Dyness battery "Internal Failure" not reported by VRM

Question

question

glennsa asked Jun 2, '23

Dyness battery "Internal Failure" not reported by VRM

My Victron Remote Console had a flashing triangle at the top. No warning or alert notification was received about a Dyness battery “Internal failure” that was eventually found in the Victron Remote Console notification screen. I believe this should have been reported by VRM.

After the warning was acknowledged (tick at the bottom of the Remote Console notification screen), the warning has been removed, but there is nothing in the VRM Alarm Log (other warnings like Overloads and High DC Ripple are listed).

Warnings and alarms are enabled, and do routinely display on my iPhone for other issues, however:

• No notification was received on VRM or via email.

• The Dyness Alarm screen in Remote Console has an OK status for all fields, and does not show any warning or fault status.

• The Dyness History screen in Remote Console is empty and does not show any warning or fault.

• There are no VRM alarm or event logs that correlate with the notification.

This occurred on an EasySolar-II 48/5000/70-50 MPPT 250/100 GX (PN: PMP482507010) and Dyness 9.6kWh Powerbox F-10.0 with the following firmware:

• EasySolar-II GX: Firmware v2.94

• MultiPlus-II 48/5000/70-48: Firmware v502

• SmartSolar MPPT VE.Can 250/100 rev2: Firmware v3.13

• Dyness 9.6kWh Powerbox F-10.0: Firmware reported by VRM Device List: v136.3

Why is this warning event not pushed by VRM or recorded in the event log? It could be an important early warning of a dangerous situation.

warnings

dyness-internal-failure-warning.png (30.0 KiB)

2 |3000

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

Modifications Space

Node Red Space

Answer 1 · 2023-06-02T09:12:45Z

nickdb answered · Jun 02, 2023 at 09:12 AM

Does your default alarm rules have warnings and alarms selected or only alarms?

This is only a warning and on a default rule it will only alert for alarms.

I am not sure that this specific BMS parameter is part of the default set.

You will need to add a manual alarm rule for that BMS and select "internal failure" and warning level, then it will notify you.

screenshot-2023-06-02-at-111311.png (27.9 KiB)

2 |3000

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

Answer 2 · 2023-06-02T09:15:08Z

glennsa answered · Jun 02, 2023 at 09:15 AM

Warnings and alarms is selected.
I do receive other warnings, but not this battery warning.

6

2 |3000

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

nickdb ♦♦ commented · Jun 02, 2023 at 09:23 AM

In the meanwhile add your own rule for that BMS, it is easy to do. According to docs (which are a little generic), This is a monitored event for CAN batteries. Will see if I can get a better answer.

0 ·

glennsa nickdb ♦♦ commented · Jun 02, 2023 at 10:11 AM

If you can suggest which parameter(s) to monitor, I can create a custom rule. I'm keen to watch as many parameters as possible that won't trigger an excessive number of warnings, until Dyness can confirm what this event means.

It has also been mentioned that this "internal error" may be a communication failure. Can you comment on that?

0 ·

nickdb ♦♦ glennsa commented · Jun 02, 2023 at 10:53 AM

When you try add the rule it will list all the parameters available to you. Not all will apply to the battery but many will, like internal alarm. Most other events like voltage etc should be covered already.

Dyness is a pylontech effectively, internal errors aren't uncommon, just do a search here and you will find a good thread.

It has nothing to do with CAN comms.

0 ·

glennsa nickdb ♦♦ commented · Jun 02, 2023 at 11:01 AM

Thanks for the help @nickdb - a number of rules have been set up and are being monitored. Silence thus far...

0 ·

nickdb ♦♦ glennsa commented · Jun 02, 2023 at 11:09 AM

https://community.victronenergy.com/questions/182659/pylontech-us3000c-us5000-internal-failure-problem.html

Might be relevant for you.

0 ·

glennsa nickdb ♦♦ commented · Jun 02, 2023 at 11:35 AM

Interesting thanks. Dyness does not support self-updating firmware so are going to connect remotely and update it and check the BMS/battery data. I'll send the above link too and ask if it has any relevance in this case.

0 ·

Answer 3 · 2023-06-04T10:55:07Z

glennsa answered · Jun 04, 2023 at 10:55 AM

Feedback: DYNESS has been in direct contact about the "Internal Failure" warning in the Victron Remote Console notification screen for a DYNESS Powerbox F-10.0 9.6kWh battery.

The DYNESS tech connected remotely to VRM, and confirmed that the Victron firmware settings for the DYNESS battery are correct, the battery communication is normal and a DYNESS firmware update is unnecessary.

DYNESS informed that recent comm failure and charge/discharge reports that they've been receiving have disappeared after downgrading the Victron firmware to v2.90, and that the warning was a false alarm from the inverter, perhaps due to a protocol change in later Victron firmware. DYNESS will discuss this with Victron.

For now, the Victron firmware has been left at v3.00 (not downgraded to v2.90), and multiple battery monitoring rules have been created in VRM -> Settings -> Alarm Rules, which will be monitored for any further warnings in (1) Alarm Logs (2) Event Logs (3) Remote Console -> Notifications.

DYNESS provided a detailed installation manual named "Dyness Battery and Victron inverter setup manual 20211222.pdf".

Dyness Battery and Victron inverter setup manual 20211222.pdf

dyness-battery-and-victron-inverter-setup-manual-2.pdf (3.7 MiB)

4

2 |3000

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

nickdb ♦♦ commented · Jun 04, 2023 at 11:07 AM

I think they are giving a bit of a bum steer. The alarms aren't generated by the inverter, it is the inverter reporting what it is sent via a CAN broadcast.

