AOlithium 48v and Victron Multiplus II frustration!

[ctid]

I have a new setup, full victron including Multiplus 5000, cerbo gx, smart shunt and smart solar mppt, lynx distributor, and currently my system is happily funneling grid through the Multiplus II 5000 system while I wait for the solar install.

I have also purchased some AOlithium 48v and connected them to the lynx distributor. Where I am struggling is getting the Multiplus to charge them. I have manually added the SOC into the shunt, and I have a canbus lead from the battery to the cerbo, and the Multiplus is correctly identifying them, it sees them as Pylons.

What is weird, even though I have cycled through either the shunt or the battery BMS to be the battery monitor, the Multiplus is reporting an incorrect SOC and it looks like it is drip charging the batteries as it sees them at 95% instead of the real 22% SOC. I have tried the Multiplus in full mode and also just charging mode. In inverter and charging mode, it shuts down with a low battery alarm.

I have attached some screen shots (from the nice new GUI!).

Does anybody have any idea what I am doing wrong?

 

[Detritalgeo]

Hmmm, I am not fluent in Victron but perhaps there is a way to configure the AOlithium master to Victron protcol? He manual suggests that all DIPs down means Victron on CAN bus or Pylon on RS485...

 

[ctid]

I've been through the manual and I have done all the different variations. Today I changed the network cable to see if it is that, and still no change.
One update, the PV panels were finished today and it is producing some power, but again nothing going into the batteries.

 

[Detritalgeo]

Just to be sure, the BMS needs to be power-cycled for a new DIP setting to take effect.

By chance are these V2 AOlithium packs with blue DIP switches? or V3 BMS's with Red DIP switches?

 

[Adam De Lay]

Can you share a picture of how you have things wired into the distributor as well as the cerbo?
A couple questions come to mind:
- Is your CAN cable from your battery plugged into the CAN port or the BMS CAN port?
- Do you have DVCC enabled?

@sunshine_eggo any other ideas?

 

[sunshine_eggo]

Pylon is typically auto-detected and forces DVCC enabled.

Your battery monitor/charger settings for the MP are interfering with the systemwide SoC. This is something of a bug in the Victron ecosystem IMHO.

Since your MP charging voltage is controlled by the BMS, it's setting is essentially irrelevant. I recommend the following:

• Set battery monitor to BMS.
• Get CVL value from the BMS.
• Set MP charger absorption voltage to 0.4V above CVL value. It should never trigger the 95% when bulk finished criterion on the MP battery monitor.

Another confounding factor - Pylon DVCC forces OFF SVS, so BMS measured voltage can't be passed to the system. Thus the MP is using its own voltage measurement, which may differ significantly from the BMS. Recommend you connect the MP V-sense terminals to the battery for a direct voltage measurement.

The 95% should eventually revert from 95% to the BMS fed value. Provided you stay below the MP charger absorption voltage, it will never influence it again.

In an effort to explain it better, and I recommend the OP do this, I have a custom widget pulling SoCs from multiple sources. The tricky one is the "VE.Bus hidden SoC" - this is the MP battery monitor "when bulk finished" SoC value. If you plot it over time, I suspect you'll often see it cycling between 95% and actual.



For my system:

258 is the BMV-702 SoC
512 is the BMS SoC
"System - Battery SoC" is the SoC fed to the system from either 258 or 512 based on my battery monitor setting.
"hidden SoC" is the inverter battery monitor generated value.

 

[ctid]

I've got the v2 with the blue switches.

Yes, I always power cycle every time they were changed.

 

[sunshine_eggo]

56.78V at 0A charge, is absolutely, positively 100% SoC.

Are you using ESS?

 

[740GLE]

Pylon is typically auto-detected and forces DVCC enabled.

Your battery monitor/charger settings for the MP are interfering with the systemwide SoC. This is something of a bug in the Victron ecosystem IMHO.

Since your MP charging voltage is controlled by the BMS, it's setting is essentially irrelevant. I recommend the following:

• Set battery monitor to BMS.
• Get CVL value from the BMS.
• Set MP charger absorption voltage to 0.4V above CVL value. It should never trigger the 95% when bulk finished criterion on the MP battery monitor.

Another confounding factor - Pylon DVCC forces OFF SVS, so BMS measured voltage can't be passed to the system. Thus the MP is using its own voltage measurement, which may differ significantly from the BMS. Recommend you connect the MP V-sense terminals to the battery for a direct voltage measurement.

