Jump to content

Are my batteries messed up? Help with Axpert settings!


Nuno

Recommended Posts

Hi folks,

My solar system is acting really weird. I got 8 sealed lead-acid 12v 260ah (2 groups of 4 for a 48v system) in order to have a good safety buffer for cloudy days. It turns out the performance of these batteries is like 10% of what is advertised. I'm not sure if one of them might be faulty and is harming the whole system.

I tried measuring each one with a multimeter, but the only difference in voltage I get in a particular battery is only of 0.05 - 0.15 V (depending on when I measure). Is this significant? All others have only a 0.01V difference (more or less) between them. Of course this is with all the batteries still connected together.

After contacting the company that sold me the kit, they sent me settings for the Axpert inverter, but after changing to their advised values, I got even worst performance starting the day at 7am with 47.3v while the days before I used to have 49.5v after the night.

Some relevant settings as of NOW:

02 - 60A
05 - USE
16 - SNU
25 - FdS
26 - 57.6V
27 - 53.4V
29 - 44.0V
 

PS - extra system info: I have 6 x 320w panels (2 strings in parallel, each with 3 in series). Totally off-grid, no power from anywhere else coming in.

 

Edited by Nuno
Link to comment
Share on other sites

you cannot measure the batteries all connected up reliably, they should be safely disconnected and allowed to rest for best results.

you are probably chronically under charging your batteries, do you have a bmv?

 

before others ask, what is your inverters firmware version?

Edited by Dex_
Link to comment
Share on other sites

1 hour ago, Nuno said:

I got even worst performance starting the day at 7am with 47.3v while the days before I used to have 49.5v after the night.

Hi Nuno. 

We will need more info to be able to help you. 

Can you please tell us the following:

  1. Your average consumption During the day measured in Kwh
  2. Your average consumption During the night measured in Kwh
  3. A little more info on your loads.

What i can see is that your system is not balanced, meaning, You have way to much battery power compared to the total of panel power to recharge them. That is why its reading less voltage as you go along. 

You have 24.96kwh of battery power , if you discharge that to 50 percent, you will have used 12.48kwh. You also have 1.98kwp of panels and you can only expect to get 1.58kwh from them after losses to re-charge your night time usage as well as to run your daytime loads. That will give you 1.58 x 6hours = 9.5kwh to put back in the battery bank. this is without any daytime loads. 

I dont think your batteries ever gets fully recharged and if they dont get to 100% at least once a week, it will result in permanent damage soon. 

I believe that you need more panels to make this system preform correctly. 

I also suggest you install a Battery monitor to be able to know exactly at what state of charge the batteries are so that you would be able to monitor and manage this situation correctly. Batteries are pretty expensive and a battery monitor can help you to protect your investment. 

Link to comment
Share on other sites

Hi, thanks for the replies!

Dex_, sadly I dont own a BMV :( About the firmware, this is what I get in the screen

U1: 20 04
U2: 02 02

Does it help?

Jaco, my load is on average 5-6kwh per day. I estimate the night load ("night" being the period without PV production) to be about 1-1.5kw.

Your numbers make sense, I dont get why the inverter reports the system fully charged (stable LED) so early when I get a sunny day (usually by 1pm).

Any advice on the float/bulk charge settings?

 

Link to comment
Share on other sites

1 minute ago, Nuno said:

I dont get why the inverter reports the system fully charged (stable LED) so early when I get a sunny day

The inverter is working on battery voltage. Voltage is not an accurate indication of the State of charge. Voltage varies with load, so you can not trust the indication on the inverter. For you to really get to the bottom of this i will still recommend something like the Victron BMV 702 

 

5 minutes ago, Nuno said:

Any advice on the float/bulk charge settings?

For that we would need to know the type of batteries you have to be able to look at the Data sheet too see these values. 

 

Link to comment
Share on other sites

1 minute ago, Jaco de Jongh said:

For that we would need to know the type of batteries you have to be able to look at the Data sheet too see these values.

I can't find the datasheet anywhere! These batteries seem to be of some kind of "local" brand, and so its harder to get things online. I'll contact the distributor to see if I have some luck...

Link to comment
Share on other sites

1 hour ago, Nuno said:

I can't find the datasheet anywhere! These batteries seem to be of some kind of "local" brand, and so its harder to get things online. I'll contact the distributor to see if I have some luck...

post a picture of the batteries, maybe someone can identify them.

Link to comment
Share on other sites

5 hours ago, Nuno said:

Any advice on the float/bulk charge settings?

I would increase the float voltage setting (setting 27) to at least 54.0 V (13.5 V per 12 V module). It looks like from the photo that your maximum absorb voltage is 14.5 V, so your bulk/absorb voltage setting (setting 26) is about right.