BMS comms aren't bi-directional, the BMS sends compliant broadcasts that conform to the victron standard, these are then used for DVCC, alerting etc.

In short the inverter won't do anything unless it is receiving bad data from the BMS.

They are an unsupported brand, so it isn't unsurprising that issues are encountered as technology moves forward.

0 ·

glennsa nickdb ♦♦ commented · Jun 04, 2023 at 01:09 PM

I understand that what Dyness says is the battery is sending the same CAN messages (from the CAN Bus node) as it always has, but a change in the Victron firmware, perhaps in how the CAN frames are interpreted (i.e. logic) has resulted in a warning being displayed in the console. Alternatively perhaps, Dyness is not sending these in a format that Victron wants, which may be what Dyness has to change to appear on the battery compatibility list.

I didn't think that Dyness was trying to deflect a battery error. Anyway, I speculate, so will monitor if anything is communicated by Dyness and/or Victron on the topic.

0 ·

nickdb ♦♦ glennsa commented · Jun 04, 2023 at 01:21 PM

It's a common framework that all batteries use (supported or not) and there really hasn't been anything changed in modern times (the release notes are publicly available).

This sounds like an incompatibility/bug in their BMS code.

They may be sending the wrong info or the BMS is incorrectly interpreting an internal event.

If anything, v3 (and 2.94) has a fix for a rare DVCC (CAN-related) event from managed batteries, so you would really want to be current.

I can't understand why they would advise a downgrade.

But anyhoo. What is always common amongst all battery vendors, they always blame the inverter ;)

1 ·

glennsa nickdb ♦♦ commented · Jun 04, 2023 at 03:38 PM

Dyness did not recommend a downgrade, but reported that the internal failure warnings disappeared after downgrading the Victron firmware to v2.90 - they meant in their testing. Hence the Victron firmware was left as v3.00 and additional custom monitoring rules created - the latest firmware is normally a good thing and invariably brings improvements, although not always without unforeseen consequences.

An installer nearby reports the same issue with Dyness batteries, and that a firmware downgrade fixed/stopped this.

Strange that this and other Dyness batteries have run for years without an issue, and then a battery internal failure warning appears after the Victron firmware is upgraded to v2.94, the same one with the fix for the rare CAN-related DVCC issue. Clearly, they are not all failing simultaneously.

If the Dyness BMS had always been sending internal failure messages that weren't being interpreted by the Victron firmware until post-v2.94, then by now there would likely also be a pattern of battery failures.

It is feasible that a change in the Victron firmware could be causing the internal failure warning to manifest, rather than multiple Dyness batteries, at multiple installations, all simultaneously failed or started incorrectly interpreting an internal event and sending the wrong info after the v2.94 update.

It appears there are two common denominators - Dyness and the firmware update. If internal failure warnings are not manifesting in other brands of batteries, then there's something peculiar to the communication between this combination that must be identified. If they are, then the probability leans towards it being the firmware update alone. Time will tell.

0 ·

Answer 4 · 2023-06-05T21:17:35Z

mvader (Victron Energy) answered · Jun 05, 2023 at 09:17 PM

Hi @GlennSA and @nickdb .

We looked into this, and noticed that a reboot of the GX device happened at 11:28 UTC (13:28 SAST), which is the same time as the alarm.

The way how the GX works internally makes for there existing a short window of a few seconds in which its possible for an alarm or warning to show in the gui but not be sent to VRM. This can only happen if the alarm is raised and them cleared super quickly.

Similarly, if an alarm is only present for one or two seconds, and that happens earlier i the boot sequence, it won’t be visible or logged anywhere. That just how it is.

Another Internal Failure alarm is logged on VRM for the 2nd of June, so the mechanism clearly works.

Looking through the previous alarms on VRM, this alarm does sometimes clear after a single second, which fits in with the theory of why that one never made its way into VRM.

See for example 2022-12-06 06:40:53.

Hope that explains it. While looking at it we did see some ambiguity with the name for that same alarm being both Internal error and Internal alarm, depending on where you look. We’ll consolidate that to one and the same: Internal alarm.

Other than that I don’t think there is anything to do here.

All the best, Matthijs

2 |3000

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

Answer 5 · 2023-06-06T06:19:23Z

glennsa answered · Jun 06, 2023 at 06:19 AM

Thanks Matthijs and Nickdb for helping work through why the battery internal failure alarm is not pushed by VRM. It's good to learn more about the GX, but I'm still curious why the alarm was triggered by the Dyness in the first place, if that's where it originated.

I'm also curious how Victron figured out which inverter's logs to review, since I'd not posted a VRM ID or any serial numbers. I'm quite happy with the remote analysis, but curious... Unless it came from the Dyness technician who I had added as an admin on VRM temporarily, and who said he'd take up the matter with Victron? Nonetheless, thanks for the feedback Matthijs.

Any idea when the two alarms will be consolidated in a firmware update?

Thanks

1

2 |3000

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

mvader (Victron Energy) ♦♦ commented · Jun 06, 2023 at 08:10 AM

Hi Glenn,

Searching for your name and Dyness resulted in only one system.

The change wrt the alarms will be no firmware update, its a rename on VRM, changing Error into failure.

I’m sure it originated at the Dyness; and the same thing happened before. If you search your alarm log you’ll find it.

Sorry to be a bit Dutch perhaps, but here it ends for me.

To learn further, speak with Dyness. And if nobody there believes that its really the Dyness sending an Internal failure over the canbus, then someone will need to install a canbus datalogger, wait for it to happen again, and prove to me that indeed its not in the those canbus logs and somehow the Victron made it up.

0 ·

question