question

Trevor Bird avatar image
Trevor Bird asked

Quattro shutting down - Lithium Batteries

Hello,

We have installed many Quattros but usually with batteries we are familar with and have had no issues. We recently installed a 24/5000/120 Quattro on a boat with very recently installed by owner 10 x 130Ah Giant Brand 12 volt lithiums in series parallel to provide 5 x 24 volt series strings giving 650 Ah at 24 volts.

We normally set the absorption voltage at 29.2 volts with float at 27.2 volt for the Batteries we sell. We set lithium mode and configure Smartshunt accordingly.

On this installation, we checked battery recommendation. Sticker on battery says "Bulk charge 14.4 - 14.7 volts." "Float charge 13.6 - 13.8 volts." On that basis we thought 2 x 14.7 volt = 29.4 volts so our setting of 29.2 volts should be ok.

Soon after installation with these parameters the Quattro would shut itself off and then spontaneously restart. Occasionally the Quattro would provide a "DC Ripple" warning.

A DC Ripple warning denotes 100 hz AC on the DC buss resulting from slightly high resistance supply cables to the Quattro. This would surely normally be in sympathy with high DC load such when the Quattro is inverting and supplying a heavy AC load. In our case this was not so. The "DC Ripple" warning appeared with no easily observable correlation with High DC current. Exhaustive checks were made to ensure that the DC cabling to the Quattro was VERY low resistance in keeping with the conventional wisdom of resolving DC Ripple warnings..

We enlisted help from our Australian Victron supplier who logged onto the VRM and very helpfully noted an unusually high variation on the Battery and Quattro DC voltages. With very little current variation quite radical voltage variations were apparent. It was about 1.5 volts of variation versus a "normal" at rest system that may exhibit say 100mV of variation.

We tested the Quattro on an AGM engine start bank on the boat and proved the Quattro was stable when using "normal" batteries.

We then investigated the "Daly 100 amp BMS" that was individually installed within each battery. It has an overvoltage sense at 3.75volt per cell. Four cell per battery gives us an overall battery voltage of 15 volts if they are all balanced before overvoltage sensing starts disconnecting cells or the battery.

12 volt batteries in parallel may not be a problem but two in series may allow one battery or the other on many strings to be slightly out of balance enough to cause havoc. No load balancing between batteries is employed.

We reduced the Absoption voltage to 27.6 volts and the Float voltage to 27.0 volts. We reduced the bank from 10 batteries to 6 just for testing and reduced charge current to 40 amps instead of the original 120 amps.

It has now run for two days without incident. Next week we will bring the other 4 batteries back into the bank and set the charge current higher. Hopefully the issue is resolved.

As I write this note it all seems very logical and easy to resolve but I confess it has taken some effort to get here.

I make the following observations that may or may not be correct. I would be appreciative of any comments to support or debunk the following theory.

These batteries in series/parallel reacted very strangely and caused a DC Ripple warning that was clearly not really DC Ripple. It also caused the Quattro to simply stop working (Turned off) and then restart and on two occasions the Quattro entered a "Fault" state and did not spontaneously turn back on. It looked very much like we may have had a Quattro fault as the previous Magnasine inverter charger did not have this problem.

It seems to me the following may help others who may experience similar.

1. Battery BMS systems can cause variations in terminal voltage that the Quattro reports as "DC Ripple". It may be a variation in terminal voltage but it is not "DC Ripple" if DC Ripple is defined as an AC waveform in sympathy with the AC output of the inverter. The DC Ripple warning in this case does not correlate with high DC current. In this case the DC Ripple warning has nothing to do with the integrity of the battery cabling.

2. The BMS induced voltage variation can cause the Quattro to simply stop with no error warning. It restarts and the cycle begins again. It would be wonderful to have a Error log saying why the Quattro turned off.

3. Occasionally the BMS induced problem causes the Quattro to shut down with a fault condition. Once again it would be nice to have an error warning relating to the fault condition.

4. Lowering the Absorption voltage allows the BMS to remain dormant and not cause terminal voltage variations. I hope this has resolved the issue.

5. I imagine problem is increased when creating 24 volts from 2 x 12 volt batteries in series. We don't typically do that but the client had already installed the bank.

6. It may be useful for Victron software to bandpass limit the DC Ripple parameter so real "DC ripple" can be separated from "DC variation". DC variation may become more common with the greater use of cheap BMS controlled Lithium batteries.

7. This problem would be EXTREMELY difficult to diagnose without VRM.

Maybe this may help someone. Any comments most welcome.

Trevor










MultiPlus Quattro Inverter Charger
1 comment
2 |3000

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 ·

Interesting. I guess this is why, as battery tech gets more complex and semi-intelligent, choosing compatible/proven/tested combinations becomes more relevant.

Those that have the biggest problems are usually those that go off piste, or consciously choose to ignore Victron recommendations.

There seems to be a fine line between what works and what doesn't.

It would be helpful if error reporting was more relevant and accessible.

0 Likes 0 ·
2 Answers
Trevor Bird avatar image
Trevor Bird answered ·

Just an update to the issue earlier raised in this post. The system is stable and charge current has now been increased to 120 amp and all batteries have been connected back into the house bank.

The reduction of the absorption voltage from 29.2 volts to 27.6 volts has presumably stopped the internal Battery BMS units activating the cell overcharge protection. As a matter of record, 29.2 Volts is perfectly within the battery suppliers specified voltage level. I have included a relevant VRM screen shot to indicate that prior to the reduction, the terminal voltage was bouncing all around the place. After the reduction we have established relative calm regarding battery terminal voltage.

We can also see that while the BMS units were creating chaos, the DC Ripple warning would occasionally appear. While this manifests as a DC Ripple warning it is actually DC voltage variation, a subtle but important difference.

While this maybe basic for seasoned lithium installers, I was interested to see that 1.6 volts of difference in absorption voltage setting is the difference between calm and absolute chaos and mayhem. It caused the Quattro to cycle continuously from "off" to "bulk" to "absorption" and "off" again. Occasionally the Quattro would lock itself out in "Fault" mode. All of this without any fault notification except for an occasional wrongly attributed DC Ripple warning.

With more cheap lithum batteries being installed maybe this may become more common. It could perhaps be something Victron may be able to identify as an excessive "DC voltage variation" rather than a "DC Ripple" warning. It that warning were implememted it may help with troubleshooting if that type of warning is appearing a reduction in absorption voltage could be tried to redcue BMS actvity when battereis are fully charged.

screen-shot-2022-04-11-at-121258-pm.png


2 |3000

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

klim8skeptic avatar image
klim8skeptic answered ·

Aussie solar + giantz batteries = doomed to failure.

Op, give your client their money back, after you uninstall your components.

9 comments
2 |3000

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

Trevor Bird avatar image Trevor Bird commented ·
We were requested to supply a Quattro. We supplied it. Should we remove it do you think?
0 Likes 0 ·
nickdb avatar image nickdb ♦♦ Trevor Bird commented ·

Inevitably your product will be blamed for the problems resulting from the DIY work, just make sure your T's and C's protect you, or you may end up regretting it and spending more time and effort than you anticipated.


0 Likes 0 ·
Show more comments