diy solar

diy solar

Growatt - Odd behavior

Thanks for the update and glad it is working out. What voltage setting did you end up using for 19 and 20?
 
The system would shut off the panels until the batteries were drained to their switch to utility level . This could take hours. I had tried heavier loads - I mean 25% was to me a reasonable load that would show me something.. and it had no effect. Going to USE setting for battery type hit the mark perfectly and the unit is behaving as hoped.

I agree that the load should be more than reasonable. I still have the older firmware with only 1 USR option. I have seen others indicate that there is some undesirable behavior with the newer firmware, so I haven't updated mine yet. I'm kind of curious as to whether the USR1 setting on the update is a new one, or whether the USR2 is.

But in any event, I think I jinxed myself with that long-winded post. After not seeing the solar portion shut down for a long time, it started doing that behavior again late in the day yesterday, after applying a 1300-1700 watt load for about a half hour. A man's gotta eat :)

I also tried shutting the unit down entirely and re-starting to see if that made any difference. It really didn't. The solar panels kicked in, but within a few seconds, it shut them off again and continued with the "interesting" behavior of pulling only from the battery, even though there was a reasonable amount of daylight left. Both of the upper side vents of the system were quite warm, with the higher temps coming from the left side.

At this point, I'm starting to lean toward overall system temperature as the reason. Given that the conversion from solar generates heat, and it's really the only portion of the system it can safely shut down without interrupting the load, it seems to make sense. My observation of the unit kicking the solar in and out at varying intervals ( 5-60 seconds apart) would seem to back this up, as it smacks of cycling based on the readings of a thermocouple. There may be other reasons for it to do it as well, but you'd need to be inside the mind of the system programmers to really know. And now that I think about it, the first time I saw this behavior was shortly after turning it on for the first time and intentionally "beating on it" to see just what it was capable of.

But at least in my case, it didn't seem to have any real negative effects on battery levels going into the evening, and this morning it happily topped off my batteries and the connected load with everything available, without returning to the solar cycling behavior.

I'm glad to hear that your system is working the way you want it to now. But if you see it again, check the temperature of the unit. I'm really curious now if that is the primary cause.
 
One item I wanted to add to the discussion. I reached out directly to growatt corporate using their contact page and they redirected me to US Support. I am including the thread from US support below. I have switched to use2 for a test today for them. but I am not optimistic. Again the system works with the settings use and 19 and 20 set to reasonable values. But they seemed adamant that use2 is "For Lithium" . additionally they seem to be conflating "Stopping charge of the battery" with the complete shut down of solar utilization for the powering of the load...

Hello George,

Thank you for your input.
The team would like to clarify that the difference between USE and US2 is that USE is only for lead acid battery and US2 is for lithium battery.
The team knows that the solar will cutoff when the battery is full in order to protect the battery from overcharging.
If the battery type is set to USE the solar power will not cut off and it can’t protect the batteries from over charging.


Best
Kevin L

From: George Bridges
Sent: Friday, May 28, 2021 10:15 PM
To: Email address redacted by poster
Subject:
RE: RE: I have an issue on an all in one unit



Hello Kevin and Marco - Thanks for the assistance. I’m afraid I do not believe this had the intended effect. The setting you guys changed resulted in the unit charging the batteries but as soon as the sun went down the unit stopped using the batteries and went to utility mode. This was with a fully charged battery. I’m not trying to tell you guys your business of course.



From the manual –

Growatt_Manual_SNIP.jpg

With the settings configured for SOL as setting 01 the unit will not power from battery during the night as this satisfies one of the criteria. When the system is set to sol and any 1 of the conditions evaluates to false – IE Solar energy is not available – the unit kicks to utility bypass mode.

I believe I discovered the problem. The issue seems to be use2 setting in battery type (5). When set to that (USE2) the battery bulk and float appear to be locked together. At least on my unit if I set to USE2 the unit would not allow me to set float to anything other than what bulk was set to (19) . I could not set 20 to anything different than 19 (in USE2)

I, based on forum posting, tried just USE as setting 5. That appears (based on the last 2 days observations. To have allowed the system to charge the batteries and run the load during the day and once the batteries are charged to power the load from solar and augmenting with batteries as needed.

Can you advise what the difference between USE and USE2 is? Additionally am I on the latest firmware?


From: Email address redacted by poster
Sent:
Friday, May 28, 2021 3:58 PM
To: George Bridges
Subject: RE: RE: I have an issue on an all in one unit



Hello George,

Thank you for emailing Growatt about the issues you are experiencing.
You PV system is shutting off PV power because the AC charge priority is set to battery first. This means that it will prioritize using energy from the battery to supply power to your load first.

