Barely 30mins. That's how long my 24V-2kW 200Ah system lasts with a 300W(ish) load before shutting down.

System spec:

• RCT Axpert V-PF1 24V 2kW inverter (bought new, about 6 months old) (Firmware main v9.52, secondary v3.39)
• GEB FT12-200 12V 200Ah Gel batteries (bought new, about 6 months old - the only reason why I have these is because I got them very very cheap and a friend has had good experiences with them)
• Victron battery balancer
• Victron battery cut-off switch, fuse holder and 250A fuse
• 50mm2 cables

Settings as per Watchpower (confirmed to be the same on inverter as well):

• Type User
• Bulk charge 29.2V (battery spec 14.6 - 15.0)
• Float charge 27.4V (battery spec 13.6 - 13.7)
• Back to discharge 29V
• Back to charge 24V
• Battery cut-off 23.9V
• Max charging current 20A
• No equalization

Other observations:

• No loadshedding since last week Thursday, system was on the whole time and charging, so the batteries should have been full.
• We had just started shedding when I got home from work yesterday and the system had been off for maybe 15mins. I checked the battery voltages at that stage with the battery cut-off disengaged and inverter off: 12.75V per battery (so 25.5V combined?!? but cut-off voltage at that stage was set to 24V? (I've since changed it to 23.9V))
• Just after the power came back, I checked the battery voltages again and they were 13.64V each. The 27.28V combined is close enough to the float charge voltage but shouldn't this be closer to 29.2V?
• Perfectly identical voltages between the batteries, in both instances of measurement, so I think they're both OK.
• Load was previously measured using a kill-a-watt clone from Geewiz, over a period of 24h, before I put the system together. It would probably have been useful to measure average load yesterday while shedding but I don't see it varying greatly from the previously measured value. I will however reinstate the kill-a-watt and check the measurements in a few days.
• I know the only way to truly know the state of the batteries is to invest in a BMV and monitor the situation properly, but the WAF of this project has already gone from somewhat good to very very bad in a hurry, so that is not an option for the moment.
• I've read this forum and others, and even made a spreadsheet comparing the various recommended settings I could find for the Axpert inverters, so I think those are OK?

TL;DR: the system doesn't provide the expected capacity at all - according to my calculations, with our little 300W load on it and a 50% DoD limit on the batteries, we should have 2.4kWh, i.e. 7-8 hours of standby. At the moment it barely manages 200Wh...

Any ideas or inputs on this from anyone would be very much appreciated!

1 hour ago, selenium said:

Battery cut-off 23.9V

Well, I don't like these lead money wasting items, but if you look at your manual, assuming you discharge at .2C or below, then the Battery cut off should be 1.75V * 6 (cells per battery) * 2 (batteries in series) = 21V

There's a difference tension wise between a cell with 0 load and a cell with a load and your cables also come into play, depending on the cable dimensions and the connector quality and crimp quality between the cable(s) and the lugs/connectors you could have some extra resistance and thus Voltage drop, once current starts flowing...

Edited November 1, 20223 yr by Kalahari Meerkat

36 minutes ago, Kalahari Meerkat said:

Well, I don't like these lead money wasting items, but if you look at your manual, assuming you discharge at .2C or below, then the Battery cut off should be 1.75V * 6 (cells per battery) * 2 (batteries in series) = 21V

There's a difference tension wise between a cell with 0 load and a cell with a load and your cables also come into play, depending on the cable dimensions and the connector quality and crimp quality between the cable(s) and the lugs/connectors you could have some extra resistance and thus Voltage drop, once current starts flowing...

Oh how I remember cursing them 🤣

I would also bring that cut-off voltage down by at least another 1V, maybe even 1.5V at the same time I would definitely decrease the charging amps to 15 amps to reduce the wear on the cells. These batteries prefer nice slow steady charging to keep them healthy.

1 hour ago, Kalahari Meerkat said:

Well, I don't like these lead money wasting items, but if you look at your manual, assuming you discharge at .2C or below, then the Battery cut off should be 1.75V * 6 (cells per battery) * 2 (batteries in series) = 21V

There's a difference tension wise between a cell with 0 load and a cell with a load and your cables also come into play, depending on the cable dimensions and the connector quality and crimp quality between the cable(s) and the lugs/connectors you could have some extra resistance and thus Voltage drop, once current starts flowing...

So THAT's what that table is for - thank you! Bloody hell. I will update the cut-off voltage and report back but it really helps to now know what that table is for and how to use it. Thanks again!

I will also check the voltage at the inverter vs. the batteries to get a feeling for what impact the cables might have.

