Hi,

I have a Voltronic 24V 3kVA 3kW PWM inverter. Regardless of charging settings the bulk charge levels off at around 27.6V it should go up the set level, normally around 28 or 29V. Everything else is fine, I can change the float voltage, charge current etc. enable equalisation and see it push charge voltage up above 29V to the set level. So it looks to me like a firmware problem as the hardware is doing everything it should. I don't seen any hardware reset except switching it off and disconnecting the battery which I have tried. 

I want use Hubble S120 lithium batteries and need to know they will be charged properly. Any bright ideas? 

* update: Having left everything off overnight I woke it up this morning and took it through a charge cycle. It did in fact push up to the bulk charge voltage so either there is some timer built in that it only does that once in a certain period or its intermittent. I cycled it through a discharge and charge again after that and it did the same old trick of levelling off at 27.6V then dropping to float of 27V

Edited March 25, 20224 yr by RobM
Update

22 hours ago, RobM said:

Any bright ideas? 

Dunno about bright, but it sounds like the premature float bug to me. That explains why it occasionally goes to the required voltage ,and why equalisation works as it should. You need at least one cloud to trigger the bug; on a totally clear day, or with utility charging and normal settings, if it's only the premature float bug, the inverter will charge the battery fully.

Since there is no patched firmware for PWM models, your only option is to use either equalisation (you can use the same equalisation voltage as the bulk/absorb voltage), or timed absorb, to try and work around the problem.

22 hours ago, RobM said:

I want use Hubble S120 lithium batteries

Those presumably have a BMS, so you hopefully can't over-charge the battery with too much absorb or equalisation charging. So that might be an adequate workaround.

BTW, Voltronic don't believe that the premature float bug is real. But it's pretty easy to convince yourself it is. See the "How to demonstrate" section of this old post.

Hi @Coulomb. Thanks for your helpful reply. Yes you may be right about the premature float bug. I had considered using equalisation as a workaround. Reading the old post on how to demonstrate it suggests that there is a combination of max charge and util charge current that would NOT trigger the bug. In my case I am only using util charging for this application hence PWM - no solar needed. My charge settings were as per default:

Max charge current: 50A

Util charge current: 25A

I will try reducing the max charge current to 40A, kicking it out of float mode and see if it charges properly. This may explain why a set of gel batteries are kaput in one year on max 200 deep cycles. 

28 minutes ago, RobM said:

will try reducing the max charge current to 40A, kicking it out of float mode and see if it charges properly. This may explain why a set of gel batteries are kaput in one year on max 200 deep cycles. 

Nope - still doing it... levels off at 27.6V and goes to float. It definitely seems to be some version of the premature float bug. 

Usually the premature float bug doesn't bite when using utility charging, as there is no "cloud" to trigger the bug. That's why you need the unusual settings to demonstrate the bug with utility charging.

24V models often have a separate utility charger, instead of using the inverter in reverse. Maybe it has a voltage limit somehow. But then why did it charge properly that time? Maybe an intermittent fault. 

Perhaps check temperature. It might be derating due to overheating. Of course, the fans will be blowing down. Perhaps try some temporary extra cooling. 

Edited March 25, 20224 yr by Coulomb

Hmm looks like there is a problem with the charger which is a total pain in the butt. Equalisation charge works fine. Its just the weird 27.6V transition which remains regardless of what the float voltage is set to. The other annoying thing is the equalisation workaround will only work at the most once a day as the immediate setting disables itself after a single run. The supplier is also not very knowledgable or helpful.

Note to self - you get what you pay for. 

 

I have the same problem on my 2.4kw 24V clone. My bulk is set to 28V but does not reach it.

I just have to add that the S-100 need not be charged to 29V as the OP would like to happen. At this point the BMS should have disconnect the battery.

I am running with S-100 lithiums. I just need to add that once charged to 27.6V I get more than 2.4kwh when I discharge so the batteries were in fact fully charged.This 2.4kwh is measured up to a discharge of 24.8V which is my disconnect level. There is so little power available from 24.8 down to 24V or lower.

I thus do not see it as a big problem as it prolongs life of lithium not to charge up to 100%.

My max grid charge of 30A does not allow charging more than 18A. Just unfortunate but I accept it coming from a cheap clone.

Edited March 26, 20224 yr by Scorp007

So here is more: If I connect two of my panels up to the PV input and set charging as PV only or PV priority it charges up to bulk voltage. Setting PV and utility charging the same old 27.6V threshold re-appears so its something to do with the utility charger. If I allow pv to charge up to above the 27.6V threshold and then switch on utility it then charges with utility to bulk voltage. This confirms the charger, either with PV or Utility can charge to bulk voltage, so unlikely to be a hardware fault. This really looks like some firmware bug. This setup does not have PV in its installed environment so PV is not an option for absorption. Looks like the only option is the equalisation workaround.

