FlexBoss21 / GridBoss Issue List

[skimandan]

Howdy folks,
First solar setup. Everything from the panels, mounts, j-boxes, disconnect switches, wallmount batteries, etc worked exactly as planned with exception of the GridBoss/FlexBoss combo. I've had my gridboss / flexboss system in production for a week, trying to fine tune my system so I don't need to touch it again for a decade or so. I have encountered a few features that do not work per spec, or in the way advertised, or are just missing completely. Signature Solar has guided me through a number of undocumented workarounds to some of the issues, what I lovingly call "hacks" that can resolve some of them.

Here is my list. Let me know if these are resolved. Please link to other issues so future buyers are aware of the current status of these issues. I'll try to stay on top of the thread and will update this post as needed. I want these products to succeed and be relevant for many years.
Sincerely,
Your humble unpaid beta tester.

Software Config / Performance –

• FlexBoss21 Manual vs. Software settings vs. Signature Solar settings. Many of the settings in the software do not do what you would expect. In order to keep my battery throughout the night I have to Activate A/C charge, then give a time period, then trickle charge the battery for the time period. Seriously? What is the point of having a manual if none of the settings in the software work the way they should. Every modification to settings practically requires a call to Signature Solar because the settings do not do what you think they should do. I can only guess that someone thought it better to staff with lots of technical support techs instead of writing an accurate manual and quality software. The problem is when I need to make a change on a weekend or if EG4 goes bankrupt, or I no longer have the latest and greatest inverter and no one is familiar with it. PLEASE UPDATE YOUR SOFTWARE AND MANUALS SO A SKILLED HUMAN WITH ELECTRICAL KNOWLEDGE CAN OPERATE IT WITH CONFIDENCE!
• FlexBoss21 Optimal string voltage is claimed to be 120V-440, but does not appear to work even when string voltage is higher than 140V. This is costing me production I was planning on. https://diysolarforum.com/threads/flexboss21-error.98378/
• GridBoss may not be able to handle high pass through current from the Grid. Awaiting resolution. Resolved in beta firmware provided to SlowRoller 2/2/2025. https://diysolarforum.com/threads/gridboss-with-flexboss-not-handling-21kw-pass-through.97560/
• Pressing RSD button on FlexBoss21 shuts down GridBoss backup load even though the grid connection remains active. Not user configurable. 2/5/25 EG4 Tech confirmed this is intended. There is no mention in the manual of main grid shutdown during RSD.
• GridBoss Smart Ports - Signature Solar Tech warned me the Smart Ports may not work in a grid down scenario when I was asking how to set it up with my EV charger. I asked three times to confirm. They said they double checked with level 2 tech. This may be a training issue with that poor tech, but I'm putting it here until proven otherwise. Ports verified to work 1/31/25 with Grid down. See this thread. https://diysolarforum.com/threads/gridboss-smart-loads.98598/
• Time of Use settings described on page 45 of the FlexBoss21 settings do not exist in the inverter settings. Requires workarounds to enact and the techs were unable to create enough workarounds to meet my 3 pricing periods. Also unable to handle weekend pricing. https://diysolarforum.com/threads/flexboss-21-time-of-use-settings.98597/

Cosmetic – simple workmanship issues that make me not want to show normal people my system:

• Web monitor does not provide picture of Flexboss21 when linked to Gridboss – its just a nebulous square. Upload a nice selfie please!
• Indoor Wallmount conduit box top cutouts do not line up with GridBoss punchouts.
• Copper only - this was well documented but nobody uses copper at 200A!

@EG4TechSolutionsTeam

 

[skimandan]

New issue identified 1/29/25
Quick charge button in phone app sets 1 hour AC charge time, but does not change your charge rate. My charge rate is set to .1 kw from another hack to hold battery charge overnight due to TOU rates. I would like the charge rate to change to max for the quick charge duration.
1/30/25 See Post 18 - The Gridboss is hold warning notice that won't clear "MIDBOX_WARNING_015: GEN voltage and frequency abnormality" - According to SS it is a known issue. EG4 confirmed upcoming fix 2/5/25.

Update 2/5/25 see post 30 for details. Essentially no fixes, confirmed AC Gen error fix upcoming

Update 3/4/25 Grid Sellback appears to be limited to 12kW even though Grid Sell Back Power(kW) is set to 16 kW. Limits PV production if no local loads are available. See posts starting at 86. Will start a separate thread to discuss, time permitting.

 

[mciholas]

PLEASE UPDATE YOUR SOFTWARE AND MANUALS SO A SKILLED HUMAN WITH ELECTRICAL KNOWLEDGE CAN OPERATE IT WITH CONFIDENCE!