As an aside: You have my complete agreement on the lead money wasting items - like I said, I got them very very cheap. The idea is to sell them once I have a better grip on the systems and terminology, and then move to LiFePO4.

But you have to learn somehow, and this is a good, not too expensive school, with lots of very knowledgeable people willing to share their experience and know-how.

1 hour ago, WannabeSolarSparky said:

Oh how I remember cursing them 🤣

I would also bring that cut-off voltage down by at least another 1V, maybe even 1.5V at the same time I would definitely decrease the charging amps to 15 amps to reduce the wear on the cells. These batteries prefer nice slow steady charging to keep them healthy.

Cursing is about right - lots of adult language has been spoken in the proximity in the last few weeks...

Thank you for your input too - I will update the cut-off voltage and report back.

1 hour ago, WannabeSolarSparky said:

cut-off voltage down by at least another 1V, maybe even 1.5V

If you change the cut-off voltage then you should reduce the Back To Charge voltage as well else it is not going to make much diffs

Edited November 1, 20223 yr by WannabeSolarSparky

1 minute ago, WannabeSolarSparky said:

If you change the cut-off voltage then you should reduce the Back To Charge voltage as well else it is not going to make much diffs

Thanks, will do.

For interest's sake, what are those settings for? Are they more related to a system where you have PV input as well?

7 minutes ago, selenium said:

For interest's sake, what are those settings for? Are they more related to a system where you have PV input as well?

As far as I know If you are not using solar panels then all the settings in red are the important battery ones to look after and set correctly.
The one in green is total charging from SOLAR+Grid

9 minutes ago, WannabeSolarSparky said:

As far as I know If you are not using solar panels then all the settings in red are the important battery ones to look after and set correctly.
The one in green is total charging from SOLAR+Grid

Thank you - every day a little more knowledge 😀

FYI, there is a premature float bug on some of the Axperts which causes it to never reach the bulk charge voltage, it just charges to the float voltage and stays there so the batteries never fully charge. Just make sure your inverter is actually reaching the bulk voltage. You can overcome this by using the equalization settings to run daily at the same voltage as your bulk setting.

On 2022/11/02 at 2:14 PM, jumper said:

FYI, there is a premature float bug on some of the Axperts which causes it to never reach the bulk charge voltage, it just charges to the float voltage and stays there so the batteries never fully charge. Just make sure your inverter is actually reaching the bulk voltage. You can overcome this by using the equalization settings to run daily at the same voltage as your bulk setting.

It certainly seems to be the case with mine - it just switched off again after about 30mins. 🤬

Checked the battery voltages: 12.9V/12.88V respectively. WTaF. The previous round they shut off at 12.75V each.

And this is after I changed the cut-off and back to charge voltage settings as recommended by @WannabeSolarSparky and @Kalahari Meerkat

Obviously can't check exact settings right now, but if I recall the cut-off is now 22.5V... Which is 11.25V per battery vs. the 12.9v I'm currently measuring.

Surely a 300W-ish load cannot make the battery voltage sag that much?

Is my unit's internal shunt possibly defective that the readings vs. settings are so vastly different? Do I need to send it back? Or would a complete reset help/be possible? @Coulomb, please chime in with your thoughts?

Please also give recommendations on how long do I need to set the equalization to run? I'd like make sure the batteries get a decent charge without tipping over into overcharging? 60mins? 90mins? I'm guessing there's a magic formula to calculate it?

Sorry for all the questions - if any of you are ever in the vicinity, I will happily buy you a drink in gratitude for all the help already given.

5 hours ago, selenium said:

Surely a 300W-ish load cannot make the battery voltage sag that much?

If all the cells are not fully absorbed and balanced at the top end and all in good health then that could quite easily happen no matter what the load is.
The BMS may be triggering a shutoff as soon as a single cell drops below the BMS safety cut-off voltage even if all the other cells are still quite full.

Relying on the bms on its own to get all the cells balanced is a long painful process (BMS Dependent)and can take from a day or two to several days/weeks to get fully balanced, especially if you are using the batteries in between.

The only real way to get the cells all nicely top balanced is to get them all connected in parallel and then do a nice slow 0.1C or 0.2C charge up to the top and then letting the batteries absorb till that take no more amps.

Other people may have more solutions, but this is the only sure-fire way I know to get good results.

NOTE: If the batteries are still under warranty then get the supplier to check them.

WRONG THREAD AND TOTALLY WRONG CHEMISTRY - This answer is not related to the OP Question.

