Jump to content

Another Axpert SOC Thread


iops

Recommended Posts

Hope everyone's safe and healthy, here's another Axpert SOC thread. Sorry.

 

I'm aware of the Victron BMV's, I'm trying to automate things myself with some calculations and test my batteries at the same time. I've spent most of today reading through various threads on the subject. I think as a last resort I'd want to avoid purchasing a BMV since my load is fairly consistent. After the first test I have some questions though, I'm starting to question the health of my batteries.

 

I'm using the below estimate to calculate SOC:

On 2015/09/10 at 12:26 AM, Alex said:

How I calculate DOD in it's simplest form.

----------------------------------------------

Formula: Volts x Amps/H = Watts/H

----------------------------------------------

In my case this is 24V x 150Ah = 3600 Watts/Hour

so if I pull a average 600W load that will give me 3600W / 600W = 6 Hours

but I only want to use 50% DOD so 6 H / 2 = 3 Hours

with this calculation I then run a test to see what my voltage is after 3 Hours under an average 600W load.

These calculations assume the inverter runs at 100% efficiency which it doesn't so I always take off 10Ah from my calculation to compensate for energy loss to be safe.

A little bit about my system:

Axpert Synapse 3.0+ 24v inverter

8x Deltec BR-12V100 AGM batteries. Specs here.

6x 330W Enersol solar panels in 3S3P

Settings changed below, everything else is stock.

01 SBU
02 30A
03 APL
05 AGM
11 20A
12 23.0V to test
13 27v
16 CSO

 

I've had the system installed so going off the installers word, my knowledge is quite limited.

With my 8x 12V batteries I have a total of 400aH. My test tonight was calculated as follows with some safeguards like deducting 50aH for inefficiencies:

24V * 350aH = 8400 Watts/hour

8400 / 800W avg load = 10.5 hours

10 hours / 2 = 5 hours to have a battery SOC @ 50%

I reached 3 hours and had the Axpert report 23.1V, it beeped for a while and went into bypass mode. My average load over these 3 hours was 681W, peaked at 1131W for 3 minutes.

Before I started the test the batteries were in float according to the battery charge LED on the Axpert. Within an hour the inverter reports DC voltage as 25.8V.

 

I'm a little confused, loadshedding sessions have lasted longer before and having 400aH surely shouldn't last 3 hours? Should I change setting 29 to less than 21V? I've heard that this skews readings even further, or should I have my batteries checked?

 

I have a Raspberry Pi setup that queries the inverter, I pipe that data into InfluxDB and visualise it in Grafana along with some scheduled tasks (via cron) to set the inverter to Utility of SBU based on my home automation. What I'm hoping to achieve is something like this:

5 hours before sunrise set the inverter to SBU along within a specific specific voltage for setting 12 and have the panels charge the batteries through the day when the sun rises

3 hours before sunset change setting 12 to a specific voltage, once reached set setting 1 to UTL

 

If I had to buy the battery monitor, how does it interface with the inverter to not simply change back to grid/batteries based on voltage?

 

Link to comment
Share on other sites

Interesting, tested with the charge settings today using solar only (OSO) with a battery cut-off back to grid at 23.5V.

It stopped using the batteries when it reached that voltage, but without charging from the grid or solar this evening the voltage went back up to 24.5V with the capacity being reported on the inverter at 82%.

Any ideas?

Link to comment
Share on other sites

7 hours ago, iops said:

the voltage went back up to 24.5V with the capacity being reported on the inverter at 82%.

The SOC guess-o-meter calculation can be different when in line mode; using SOL output charge priority mode in the evening means you would be in line mode.

In battery mode, I believe that the calculation would result in 2% + 2 x (245 - 235) x 1% = 22%.

Oops, I just checked and it seems that if you're in line mode but not utility charging, then it should be as per the battery mode. Perhaps the 24 V models guess differently.

There is no way that 12.25 V per 12 V battery module represents 82% SOC.