The 95% should eventually revert from 95% to the BMS fed value. Provided you stay below the MP charger absorption voltage, it will never influence it again.

In an effort to explain it better, and I recommend the OP do this, I have a custom widget pulling SoCs from multiple sources. The tricky one is the "VE.Bus hidden SoC" - this is the MP battery monitor "when bulk finished" SoC value. If you plot it over time, I suspect you'll often see it cycling between 95% and actual.

View attachment 270867

For my system:

258 is the BMV-702 SoC
512 is the BMS SoC
"System - Battery SoC" is the SoC fed to the system from either 258 or 512 based on my battery monitor setting.
"hidden SoC" is the inverter battery monitor generated value.

Seems realy complicated vs running no BMS comms and just tracking SOC via the smart shunt he already has.

I still don’t see all the need for full blown closed loop with BMS when you have a smart shunt.

 

[sunshine_eggo]

Seems realy complicated vs running no BMS comms and just tracking SOC via the smart shunt he already has.

It's really not. That's just how I explain things.

It's literally one setting, and it's usually the user inducing it.

Oddly enough, for this user and the claimed battery voltage, the 95% is WAY more accurate than 22%.

I still don’t see all the need for full blown closed loop with BMS when you have a smart shunt.

I mostly disagree. Having the BMS say:

If voltage too high, current too high, or temp too low/high, then set charge current to 0A.
If voltage too high, current too high, or temp too low/high, then disable all chargers (specific to my Batrium, not a thing with Pylon)
If voltage too low, current too high, or temp too low/high, then turn off inverter.

And have the system respond accordingly is more desirable than having a battery cut in and out of the circuit.

 

[ctid]

Many thanks for the help guys, I really appreciate you taking the time.

I've attached some pictures of my setup, hopefully my hardware installation is ok and it is the software that needs tweaking. I'm going to take a look at your suggested settings sunshine_eggo and I will feedback.

Before I do, to answer a couple of queries. Pylontech batteries has been auto detected and DVCC is forced on EDIT It was forced on, just looked now and it is on but can to turned off. The battery is connected into the can bus RJ45. I've tried the BMS port as well. I have a type A and a type B cable.

22% is definitely the SoC, the batteries have a display and this is the reading. They are new and were only delivered just before Christmas. They have never had a charge or any load put on them. I initially had all my batteries connected, but when I ran into this issue, I kept it simple with just one battery. To check if it is an issue with a particular battery, I have tried all 6, none of them will charge.

I'm not using an ESS, no assistants have been installed.

As you can see from the pictures, the MS are in absorption mode, but I think because the MS see 95% SoC, they are trickle charging. The battery has been connected for nearly 5 days now.

No amps are being read from the battery BTW, I don't know if this is an issue?

 

[ctid]

ok, I've been busy @sunshine_eggo but still no success unfortunately.

I have attached some screen shots of the settings. I have followed your advice and...

1, BMS is the battery monitor. It was before so this is unchanged
2. Checked the CVL from the BMS, this is 57v.
3. Changed the absorption amount to 57.4v
4. I have connected the multiplus voltage sense to the battery.

no noticeable change yet and battery still not charging.

 

[Adam De Lay]

Can you try turning off “bms control” at the bottom of the DVCC page?

 

[sunshine_eggo]

New batteries almost always have incorrect SoC as they take an initial guess based on voltage.

What I find interesting is that you initially reported a voltage of 56.78; however, the battery display says 52.5V, which is consistent with the lower reported SoC.

Batteries MUST be on the BMS-CAN port UNLESS you have modified the VE.CAN port to 500kbps.

The evidence thus far says:

1) there is an open circuit between the charging sources and the battery.
2) the BMS is in charge protection.
3) the BMS is commanding a 0A charge current.

Do you have any assistants installed or virtual switch active?

Have you tried disconnecting the data cable and rebooting the cerbo (open loop)? The MPPT will need to be reset in VictronConnect to eliminate BMS control.

 

[740GLE]

New batteries almost always have incorrect SoC as they take an initial guess based on voltage.

What I find interesting is that you initially reported a voltage of 56.78; however, the battery display says 52.5V, which is consistent with the lower reported SoC.

Batteries MUST be on the BMS-CAN port UNLESS you have modified the VE.CAN port to 500kbps.

The evidence thus far says:

1) there is an open circuit between the charging sources and the battery.
2) the BMS is in charge protection.
3) the BMS is commanding a 0A charge current.

Do you have any assistants installed or virtual switch active?