The amount of returns, customer frustration, support calls, mistakes that are caused by poor documentation is astounding to me for Chinese products like inverters. Apparently, there is a culture in China of simply not providing manuals or providing lame ones. This really hurts their product reputations.

If they would put some effort into making precise and clear manuals, they would easily make back the investment in reduced support and returns. But no, they can't seem to do that for some stupid reason.

FlexBoss21 Optimal string voltage is claimed to be 120V-440, but does not appear to work even when string voltage is higher than 140V. This is costing me production I was planning on. https://diysolarforum.com/threads/flexboss21-error.98378/

This one is not an issue with the unit, you just didn't understand how they work and what compromises you are making with your highly disparate string voltages. The manuals do say you need to get to 250 volts for full MPPT power. There's nothing here for EG4 to fix other than perhaps being a bit more clear in one sentence that you feel mislead you.

Pressing RSD button on FlexBoss21 shuts down GridBoss backup load even though the grid connection remains active. Not user configurable.

That's the way it should work. Hitting RSD should shutdown the entire "solar system" and the backup outputs from GB are part of that. If the grid to backup load relay stayed active, then the FB AC voltages would remain present in RSD.

If you want the backup loads to work, put them in bypass before you RSD the inverter.

GridBoss Smart Ports - Signature Solar Tech warned me the Smart Ports may not work in a grid down scenario when I was asking how to set it up with my EV charger. I asked three times to confirm. They said they double checked with level 2 tech. This may be a training issue with that poor tech, but I'm putting it here until proven otherwise.

I'm confused what your complaint was. Do smart ports work grid down or not? If they don't, then that is a problem. If they do, what is the complaint?

Mike C.

 

[Brucey]

This one is not an issue with the unit, you just didn't understand how they work and what compromises you are making with your highly disparate string voltages. The manuals do say you need to get to 250 volts for full MPPT power. There's nothing here for EG4 to fix other than perhaps being a bit more clear in one sentence that you feel mislead you.

I'm sorry but how is one to know that the 120V mentioned in the specifications is actually nonsense and the actual minimum voltage for full production is 250V, over double the spec?

Does this also apply to the 18kpv or is this something new?

@FilterGuy not sure if you can give some background on design decisions here?

 

[CaySol]

Smart ports work with grid down for me. And the manual is a bit confusing but it’s accurate. I haven’t had a problem with any of the settings yet. One quirk is that ac charge takes a while to “spin up” after you activate. Maybe more than a minute

 

[NC_hydro]

trying to fine tune my system so I don't need to touch it again for a decade or so.

When you say words like this, that's when you skip EG4 BetaBoxes and go with Victron, SMA, Schneider tier products.

 

[CaySol]

Sizing your string to just have enough power to start the mppt when you have full sun is probably a mistake. Why cut it so close?

 

[mciholas]

I'm sorry but how is one to know that the 120V mentioned in the specifications is actually nonsense and the actual minimum voltage for full production is 250V, over double the spec?

From the manual:



You have to get over 140 volts to start, it will track to 120, but full power requires 250, and optimal is 360.

If you wire your system to have 440 and 120 MPPT voltages, it will suck, and it did. That's about as bad as it gets for PV setup.

MPPTs work this way on a wide variety of systems. There is nothing to fix on the unit here.

Mike C.

 

[42OhmsPA]

Sincerely,
Your humble unpaid beta tester.

Love your honesty.

Good luck on getting things resolved and working optimally so you can enjoy your system.

From the manual:

Why wouldn't they make the spec sheet @Brucey shared the same?
I know I know RTFM but in my mind a spec sheet should have critical parameters and match the manual.

 

[skimandan]

This one is not an issue with the unit, you just didn't understand how they work and what compromises you are making with your highly disparate string voltages. The manuals do say you need to get to 250 volts for full MPPT power. There's nothing here for EG4 to fix other than perhaps being a bit more clear in one sentence that you feel mislead you.

Yes Mike we discussed this in the linked thread. I still don't understand you defending it. I'm not interested in the electrical engineering, I paid EG$ to figure that out. It just doesn't work the way that their WORDS say it should, and I shouldn't have to install the device in order to learn this. I accept the hardware probably can not change to meet my string design, but THE WORDS THEY USE TO DESCRIBE IT can change.

Smart ports work with grid down for me. And the manual is a bit confusing but it’s accurate. I haven’t had a problem with any of the settings yet. One quirk is that ac charge takes a while to “spin up” after you activate. Maybe more than a minute

Probably a training issue with Signature Solar Tech. Signature solar has reached out. I will confirm.

