EG4-6500EX remote shutdown options?

[fmeili1]

Hi,

I'm trying to find a "clean" solution to remote shutdown my parallel EG4-6500EX AIO's (e.g. in case of an emergency).

I've already installed contactors/relays to shutdown the whole system by disconnecting all possible energy sources of the AIO's (AC-in, batteries and PV via Tigo PVRSS) in case of an emergency (e.g. smoke detector activation or pushing one of the emergency buttons or remote controlled by intention). Usually the AC-in contactor remains disconnected until I really need to use the grid to cover the power needs of the house (all of these disconnects could be controlled remotely via MQTT from a SmartHome system. Also all SA features are available in the SmartHome system for monitoring and controlling).

But I fear that if the AIO's are running as usual in inverter mode under high load it may not be good for the AIO's electronic components to suddenly just cut all sources of energy which will in fact be a "hard" shutdown for the AIO's.

Now I think about how to make such an emergency shutdown more friendly (soft) for the AIO's. After studying the AIO manual, the Modbus protocol manual and the SolarAssistant features, I was not able to find a clean solution so far.

First idea was to send an MQTT command via SA to switch to standby mode. But I've learned the EG4 Modbus protocol does not implemented such a feature, therefore it's not possible to be implemented in SA as well. Until now, it looks like that the AIO can only be switched to standby via releasing the power button manually by pressing it.

Second idea was (because of a hint in the SA power management automation description), to raise the program 29 setting (Low DC cut-off voltage) to a very high voltage to let the inverter detect suddenly a low voltage condition which would initiate immediately a shutdown (regarding the description in the manual). But unfortunately the highest voltage will be 48V for this setting. My batteriy voltage will usually always be over 48V but for this solution to function I need to set it e.g. to 60V to always trigger a shutdown via this low voltage cut-off feature. I'm not using closed loop communication and so I need to rely on voltages for this setting but anyway even with closed loop BMS communication I would not be able to set this value higher as 90% which would also not be enough to initiate an emergency shutdown because it have to work even if the batteries are at 100% SOC.

Third idea was to limit the "Maximum discharging current" to zero with program 41 to stop energy usage from the batteries, but the lowest possible value for this program is 30A.

Fourth idea was to play around with program 1 "Output source priority". Usually my system runs always in "SBU" mode. So I thought about the following scenario in case an emergency shutdown would happen:

1. Activate AC-in contactors to prepare using the "USB" mode
2. Set AIO's to "USB" mode (via SA and MQTT commands) and wait until all changed to "Grid" mode
3. Disconnect the batteries from the AIO's via the battery relays
4. Disconnect the PV energy by initiating a PVRSS via Tigo
5. Now disconnect AC-in via the contactors

The idea for this fourth variant is, that while the AIO's are in bypass (grid) mode (while no battery and no PV connected), it would be less stress for the AIO electronic to do a "hard" shut down which in fact will happen after disconnecting the AC-in.

But I still hope for a better solution...

Maybe someone will have other ideas...

 

[Crowz]

I roamed around the net and wasn't able to find anyone doing an emergency shut down setup on that inverter anywhere. I guess its more for the newer ones that have to meet that requirement with the changes that were added to the rules.

 

[Adam De Lay]

Fourth idea was to play around with program 1 "Output source priority". Usually my system runs always in "SBU" mode. So I thought about the following scenario in case an emergency shutdown would happen:

1. Activate AC-in contactors to prepare using the "USB" mode
2. Set AIO's to "USB" mode (via SA and MQTT commands) and wait until all changed to "Grid" mode
3. Disconnect the batteries from the AIO's via the battery relays
4. Disconnect the PV energy by initiating a PVRSS via Tigo
5. Now disconnect AC-in via the contactors

The idea for this fourth variant is, that while the AIO's are in bypass (grid) mode (while no battery and no PV connected), it would be less stress for the AIO electronic to do a "hard" shut down which in fact will happen after disconnecting the AC-in.

Aside from some kind of hard disconnect, I can't think of any other way to accomplish this. The only thing that concerns me is that you're having to rely on an outside device sending the proper programming commands to trigger this type of shut down.

I guess the only other thing that comes to mind is if you could somehow wire a relay in series with the 6500 "power button" and then interrupt that switch when you need to do the emergency shutdown. That should put the inverter in standby mode. I will say I'm unaware if the switch is wired in or soldered to a pcb.