Link to comment
Share on other sites

Does anyone have suggestions on settings for a BMV 700 connected to a Raspberry Pi running ICC that will result in my Trojan T105 battery bank of 2 parallel 48V strings charging fully. In desperation I have set setting 32 on the Axperts (2 in parallel) to 90 minutes which kind of works depending on the SOC. The BMV settings are as recommended in the Victron video but it seems the BMV does little if any control and switches to float irrespective of the time setting for bulk charge time. That is when setting 32 on the Axperts is set to Auto and time factor on BMV to 90 minutes. @Coulomb and @plonkster  you are probably amongst the right people to ask. Perhaps I will with your help solve this ongoing problem before the grim reaper comes to visit. 

Link to comment
Share on other sites

Not sure if I'm overcomplicating this with my home automation tie-in. Are my panels enough to charge my 8x 100aH batteries?

It's a sunny day today, load on the inverter is 390W with solar generating 640W. My understanding is that 250W is going to the batteries? PV amperage is between 20 and 25 Amps.

 

Ran a test this morning and saw less substantial voltage drops, my batteries had been at 27V for about 4 hours before switching to DC at 05:15. At 10:15 I was down to 24.3 on the batteries with an average of 400W load. So that's roughly 30% DOD?

Link to comment
Share on other sites

That is one of the reasons the BMV is recommended. The data that you have is very limited and therefore almost impossible to judge from. As I remember the Axpert did report the battery amps charge / discharge. You could use that to determine the AH drawn and charged and that could , in a way  , give you an idea as to if you are at least putting back what you have used.

Have a look at the graph below. You will see that the batteries take as much as there is available until about 10 then start to drop away even though there is still plenty to offer from the panels.  You could look at you data and check if the same happens on your system. If you get a similar curve you could assume that your panels gives enough juice.

But again there is a huge amount of could , guess , might in there  ....

 

2020-05-02_134032.thumb.jpg.5a115056053a69845dc0109bffa387bf.jpg 

 

Link to comment
Share on other sites

1 hour ago, PaulF007 said:

You could use that to determine the AH drawn and charged and that could , in a way  , give you an idea as to if you are at least putting back what you have used.

Right, it is possible to work this out in software. If you take samples of the amps, and multiply it by the time between the samples, and add them all up, you get an Ah figure.

This is equivalent to integrating the area under the chart, where the chart shows ampere on the vertical scale and time on the horizontal. The process of splitting this into little time slots with a sample point, multiplying together (essentially working out the area of the rectangle this forms) and adding them up is called newtonian integration. And it is exactly what the BMV does internally... it just has a very good current shunt while the Axpert estimates the DC value from the AC values (as far as I remember it has this in common with the Multiplus).

Link to comment
Share on other sites

Thanks @PaulF007, it does make sense - I was seeing the BMV as something to action off as opposed to just monitoring, but it makes complete sense in situations like this.

I think I need to get the installer out so they can run some measurements. Even on sunny days it looks like the battery is being used constantly throughout the day. My panels only ever seem to cover the load and not really charge the batteries. I've included a screengrab from my Grafana graph yesterday as an example.

I'm not opposed to changing my solar wiring, it's currently 3S3P but I'd need an external MPTT charger since the 3kVa Axpert seems to have quite limited solar input.

 

Thanks again for everyones input, I'm learning slowly.

 

I'll need to double check the DC charge metric tomorrow, I'm fairly sure I've seen that PV has an input into DC.

Screenshot 2020-05-02 at 15.24.32.png

Link to comment
Share on other sites

I assume the blue line is the solar output , if so then yes you have a challenge as you have a almost perfect bell curve witch means that you are using your system optimal but also that your batteries will most likely not get charged well.Also you base load is quite high (500 w) during night time , but that is just a point of view...

Maybe show a graph with you solar vs battery watts , this should also give an indication.