Edited November 8, 20223 yr by WannabeSolarSparky
Note added - Correction added - fact checked :)

35 minutes ago, WannabeSolarSparky said:

If all the cells are not fully absorbed and balanced at the top end and all in good health then that could quite easily happen no matter what the load is.
The BMS may be triggering a shutoff as soon as a single cell drops below the BMS safety cut-off voltage even if all the other cells are still quite full.

Relying on the bms on its own to get all the cells balanced is a long painful process (BMS Dependent)and can take from a day or two to several days/weeks to get fully balanced, especially if you are using the batteries in between.

The only real way to get the cells all nicely top balanced is to get them all connected in parallel and then do a nice slow 0.1C or 0.2C charge up to the top and then letting the batteries absorb till that take no more amps.

Other people may have more solutions, but this is the only sure-fire way I know to get good results.

NOTE: If the batteries are still under warranty then get the supplier to check them.

Nothing related to lithium will apply as they are Gel. No cut off from BMS.

My 2c would be if those batteries jump to over 13.5V shortly after the power came back on from 12.75V they no longer have anything close to 100Ah. It seems due to cycling even if it is only 5 months old. It does also seen the voltage is collapsing under the 15A load.

On 2022/11/02 at 2:14 PM, jumper said:

FYI, there is a premature float bug on some of the Axperts

Ok, maybe expand and mention which "some of the Axperts" are, does this apply to the 24V inverters as well?

Failing an answer that is useful in this case, maybe @Coulomb can help us out here, he is probably the most knowledgeable dude around as far as Axpert type inverters are concerned. Could you comment on the premature float bug, how does it occur and on which models and how can one determine whether this is happening here. @selenium is running a RCT Axpert V-PF1 24V 2kW inverter without solar panels and his GEB FT12-200 12V 200Ah Gel batteries are not exactly holding up their end after 6 months. With the amount of loadshedding, I suspect they have had a rough time of it and are likely somewhat the worse for wear by now.

As for the Voltage dropping quickly under load etc. this could point to the Lead Acids having a pretty high internal resistance by now aka stuffed batteries, basically as soon as a load is applied the Voltage drops more than one would expect and this is likely due to higher internal resistance on the cells by now.

 

3 hours ago, selenium said:

It certainly seems to be the case with mine - it just switched off again after about 30mins. 🤬

I think the first thing to do is to diagnose your inverter to see if it definitely has the bug. Best to do this when the batteries are drained, not when they are full as the voltage climbs too quickly.

You'll need to watch the batteries charge and be there when they reach float voltage and check the charging light on the inverter. It should be flashing green while charging and keep flashing as it passes float voltage and continues to bulk voltage and then it will go solid green when it reaches bulk voltage to indicate it is floating and then the voltage should drop down to the float voltage... this is all if it is working correctly without the bug. If the light goes solid green when you first reach float voltage and the voltage doesn't increase to the bulk voltage, then you have the bug, the inverter will just sit there at float.

If you have the bug then you can use the equalize settings to work around the issue by setting the equalization voltage to the same as the bulk voltage and then set it to 60mins. Some inverters like my older 24V MKS don't have the equalize settings, so I have to change my float voltage to the same as the bulk voltage and then go and physically change it back an hour later (I only do that once a week or so). Others use an external charger on the batteries to take care of the bulk charge and let the inverter just do the float.

I have the bug on my 24V original inverter and also on a 48V clone, it is all over the place. The problem is that the only proper fix is the patched firmware from @Coulomb, but that is only for 48V low voltage models, there is no patched firmware for 24V models at all.

 

Edit: here's the post from Weber on the Aussie forum, with a better explanation: https://forums.aeva.asn.au/viewtopic.php?title=pip4048ms-inverter&p=59896&t=4332#p59896

Edited November 8, 20223 yr by jumper

2 hours ago, Scorp007 said:

Nothing related to lithium will apply as they are Gel. No cut off from BMS.

My 2c would be if those batteries jump to over 13.5V shortly after the power came back on from 12.75V they no longer have anything close to 100Ah. It seems due to cycling even if it is only 5 months old. It does also seen the voltage is collapsing under the 15A load.

lol yeah, my bad, I confused this thread with one of my private threads about the same scenario of sudden voltage sag...

Apologies to the OP and any confusions created 😇

2 hours ago, WannabeSolarSparky said:

lol yeah, my bad, I confused this thread with one of my private threads about the same scenario of sudden voltage sag...

Apologies to the OP and any confusions created 😇

No worries, most days I don't know my arse from my elbow🤣🤪

9 hours ago, Kalahari Meerkat said:

@selenium is running a RCT Axpert V-PF1 24V 2kW inverter without solar panels and his GEB FT12-200 12V 200Ah Gel batteries are not exactly holding up their end after 6 months.