 

[smr.ruby]

When I bought an MPP LVX6048 from Ian at Watts247 and I wanted to do this and he told me that there was a pair of wires in the RJ45 connector for the display that were connected to the state of the push button. I still haven't tried this but he said you could remote control on/off by connecting that pair. I always assumed I could figure out which pair tracing the board or pushing the button and testing wires. Since your EG4-6500EX is a very similar Voltronic unit maybe it has the same feature?

Does anybody else know about this? I searched the internet and couldn't find anything. I had asked Ian what pins they were but he didn't respond and I didn't bug him - nice guy though

 

[Quattrohead]

Yes the on-off button is just a pair of wires from the control panel back to the guts, that little two pin one that likes to fall out when you dangle the cover from the wires.

 

[smr.ruby]

Yes the on-off button is just a pair of wires from the control panel back to the guts, that little two pin one that likes to fall out when you dangle the cover from the wires.

I'm not following. It makes sense that the button would be connected to a pair of wires from the display back to the main box. But if I let the face plate dangle from the wires it is dangling from a single multi-conductor cable with ethernet type RJ45 connectors. Which "little two pin" one? We're talking about this panel right? Power button is (6) and the connector we assume the power button wires are exposed in is (11). I can't test now as I'm a couple hours drive from my inverters.

 

[Adam De Lay]

I'm not following. It makes sense that the button would be connected to a pair of wires from the display back to the main box. But if I let the face plate dangle from the wires it is dangling from a single wire with ethernet type RJ45 connectors and a single multi-conductor cable. Which "little two pin" one? We're talking about this panel right? Power button is (6) and the connector we assume the power button wires are exposed in is (11). I can't test now as I'm a couple hours drive from my inverters.
View attachment 185838

He’s talking about when you remove the bottom cover to access all the wiring, not the display itself. When you remove the front cover, there’s 2 or 3 sets of wires you need to disconnect.

 

[smr.ruby]

Okay that is where my 6000EX isn't the same as the 6500EX it appears, didn't realize that. The display panel connector is the same RJ45 but I guess you are talking about these connectors? I don't have these on mine, only the RJ45 goes from the cover to the main chassis. Still even in the 6000EX the power button is probably a pair of wires in the RJ45 connector to the control panel. Same with OPs 6500EX

 

[Quattrohead]

The RJ45 and the three connectors plug into a little interface board, that may be mounted inside the 6000 unit instead of in the control panel like the 6500.

 

[smr.ruby]

First idea was to send an MQTT command via SA to switch to standby mode. But I've learned the EG4 Modbus protocol does not implemented such a feature, therefore it's not possible to be implemented in SA as well.

@fmeili1 : while I agree that finding a digital input option for shutting down the inverter should be the goal here... since you mentioned the Modbus interface have you had success writing any values over Modbus and changing inverter settings? I was excited to see modbus for the 6000EX as well as I'm still using the MPP/Voltronic protocol (via this) to interrogate the inverter - which works great but all of the write commands that should exist fail. If somebody had success writing over Modbus then I've got a compelling reason to stop using the original protocol.

 

[fmeili1]

Aside from some kind of hard disconnect, I can't think of any other way to accomplish this. The only thing that concerns me is that you're having to rely on an outside device sending the proper programming commands to trigger this type of shut down.

yeah, that's right. It has to be implemented reliable because it's external. In my case I already use a self made controller with a mini UPS which has to be reliable anyway. If my controller breaks, the whole solar system would stop working because the PVRSS, the battery contactors and the AC-in contactors are all normally open (NO) driven so everything depends on this working controller...

I guess the only other thing that comes to mind is if you could somehow wire a relay in series with the 6500 "power button" and then interrupt that switch when you need to do the emergency shutdown. That should put the inverter in standby mode. I will say I'm unaware if the switch is wired in or soldered to a pcb.

That's a really good idea - as a last resort!
I'll try to open one of these display modules and will take a look how this button could be modded to be "smart"...

 

[fmeili1]

@fmeili1 : while I agree that finding a digital input option for shutting down the inverter should be the goal here... since you mentioned the Modbus interface have you had success writing any values over Modbus and changing inverter settings? I was excited to see modbus for the 6000EX as well as I'm still using the MPP/Voltronic protocol (via this) to interrogate the inverter - which works great but all of the write commands that should exist fail. If somebody had success writing over Modbus then I've got a compelling reason to stop using the original protocol.

I never programmed the Modus interface for the EG4-6500EX by myself (I've only studied the Modbus spec of the EG4-6500EX). I just use SA to do the Modbus communication with the inverters and batteries. My smart home (OpenHAB) communicates via MQTT with SA and SA converts these MQTT messages to Modbus protocol.

All values which SA offers for read/write I'm able to change.