We have changed your AC priority to solar first which will prioritize solar energy to supply power to home loads.
We will continue to monitor your inverter system to see if this settings fix will solve your problems.
Please let us know if you have any other questions.

Best
Kevin L
 
This thread has been a real pleasure to read, as I'm having the same issue with the Growatt SPF 3000TL LVM-ES.

Keep it coming.
what is the settings you are using for 5, 19, 20 and what is the battery type and configuration as well as solar panel configuration?
 
Tried growatt's recommended setting of USE2. This was the result. Reduced the overall load last night to start the batteries off with a higher charge this am. The unit was set to USE2 - at around 9:55am sun fully up the unit just turned off the solar panels when the battery was full. I left it this way for an hour wondering if it might , dunno , rest and reset. Finally put the setting back to USE and other than some cloud interaction unit functioned fine for the rest of the day...

Growatt_Solar_clipping_still.jpg
 
what is the settings you are using for 5, 19, 20 and what is the battery type and configuration as well as solar panel configuration?
Solar Panels: 20 panels of 100 watt each, configured 10s2p for a total of 2KW
Batteries: 16 LifePo4 CALB cells 100ah, configured 48v 5KWh
BMS: Overkill Solar
All in one: Growatt SPF 3000TL LVM-ES

My system is at my off-grid cabin so I only get to test over the weekend. I'll be back on Monday with observations based on my readings from here.

Oh and here's a stupid question: Can I get the Growatt monitoring app without wifi connectivity? Doesn't seem like it. At least I have my Bluetooth BMS (Overkill Solar) for some data.
 
I agree that the load should be more than reasonable. I still have the older firmware with only 1 USR option. I have seen others indicate that there is some undesirable behavior with the newer firmware, so I haven't updated mine yet. I'm kind of curious as to whether the USR1 setting on the update is a new one, or whether the USR2 is.

But in any event, I think I jinxed myself with that long-winded post. After not seeing the solar portion shut down for a long time, it started doing that behavior again late in the day yesterday, after applying a 1300-1700 watt load for about a half hour. A man's gotta eat :)

I also tried shutting the unit down entirely and re-starting to see if that made any difference. It really didn't. The solar panels kicked in, but within a few seconds, it shut them off again and continued with the "interesting" behavior of pulling only from the battery, even though there was a reasonable amount of daylight left. Both of the upper side vents of the system were quite warm, with the higher temps coming from the left side.

At this point, I'm starting to lean toward overall system temperature as the reason. Given that the conversion from solar generates heat, and it's really the only portion of the system it can safely shut down without interrupting the load, it seems to make sense. My observation of the unit kicking the solar in and out at varying intervals ( 5-60 seconds apart) would seem to back this up, as it smacks of cycling based on the readings of a thermocouple. There may be other reasons for it to do it as well, but you'd need to be inside the mind of the system programmers to really know. And now that I think about it, the first time I saw this behavior was shortly after turning it on for the first time and intentionally "beating on it" to see just what it was capable of.

But at least in my case, it didn't seem to have any real negative effects on battery levels going into the evening, and this morning it happily topped off my batteries and the connected load with everything available, without returning to the solar cycling behavior.

I'm glad to hear that your system is working the way you want it to now. But if you see it again, check the temperature of the unit. I'm really curious now if that is the primary cause.
just wanted to follow up - my issue was not temperature. Sorry not seeing a temp issue on my unit - it's all related to the usage of us2 setting. I'll post in the broader thread about the latest email from growatt
 
Solar Panels: 20 panels of 100 watt each, configured 10s2p for a total of 2KW
Batteries: 16 LifePo4 CALB cells 100ah, configured 48v 5KWh
BMS: Overkill Solar
All in one: Growatt SPF 3000TL LVM-ES

My system is at my off-grid cabin so I only get to test over the weekend. I'll be back on Monday with observations based on my readings from here.

Oh and here's a stupid question: Can I get the Growatt monitoring app without wifi connectivity? Doesn't seem like it. At least I have my Bluetooth BMS (Overkill Solar) for some data.
you can use a computer and connect from the computer to the growatt using a usb cable - (mine came with a USB cable but it's just a usb cable - nothing special) the tool I use is pvkeeper2 and it's hosted on ian's watts247 site https://watts247.com/manuals/gw/SoftwareCD/PVKeeperV2.0/

the data from pvkeeper is very rich. But the computer has to be on to capture it. but it captures everything.
 
so in the ongoing conversation with Growatt this is the latest response.



