix. So.. there is that fact.

I doubt there will be a fix, the 3P configuration most likely had voltage drop under load and the reason why the SCC would cut out. I'm quite certain the panels on the camper are flat mounted and again, the PV voltage will not stay up under the load of the SCC especially with flat mounted panels, low sun horizon at this time of year. 3S might keep the voltage up high enough.

This is not a new problem. It's just a case of PV voltage/yield is not high enough when the SCC loads it down.

 

[Off Gridin' It]

I doubt there will be a fix, the 3P configuration most likely had voltage drop under load and the reason why the SCC would cut out. I'm quite certain the panels on the camper are flat mounted and again, the PV voltage will not stay up under the load of the SCC especially with flat mounted panels, low sun horizon at this time of year. 3S might keep the voltage up high enough.

This is not a new problem. It's just a case of PV voltage/yield is not high enough when the SCC loads it down.

I had a monitor on it continuously for two weeks. The lowest recorded voltage was 36 volts on a few brief occasions but typically it remained at an average of 45 volts. The PV input limit is between 30 volts to 115 volts. I also tested the voltage at the PV inputs on a very dark rainy day with the voltmeter and even with that it was still in the upper 30s to low 40s. So in all cases per the Growatt specs it was within specs.

Remember I was able to reproduce this on the bench. That bench test in full sun saw voltage at 47 volts. That was 17 volts higher then the 30 volt min and the problem persisted.

Signature Solar and Growatt agreed with this assessment and testing. I also redid many of these tests at both Signature Solar and Growatt request. As I said, they believe it to be a firmware issue and they are working on a fix.

So, yeah if the voltage dropped lower then 30 then I would expect this to happen but not if it is within specifications.

 

[feman43]

I'd be interested in knowing what the panels are. It certainly has the earmark of voltage drop under load. Kind of. The strange part is the period of the cycle which has the earmark of a timer in the software periodically checking the input voltage and disconnecting. Disconnecting the SCC then allows the Voc to climb and restart the SCC. That the fix is to raise the Voc, 2p to 2s, implies this is an input voltage issue. I just bought one of these units and this behavior makes me question my decision. It's funny that this would be the only example of this occurring though. That two different boxes have the same result does not imply bad boxes but rather something on the system side is at fault. A third and forth box from perfectly working systems would likely have the same result and then we'd want to know what makes the working system different from this system, hence my interest in the type of panels. I'd hook my recording oscilloscope up to it watch the PV voltage during the event. Depending on the recording device, the recorded low voltage may or may not be the actual lowest voltage and the voltage may be falling far enough to cause the SCC to drop off. This would be evident in an Oscope trace and you'd likely see a saw tooth shaped wave form. Then again, the fix seems to be to simply have enough input voltage ie running the 2s configuration which seems to be an acceptable solution.

I honestly don't know squat about these solar array things and I am here to learn about them. So I went and looked at some 385W solar panel specs to see what I could determine and found what I would conclude as consistent with this "problem". While these panels have a Vocc of 40Vdc the max power Vmpp often is only 31.2 Vdc. This is woefully close to the cutoff voltage of the GW MPPT. It is very likely that the MPPT voltage loading sweep is in fact drawing the PV voltage below the min 30Vdc which is causing the shut down to occur. It may or may not something a software change can fix however it is totally clear that increasing the Vmpp of the PV array will fix it. My conclusion is that the GW is not the issue.

 

[Mattb4]

... So I went and looked at some 385W solar panel specs to see what I could determine and found what I would conclude as consistent with this "problem". While these panels have a Vocc of 40Vdc the max power Vmpp often is only 31.2 Vdc. This is woefully close to the cutoff voltage of the GW MPPT. It is very likely that the MPPT voltage loading sweep is in fact drawing the PV voltage below the min 30Vdc which is causing the shut down to occur. It may or may not something a software change can fix however it is totally clear that increasing the Vmpp of the PV array will fix it. My conclusion is that the GW is not the issue.

