EG4 6500 near miss: fans stopped when switching SBU->SUB->SBU relatively quickly

[solarhombre]

Howdy!

I was testing a dropdown helper in homeassistant that would allow me to switch the output source priority in my 2 EG4 6500 (it works!).

The sequence of events is as follows, starts with inverters in SBU output source priority:
18:09:51 POP01 (SUB) command sent to inverter 1 -> ACK
18:09:52 POP01 (SUB) command sent to inverter 2 -> ACK

A few seconds later the UPS battery in my office clicked, signaling the switch from SBU to SUB. All good. I see the graphs in homeassistant and the fans are going up as batteries start charging (40A per inverter). Then I switched the dropdown back to SBU.
18:10:16 Batteries start charging
18:10:36 POP02 (SBU) command sent to inverter 1 -> invalid CRC in response
18:10:36 POP02 (SBU) command sent to inverter 2 -> ACK
18:10:44 Batteries stop charging
18:10:56 Fans go down to 30% speed. Inverter temperature starts going up fast (47C, 50C)
18:11:09 AC kicks in, taking ~2kWh from each inverter. Fans do not move up.
18:16:17 Temps reach 65C and 59C



At this point I went to the garage, noticed the fans were too "quiet" for what they should be and killed utility input, turned off inverters and let them power off. The load was about <400W when I turned them off. Starting them a few minutes later did the trick and now the fans move up with load (not temp?!) as usual. Temperatures are normal and both inverters are in SBU mode without me changing anything after the automated commands were sent.

Anyone else experienced anything similar? Is there supposed to be a safety delay in switching output source priority?

 

[Adam De Lay]

Can't say I've ever heard of this happening.

From my understanding, all HomeAssistant is doing is sending the command to the inverter to make the switch (just like WatchPower or SolarAssistant does). The inverter's actually the thing making the change between modes.

I do know that I've seen the inverters take their time between switches when I've performed them in the past. Sometimes it's not instant.

 

[solarhombre]

Can't say I've ever heard of this happening.

From my understanding, all HomeAssistant is doing is sending the command to the inverter to make the switch (just like WatchPower or SolarAssistant does). The inverter's actually the thing making the change between modes.

I do know that I've seen the inverters take their time between switches when I've performed them in the past. Sometimes it's not instant.

Yeah, it takes a few seconds, probably longer than what I waited to switch it again. AC starting right in between probably didn't help. It hasn't happened again.

Now my only problem is that some values (max charge amps) don't stick even when I change them on the inverters. I'll keep playing until I figure out what happens. I also don't like that the display is somehow powered. It keeps a mini fan I have plugged in a USB port to the Pi running even when the Pi and the inverters are off. I might switch to serial cables instead of the OTG port.

 

[PreppenWolf]

I place a large portion of my systems stability (6500s) on the fact I do not use any 3rd party software or apps to constantly tinker with them. If I want to change something, I walk down to the inverters and do it on the panel.

 

[Adam De Lay]

Yeah, it takes a few seconds, probably longer than what I waited to switch it again. AC starting right in between probably didn't help. It hasn't happened again.

Glad it hasn't happened again, but I'm not a big fan of anomalies...

Now my only problem is that some values (max charge amps) don't stick even when I change them on the inverters. I'll keep playing until I figure out what happens.

I know there were a few firmware versions that had issues with settings not saving properly. Which version are you running?

I also don't like that the display is somehow powered. It keeps a mini fan I have plugged in a USB port to the Pi running even when the Pi and the inverters are off. I might switch to serial cables instead of the OTG port.

Yeah this was one of the reasons I didn't like using the OTG port for the Solar Assistant communication. The fact that the pi kept back-feeding power to the display if I ever had to power the system down was annoying.

 

[solarhombre]

Glad it hasn't happened again, but I'm not a big fan of anomalies...

Same here. And the glitch with max charge not sticking started after that.

I know there were a few firmware versions that had issues with settings not saving properly. Which version are you running?

Latest and greatest. 79.11+61.13.
I use your resources thread to know when a new one comes up. Thank you!

Yeah this was one one of the reasons I didn't like using the OTG port for the Solar Assistant communication. The fact that the pi kept back-feeding power to the display if I ever had to power the system down was annoying.

I had a Pi4 dying (8 red flash lights, memory error) after cutting power to the inverters while they were on. I blame a surge sent thru USB to the Pi but can't prove anything.

 

[Adam De Lay]

Same here. And the glitch with max charge not sticking started after that.

Does this happen even if you change the value directly on the inverter?

 

[solarhombre]

Does this happen even if you change the value directly on the inverter?

Yes, that's how I've been doing it for the past few days at least once a day to no avail. It resets back to 120A every time.

When testing by hand before adding the automation to HA, tried the MCHGC and MUCHGC commands and I was getting an ACK but nothing changed when I looked on the inverter itself.

 

[Adam De Lay]

Yes, that's how I've been doing it for the past few days at least once a day to no avail. It resets back to 120A every time.

When testing by hand before adding the automation to HA, tried the MCHGC and MUCHGC commands and I was getting an ACK but nothing changed when I looked on the inverter itself.

So that means the inverter isn't allowing the change. I've had this happen several times but there hasn't been any consistency.

The last time it happened, it was resolved after I performed the next firmware update. I don't think it was the actual update that fixed the issue though. I think it was either the full reboot of the inverters OR just the re-application of a firmware update. SS has said sometimes the updates don't process fully (even though they look like they do), so re-apply the update might resolve the issue.

 

[solarhombre]

So that means the inverter isn't allowing the change. I've had this happen several times but there hasn't been any consistency.

The last time it happened, it was resolved after I performed the next firmware update. I don't think it was the actual update that fixed the issue though. I think it was either the full reboot of the inverters OR just the re-application of a firmware update. SS has said sometimes the updates don't process fully (even though they look like they do), so re-apply the update might resolve the issue.

Well, that gives me hope. I was tempted to reinstall the current version, but I'll wait for the next update.

Lucky me, I just did a full reset of all control parameters (see reply to https://diysolarforum.com/threads/factory-reset-for-eg4-6500ex-48.55292/) just when the batteries reached 95% so setting #2 changed to 10A until full. I'll do another test of setting max current once I get the batteries below 95% again.

Thanks!

 

[Adam De Lay]

Well, that gives me hope. I was tempted to reinstall the current version, but I'll wait for the next update.

Lucky me, I just did a full reset of all control parameters (see reply to https://diysolarforum.com/threads/factory-reset-for-eg4-6500ex-48.55292/) just when the batteries reached 95% so setting #2 changed to 10A until full. I'll do another test of setting max current once I get the batteries below 95% again.

Thanks!

Good luck!

 

[solarhombre]

Found the "issue". From the manual:


It seems that setting #2 is controlled by the BMS. I have the 2P1 inverter set to EG4 and the 2P2 set to USE to avoid an error code. I would swear I changed setting #2 before, but oh well.

 

[Adam De Lay]

Found the "issue". From the manual:
View attachment 156273

It seems that setting #2 is controlled by the BMS. I have the 2P1 inverter set to EG4 and the 2P2 set to USE to avoid an error code. I would swear I changed setting #2 before, but oh well.

Ah yeah. I guess I should have asked if you were using the EG4 option for your batteries. When you're using "closed loop communication", the BMS takes over several settings in the inverter. Unfortunately those values don't get updated so you really don't know what the actual values are. They're just placeholders.