Hello George,

The system for US2 is working as needed. As the lithium battery without a BMS, if you need to use solar power during the battery voltage is full, please ask the battery supplier to integrate the BMS of the inverter into the battery to make sure the BMS of the battery is compatible with the inverter.

Best
Kevin L


I would note that in my multiple responses they are aware that the system batteries have Overkill Solar BMS. Now that's probably not what they want as I have no way of knowing if overkill solar BMS (https://www.resacs.cz/user/related_files/8s-100a-6.pdf) will integrate with or communicate with the Growatt. I suspect not at this time. Who knows about the future. But I would also note - at no time have they challenged my observations or understanding of the US2 settings impact (solar power clipping off as batteries reach a full charge as indicated by setting 19).

I'm not terribly sure what else to do other than use setting USE since it appears to work. hopefully it isn't doing any damage to the LifePo4 cells - I don't know how it could be but I'm not a solar or battery engineer. The BMS' are not locking the battery down IE they are not going overvolt on cells or pack. And at least seemingly the unit charges the batteries and then powers the load from the panels if in use mode. You know like you'd expect a solar inverter to do. Not so much in us2 (which Growatt tech seems to be intent on us using). I mean if the thing is just going to shut off the solar in the middle of the day isn't it really just a acting like a solar Charge controller and a battery powered inverter in one box.. a solar inverter that shuts off solar panels halfway through the day is not, in my opinion, very useful. I would be interested if this behavior is seen in any of the other "all in one" s...
 
isn't it really just a acting like a solar Charge controller and a battery powered inverter in one box.

That's exactly what this system seems to be. Well, that and smart power switch to revert to Utility, if so configured.

As I indicated before, the battery is never out of the equation. I believe that the issue is that the Growatt can't assume that in USER2 mode, the lithium batteries connected to the system are protected by an external BMS, so it plays it ultra safe when charging. If I am interpreting your latest graph correctly, and am correct in my understanding of what the system is doing, energy production followed the needs of the battery and then shut off. As lithium batteries don't require a float charge, and the system could not sense a significant enough drop in the battery SOC, solar production did not resume and it continued to power the load from the battery. When you switched to USER1, presumably with a safe float charging level set, the solar kicked back in and there is a small spike for a short period of time before dropping to almost exactly the point at which the unit shut off solar on USER2. Because the unit must supply a float charge in USER1, and in the absence of a Utility connection, it does so with solar and when it's enabled, it will also power your load.

So back to USER2, and why it shuts down for so long. Compared to SLA, LiFePo4 has a very flat discharge curve. Depending on the size of the battery bank and the loads being powered, it might take quite a while before the system decides that the SOC has changed significantly enough to again start charging the batteries. Remember, with the Growatt, there are only two operational states the battery can be in. Charging or Discharging. Solar, on the other hand, has three operational states, and those are OFF, charging the battery while powering the load, and assisting the battery to power the load while it is discharging. The battery cannot be bypassed to power loads entirely by solar, so without an integrated BMS which the Growatt understands, it really can't see what is happening at the battery in fine enough detail to make better decisions for charging and the solar is turned off. I.e. if the load isn't dropping the voltage on the battery, the system seems to be thinking solar is required neither for charging or assisting, and without the ability to see what is happening at the individual cell levels, it could take a long time before the system is triggered to re-enable those functions.

As you found, forcing the system to supply a float charge via USER1 mode, seems to be the answer. So long as the batteries are protected by a BMS, and the float charge is low enough, the solar should stay engaged and be ready to power your loads as it has no other way to supply the float charge to the batteries. In this mode, your BMS (if you trust it) acts as the battery shut-off, so even if the unit thinks it is charging the battery, it may not actually be doing so, depending on the battery's SOC. As the Growatt rep has indicated, I would not recommend doing this with a "dumb" lithium battery, as there is no external mechanism to prevent overcharging, and it could be dangerous.

This probably isn't the only reason the unit cuts off solar, but it's the best my analytical capabilities can muster at this time.

*Edit* An additional thought. As alluded to by the rep, the Growatt system MAY have the ability to do what you are asking, but only if a compatible BMS is used. This BMS is likely the only way the Growatt can remove the battery from the system and allow solar to remain engaged. Essentially, the same function as provided by a 3rd party BMS with float charging enabled on the Growatt, but with finer control over the charging behavior through more information from the battery and without the float requirement.
 
