battery charge level discrepancy between Victron Charge Controller and SmartShunt on one setup, the other works fine

[mrfrozenpotatohead]

Okay, have 2 setups at home both use the same Eco-Worthy 50Ah LiFePO4 bateries, both have Victron Charge controllers and SmartShunts. I am having problems with the battery box version for mobile usage, primarily ham radio. I've swapped the batteries between the systems and the result is the same on the "mobile" version.

Basically the CC seems to be happy and gives the green steady light and shows 13.49v, same as on the home system, However on the battery box,. the Shunt is always showing 12.85v which matches the 12.9v that the onboard volt meter shows. The battery box is from AllTopUSA, from the point of view of the Shunt, the battery ISN'T being charged up, but the CC seems to think it is. The settings on both CC''s and shunts are configured the same with the exception that max changing amps is 25 on home system and 15 on the mobile system.

Home system
- 405W panel
- Victron MPPT 100-30
- Victron 500A/500mV shunt

Mobile system
- Have tried multiple panels with same results - 40w and 100w Coleman (Canadian Tire) rigid panels, 200W AllPower folding panel and a Dokio 100w flexible panel.
- battery box is https://www.alltopusa.com/collectio...es/products/b097245wv6?variant=47278013186362
- Victron MPPT 75-15
- Victron IP65 500A/500mV shunt

You can see a photo below of the 4 Victron items in use

Before I pull the battery from the box and run another external charge, any suggestions on what I should be looking at? I had previoulsy done an external charge using the Noco Genuis 10 in Lithium mode, after noticing that my Impulse Electronics 116W DC lithium charger was no longer working with any of my lithium batteries, it would light up green indicating full charge but an inline power meter showed no voltage or current flowing at all. This was with a 8Ah, 18Ah and one of the 50Ah batteries noted above. Noco was the one higher powered charger I had at hand.

Once fully changed, I took the battery box out in the field and tested it running a fridge and an air compressor drawing the battery down to 93%. Fridge was connected via the cigarette lighter socket and the compressor was attached to the external direct connections to the battery which are un-fused to allow high current draw. And that's where it's stuck from the point of view of the Shunt. Went away for a couple sunny days expecting to come back and see state of charge back to 100% and 13.49V but no change whatsoever. This has been hooked up about a week now of sunny days cycling through the various panels

The CC is hooked up to one of the 2 external 50amp Anderson connectors while the Shunt is wired up inside the battery box.

The relevant Victron hardware and data are underlined in red.

 

[740GLE]

Smart network, create two of them, one SCC and one shunt per network

 

[mrfrozenpotatohead]

Hmm, the battery box is designed to leave the house and go into the back country, I am not sure I understand what you are suggesting, can you clarify further? Reading here

1. Introduction

www.victronenergy.com

I am not seeing how this helps the SCC and Shunt that are in/attached to the battery box helps keep them in sync like the permanent home setup.

 

[HRTKD]

Is it possible that the BMS on your mobile system is giving the solar charge controller the middle finger because there is a cell that has hit the high voltage disconnect?

Are the settings on the solar charge controller correct for the battery chemistry and capacity?

 

[mrfrozenpotatohead]

How would I test the per cell level or reset each?

Same config with both setups. I'll confirm SCC config in morning when the sun is back on ;-)

From memory, one of the Victron lithium presets which match the Eco-Worthy charging requirements.

 

[mikefitz]

The relevant Victron hardware and data are underlined in red.

The Victron charger in in float mode but the battery BMS has disabled the charge path. With disabled charge FETs, the battery will indicate a voltage into a voltmeter that's lower than the 'true' series cell volts, due to a volt drop across the disabled charge FETs.
I suspect the battery is reaching near fully charged and cell overvolts is causing BMS protection, the typical volt drop will be 0.6 volts, thus 12.85 + 0.6 = 13.45, a fully charged battery.
Applying a load of a few amps should allow BMS recovery.

The cell balance should inprove over time. To speed up the cell balance process, perhaps lower the charge volts , as a test, to down to 13.8 in 0.1 steps, to reduce the volts below the protection level. Once you find a charge volts that doesn't cause protection , say 13.8 volts, (protection will be indicated by a sudden charge current fall to zero). , set up this as absorbtion and float volts. Leave the battery charging under these conditions for a few days. Revert to the default lithium charge voltages, 14.2 absorbtion, 13.5 float, once cell balance has Improved.