Problem with your conclusion is it does not account for the reported PV voltages that Off Gridin'it had. He said that measured PV voltage never dropped below 30vDC.

 

[feman43]

Except how he measured that voltage could be part of the confusion. The small little voltage display devices have large filters in them to limit the display fluctuations which would cause consternation by the person watching them. They basically time average the voltage on a long period.
What this translates to is the volt measurement display device isn't optimum to catch this kind of problem which is likely triggered in a few milliseconds, hence why I suggest using an oscilloscope. It also depends upon where the voltage is measured. The GW spec says the low cutoff is approximate 30 Vdc and the panel specs says 31.2Vdc @ its optimum MPP. That's a small 1.2 Vdc margin. A small voltage loss across wires and slightly higher cutoff at the SCC that 1.2 Vdc margin goes away in a hurry. The MPPT does a power sweep to find the optimum power which means it is sweeping the input current to find the best power. This means the load on the PV is varied and it is highly likely that the load pulls the voltage down below the threshold at some point causing a shut down disconnect. All of the evidence supports this, especially the fact that increasing the PV array voltage, margin above cutoff was increased significantly, fixed the problem. By changing the PV array to 2S puts the operating voltage in the middle of the range. Seems the problem is solved.

 

[Mattb4]

Except how he measured that voltage could be part of the confusion. The small little voltage display devices have large filters in them to limit the display fluctuations which would cause consternation by the person watching them. They basically time average the voltage on a long period.
What this translates to is the volt measurement display device isn't optimum to catch this kind of problem which is likely triggered in a few milliseconds, hence why I suggest using an oscilloscope. It also depends upon where the voltage is measured. The GW spec says the low cutoff is approximate 30 Vdc and the panel specs says 31.2Vdc @ its optimum MPP. That's a small 1.2 Vdc margin. A small voltage loss across wires and slightly higher cutoff at the SCC that 1.2 Vdc margin goes away in a hurry. The MPPT does a power sweep to find the optimum power which means it is sweeping the input current to find the best power. This means the load on the PV is varied and it is highly likely that the load pulls the voltage down below the threshold at some point causing a shut down disconnect. All of the evidence supports this, especially the fact that increasing the PV array voltage, margin above cutoff was increased significantly, fixed the problem. By changing the PV array to 2S puts the operating voltage in the middle of the range. Seems the problem is solved.

Uh, not really. You assume there is a load. All the MPPT sweeps would do nothing to drop voltage if there is no load from batteries being charged. That is even allowing for your assumption without testing in the real world. While I agree that higher PV panel voltage stops the odd behavior of cyclic AIO power on self testing this cyclic testing is a poor programing choice.

Your theory sounds good but your supporting data is lacking.

 

[Off Gridin' It]

I'd be interested in knowing what the panels are. It certainly has the earmark of voltage drop under load. Kind of. The strange part is the period of the cycle which has the earmark of a timer in the software periodically checking the input voltage and disconnecting. Disconnecting the SCC then allows the Voc to climb and restart the SCC. That the fix is to raise the Voc, 2p to 2s, implies this is an input voltage issue. I just bought one of these units and this behavior makes me question my decision. It's funny that this would be the only example of this occurring though. That two different boxes have the same result does not imply bad boxes but rather something on the system side is at fault. A third and forth box from perfectly working systems would likely have the same result and then we'd want to know what makes the working system different from this system, hence my interest in the type of panels. I'd hook my recording oscilloscope up to it watch the PV voltage during the event. Depending on the recording device, the recorded low voltage may or may not be the actual lowest voltage and the voltage may be falling far enough to cause the SCC to drop off. This would be evident in an Oscope trace and you'd likely see a saw tooth shaped wave form. Then again, the fix seems to be to simply have enough input voltage ie running the 2s configuration which seems to be an acceptable solution.