I agree that you need more panels. You are probably suffering from the premature float bugs; unfortunately there is no patched firmware for your inverter.

You haven't said what model your inverter is. Does it have the "bulk" charge time setting (really absorb time, setting 32)? If so, you can blunt the effects of the premature float bugs by using a longish (say 120 minute) absorb time. Because of one of the premature float bugs, that 120 minutes will include time when the battery is below the absorb voltage, so you won't be getting all of that 120 minutes.

Running your battery overnight below 48 V is not good; it might already be too late for the battery.

What were your settings before the change (just the changes)?

Finally, how old is the battery?

 

Link to comment
Share on other sites

16 minutes ago, Coulomb said:

I would increase the float voltage setting (setting 27) to at least 54.0 V (13.5 V per 12 V module). It looks like from the photo that your maximum absorb voltage is 14.5 V, so your bulk/absorb voltage setting (setting 26) is about right.

Yes I already did this today morning :)

Quote

I agree that you need more panels. You are probably suffering from the premature float bugs; unfortunately there is no patched firmware for your inverter.

You haven't said what model your inverter is. Does it have the "bulk" charge time setting (really absorb time, setting 32)? If so, you can blunt the effects of the premature float bugs by using a longish (say 120 minute) absorb time. Because of one of the premature float bugs, that 120 minutes will include time when the battery is below the absorb voltage, so you won't be getting all of that 120 minutes.

I'm not sure of the model because they are all so similar to each other! It does have settings 30, 31, 33, 34, 35, 36. Doesnt have 32 and doesnt have 04 either. Does that give a clue? (I attach a photo of it, maybe it helps to identify the right model). Oh, it also has a USB and Ethernet port, no serial.

Quote

Running your battery overnight below 48 V is not good; it might already be too late for the battery.

What were your settings before the change (just the changes)?

Finally, how old is the battery?

The only significant change was the float voltage, that I had set up higher (not sure if 54 or 54.4).

The batteries are around 1yr old, but only 2months of more intense use. Before that they were mostly used for tools once in a while. Now I'm living fulltime off-grid, so thats the big difference :)

thanks!

20190924_131844.jpg

Edited by Nuno
inverter details added
Link to comment
Share on other sites

Just one more question...

I understand, by the numbers, that I may have not enough panels and the batteries never get fully charged... BUT, how then can we explain that almost every sunny day the inverter stops charging the batteries?

Usually by 1-2pm, the PV power drops to 150w and keeps it at that level, unless there is a load. So, if I turn a 1000w pump on, the PV power goes back to 1000w and then as soon as I turn it off (or some minutes mores) it drops back again to 150w. I always understood this to be a sign of having the batteries at full charge

Edited by Nuno
typo fix
Link to comment
Share on other sites

14 minutes ago, Nuno said:

I always understood this to be a sign of having the batteries at full charge

If this is correct and your batteries are fully charged, then there is no way that an 520Amp hour battery bank is sitting at 47.3 volt after using only 1.5kwh out of it during the night. 

Your suspicion that the batteries might be damaged, may be right, but then again, what could have caused it to get to this stage after only one year of usage. I saw on the internet that this manufacturer claims 1400 cycles at 80% discharge, If that is true, there is now way your consumption could have caused the batteries to fail within a year.  

Please also remember, batteries that are damaged, more than often shows fully charged quite quickly, it might be correct, it does not take anymore charge, but its not known what the real capacity of the bank is any more. 

Link to comment
Share on other sites

2 hours ago, DaveSA said:

Have you set the Inverter to Equalize charge ?  And are any of the batteries warmer than the rest ?

The inverter is set to do an equalization charge every 60 days or so... should I do that more often?

I checked the temperatures a few time, didn't see anything different. The only difference was only the ~0.1v in a particular battery. But I guess thats not a significant number...

Link to comment
Share on other sites

22 hours ago, Nuno said:

I'm not sure of the model because they are all so similar to each other!

Yes, but you (should) have the advantage of the sticker on the side, with a reseller's model number/name.

Quote

It does have settings 30, 31, 33, 34, 35, 36. Doesnt have 32 and doesnt have 04 either. Does that give a clue?

That means it's possibly a VM 5000-48 (not a VM II). But the settings information above is what I was after (see below).

Quote

Oh, it also has a USB and Ethernet port, no serial.

I think you'll find that it has a serial port, they use an RJ-45 connector which is usually reserved for ethernet. You should have been provided with an adapter cable that has an RJ-45 plug at one end and a 9-pin RS-232 connector at the other.