 

[740GLE]

Hmm, the battery box is designed to leave the house and go into the back country, I am not sure I understand what you are suggesting, can you clarify further? Reading here

1. Introduction

www.victronenergy.com

I am not seeing how this helps the SCC and Shunt that are in/attached to the battery box helps keep them in sync like the permanent home setup.

The SCC is stationary, and the battery is mobile with shunt installed? You only connect the SCC to the battery when charging?

The smart network between SCC and shunt is only needed when charging and if they are within Bluetooth range. That higher voltage reading of the SCC will be replaced with shunt battery voltages, and force the SCC to continue charging until true battery voltage reaches charging parameters.

IMO on a 12v system, a smart network is key for getting that battery voltage/SOC up to 100%. Such little voltage drop between SCC to battery can mean a huge hit to SOC.

Can you post a picture of the cable connections between battery and the SCC?

I run a 100/15 charger, it is protected via automotive 20a ATC fuse and about 5 feet of 10awg. I typically see about 1-2v drop at 15a between terminals of the SCC and the battery terminals. The smart shunt tells the SCC to continue charging raising charging voltage, negating any voltage drop of the cabling.

 

[mrfrozenpotatohead]

How would I test the per cell level or reset each?

Same config with both setups. I'll confirm SCC config in morning when the sun is back on ;-)

From memory, one of the Victron lithium presets which match the Eco-Worthy charging requirements.

 

[mrfrozenpotatohead]

That's the home based SCC settings, only difference with battery box is the max charge current which is set to 15amps.

 

[mrfrozenpotatohead]

The SCC is stationary, and the battery is mobile with shunt installed? You only connect the SCC to the battery when charging?

The smart network between SCC and shunt is only needed when charging and if they are within Bluetooth range. That higher voltage reading of the SCC will be replaced with shunt battery voltages, and force the SCC to continue charging until true battery voltage reaches charging parameters.

IMO on a 12v system, a smart network is key for getting that battery voltage/SOC up to 100%. Such little voltage drop between SCC to battery can mean a huge hit to SOC.

Can you post a picture of the cable connections between battery and the SCC?

I run a 100/15 charger, it is protected via automotive 20a ATC fuse and about 5 feet of 10awg. I typically see about 1-2v drop at 15a between terminals of the SCC and the battery terminals. The smart shunt tells the SCC to continue charging raising charging voltage, negating any voltage drop of the cabling.

Yes and no.... The shunt is installed inside the battery box and I plan to install the charge controller in the battery box also so permanently usable, since I commonly take panels with me for field usage and the Allpower 200W folder will be my primary panel until I find a good price on their 600W folding unit and of course upgrade the SCC.

Right now it sits on top of the box with final leg from PV and into the box has 12awg, internally in the box, my recollection is it came with 8awg, though when I added cabling for the shunt I used 6awg which is what I have handy. Will have to open and make a drawing of cabling, will do that later today. Though you can see their internal wiring diagram, of a sort, here:

Portable Battery Box - Manual

Update Date: Dec/05/2023

www.alltopusa.com

 

[740GLE]

solution to all your problems

 

[mrfrozenpotatohead]

I'll watch the video and revert with any questions, what's surprising me is the synced charging "just worked" with my dedicated home setup, I didn't have to anything special which is why I was surprised when it didn't "just work" with the battery box setup.

Previous events I attended, eg

https://www.seapac.org/seminars/2023/SEAPAC_2023%209%20STEPS%20TO%20SOLAR%20SUCCESS%20-%20KB7RHI.pdf

showed it "just working" also.

 

[mrfrozenpotatohead]

I watched the video and that all makes sense when you have an environment with multiple SCC and panels as part of a system.

@740GLE - are you suggesting I do that? Assuming that's the case, my take is as follows, please confirm this is your recommended path

1) leave my current home system alone, that just works and is all in sync and don't have battery box as part of that network since it leaves the house pretty regularly.
2) for the mobile battery box, setup a new VE network, joining it's SCC and SmartShunt in that network and allow then to sync properly.
3) something different???