Link to comment
Share on other sites

3 hours ago, PaulF007 said:

I assume the blue line is the solar output , if so then yes you have a challenge as you have a almost perfect bell curve witch means that you are using your system optimal but also that your batteries will most likely not get charged well.Also you base load is quite high (500 w) during night time , but that is just a point of view...

Maybe show a graph with you solar vs battery watts , this should also give an indication.

I've uploaded some additional graphs, I don't seem to have a battery watt measurement so I'm calculating whatever the Amp reading is by 24V. If I look at the battery amp charge vs discharge it looks like my batteries work hard during the day hardly getting charged at all.

Night usage is high because I have a small homelab here with some server and network gear, it's a fairly constant 350W draw 24/7. Going to charge by my batteries overnight and see how long they last tomorrow evening. If the batteries are OK, what are the next steps - more panels, inverter settings change or look to maximise my 6 panels?

 

Maybe I'm expecting too much from my system, what I'm hoping to do is charge the batteries to 100% and cover load where possible with the panels and have the battery run either early morning or in the evenings to cover that load.

Screenshot 2020-05-02 at 22.06.00.png

Screenshot 2020-05-02 at 22.06.11.png

Screenshot 2020-05-02 at 22.11.27.png

Link to comment
Share on other sites

If I read the 'Usage' chart in your screenshot correctly, then your PV only barely covers your load between 10h00 and 15h00. This would explain why your batteries are not really being charged as there is basically no excess PV available beyond your load

What I find strange in your chart sis that your PV barely gets to 600W (except for 2 short peaks). You write that you have 6x 330W panels, which means that you should be getting significantly more PV from your panels (the bell-curve shape of your PV makes me believe that it was a relatively cloud-free day, correct?). So I would start with checking why you are only getting so little out of your panels

One other note: You write that your panels are wired 3S3P - but that would mean having 9 panels...

Edited by wolfandy
Word missing...
Link to comment
Share on other sites

14 minutes ago, wolfandy said:

If I read the 'Usage' chart in your screenshot correctly, then your PV only barely covers your load between 10h00 and 15h00. This would explain why your batteries are not really being charged as there is basically no excess PV available beyond your load

What I find strange in your chart sis that your PV barely gets to 600W (except for 2 short peaks). You write that you have 6x 330W panels, which means that you should be getting significantly more PV from your panels (the bell-curve shape of your PV makes me believe that it was a relatively cloud-free day, correct?). So I would start with checking why you are only getting so little out of your panels

One other note: You write that your panels are wired 3S3P - but that would mean having 9 panels...

I've also been thinking about that, it's a lot to wrap my head around so I've probably been Googling too much as opposed to just asking for help and getting the installer back out. Worryingly that is what how they said they'd wired it up. I'll follow up on Monday and have them come out to check. Thanks for your help!

Any recommendations on how to have them wired up? My inverter MPTT seems quite limited, specs below:

Max power: 1500W

System voltage: 24VDC

Operating voltage range: 30~115VDC

Max solar voltage: 145VDC

 

The more I look at it, the more I seem to see the short falls of the Axpert. Maybe it's worth getting a Voltronic MPTT charge controller and then switch to one of their inverters later.

Link to comment
Share on other sites

8 minutes ago, iops said:

Any recommendations on how to have them wired up?

You would need to post the detailed specs (VOC, etc) of your panels for us to help you better

11 minutes ago, iops said:

The more I look at it, the more I seem to see the short falls of the Axpert. Maybe it's worth getting a Voltronic MPTT charge controller and then switch to one of their inverters later.