That's the way it should work. Hitting RSD should shutdown the entire "solar system" and the backup outputs from GB are part of that. If the grid to backup load relay stayed active, then the FB AC voltages would remain present in RSD.

Why should my backup loads be shut off when the grid is still active? Are there not relays that isolate the inverter from the GridBoss? I thought the intent was bring generation voltages below 100V and to deactivate the batteries, not shut down power to the entire house when grid is active. Yes, if I am home I can flip a switch, but why?

When you say words like this, that's when you skip EG4 BetaBoxes and go with Victron, SMA, Schneider tier products.

I don't think any of them would have been able to provide a product that meets my needs within a budget, and I am too far into this to change now. I also think nearly all of the Flexboss/GridBoss issues are fixable. I do wonder if they will invest in fixing them or just move on to the next product.

Sizing your string to just have enough power to start the mppt when you have full sun is probably a mistake. Why cut it so close?

See the thread discussing it for more details.

 

[mciholas]

I still don't understand you defending it.

If you bought a car that said "30 miles per gallon" and then drove it at 120 miles per hour, you aren't going to get 30 miles per gallon. Whose fault is that?

It works exactly the way the manual says it works.

Mike C.

 

[skimandan]

MPPTs work this way on a wide variety of systems. There is nothing to fix on the unit here.

Mike, again you are not understanding my complaint. Please start your own thread to opine on optimal inverter theory. I'm talking about the MARKETING and WORDS, not hardware. Communication of what the thing I bought will do!

• Three MPPT solar charge controllers support a PV input of 600V with an optimal range of120 – 440 VDC. The three MPPTs allow for flexibility in paralleling strings. MPPT 1 and 2allow up to 26A each, while MPPT 3 allows 15A, for a total of 21kW of utilized solar power

Nobody is criticizing my 480V strings that are also out of optimal.

 

[EG4TechSolutionsTeam]

Howdy folks,
First solar setup. Everything from the panels, mounts, j-boxes, disconnect switches, wallmount batteries, etc worked exactly as planned with exception of the GridBoss/FlexBoss combo. I've had my gridboss / flexboss system in production for a week, trying to fine tune my system so I don't need to touch it again for a decade or so. I have encountered a few features that do not work per spec, or in the way advertised, or are just missing completely. Signature Solar has guided me through a number of undocumented workarounds to some of the issues, what I lovingly call "hacks" that can resolve some of them.

Here is my list. Let me know if these are resolved. Please link to other issues so future buyers are aware of the current status of these issues. I'll try to stay on top of the thread and will update this post as needed. I want these products to succeed and be relevant for many years.
Sincerely,
Your humble unpaid beta tester.

Software Config / Performance –

• FlexBoss21 Manual vs. Software settings vs. Signature Solar settings. Many of the settings in the software do not do what you would expect. In order to keep my battery throughout the night I have to Activate A/C charge, then give a time period, then trickle charge the battery for the time period. Seriously? What is the point of having a manual if none of the settings in the software work the way they should. Every modification to settings practically requires a call to Signature Solar because the settings do not do what you think they should do. I can only guess that someone thought it better to staff with lots of technical support techs instead of writing an accurate manual and quality software. The problem is when I need to make a change on a weekend or if EG4 goes bankrupt, or I no longer have the latest and greatest inverter and no one is familiar with it. PLEASE UPDATE YOUR SOFTWARE AND MANUALS SO A SKILLED HUMAN WITH ELECTRICAL KNOWLEDGE CAN OPERATE IT WITH CONFIDENCE!
• FlexBoss21 Optimal string voltage is claimed to be 120V-440, but does not appear to work even when string voltage is higher than 140V. This is costing me production I was planning on. https://diysolarforum.com/threads/flexboss21-error.98378/
• GridBoss may not be able to handle high pass through current from the Grid. Awaiting resolution. https://diysolarforum.com/threads/gridboss-with-flexboss-not-handling-21kw-pass-through.97560/
• Pressing RSD button on FlexBoss21 shuts down GridBoss backup load even though the grid connection remains active. Not user configurable.
• GridBoss Smart Ports - Signature Solar Tech warned me the Smart Ports may not work in a grid down scenario when I was asking how to set it up with my EV charger. I asked three times to confirm. They said they double checked with level 2 tech. This may be a training issue with that poor tech, but I'm putting it here until proven otherwise.
• Time of Use settings described on page 45 of the FlexBoss21 settings do not exist in the inverter settings. Requires workarounds to enact and the techs were unable to create enough workarounds to meet my 3 pricing periods. Also unable to handle weekend pricing. https://diysolarforum.com/threads/flexboss-21-time-of-use-settings.98597/

Cosmetic – simple workmanship issues that make me not want to show normal people my system:

• Web monitor does not provide picture of Flexboss21 when linked to Gridboss – its just a nebulous square. Upload a nice selfie please!
• Indoor Wallmount conduit box top cutouts do not line up with GridBoss punchouts.
• Copper only - this was well documented but nobody uses copper at 200A!