 

[sunshine_eggo]

This isn't a sync issue. The battery and MPPT are electrically disconnected. The shunt is reading the terminal voltage. The MPPT is reading its float voltage because it can't read a voltage at its terminals.

A VE.Smart network won't fix the issue, but in most cases, it performs better, and you should do it anyway.

This post is likely what's going on:

Post in thread 'battery charge level discrepancy between Victron Charge Controller and SmartShunt on one setup, the other works fine'

Sep 3, 2024

The relevant Victron hardware and data are underlined in red.

The Victron charger in in float mode but the battery BMS has disabled the charge path. With disabled charge FETs, the battery will indicate a voltage into a voltmeter that's lower than the 'true' series cell volts, due to a volt drop across the disabled charge FETs.
I suspect the battery is reaching near fully charged and cell overvolts is causing BMS protection, the typical volt drop will be 0.6 volts, thus 12.85 + 0.6 = 13.45, a fully charged battery.
Applying a load of a few amps should allow BMS recovery.

The cell...

• mikefitz

• 

When a LFP battery reads something like 12.6-12.9V, and you're reasonably confident it should be fully charged, it's likely reporting a false voltage while it's in over-volt protection mode. If you load the battery with just a few amps for a few minutes (maybe need to disable the MPPT, so it doesn't power the load), and the voltage JUMPS, then that's confirmation. At that point the shunt and MPPT will read very similar voltages.

The only "fix" here is to get the cells balanced. That's going to take time and cycles.

 

[mrfrozenpotatohead]

Okay, I'll start with direct charging the battery so I am 100% certain it's fully charged and Shunt is reading same voltage as Scc. Then put it back in battery box, hook up a load and then see if they stay in sync or not. Does that make sense?

 

[sunshine_eggo]

Okay, I'll start with direct charging the battery so I am 100% certain it's fully charged and Shunt is reading same voltage as Scc. Then put it back in battery box, hook up a load and then see if they stay in sync or not. Does that make sense?

Instead, turn off the MPPT and give it a small load for a few minutes. The voltage should jump and the MPPT and shunt will read the same voltage (within error margins).

 

[mrfrozenpotatohead]

Wait, turn off MPPT, so disconnecting is good enough, then put load on battery, check Shunt to detect that battery level is falling, then re-attach the MPPT? I can plug a load into the 2nd 50A Anderson connector while disconnecting the SCC from the other 50A Anderson connector.

Not sure I get the "voltage should jump" part, expecting it to decrease.

 

[sunshine_eggo]

Wait, turn off MPPT, so disconnecting is good enough, then put load on battery, check Shunt to detect that battery level is falling, then re-attach the MPPT? I can plug a load into the 2nd 50A Anderson connector while disconnecting the SCC from the other 50A Anderson connector.

On second thought, apply a load first.

Not sure I get the "voltage should jump" part, expecting it to decrease.

If the shunt is indicating 12.XV and the MPPT is indicating 13.5V as you showed above, once a load is applied for several minutes, the battery voltage should increase, and the shunt and MPPT will report about the same value.

 

[mikefitz]

I am 100% certain it's fully charged and Shunt is reading same voltage as Scc.

That may not happen if the BMS enters protection mode. It's very probable the cells are not balanced so at near full charge the BMS, in effect, disconnects the battery from the charger and the charge current falls to zero. Thus the charge controller may indicate a charge volts, say 13.5, but the indicated battery volts are dependent on a 'leakage ' path through the 'off' BMS.

It's confusing that the charger indicates a higher voltage than the shunt. But with the battery not accepting current, the charger enters float mode and sets its output to a constant float volts, 13.5. The shunt with its high impedance voltage measuring circuit, detects a voltage 'leaking' out of a 'turned off' battery.

When you apply a load the BMS recovers, ( as the cell volts fall below protection level), and you see the true battery volts.

 

[740GLE]

This isn't a sync issue. The battery and MPPT are electrically disconnected. The shunt is reading the terminal voltage. The MPPT is reading its float voltage because it can't read a voltage at its terminals.