I am running a 5kVA Axpert and have also added a Victron MPPT (and taken the panels completely off the Axpert). Mostly because I wanted to add more panels, but also because my Axpert was running quite hot under full PV. The Victron MPPT allowed me more PV and at the same time it takes strain off the Axpert (I am hoping that will prolong its life). I also thought about going fully Victron, but (a) I read somewhere that the inverter part of the Axpert is actually quite decent and (b) I would then not just require the inverter but also a GX device for system monitoring/management (which further adds costs). I have therefore decided to run my Axpert until it dies and only then do the full switch to Victron

Link to comment
Share on other sites

Ok a couple questions:

What is you max expected solar output :
Is all your panels looking in the same direction :
What direction is your panels looking at:
What is the tilt of your panels:

These should give us a understanding of what your setup looks like. My brother in-law has a system installed. (Quite nice) but as he is living in one of those townhouses roof space is a challenge. They ended up putting panels on both sides of the roof this resulted in him not getting full potential from them as they will be exposed to different solar "strengths"  not much that could be done about it , apart from adding another mppt. But I agree , your output does not seem to match your description of the system..

Also a couple of other questions 

On 2020/04/30 at 9:32 PM, iops said:

With my 8x 12V batteries I have a total of 400aH

What is the size of your batteries : 
How are they configured :

Reason for asking is that I have came across it where people switch series and parallel calculations . So 2  x 12v 50ah batts in series will give you 24v 50 ah whereby 2 x 12v 50ah in parallel will give you 12v 100ah ( Quite a difference in ah and "capacity" if you get it wrong) 

Last thought - There is a possibility that some of the batts could already be shot. Seeing that you are discharging them heavily , if they dont get charged properly they will not last long at all. 
You should maybe consider charging them for at least a full day without load just to make sure they are fully charged. A good indication will be if they cannot charge more than 1or 2 amps in the middle of the day ..    


 

Link to comment
Share on other sites

19 hours ago, iops said:

Not sure if I'm overcomplicating this with my home automation tie-in. Are my panels enough to charge my 8x 100aH batteries?

It's a sunny day today, load on the inverter is 390W with solar generating 640W. My understanding is that 250W is going to the batteries? PV amperage is between 20 and 25 Amps.

250 W into a 400 Ah battery is very small, only 250/50 = 5 A, or 5/400 C = .0125C (C/80). So it would take 80 hours of sunny days like this one to fully charge the battery from dead flat, or "only" at least 16 hours from 80% SOC to 100% SOC.

Bluntly: no, your panels are not enough.

Link to comment
Share on other sites

21 hours ago, ebrsa said:

the BMV does little if any control and switches to float irrespective of the time setting for bulk charge time.

Sorry, I know next to nothing of ICC.

21 hours ago, ebrsa said:

Perhaps I will with your help solve this ongoing problem before the grim reaper comes to visit. 

Is the reaper coming for your battery or for you? 😮  (Or both?)

Link to comment
Share on other sites

@iops, you cannot simply divide the Wh by 2 in order to get 50% DoD. You have to take the Peukert component into consideration when you use LA batteries. For that it is very easy to use the BMV.

Link to comment
Share on other sites

@Coulomb the problem is the BMV700 settings I think. It tells ICC when it detects that the batteries are fully charged and ICC then only tells the Axperts to switch the charging voltage to float. So the BMV700 settings is the issue. I have increased that to 58.4V and not 0.2V less than bulk charge voltage as recommended by Victron. Will see how that works out over the next few days.

Since I hit 82 during the past week and the batteries are about 3 years old, it will probably not be them that falls victim to the reaper first if I can get the charging working correctly. But I live in hope. 

Link to comment
Share on other sites

18 minutes ago, wolfandy said:

He states that he has 6x 330W panels installed - yet he barely gets more than 600W out of them (and hence the low W into the batteries). Does that not sound strange to you? 

Ah. I read so many posts... Yes, 1980 W nominal is far more reasonable, and one would expect at least 1500 W on a good solar day a few hours either side of noon, assuming reasonable angles, negligible shading, etc.

Link to comment
Share on other sites

Huge thanks for everyone's input so far, it's massively appreciated!

13 hours ago, wolfandy said:

You would need to post the detailed specs (VOC, etc) of your panels for us to help you better

Panel specs attached, turns out they're actually 300W, not 330W.