@EG4TechSolutionsTeam

Thank you for your feedback! We appreciate you taking the time to share your experience. We will investigate the issues you've mentioned further, as we are committed to continuously improving our products to make them easier and more efficient for our customers to use.

Can you please post screenshot of your settings for us to review? Thank you!

 

[Brucey]

From the manual:

View attachment 274372

You have to get over 140 volts to start, it will track to 120, but full power requires 250, and optimal is 360.

If you wire your system to have 440 and 120 MPPT voltages, it will suck, and it did. That's about as bad as it gets for PV setup.

MPPTs work this way on a wide variety of systems. There is nothing to fix on the unit here.

Mike C.

This is not the way my Victron mppts work. Now they are separates versus an AIO, but my Victron rs450/100 and 150/35s can deliver full performance at their spec sheets listed lowest startup voltage. For the 150/35s that's a startup voltage of +5V over battery voltage, and +1V to continue charging. For the 450/100 its 120V startup and 80V operating minimum. There's no need to dig in the manuals to figure out what the true minimum voltage for full production is.

 

[Brucey]

Mike, again you are not understanding my complaint. Please start your own thread to opine on optimal inverter theory. I'm talking about the MARKETING and WORDS, not hardware. Communication of what the thing I bought will do!

• Three MPPT solar charge controllers support a PV input of 600V with an optimal range of120 – 440 VDC. The three MPPTs allow for flexibility in paralleling strings. MPPT 1 and 2allow up to 26A each, while MPPT 3 allows 15A, for a total of 21kW of utilized solar power

Nobody is criticizing my 480V strings that are also out of optimal.

Optimal range should be listed as starting at 250V, to match what's in the manual.

 

[mciholas]

• Three MPPT solar charge controllers support a PV input of 600V with an optimal range of120 – 440 VDC.

If they had said "an optimizing range of 120 - 440 VDC", that would have been more accurate since the MPPT will do what it can over that range. But that's not the "optimal" voltage to give the MPPT.

So you are upset about one word that is slightly incorrect and better details are given in the manual.

Mike C.

 

[skimandan]

I sent EG4 my settings.

Disappointed is all.

The proper auto analagy is being sold a car that accepts 85 octane, so you buy a tanker full of 85, and then trolls on the internet tell you your car is going to run like garbage on 85 and you need 91, and that you you should have known better. But all you have is that tanker of 85. So it sputters instead of purrs. And the trolls continue to troll, despite politley being asked to leave.

In the meantime, I'm going to go out and yell at the clouds. It's full time job here in Portland!

 

[obsoletion]

I have a Gridboss and 3 Flexboss inverters. Been running about a week now. They are working but it does take some time to understand the manual and how to set them all up. I agree the manuals and process could be better, but so far the product is solid in operation. That said, there are some issues that need to be resolved.

1) The time in the Gridbos and Flexbosses don't sync. They really should have a way to get and sync time from a time server.
2) The app and the web monitor are not the same and even the setting are different - that can cause some issues
3) The Gridboss is not working with the mobile app for monitoring, but you can sometime change the settings.... sometimes
4) The Gridboss is hold warning notice that won't clear "MIDBOX_WARNING_015: GEN voltage and frequency abnormality" - According to SS it is a known issue.

It's a new product and has great potential and looking forward to these issues being resolved with software updates

 

[skimandan]

1) The time in the Gridbos and Flexbosses don't sync. They really should have a way to get and sync time from a time server.

Mine appear synced on the web browser version. I think Signature Solar did some undocumented magic on my first call and linked the system like this. This is the "nebulous square" that should be a FlexBoss21 image.

4) The Gridboss is hold warning notice that won't clear "MIDBOX_WARNING_015: GEN voltage and frequency abnormality" - According to SS it is a known issue.

I have this too since initial setup. SS said it would clear if I powered down for a while but it did not. Now I'm driving with a warning light. I added this to the list above.

I have pretty much given up on the app and just use the browser version. It can be accessed from any device for consistency and the only useful feature in my opinion, Quick Charge Button, doesn't work for me.