I honestly don't know squat about these solar array things and I am here to learn about them. So I went and looked at some 385W solar panel specs to see what I could determine and found what I would conclude as consistent with this "problem". While these panels have a Vocc of 40Vdc the max power Vmpp often is only 31.2 Vdc. This is woefully close to the cutoff voltage of the GW MPPT. It is very likely that the MPPT voltage loading sweep is in fact drawing the PV voltage below the min 30Vdc which is causing the shut down to occur. It may or may not something a software change can fix however it is totally clear that increasing the Vmpp of the PV array will fix it. My conclusion is that the GW is not the issue.

" It's funny that this would be the only example of this occurring though"

This isn't the only example of this happening. I run the largest Growatt group on Facebook. On that group there are many examples of this happening and no one found a fix until I figured this out.

As I stated in my update to the solution Both Signature Solar and Growatt has unanimously agreed that this is a problem and may require a firmware update.

There was no notable voltage drop. The lowest recored voltage during testing was just over 36 volts which was within the operating voltage of the MPPT and there where no loads connected to either inverter at the time.

So until they correct the issue in the firmware or update the specification you have to keep that voltage higher to avoid the issue. Although I didn't have a way to find out where the voltage threshold is it did work perfectly with 96 volts.

I may do some additional testing to test different voltage inputs into the PV side to see where the cutoff is to see what the true minimum MPPT voltage is currently in the future. But for now we know that isn't 30 volts.

 

[Zwy]

Problem with your conclusion is it does not account for the reported PV voltages that Off Gridin'it had. He said that measured PV voltage never dropped below 30vDC.

I watched his video, the panels were in shade and flat. Look at 4th video.

2.4A is charging current, about 60W from a 395W panel.

Isn't hard to figure out what is occurring. The MPPT is searching for max power point with the low yield and that is when the shutdown occurs.

 

[Off Gridin' It]

I watched his video, the panels were in shade and flat. Look at 4th video.

2.4A is charging current, about 60W from a 395W panel.

Isn't hard to figure out what is occurring. The MPPT is searching for max power point with the low yield and that is when the shutdown occurs.

This was already solved. Between me, signature solar and growatt they determined it to be a software bug. As I reported the voltage never dropped below spec. Even when the voltage and current were strong in full sun the problem occurred. The fix was to change the panel configuration until the bug could be corrected. The customer ended up just having me install additional panels.

 

[Zwy]

This was already solved. Between me, signature solar and growatt they determined it to be a software bug. As I reported the voltage never dropped below spec. Even when the voltage and current were strong in full sun the problem occurred. The fix was to change the panel configuration until the bug could be corrected. The customer ended up just having me install additional panels.

I already had told you at the beginning of the thread why it was doing it. https://diysolarforum.com/threads/growatt-spf-3000tl-lvm-24p-bypass-error-issue.51247/post-651805

I never changed my panel configuration because I ran 2S3P on my truck camper due to possible shading issues. At some point it will be 4S2P when the roof air comes off.

It isn't a software bug, it is the MPPT searching for max power point. The PV voltage dropping low due to low yield combined with the power to run the panel. 60W from a 395W panel is 15% yield and shows the low light condition. At 60W, there isn't enough yield to charge the battery and run the display panel, that is why you see some charging occur even when the panel is off. It is when the MPPT searches, that you see the panel shutdown.

 

[Off Gridin' It]

I already had told you at the beginning of the thread why it was doing it. https://diysolarforum.com/threads/growatt-spf-3000tl-lvm-24p-bypass-error-issue.51247/post-651805

I never changed my panel configuration because I ran 2S3P on my truck camper due to possible shading issues. At some point it will be 4S2P when the roof air comes off.

It isn't a software bug, it is the MPPT searching for max power point. The PV voltage dropping low due to low yield combined with the power to run the panel. 60W from a 395W panel is 15% yield and shows the low light condition. At 60W, there isn't enough yield to charge the battery and run the display panel, that is why you see some charging occur even when the panel is off. It is when the MPPT searches, that you see the panel shutdown.

You're incorrect. You're being stubborn and not reading the whole thread. And GROWATT (The actual company) Said it IS a software bug. And it was solved. Why do you not want to listen?