20 hours ago, Nuno said:

BUT, how then can we explain that almost every sunny day the inverter stops charging the batteries?

It could be the premature float bug. A cloud passes by, the PV power goes low, the inverter sees the combination of a suitable battery voltage and low current for a certain period of time... bingo! The battery (it thinks) is full! Time to go to float mode, were the battery hardly charges at all (it could take a month to charge your big battery at float voltages). The problem is with the "suitable battery voltage". For reasons best known to the cosmos, Voltronic Power seem to believe that the appropriate voltage is the float voltage, or even up to half a volt lower than that setting. Well, that could happen at quite low states of charge. At first,I thought it was a typo in their source code, but more and more evidence points to the fact that they think that they are right. This float bug has existed since at least 2013, about the time of the first models produced.

Well, they are wrong. You have to see the low current at nearly the absorb/bulk voltage, not the float voltage, before the battery can be declared full. On a completely cloudless day, a Voltronic Power inverter with factory firmware will charge the battery fully. But as soon as the first cloud or shade appears, as long as the battery isn't dead flat, it will stop charging. You can even demonstrate it with utility charging (see the link I posted to the premature float bugs post). For some of the 48 V MKS models, we have produced patched firmware to correct the premature float bugs, but sadly your inverter isn't one of these.

The fact that you have settings 30-36 (except 32) means you have equalisation capability. The equalisation is quite flexible; it looks like it can be applied every day, and the voltage is settable. If (and it's looking to be a bit of a big IF at this stage) your only problem is the premature float bug, you could perhaps work around the premature float bug by using equalisation every day. NOTE: set the equalisation voltage (setting 31) to the same, or possibly even a tiny bit less (if it will let you) than the absorb voltage. The absorb voltage should be 56.8 V to 57.6 V (14.2 - 14.4 V per 12 V module), and really should be temperature compensated. So as we head towards summer, that voltage should go DOWN. How much, I leave to you or the lead acid experts; a quick Google seems to indicate -3 mV/°C/cell, or -72 mV/°C for the whole battery, probably with 25°C as a base point. So when the average temperature while charging is about 28 °C, you should cut down the absorb voltage by some .072 x 3 ~= 0.2 V, or 57.4 V max. In winter, you can increase the voltage by a similar amount for 22 °C operation. I would start with say 60 and 120 (1 and 2 hours) for settings 33 and 34 (the defaults). Actually, 14.4 V per 12 V module is the gassing voltage at 25°C, so stay below this by a tenth of a volt or two. This will force your battery to be just below the gassing voltage for at least 60 minutes per day, even after it thinks that the battery is full, forcing it to be charged more. If this gives you more run time at night, then you know you're making progress.

NOTE: don't do this every day forever; this is only to overcome what I suspect is a chronic undercharging of the battery. After that, I'd either disable automatic equalisation, or do it every 60 or so days (maybe play with that value), and use setting 36 (equalisation activated immediately) when it seems to need it. I almost hesitate to suggest this, because it would be so easy to overcharge your battery, and it's going to be difficult to tell when you're doing this. Overcharging a sealed battery like yours will cause it to lose water, and that can't be replaced (well, not without a lot of effort; Google this if you think you need to). So be listening for gassing sounds (very soft bubbling / popping sounds), and stop charging immediately if that happens (use setting 30 to disable equalisation; don't turn off the rocker switch; this will just  turn off the inverter, removing all loads, and leave the Solar Charge Controller running).

A bit to absorb there (so to speak).

 
Link to comment
Share on other sites

Hi everyone:

        I'm trying to help my neighbor (70yo) that got screwed, He has a 48v setup but the inverter shuts off (He's Offgrid) at about midnight (42v at 6PM), He's got 8 12v 300AH Flooded batteries (8 batteries in series and parallel) but I checked the voltage on each after disconnecting and some have 8.02v, 12.9v, 10.5v, 10.2v all over, he's got a 12v 2A charger, so I'll help him charge to full each of them, then connect all of them in parallel to balance and reconnect, could the batteries be defective for them to charge unevenly? Is there a BMS for Flooded lead acid batteries? ANY answer would be appreciated and thank you.

Link to comment
Share on other sites

10 hours ago, LuisMiguel said:

could the batteries be defective for them to charge unevenly?

Yes. The ones around 10 V probably have one collapsed cell, and the one around 8 V probably has two collapsed cells. (Each cell is 2.0 V nominal). There is almost never a recovery from a collapsed cell, unfortunately. Lead acid is wickedly unforgiving like that. This is part of the reason that LFP boxed batteries with built-in BMS are becoming so popular. They cost more per Ah to install, but you get to use much more of the available capacity, they last much longer, and they don't tend to fail catastrophically if you don't maintain them.