 

[obsoletion]

I am struggling with an issue on my Flexboss setup. When using closed communications, whenever I have the system export to grid from the batteries, it has a communication error (W000) and stops working for 5 minutes. It then clears the comm error and resumes for some random amout of time, 10 mins to 90mins, then repeats with the error and reset. It only happen when discharging to the grid. It may also when just operating on the batteries when the grid is down, but i have not tested that. Signature solar has checked it and the only solution at this point is to set it to Lead Acid. Hoping its a firmware issue that gets resolved.

 

[mciholas]

When using closed communications, whenever I have the system export to grid from the batteries, it has a communication error (W000) and stops working for 5 minutes. It then clears the comm error and resumes for some random amout of time, 10 mins to 90mins, then repeats with the error and reset.

This profiles as a bit error on the communication channel caused by electrical noise which then freaks out the firmware.

Ideally, the firmware should be robust enough to tolerate such errors as long as they aren't too many, but it is hard for developers to handle the cases they don't see in testing. Your setup may have exposed such a case.

Suggestions:

1. Use good quality cables for the comms channel. Cat6 is best.

2. Double check your battery is properly grounded to the inverter and the inverter is grounded properly.

3. For RS485, many folks don't include a ground connection in the comm cable, just wire A and B. I advocate you do include a ground wire to get common mode noise excursion on the comm cable down to a minimum.

4. Don't route the comm cable next to AC or DC power lines. Try to physically keep it separate. Running it inside the same conduit is particularly bad. Maybe temporarily run a comm cable out in the open to test this theory.

5. On the FB21, there is a set of DIP switches that control "RS485 balance resistance" which I interpret to mean it selects RS485 resistive termination. I would ask EG4 about this and maybe you need to flip those switches (or they got flipped the wrong way at one time). The manual does not explain these switches at all.

6. Based on your prior post, you have multiple inverters, so the communication between them and to the GB could all be part of the problem, it might not be the battery comms that is really at fault. To isolate, do go to lead acid briefly and see if that changes the behavior.

You left out useful details like what battery you have, what communication protocol are you using, how long is the cable, and so forth. Details matter for debugging things.

Mike C.

 

[Quattrohead]

Yes definitely double check your termination settings.

Also it is preferable to run your data away from your AC power but not absolutely necessary, but local conditions may dictate otherwise.

I don't actually know if it is bog standard RS485 or what running between the inverter and grid boss. Possibly RS485 with a separate 60hz frequency standard, not sure at all.

If you are going to run shielded RS485, only connect it to one end.

 

[CaySol]

Built shielded cat6a cables to run with the Flexboss grid ac wires back to the grid boss. Only grounded one termination(at the Flexbosses). About an 80ft run and seems to be working fine so far.

 

[obsoletion]

This profiles as a bit error on the communication channel caused by electrical noise which then freaks out the firmware.

Ideally, the firmware should be robust enough to tolerate such errors as long as they aren't too many, but it is hard for developers to handle the cases they don't see in testing. Your setup may have exposed such a case.

Suggestions:

1. Use good quality cables for the comms channel. Cat6 is best.

2. Double check your battery is properly grounded to the inverter and the inverter is grounded properly.

3. For RS485, many folks don't include a ground connection in the comm cable, just wire A and B. I advocate you do include a ground wire to get common mode noise excursion on the comm cable down to a minimum.

4. Don't route the comm cable next to AC or DC power lines. Try to physically keep it separate. Running it inside the same conduit is particularly bad. Maybe temporarily run a comm cable out in the open to test this theory.

5. On the FB21, there is a set of DIP switches that control "RS485 balance resistance" which I interpret to mean it selects RS485 resistive termination. I would ask EG4 about this and maybe you need to flip those switches (or they got flipped the wrong way at one time). The manual does not explain these switches at all.

6. Based on your prior post, you have multiple inverters, so the communication between them and to the GB could all be part of the problem, it might not be the battery comms that is really at fault. To isolate, do go to lead acid briefly and see if that changes the behavior.

You left out useful details like what battery you have, what communication protocol are you using, how long is the cable, and so forth. Details matter for debugging things.

Mike C.

This is awesome feedback. I am using 12 Basen Green 314ah batteries and they have JBD BMS's. It says it communicates with LUX inverters over CAN bus only, RS485 is not supported.

I am going to switch to a shielded CAT 6 from the orange cable that came with the inverters and also add ground wires to the batteries which i didn't do...... will see how that goes and post some more.

 

[mciholas]

It says it communicates with LUX inverters over CAN bus only, RS485 is not supported.

CANbus is less tolerant to noise than RS485. Just the nature of its dominant/recessive bit encoding scheme.

A shielded cable might help.

Do check the grounding.

Mike C.