question

tylera avatar image
tylera asked

CCGX crash and alternating device IDs

I have a rather complicated set-up(*).


Today my CCGX went through periods of rebooting continually. There is a single BMV connected to the CCGX. During the crashes, it seemed that the CCGX was assigning the BMV two different IDs (289 to start, and 288 to end). What would cause this?



screen-shot-2021-04-14-at-62402-pm.png



(*) I have a 48V and a 24V system. They are connected by an MPPT (not connected to the CCGX).

There is a single CCGX

The 48V and 24V each have a MPPT connected to solar panels (each connected to the CCGX).

The 48V has a third-party BMS connected via CAN to the CCGX.

The 24V has a Victron BMV connected to the CCGX.

There is a Quattro connected to the CCGX.

There are two Oceanvolt motors connected via CAN to the CCGX.


CCGX Color Control
screen-shot-2021-04-14-at-62402-pm.png (341.5 KiB)
2 |3000
Toggle Comment visibility. Current Visibility: Viewable by all users

Up to 8 attachments (including images) can be used with a maximum of 190.8 MiB each and 286.6 MiB total.

2 Answers
mvader (Victron Energy) avatar image
mvader (Victron Energy) answered ·

Hi tylera,

if you update to v2.65 or later, that VRM instance number will not change again.

Possibles causes for the reboot could be CPU overload or memory exhausted. That would be visible on the diagnostics page on VRM, as the last boot reason.

Also, there is a Dbus roundtrip time number on that same page. Should be below approx 200ms. Lower even. If above, systems is (too) busy, and might eventually rebooted.

The only other reason for reboots that I can think of is loose wire or too low voltage in the power supply.

Btw are you sure the CCGX rebooted? One other option is that just that usb port enabled/disabled itself due to EMI issues. See manual and beads recommended, in that case.

If thats it, our repairs department could ship you a bag if needed.


5 comments
2 |3000
Toggle Comment visibility. Current Visibility: Viewable by all users

Up to 8 attachments (including images) can be used with a maximum of 190.8 MiB each and 286.6 MiB total.

tylera avatar image tylera commented ·

Thanks for the response.


I repeatedly saw the blue start-up screen and took that as an indication of a reboot (followed by not having any data, followed by data collection and normal operation). I'm running v2.6, I'll talk to Oceanvolt as they are pesky about updates for some reason (I haven't learned yet).


Where is the diagnostics page on VRM? I don't see that.

0 Likes 0 ·
tylera avatar image tylera tylera commented ·

@mvader (Victron Energy) can you help me with where to find the "diagnostics page on VRM"?

0 Likes 0 ·
seb71 avatar image seb71 tylera commented · 
https://vrm.victronenergy.com/installation/(your_installation_ID)/diagnostics


0 Likes 0 ·
tylera avatar image tylera seb71 commented ·

Thanks @Seb71, I hadn't seen that page before!

0 Likes 0 ·
vu360 avatar image vu360 commented ·

I have the same problem with CCGX (v2.72) rebooting with an Oceanvolt motor installation, 4 MPPT controller, Valence BMS, and a BMV for house batteries. Looks like the system is bogged down and reboots as per diagnostic info (Watchdog reboot). My guess for some reason the dbus gets bogged down (D-Bus Round-trip time: 7832 ms), but every time I look at the CPU of the device, it's usually 72% or below. The device continues to run before rebooting cycle starts.

Is there a way to prevent the watchdog rebooting or see why the DBus is so slow?

0 Likes 0 ·
mvader (Victron Energy) avatar image
mvader (Victron Energy) answered ·

Hi, Continuous 72% CPU is quite high. And as I'm (a) not aware of any Oceanvolt-related CPU issues, and also not any other, and (b) you have five VE.Direct devices connected (which is the official maximum for the CCGX) and then even more (Valence, Oceanvolt motor) I'm quite sure the cause for this is simple: system is overloaded.


The only solution for this would be to disconnect some devices, or to replace the CCGX by a Cerbo GX (but I don't know how possible that is, since there might be special Oceanvolt code running, please contact them for that).


To mitigate, prevent having the main screen of the CCGX on the flow-overview, since thats rather CPU intensive. And disable any extras, such as Local and remote MQTT services as well as modbustcp.

2 |3000
Toggle Comment visibility. Current Visibility: Viewable by all users

Up to 8 attachments (including images) can be used with a maximum of 190.8 MiB each and 286.6 MiB total.