@Scorp007, you are right this is less of an issue with Lithium batteries with their own BMS. I'm replacing the gels with S-120's so perhaps its not a big issue and not worth the inconvenience and cost of taking the thing back to the supplier where they will probably simply replace the utility charger.  

On 2022/03/27 at 11:32 AM, RobM said:

So here is more: If I connect two of my panels up to the PV input and set charging as PV only or PV priority it charges up to bulk voltage. Setting PV and utility charging the same old 27.6V threshold re-appears so its something to do with the utility charger. If I allow pv to charge up to above the 27.6V threshold and then switch on utility it then charges with utility to bulk voltage. This confirms the charger, either with PV or Utility can charge to bulk voltage, so unlikely to be a hardware fault. This really looks like some firmware bug. This setup does not have PV in its installed environment so PV is not an option for absorption. Looks like the only option is the equalisation workaround.

@Scorp007, you are right this is less of an issue with Lithium batteries with their own BMS. I'm replacing the gels with S-120's so perhaps its not a big issue and not worth the inconvenience and cost of taking the thing back to the supplier where they will probably simply replace the utility charger.  

I normally charge my S-100 only with PV. During LS I do use the grid. While charging with grid I found before the bulk level was reached - It was set tobuld to 27.8V. The weird thing was that it would stop charging and the current falls to zero for about 2 sec. Then it would charge at 15A for 2 sec and then just repeat. This all happened once over 27.2V. This is already close to the level I prefer to use so not a train smash. It is also way below the level where the S-100 BMS can come into play.

Following the post from @Chris Louw using an external MPPT I am charging trying it with the Axpert off to save the 25W that the inverter is using when it switches on during PV charging. Panels connected to the external MPPT instead of the Axpert. This standing load can be as high as 600Wh per 24hrs. This I see as a good mod to use with the current rain and miserable PV available.

I have a 24V OCI 3kva/2.4KW inverter. The first few times it recharged my LifePO4 batteries in bulk mode to 28.4V as per setting, then it suddenly got this same issue, it will only bulk charge to 27.6V and go to float. I am charging from utility power only, no panels.

After much reading and pondering..... I think for me the 27.6V thing only started after I installed WatchPower software, and used it to change a parameter on the inverter. Maybe it changed some value in the background that it's not supposed to?

Did any of you with similar issue use the software?
 

22 hours ago, DJ1 said:

Did any of you with similar issue use the software?

I did use watchpower - off the top my head I'm not sure if it was before or after this 27.6V thing started. I did reset everything to defaults using the watchpower function but this made no difference. 

This 27.6V is driving me up the wall. I borrowed 2x 100AH lead acids, still the same. Maybe I should return it, my mind can't switch off if everything is not in perfect order.

@DJI I would. Either the supplier must offer a fix or take it back. Which model do you have? If you are going to use an Axpert type inverter go for Kodak as they have proper backup and 3 year warranty. 

 

I have this one from Takealot. https://www.takealot.com/oci-vm-value-3kva-2-4kw-1kw-mppt-102-vdc-hybrid-off-grid-inverte/PLID72935564

Thanks, will have a look at Kodak models.

On 2022/05/17 at 5:28 PM, DJ1 said:

This 27.6V is driving me up the wall. I borrowed 2x 100AH lead acids, still the same. Maybe I should return it, my mind can't switch off if everything is not in perfect order.

I'm not sure why it is driving you up the wall. 27.6V while charging gives about 27V at rest. That is to me close to 99% full and being used between 0 and 100% is giving away a big chunk of life for not much gain in power used.

Mine also charges up to this level and I get close to quoted capacity by discharging to 24.6V. Again not much point in discharging to a lower level from the available power point.

On 2022/05/19 at 7:37 PM, Scorp007 said:

I'm not sure why it is driving you up the wall. 27.6V while charging gives about 27V at rest. That is to me close to 99% full and being used between 0 and 100% is giving away a big chunk of life for not much gain in power used.

Mine also charges up to this level and I get close to quoted capacity by discharging to 24.6V. Again not much point in discharging to a lower level from the available power point.

Yes you are right, I will try not to worry about it anymore.

I have the same issue with MasterPower Omega, which is another Axpert VM 3000 clone (the MPPT version of first-generation V solar inverters).

It is related to the way how they calculate absorption time. If the battery does not fall below 24.8 V during discharge, the firmware assumes there is no need to absorb, thus it goes directly to float.

If you discharge low enough, it will measure the charging time up to 24.8V and use that to calculate the absorption time (10 * T1), where T1 is the charge time until reaching 24.8V.

This is annoying when using it with LiFePO4 batteries, but I think that's the price we pay when buying the cheapest inverters on the market.