Last edited:
That's exactly what this system seems to be. Well, that and smart power switch to revert to Utility, if so configured.

As I indicated before, the battery is never out of the equation. I believe that the issue is that the Growatt can't assume that in USER2 mode, the lithium batteries connected to the system are protected by an external BMS, so it plays it ultra safe when charging. If I am interpreting your latest graph correctly, and am correct in my understanding of what the system is doing, energy production followed the needs of the battery and then shut off. As lithium batteries don't require a float charge, and the system could not sense a significant enough drop in the battery SOC, solar production did not resume and it continued to power the load from the battery. When you switched to USER1, presumably with a safe float charging level set, the solar kicked back in and there is a small spike for a short period of time before dropping to almost exactly the point at which the unit shut off solar on USER2. Because the unit must supply a float charge in USER1, and in the absence of a Utility connection, it does so with solar and when it's enabled, it will also power your load.

So back to USER2, and why it shuts down for so long. Compared to SLA, LiFePo4 has a very flat discharge curve. Depending on the size of the battery bank and the loads being powered, it might take quite a while before the system decides that the SOC has changed significantly enough to again start charging the batteries. Remember, with the Growatt, there are only two operational states the battery can be in. Charging or Discharging. Solar, on the other hand, has three operational states, and those are OFF, charging the battery while powering the load, and assisting the battery to power the load while it is discharging. The battery cannot be bypassed to power loads entirely by solar, so without an integrated BMS which the Growatt understands, it really can't see what is happening at the battery in fine enough detail to make better decisions for charging and the solar is turned off. I.e. if the load isn't dropping the voltage on the battery, the system seems to be thinking solar is required neither for charging or assisting, and without the ability to see what is happening at the individual cell levels, it could take a long time before the system is triggered to re-enable those functions.

As you found, forcing the system to supply a float charge via USER1 mode, seems to be the answer. So long as the batteries are protected by a BMS, and the float charge is low enough, the solar should stay engaged and be ready to power your loads as it has no other way to supply the float charge to the batteries. In this mode, your BMS (if you trust it) acts as the battery shut-off, so even if the unit thinks it is charging the battery, it may not actually be doing so, depending on the battery's SOC. As the Growatt rep has indicated, I would not recommend doing this with a "dumb" lithium battery, as there is no external mechanism to prevent overcharging, and it could be dangerous.

This probably isn't the only reason the unit cuts off solar, but it's the best my analytical capabilities can muster at this time.

*Edit* An additional thought. As alluded to by the rep, the Growatt system MAY have the ability to do what you are asking, but only if a compatible BMS is used. This BMS is likely the only way the Growatt can remove the battery from the system and allow solar to remain engaged. Essentially, the same function as provided by a 3rd party BMS with float charging enabled on the Growatt, but with finer control over the charging behavior through more information from the battery and without the float requirement.
thank you very much for this additional detail and your insight. Do you know if this is the same way the MPP all in ones behave? I responded back to growatt and asked...

"Is there currently a battery on the market that works with growatt in the use2 setting? Like who makes a BMS that does that?"

When you say "Float Charge is low enough" for a 100 AH 24 volt lifepo4 (using 100 AH calb cells if that matters) what would you expect a float value to be. Or do you mean a float value low enough that it doesn't push the BMS into tripping? (and I am using Overkill Solar BMS)
 
Do you know if this is the same way the MPP all in ones behave? I responded back to growatt and asked...

"Is there currently a battery on the market that works with growatt in the use2 setting? Like who makes a BMS that does that?"

When you say "Float Charge is low enough" for a 100 AH 24 volt lifepo4 (using 100 AH calb cells if that matters) what would you expect a float value to be. Or do you mean a float value low enough that it doesn't push the BMS into tripping? (and I am using Overkill Solar BMS)

I have no experience with the MPP all-in-ones, other than that they seem often to be referred to in the same breath as the Growatts. My gut tells me that they are very similar units with different firmware and possibly different construction quality. Considering that the comparable MPP units have "Support lead acid and compatible LiFePo4 batteries" in the feature list, I personally wouldn't expect them to be much different.

With regard to your question about compatible batteries, I would think that you would be looking at the Lithium setting on the Growatt, rather than USE2. The latter appears to be for lithium batteries without a BMS, thus the behavior you dislike. But don't take my word for it.

Recommended float charge settings seem to be in the range of 27.2 to 27.6, based on the sticky post here and my battery manufacturers recommendations. I would think anything between those two numbers should be fine.
 