Have you tried disconnecting the data cable and rebooting the cerbo (open loop)? The MPPT will need to be reset in VictronConnect to eliminate BMS control.

So the solution is running open loop

 

[sunshine_eggo]

So the solution is running open loop

Nope. Need to establish that the chargers will work when not under BMS control. Diagnostics are particularly challenging when the system is on another continent.

56.68V vs. 52.5V indicates a major issue is present, and since closed loop with Pylontech CAN'T pass voltage to the system, this is a true ~4V deviation in measurements. The most common cause for this is isolation of the battery from the system where the system is measuring charger voltage not battery voltage.

 

[740GLE]

Eggo is just messing with you, initial set up is always a bear.

The cerbo is reporting battery temp of 4c.

What’s the low temp charging protection setting? It’d alarm on low temp protection right?

 

[sunshine_eggo]

Eggo is just messing with you initial set up is always a bear.

The cerbo is reporting battery temp of 4c.

What’s the low temp charging protection setting?

Damn it... I missed that.

I don't know what the default is, but 5°C is very common, so that would do it, though I would expect an alarm on the battery.

@ctid

go into VRM advanced tab. Enable battery monitor "BMS charge and discharge limits"

Looks like this:



You can see that my battery was in charge protection mode this morning where CCL was 0.0A.

If yours reports 0A, that's the likely cause, and going open loop wouldn't work either.

 

[Detritalgeo]

Looks like the BMS charge cut temp is -5C and the release is 5C.

Is it possible the packs were below freezing at some time and have not gotten back to the release temp yet?.. Have you tried bringing the pack into the house for a day to get fully up to temp then see if it will charge?

 

[sunshine_eggo]

Looks like the BMS charge cut temp is -5C and the release is 5C.

Is it possible the packs were below freezing at some time and have not gotten back to the release temp yet?.. Have you tried bringing the pack into the house for a day to get fully up to temp then see if it will charge?

Only issue is I would expect an alarm.

CCL value in VRM will establish if it's in protection mode.

 

[ctid]

I was desperately hoping it was the temperature. So I put a heat pad under the battery and wrapped it up and it is now at 7C and unfortunately no change.

I rebooted the cerbo with the data lead out, reset the MPPT BMS too and the MP that was in bulk charge, turned all charging off and sat there on pass thru (I've got it set to charger only while I'm trying to solve this). It didn't feed anything to the battery.

The VE can was on 500k but I have changed the port now to BUS can to be safe.

I am able to take the BMS out of DVCC, again, no change when I did this.

I added the widget on the VRM and the screenshot is attached. I have also taken some photos of my bus bar setup just in case I have made a stupid mistake there.

 

[ctid]

OH MY ***. I may well have stumbled on the problem!

The current bus bar is made up of 3 lynx power in distributors. Between number 2 and 3, I have a smart shunt, which is installed the right way round BTW. To connect the positive terminals I have a nice thick piece of copper bar.

After the shunt I have two +ve and two -ve battery leads as eventually I will have 2 banks of 3 batteries connected.

In a desperate, clutching at straws exercise, and while I had the lids of for the photo, I moved the spare set of battery leads into the middle bus bar and powered everything up, and bingo! Battery is now charging.

I connected original leads to another battery, with the data lead, and nothing. So it is the positioning that was stopping the charge.

Would any of you gurus know why this is happening?

 

[ctid]

I'm going to remove the copper and shunt to test the setup. I am now starting to doubt the shunt, I bought it from Amazon warehouse, A1 condition but I cannot see what else it could be.

Thank you to everyone who gave me help, if nothing else, I have learnt a heck of lot from your guidance.

I will update when I have more information!

 

[sunshine_eggo]

The evidence thus far says:

1) there is an open circuit between the charging sources and the battery.

@ctid

So this appears to be the case.

What is the purpose of the smart shunt. Is it a DC energy meter for DC loads on the 3rd power-in?

Based on what you've posted and the limited labeling, I'm very concerned your system is not wired correctly. Could you please post a picture of your entire bus bar set and label what's connected where?

 

[ctid]

Yes, you were spot on with the open circuit suggestion! Thank you.

I've labelled the whole bus bar in the attached.

The system is 3 phase, so all the leads from the MP's are the same length. Currently, I only have 1 MPPT connected as I don't have all my panels yet.

I have a total of 6 batteries, 3 of them are currently connected. Once everything is settled down and running smoothly, I was planning on adding the other 3.

But you have me worried now @sunshine_eggo! I hope I haven't made a mistake.