A VE.Smart network won't fix the issue, but in most cases, it performs better, and you should do it anyway.

This post is likely what's going on:

Post in thread 'battery charge level discrepancy between Victron Charge Controller and SmartShunt on one setup, the other works fine'

Sep 3, 2024

The relevant Victron hardware and data are underlined in red.

The Victron charger in in float mode but the battery BMS has disabled the charge path. With disabled charge FETs, the battery will indicate a voltage into a voltmeter that's lower than the 'true' series cell volts, due to a volt drop across the disabled charge FETs.
I suspect the battery is reaching near fully charged and cell overvolts is causing BMS protection, the typical volt drop will be 0.6 volts, thus 12.85 + 0.6 = 13.45, a fully charged battery.
Applying a load of a few amps should allow BMS recovery.

The cell...

• mikefitz

• 

When a LFP battery reads something like 12.6-12.9V, and you're reasonably confident it should be fully charged, it's likely reporting a false voltage while it's in over-volt protection mode. If you load the battery with just a few amps for a few minutes (maybe need to disable the MPPT, so it doesn't power the load), and the voltage JUMPS, then that's confirmation. At that point the shunt and MPPT will read very similar voltages.

The only "fix" here is to get the cells balanced. That's going to take time and cycles.

A Smart network (with proper settings) will prevent the BMS from ever getting to overvotage protection in the first place.

It isn’t the end all be all for fixing the issue now but it solves so many headaches in the future.

As for SCC settings, I don’t like the stock victron LiFePo settings, it doesn’t allow for a long duration of holding battery voltage in the “balancing zone”. Which maybe the root issue the OP is seeing. I like absorbtion for 2hrs with rebulk.

 

[sunshine_eggo]

A Smart network (with proper settings) will prevent the BMS from ever getting to overvotage protection in the first place.

I don't agree with this at all. Overvoltage protection is hit because of excessive charge voltage or cell imbalance.

The battery voltage is always the battery voltage. Just because the MPPT is reading a higher voltage due to current and wiring losses, it doesn't drive the battery to a higher voltage. It shortens bulk phase and extends absorption phase causing the charge to full to take a bit longer, but it will never push the battery to a higher voltage outside of a VE.Smart network than it will when in one.

It isn’t the end all be all for fixing the issue now but it solves so many headaches in the future.

Totally agree here.

As for SCC settings, I don’t like the stock victron LiFePo settings, it doesn’t allow for a long duration of holding battery voltage in the “balancing zone”. Which maybe the root issue the OP is seeing. I like absorbtion for 2hrs with rebulk.

Yeah... they can use a little tweaking.

 

[740GLE]

We can both agree that proper settings and battery management will always avoid a BMS from hitting over voltage protection.

Part of good SCC settings is to allow proper balancing.

If you are hitting over voltage protection some part of the equation is wrong.

 

[sunshine_eggo]

We can both agree that proper settings and battery management will always avoid a BMS from hitting over voltage protection.

Except when they're new batteries. Too many arrive unable to charge to their stated charge voltage. You can configure everything exactly as you're supposed to, but you'll hit OVP.

Additionally... batteries in storage... they'll often drift out of balance.

Part of good SCC settings is to allow proper balancing.

If you are hitting over voltage protection some part of the equation is wrong.

In the vast majority of cases I've observed on this board, it's simply the battery cells being imbalanced. The issue is the battery. One must alter optimal settings to address a suboptimal battery condition until the imbalance is corrected and optimal settings can be restored.

A VE.Smart network has very little to contribute to prevention or recovery, but it is a more efficient setup, and I love it very hard.

 

[HRTKD]

A VE.Smart network has very little to contribute to prevention or recovery, but it is a more efficient setup, and I love it very hard.

One way that I could see the MPPT network helping is if the two solar charge controllers combined were exceeding the maximum amps that the battery bank can handle. Or does that require the Cerbo GX in the middle to manage the Max Amps?

 

[sunshine_eggo]

One way that I could see the MPPT network helping is if the two solar charge controllers combined were exceeding the maximum amps that the battery bank can handle. Or does that require the Cerbo GX in the middle to manage the Max Amps?

Cerbo DVCC