OP has the perfect recipe for batricide: 24V system (so high battery current for a given load, and no patched firmware), lead acid battery, and frequent load shedding. Default settings are not designed with frequent load shedding in mind.

The biggest problem is the low DC cut-off setting. This is usually around 10.5V per 12V module, which is appropriate for maximum run time in the assumed RARE condition that it's needed. When you have load shedding multiple times per week, this is way too much discharging for lead acid, and they quickly lose capacity and increase internal resistance.

With frequent load shedding, it's imperative to set the low DC cut-off (when using a lead acid battery) to 12 V per nominally 12 V module, i.e. 24.0V in this case, or 48.0V for 48V inverters.

ALL Voltronic inverters come with the premature float bugs from the factory. They just don't believe it's real. It means that when charging from solar, when about 75% charged and a cloud comes by, the inverter sees low current into the battery, and decides incorrectly that this must be because the battery is full. It only wants the battery to be within half a volt (quarter volt for 24 V systems) of the FLOAT voltage to declare it full. Of course, it should be nearly the BULK/ABSORB voltage and low battery current to declare it full. If charging mostly from solar, which is what most of us like to do because it's free once the system is paid for, this bug causes the battery to be chronically under-charged. Lead acid suffers from this, accelerating the degradation that is already bad because of the excessive discharge.

As to what can be done: charge each battery module separately with a car battery charger. Watch the battery voltage. If it shoots up immediately to say 14V, that's not a good thing; it means that the battery acid is almost water. If you're lucky, the battery module will recover a little, and the battery terminal voltage will actually go DOWN as it is charging; this is a good sign that the internal resistance is falling.

As for how long to set the equalisation charge to work around the premature float bug, it's a delicate compromise. Too long and you could over-charge the battery and boil out the small quantity of water that is in them. Too short and you're not charging it enough. 60 minutes seems like a good place to start. Sadly, you'll have to watch the battery very closely. Worse, the chances of recovering your existing battery are very slim. That's one of the reasons that lithium batteries are so popular: they're not so delicate, and also they have a battery management system protecting them. Plus, lithium chemistries don't mind frequent discharges, as long as it doesn't go below about 5-10% SoC very often. Of course, they cost a lot more up front. Also the options for 24V systems seem to be quite limited.

Edited November 9, 20223 yr by Coulomb

Thanks @Coulomb, so, I guess the premature float bug is not at issue here, since @selenium does not have solar panels as far as I understood it, its just too many cycles on the batteries that are not designed for this kind of use, really... we've been having 2 to 3 load shats per day in recent times, now 1 or 2 per day, if it happens during sunshine and you have solar panels, no worries, but hey, yesterday at noon to just after 2PM and again at 10PM until just after midnight... during the day when the fscking clouds aren't there, no worries, but at night, certainly the battery has to make do... but.... again the OP has no panels, as far as I know...

So, @selenium, I suspect your batteries, however cheap you got them, are seriously the worse for wear and you can try and baby them back to slightly better health, but the outlook is somewhat on the grim end of the scale, I'd say...

1 hour ago, Kalahari Meerkat said:

Thanks @Coulomb, so, I guess the premature float bug is not at issue here, since @selenium does not have solar panels as far as I understood it, its just too many cycles on the batteries that are not designed for this kind of use, really... we've been having 2 to 3 load shats per day in recent times, now 1 or 2 per day, if it happens during sunshine and you have solar panels, no worries, but hey, yesterday at noon to just after 2PM and again at 10PM until just after midnight... during the day when the fscking clouds aren't there, no worries, but at night, certainly the battery has to make do... but.... again the OP has no panels, as far as I know...

So, @selenium, I suspect your batteries, however cheap you got them, are seriously the worse for wear and you can try and baby them back to slightly better health, but the outlook is somewhat on the grim end of the scale, I'd say...

Damnit!

Thank you @Coulomb and @Kalahari Meerkat and everyone else who answered - I will try charging with a car charger and hopefully rescue what little capacity remains. My cut-off voltage was set at 24V until very recently, so I'm hopeful that the money-wasters aren't completely fsck'd.

 

5 hours ago, Coulomb said:

... OP has the perfect recipe for batricide: 24V system (so high battery current for a given load, and no patched firmware), lead acid battery, and frequent load shedding. Default settings are not designed with frequent load shedding in mind....

Just my rotten luck. *sigh* Live and learn, live and learn.

Update: I've decided to bite the bullet and start looking at non-moneywaster options. Thread below - please chime in!