Lead acid usually doesn't have a BMS, except perhaps something like a Victron BMV to measure Ah of discharge and converting that to a SOC (presuming a capacity). Though there are such things as battery balancers that attempt to notice when one half (or one half of one half) becomes significantly different in voltage to the other half, and sound an alarm if so. Then it's up to the owner to figure out what's wrong, and hopefully correct it before the batteries become ruined.

Link to comment
Share on other sites

Hi @Coulomb

thanks very much for the detailed answer. I'll look more closely to see if the float bug is an issue or not.

I recall clouds passing by lowering the input power, but as soon as the cloud is out of the way, the power goes up again.

(Side note, actually I dont have Setting 31 either, it goes from 30 to 33.)

I just changed the Equalization voltage to 57.6 (it was set at 58.4). The reported voltage jumps from the floating voltage (53.7 at the time), to 57.x in just 5 seconds or so, even with just 500w of PV input. Of course, if I stop the equalization process, it comes down equally fast. Is this expected?

Link to comment
Share on other sites

 

Quote

I just changed the Equalization voltage to 57.6 (it was set at 58.4). The reported voltage jumps from the floating voltage (53.7 at the time), to 57.x in just 5 seconds or so, even with just 500w of PV input. Of course, if I stop the equalization process, it comes down equally fast. Is this expected?

No. This is a sign of a battery with a high internal resistance. This could be because it is at a low state of charge, or damaged (low capacity), or both. All you can hope for is that the battery will absorb charge and slowly reduce its internal resistance, so that next time it might take 10 seconds instead of 5, and eventually many hours.

Sometimes you find that the battery voltage shoots up in a few seconds as you have seen, but then as it charges a bit, the internal resistance lowers and the voltage actually decreases.

So now my theory is that rather than the charge bugs being your main problem, it is the internal resistance. When it gets this high, the inverter will note that the battery is at a high voltage and taking low charge, and interpreting this as the battery being full. It then switches to float mode, which is exactly the opposite of what the battery needs to recover.

I wonder what you should change setting 02 to. The inverter uses that setting to determine the end of charge current limit (basically that setting divided by 5, so presently 60/5 = 12 A. But your panels even in ideal conditions will produce 320 x 6 / 50 = 38 A. Commonly only 75% of that, or about 29 A. So whenever you get a cloud that reduces output to 40%, the charge bug will kick in and stop charging. If you change the maximum charging current (setting 02) to say 40 amps, then the cloud has to reduce the PV power to about 26% of maximum before the bug kicks in. Presumably, 26% clouds are less frequent than 40% ones, so that will help. 

But then the problem is how to prevent over-charging. The charge will only stop when you get the charge current under 8 A, which on a cloudless day might never happen with such a large capacity battery.

I agree with @Jaco de Jongh that your system is severely unbalanced: way too little solar power for such a large battery. So long term, you definitely need more panels. Use a separate charge controller if needed.

Short term, get as much charge into the battery as possible. I would raise the float voltage (setting 27) to the float voltage, or maybe 0.1 V less, SHORT TERM ONLY. Once it has a bit of charge and isn't showing such high internal resistance, put the float voltage back to normal, and use hours of absorb-voltage equalisation every day to get the battery towards full. Then reduce the frequency of equalisation to perhaps 1 per week, unless the battery shows signs of not charging fully most days.

You should also try and figure out if you have some bad modules that won't recover. Be prepared for the distint and unfortunate possibility that they are all ruined. Lead acid is very unforgiving of poor charging.

Edited by Coulomb
Link to comment
Share on other sites

9 hours ago, Coulomb said:

Short term, get as much charge into the battery as possible. I would raise the float voltage (setting 27) to the float voltage, or maybe 0.1 V less, SHORT TERM ONLY.

sorry, do you mean to change the float voltage to the BULK voltage?

Right now this is what I have

02 - 40A
26 (bulk) - 58.4V
27 (float) - 54.5V
31(eq voltage) - 57.6V

(I dont seem to have a setting for the absorption voltage)

PS - I found a copy of my manual online. This one seems to have the exact settings I have available: https://slideshare.net/dennyhonen/voltronic-axpert-v-user-manual

Edited by Nuno
added manual
Link to comment
Share on other sites

14 hours ago, Nuno said:

sorry, do you mean to change the float voltage to the BULK voltage?

As a temporary measure to charge the battery as fast as possible, yes. But 57.6 V max (14.4 V per 12 V module, 2.40 VPC; this is the gassing voltage). Preferably a few tenths of a volt below that level, especially as the weather warms up.