As you found, forcing the system to supply a float charge via USER1 mode, seems to be the answer. So long as the batteries are protected by a BMS, and the float charge is low enough, the solar should stay engaged and be ready to power your loads as it has no other way to supply the float charge to the batteries. In this mode, your BMS (if you trust it) acts as the battery shut-off, so even if the unit thinks it is charging the battery, it may not actually be doing so, depending on the battery's SOC. As the Growatt rep has indicated, I would not recommend doing this with a "dumb" lithium battery, as there is no external mechanism to prevent overcharging, and it could be dangerous.

I think the issue has been everyone wants to set parameters for longer cell life below the knee area. With a flat discharge curve after the knee area, the GW has no idea what the current battery SOC is. The GW battery probably has a shunt that tells the GW thru the RS485 port the SOC of the battery.

This is a case where the knee area actually becomes an advantage. Settings should take advantage of this area to keep the GW unit operating using the solar input. As the knee area is quite steep, the settings can easily trigger the GW to resume operation off solar. I think the GW engineers don't quite understand that some users might be using very large battery packs where once the knee area is passed, the curve would be very flat on a very large bank.
 
I have no experience with the MPP all-in-ones, other than that they seem often to be referred to in the same breath as the Growatts. My gut tells me that they are very similar units with different firmware and possibly different construction quality. Considering that the comparable MPP units have "Support lead acid and compatible LiFePo4 batteries" in the feature list, I personally wouldn't expect them to be much different.

With regard to your question about compatible batteries, I would think that you would be looking at the Lithium setting on the Growatt, rather than USE2. The latter appears to be for lithium batteries without a BMS, thus the behavior you dislike. But don't take my word for it.

Recommended float charge settings seem to be in the range of 27.2 to 27.6, based on the sticky post here and my battery manufacturers recommendations. I would think anything between those two numbers should be fine.
Thanks for the response - Lithium setting on the growatt is only for use of THEIR battery - it has a comunications wire that plugs into the growatt.
 
so in the ongoing conversation with Growatt this is the latest response.



Hello George,

The system for US2 is working as needed. As the lithium battery without a BMS, if you need to use solar power during the battery voltage is full, please ask the battery supplier to integrate the BMS of the inverter into the battery to make sure the BMS of the battery is compatible with the inverter.

Best
Kevin L


I would note that in my multiple responses they are aware that the system batteries have Overkill Solar BMS. Now that's probably not what they want as I have no way of knowing if overkill solar BMS (https://www.resacs.cz/user/related_files/8s-100a-6.pdf) will integrate with or communicate with the Growatt. I suspect not at this time. Who knows about the future. But I would also note - at no time have they challenged my observations or understanding of the US2 settings impact (solar power clipping off as batteries reach a full charge as indicated by setting 19).

I'm not terribly sure what else to do other than use setting USE since it appears to work. hopefully it isn't doing any damage to the LifePo4 cells - I don't know how it could be but I'm not a solar or battery engineer. The BMS' are not locking the battery down IE they are not going overvolt on cells or pack. And at least seemingly the unit charges the batteries and then powers the load from the panels if in use mode. You know like you'd expect a solar inverter to do. Not so much in us2 (which Growatt tech seems to be intent on us using). I mean if the thing is just going to shut off the solar in the middle of the day isn't it really just a acting like a solar Charge controller and a battery powered inverter in one box.. a solar inverter that shuts off solar panels halfway through the day is not, in my opinion, very useful. I would be interested if this behavior is seen in any of the other "all in one" s...

I've been watching this thread, and suspected this was going to be their response.

I am curious what you happen if you responded with something like: "My battery supplier is interested in making their BMS compatible with the Growatt inverter. They asked for documentation of the communication protocol - could you please provide? Thanks."

I would be interested to see that documentation. Hardware that can speak RS485 is cheap, and relatively simple to engineer. I know that doesn't directly resolve the issue here, but I'd love to build something that connects between the Growatt and BMS (Overkill already has documentation on its TTL UART interface, has Arduino libraries, etc.)
 
Lithium setting on the growatt is only for use of THEIR battery - it has a comunications wire that plugs into the growatt.
Yes, this is understood. My comment was regarding the question you posed to the Growatt:
"Is there currently a battery on the market that works with growatt in the use2 setting? Like who makes a BMS that does that?"
My understanding of the USE2 setting is that it is intended for a lithium battery which does not have a BMS. So technically, all batteries on the market "work" with it, just not in the manner you desire. A BMS will likely not change that fact, unless it is one which is compatible with the unit's Lithium battery setting, which would then be used instead of USE2.
 
Back
Top