15 hours ago, elvisfox said:

If you discharge low enough, it will measure the charging time up to 24.8V and use that to calculate the absorption time (10 * T1), where T1 is the charge time until reaching 24.8V.

I note that many Voltronic inverter manuals have that diagram in them, with the (10 * T1) in small print, but it's a copy and paste error. Absorb is terminated by the battery charge current falling below a threshold for a certain time. That may not apply to other inverter models, of course.

Hi @Coulomb,

So I did some testing yesterday.

I have discharged my batteries to nearly 0% SOC on purpose (but before UV shutdown), and then switched on the grid. After that it actually went to CV (28.4V in my case), and it stayed there for about 6 hours before I stopped it manually. The charging current actually dropped below 0.2A after 15 minutes of CV (monitored through BMS), but it did not trigger the inverter back to float. Remarkably, the fan was constantly blowing at full speed even with no battery current.

So, no, it is not terminated by the charge current, it actually does that 10*T1.

Also my inverter does not report charging current through LCD. Battery current is only displayed while discharging, but when charging from grid it reports 0. So I am not sure if it is even capable of monitoring the charging current. I suspect the feedback is in hardware, not in the firmware, so it keeps CV for 10*T1 whatever the current is.

Edited January 31, 20233 yr by elvisfox

16 hours ago, elvisfox said:

So, no, it is not terminated by the charge current, it actually does that 10*T1.

I suppose that's possible, given that some 24 V models have their own mains charger (they don't use the inverter in reverse). But I'm still sceptical. 

 

16 hours ago, elvisfox said:

Also my inverter does not report charging current through LCD. Battery current is only displayed while discharging,

You may find that there is a separate data page for charge current versus discharge current. Minus signs are expensive.

It seems like i have the same issue. Mecer 3kva 24v. With Lifepo4 100Ah battery. 

 

So here is my quistion. Can i set the float voltage to 27.2v and the batteries will keep on absorbing till a full SOC. I dont see this happening for more than 12 hours anyway. As Load shedding is a reality.

 

The Equalize function would happen once a day only. And since the batteties are being cycled atleast 3 times a day. I dont think its the way to go.  And is equalize constant volts and amps? Or just CV? 

 

Same question with Float. To my knowledge it IS CV.

 

I will in the next couple of days create a new Topic to explain my setup. As i have questions regarding adding a solar aswell. This is a beat Load shedding setup. Not trying to get off thw grid. The panels will offset the efficiency losses and perhaps run my 200w current during the day.

5 hours ago, Flouw said:

Can i set the float voltage to 27.2v and the batteries will keep on absorbing till a full SOC.

For an 8S LFP battery, 27.2 V is only an average of 3.40 VPC; you'll need a bit more than that to charge the LFP battery in reasonable time. 27.6 V is equivalent to what I use for the bulk/absorb voltage. In general, yes, you can set the float voltage higher to mitigate the premature float bug, at the expense of battery life. But in a heavy load shedding environment, they will have a tough life anyway, and likely won't spend much time at high voltage.

 

5 hours ago, Flouw said:

The Equalize function would happen once a day only. And since the batteries are being cycled at least 3 times a day.

Yes, in that environment, equalisation is less useful. Timed absorb might still be worth trying, if available.

5 hours ago, Flouw said:

And is equalize constant volts and amps? Or just CV? 

Same question with Float. To my knowledge it IS CV.

Equalise, bulk/absorb/CV and float are all current limited CV stages, just with different voltage settings, and conditions for progressing to the next stage. Internally, the equalise charge is implemented as a float stage with a (potentially) different voltage setting.

On 2023/02/01 at 5:14 AM, Coulomb said:

You may find that there is a separate data page for charge current versus discharge current. Minus signs are expensive.

According to the manual, there should be one with PV charging current but it is not present in practice. Probably it appears if the solar is connected (because I only use it as UPS without PV), but I am not sure.

When operating from the mains there is no screen that would report a charging current.

44 minutes ago, elvisfox said:

According to the manual, there should be one with PV charging current but it is not present in practice. Probably it appears if the solar is connected (because I only use it as UPS without PV), but I am not sure.

When operating from the mains there is no screen that would report a charging current.

My 24V can only show a charge current(LHS) if the PV charging exceeds the actual power for loads that are on. 

NO AC charge is shown. On the right when on battery one can see the battery discharge current. 

15 hours ago, Scorp007 said:

Good Morning - I got the fivestar VL-3524M-FS model this weekend - So its the 3KVA 24v MPPT Model. I have a 100ah 24v LiFePO4 battery connected to it. However the settings on the Unit does not allow me to configure it for a 24v - If I make the battery Type USE then I cant set Floating and Cutout Charge for a 24v it only goes between 10v and 14v the manual also and menu settings of the inverter is different to the PVM unit. Can anyone advice me or should I go back to where I bought it from?
[email protected]