Quote

Right now this is what I have...
26 (bulk) - 58.4V

As mentioned above, this is too high! It will dry out your battery.

Quote

(I dont seem to have a setting for the absorption voltage)

Voltronic misname the absorbtion voltage setting; they call it the "bulk charging voltage (C.V voltage)". It's only at that voltage at the end of the bulk stage; it tries to stay at this voltage for all of the absorb stage. So this is setting 26.

Edited by Coulomb
typo
Link to comment
Share on other sites

Alright folks, I thought it would be nice to share the development of this story, in case anyone else in the future finds itself in the same spot.

Thanks again to all and especially @Coulomb for the advice!

So I changed setting 27 (float voltage) to 56.9v and logged the voltage of the batteries as reported by the inverter twice a day. One at night, before bed, and one early morning (without PV power or very little power ~30w).

First night: 49.9v
Morning: 46.9v (inverter alarm beeping with error 04)
Second night: 50.4v
Morning: 47.8v (no alarm)
Third night: 49.9v
Morning: 50.0v
Fourth night: 51.4v
Morning: 50.3v

So in general it seems to be improving. The big change was on the third night, there was absolutely no loss of voltage during the whole night (with around 1000w of consumption)
I think tomorrow I'll change setting 27 back to its default and see if it holds the same way or not.

One thing I noticed is that it only reaches 56.9v if there is a good amount of power coming in (>600w or so). Otherwise it drops to 54v 53v...52v.... and usually holds around 52 with low sun. (no loads)

I do wonder why it drops so much (from ~52v to 50v), in just a few hours (say, from 17h to 22h), and no further drop during the whole night. The consumption is about the same.

If anyone has any further insights/tips/advice, please do share :)

thanks again everyone!

Edited by Nuno
typo fix
Link to comment
Share on other sites

2 hours ago, Nuno said:

Alright folks, I thought it would be nice to share the development of this story, in case anyone else in the future finds itself in the same spot.

Thanks again to all and especially @Coulomb for the advice!

So I changed setting 27 (float voltage) to 56.9v and logged the voltage of the batteries as reported by the inverter twice a day. One at night, before bed, and one early morning (without PV power or very little power ~30w).

First night: 49.9v
Morning: 46.9v (inverter alarm beeping with error 04)
Second night: 50.4v
Morning: 47.8v (no alarm)
Third night: 49.9v
Morning: 50.0v
Fourth night: 51.4v
Morning: 50.3v

So in general it seems to be improving. The big change was on the third night, there was absolutely no loss of voltage during the whole night (with around 1000w of consumption)
I think tomorrow I'll change setting 27 back to its default and see if it holds the same way or not.

One thing I noticed is that it only reaches 56.9v if there is a good amount of power coming in (>600w or so). Otherwise it drops to 54v 53v...52v.... and usually holds around 52 with low sun. (no loads)

I do wonder why it drops so much (from ~52v to 50v), in just a few hours (say, from 17h to 22h), and no further drop during the whole night. The consumption is about the same.

If anyone has any further insights/tips/advice, please do share :)

thanks again everyone!

off the bat you are probably most active then so it is to be expected.

Link to comment
Share on other sites

8 hours ago, Nuno said:

So in general it seems to be improving.

Yes. So your battery really was chronically undercharged.

6 hours ago, Dex_ said:

One thing I noticed is that it only reaches 56.9v if there is a good amount of power coming in (>600w or so). Otherwise it drops to 54v 53v...52v....

Oops, I quoted Dex_'s quote of your post. Same text.

That's how batteries work. When charging, it takes a lot of current to push the voltage up (unless they have high internal resistance). As they charge, it take less current to maintain a higher voltage. When full, it takes rather little current to keep the voltage high; this is the sign that the battery is full, and the inverter-charger uses this signal to switch from absorb stage to float stage. Or at least, that's how it should work.

8 hours ago, Nuno said:

I do wonder why it drops so much (from ~52v to 50v), in just a few hours (say, from 17h to 22h),

I think it's "resting". It take some hours for the voltage to settle towards its resting (or in your case, near-resting) level. So your "real" battery voltage is around 50 V, or 12.5 V per module. That means its state of charge is good (12.6 V or 12.8 V would be better). The 13 V per module isn't sustainable.

It could also be related to  temperature. I'm no lead acid expert.

Quote

and no further drop during the whole night.

I think it just means that your loads aren't very high at night, relative to the capacity of the battery. So while the SOC will creep down a little, it isn't enough to see a large drop in voltage at night. Or again, it may be temperature related.

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

×
×
  • Create New...