question

dj-dulux avatar image
dj-dulux asked

Slow [or maybe not] responding DVCC with Pylontech

We have done a fair few ESS Victron system and have recently moved to mobile power. One system is causing us some minor issues, it all appears to work as intended, but the DVCC does not seem to be respected and overshoots, IE, as battery gets close to 90%, the max charge current requested over DVCC reduces, but it doesn’t seem to be respected immediately and it takes much longer for the current to reduce than I think it should. We also see when the battery says 0A it ignores and keeps pushing charge in. This is leading to overvoltage alarms and causes the battery to not balance. Batteries are Pylontech 12100 2S5P (24V 500Ah), Multiplus 2 3000, Cerbo, MPPT 100/50 rev 2. Batteries connected to Cerbo via BMS-CAN, all battery parameters from batteries populating in VRM. All latest firmware. The strange thing is, VE.Bus state in all other installs show External Control, in this installation it shows Absorption / Bulk etc. It does show external control on the Dashboard for MPPT and Multi.

Any thoughts?

Thanks,

Simon…

Pylontech
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.

4 Answers
nickdb avatar image
nickdb answered ·

First thoughts, those 12V batteries aren't on the compatibility list, so that will make it interesting.

Mobile - ie off-grid? Places a much higher demand on the batteries, overshoots happen and it can't lean on the grid, which is why the pack must be sized right to deal with them. Power has to go somewhere, DVCC can't change that and the inverters have a ramp time which the batteries don't appear to be enjoying it.

Does this happen when grid connected?

How exactly have you wired the pack?

Can you see which cells are imbalanced on vrm and what the difference is?

Have you checked that one or more are at lower SOC and considered charging them all individually?

The alarms are BMS generated, so ultimately it is down the pack, smaller batteries like drop-ins can be less tolerant of how they are used.


3 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.

dj-dulux avatar image dj-dulux commented ·

Thanks for the answer, not on the compatability list, but they seem to pass all the info back to the BMS OK with no errors. This is a mobile setup, but its currently connected to the gid for most of the time. The DVCC just doesn't seem to be respected, even with no real load or change in load. Soon as I increase the DVCC limit voltage it tries to charge even if BMS says 0A.

Battery connections good, which battery goes out of balance is random, no pattern as far as I can tell. Have manualy balanced couple times now.

Couple images below, dashboard says external, by VE Bus state still says Absorption etc, all my other jobs just say External control....

Dash.jpgvebus state.jpg

Cheers,


Simon...

0 Likes 0 ·
dash.jpg (33.7 KiB)
vebus-state.jpg (28.8 KiB)
nickdb avatar image nickdb ♦♦ dj-dulux commented ·
With that many batteries, where and how you wire them matters. There is a good section in the wiring unlimited guide that covers series/parallel packs, which, if you haven't already, may be worth checking.

That it passes data to the GX, does not mean it will work optimally, as it is untested.

the 24V US/UC/UP/Force etc batteries are all tested. It isn't easy to balance across this sort of architecture.

Personally I would have kept it simple and used their 24V US2500 which is tested, but that may not fit into your application.

Unfortunately when you go off piste, the probability of hitting a tree increases ;)

It might be worth contacting Pylon support for their help, especially if the BMS is having problems balancing.

Good luck.


0 Likes 0 ·
Show more comments
kevgermany avatar image
kevgermany answered ·

For external control, overridden by remote must be ticked in VE.Config.

Are you using the correct Comms cable between batteries and Cerbo? Is it tested?

1 comment
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.

dj-dulux avatar image dj-dulux commented ·
Are you talking about the AC1 input current limit, its not actually ticked, but can not find any reference to say it needs to be? Will give it a try anyway.


Cheers,

Simon...


0 Likes 0 ·
dj-dulux avatar image
dj-dulux answered ·

I am seeing varible behavior on how VE.BUS STATE reports in Advanced between firmware versins so I am now wondering if this is a bad sign that all is well?

Been through a few installs on my VRM and found the following:

V2.92 MP (501) shows External Control with managed batteries

V3.10 MP (498) shows Bulk with managed batteries

V3.22 MP (508) shows Absorption/ Bulk etc with managed batteries

Is this expected behaivor, the top 2 installations have been running well (running BYD or Pylontech batteries), the bottom one is the one I think has issues. MPPT seem to respect DVCC, MP does not.

Thanks,

Simon....

1 comment
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.

nickdb avatar image nickdb ♦♦ commented ·

The state display changed between versions, as did pylontech dvcc behavior. Get the gx’s and multis onto a consistent, current production version. Changes are documented in the release notes.

2.9 is outdated.

0 Likes 0 ·
dj-dulux avatar image
dj-dulux answered ·

Did a couple little experiments yesterday, again the MPPT seems to respect the DVCC, when CCL at 0A the MPPT does nothing. But if I back off the CVL in the DVCC limit settings back toward 28V the MP starts to pass current to charge the batteries, ramped up to 6A or so, then reduced to 1-2A after several minutes.

The CCL remained at 0A in Parameters so why is the MP ignoring it?

Thanks,

Simon...

1 comment
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.

nickdb avatar image nickdb ♦♦ commented ·

It will initially fluctuate but it should settle down.

CCL=0 is not a good way to stop charge, there are good reasons it can't always be honoured.

The battery ideally needs to lower its CVL to control charge.

That 1A isn't going to break anything and it should settle eventually.

If the battery isn't sending an alarm then the BMS is not in a panic.

With a series/parallel combo pack it is unlikely to ever be 100% nor would the BMS be accurately controlling all the individual batteries.

1 Like 1 ·