I'd also like the option to add more panels at some point, might take the plunge and do something similar to you with the Victron MPPT.

6 hours ago, PaulF007 said:

What is you max expected solar output :
Is all your panels looking in the same direction :
What direction is your panels looking at:
What is the tilt of your panels:

I've included some pictures of the panels on the roof, they're on a carport roof - it was determined that would be the best place for them. The sun rises where the complex is and sets behind our roof on the right hand side, looking at the panels from the gate. Probably a bad way to explain that, but it's all I know :)

 

6 hours ago, PaulF007 said:

What is the size of your batteries : 
How are they configured :

From what I've been told, the batteries are wired up to create a 24V system, so two 12v batteries to make up 24V in series for all of them and then parallel? All 8 batteries should be in Series Parallel. I'll have to double check with the installer though.

 

6 hours ago, PaulF007 said:

Last thought - There is a possibility that some of the batts could already be shot. Seeing that you are discharging them heavily , if they dont get charged properly they will not last long at all. 
You should maybe consider charging them for at least a full day without load just to make sure they are fully charged. A good indication will be if they cannot charge more than 1or 2 amps in the middle of the day

They've been on AC charge since last night 21:30, battery voltage went up to 28.2V in the first hour at a 40A charge, stayed there for 2 hours and dropped back down to 27V and has stayed there, charging at 2A since 00:30 this morning.

DJI_0114.jpg

DJI_0124.jpg

300W-EnerSolDatasheet.pdf

Edited by iops
Link to comment
Share on other sites

At that angle you are going to suffer during winter months. Summer will be fairly good in the middle of the day so that could account for the poor performance. 

Considering the batteries. Looking at your graph again I can see when you charged them at night and the curve suggest that they had a proper charge.Personally I suspect that they might then not be in the best of condition but I could be completely wrong. The only way to know for sure is to get a battery tester and test each battery under full load and check the recovery time.  Just make sure that they are all well charged before you do the test else you will not get proper results. 

If you can not test them at this stage I would not use them during night time. Being in town you are not going to lose that much in terms of costs. You could then use them for backup when load-shed arrives again.  

Link to comment
Share on other sites

3 hours ago, iops said:

Panel specs attached

Ok - I am not an expert on this, so I am hoping some more senior forum members will correct me if I am wrong

Your panels are 72 cells and 44.8V VOC. If they are connected 3S, then my guess would be that you are getting too close to your max operating voltage of 115VDC and your Axpert starts to derate

But somehow I do not think that this by itself will explain that you get so little out of your panels

3 hours ago, iops said:

I've included some pictures of the panels on the roof, they're on a carport roof - it was determined that would be the best place for them.

Do you get shading from anywhere on your panels during any part of the day?

32 minutes ago, PaulF007 said:

At that angle you are going to suffer during winter months. Summer will be fairly good in the middle of the day so that could account for the poor performance. 

+1. But do you think the angle will cause him only getting 600W out of his 1800W of panels? I simply do not have enough experience to be able to answer that...

Link to comment
Share on other sites

12 minutes ago, wolfandy said:

But do you think the angle will cause him only getting 600W out of his 1800W of panels? I simply do not have enough experience to be able to answer that...

There is a lot of variables that we just do not know. As you noted shading even on just one halve of one panel will have an effect on the production of the panels and with that flat angle there is a good chance that there will be shading of sorts. 

Link to comment
Share on other sites

5 hours ago, wolfandy said:

If they are connected 3S...

If so, disconnect them immediately! You will damage the inverter. With 25 degrees C the voltage will rise to 44,8 x 3 = 134,4 V and with that temp coeff. of 0,37% it means a Voc of 49,13 at 0 degree C per panel. You can only use these panels in a 2s3p setup. If I may give you an advise put the 2 panels next to the concrete wall in series and the 2 middle ones as well, that will give you the best results when there are shaded panels.

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...