Jump to content

Is it possible to calculate potential PV output from Axpert stats?


Hyarion
 Share

Recommended Posts

I've bought a house recently which has an Axpert inverter, solar panels and batteries.  I've programmed one of my raspberry pi's to monitor the stats on the inverter and have been watching the graphs with interest.

So obviously there are times where the inverter decides to stay on Eskom or switch back to Eskom (e.g. morning sun rising, clouds, sun setting).  What I'm curious about is whether it's possible to calculate how much power the panels could be outputting when it's running off Eskom.

To give an example:

Between 12:00 and 13:00 the PV Voltage was around 114V, PV Current was 24A and everything was running off Solar.  Approx. 1.2kW load.

At 13:00 the inverter switched back to Eskom and continued to 14:00.  PV Voltage was around 136V, current was 1A.  Approx. 1.2kW load

During that time period of 13:00-14:00, is there a way I could calculate how much the panels could have produced?  Or worded another way, could I calculate what load would have been supported?

Link to comment
Share on other sites

Thank you Calvin, I'd prefer a method that uses the actual data from the inverter though.  I'm as much interested in what the estimated output should be as much as I am in what the out currently is.

For example, how does the inverter know there's sufficient solar power to switch off from the grid?  Does it require more info than is made available over the serial connection or can it be calculated?

Link to comment
Share on other sites

@Hyarion - From my 2 years of learning my Axpert based solar setup, I can offer the following up...

1 - Your cut off / return to Eskom between 13:00 and 14:00 is not due to load... This is due to over voltage from your Solar Panels. I have the Axpert King which has the same specs as most other Axpert units in the voltage range you are mentioned... The unit operates between 60 and 115v (from 30v for the 3k version) but has a max voltage of 145V...  The units will start to clip / reduce PV production to prevent damage as it gets closer to the 115v point but over that point it will stop production. Over 145v you run the risk of blowing the MPPT.

2 - You can not predict (any more than a thumb suck) of the production you are wasting. I have tried everything to work our based on the readings, what could have been generated. Early in the morning, my voltage is up around 80v but amperage is low... during peak periods the voltage drops anywhere in the 70s but the amperage is much higher... while they do correlate, I have yet to find out a method of predicting generation without an active load connected


To solve your problem as described in point 1... You need to look at your panel specs vs the configuration... I had a long debate with my installer who wanted to install my 355w panel in a series string of 3 which would have resulted in voltages very similar to yours. In the end, I insisted that my panels were installed only 2 in series and have not looked back once.

What panels do you have and how of they configured? Also which Axpert unit do you have?

Link to comment
Share on other sites

50 minutes ago, Hyarion said:

Thank you Calvin, I'd prefer a method that uses the actual data from the inverter though.  I'm as much interested in what the estimated output should be as much as I am in what the out currently is.

For example, how does the inverter know there's sufficient solar power to switch off from the grid?  Does it require more info than is made available over the serial connection or can it be calculated?

Sadly I am not aware of any way of getting this data from the inverter.

I was also interested in getting this, and ended up making a pyranometer (using a small panel, a load resistor and a microcontroller with an analogue input) to measure it directly.

It worked surprisingly well, as long as you size the load resistor so that the panel operates in the linear part of it's curve.  The main problem is that it responds almost instantly to eg. clouds moving in front of the sun, whilst the Axpert's SCC reacts several seconds later. 

Link to comment
Share on other sites

1 hour ago, Calvin said:

Sadly I am not aware of any way of getting this data from the inverter.

This data would probably also never be the same even if it could be measured due to everyday the conditions of clouds and loads are different. I also tend to agree with Jay-Dee it sounds like over voltage from the panels.

 

 

Link to comment
Share on other sites

@Jay-Dee Thank you, that is very interesting regarding the over-voltage!  I should have someone coming to inspect the setup in the next week or 2 to quote me on a few changes, I will get him to check how the panels are configured and quote for altering the setup.  While the previous owner was meticulous in keeping documentation on everything in the house, the installer he used gave minimal details.  The invoice from the installation a few years ago just says "8 x 250W Solar Panels".  No brand or model number, no details about how they are configured or anything like that.  I'll get them checked out and hopefully can have them reconfigured in a better setup.

The inverter is an older Axpert 5Kva (Mecer Branded): SOL-I-AX-5M

 

@Calvin Thank you for the update, it's a pity it's not doable using the stats from the inverter.  A pyranometer sounds pretty interesting, sounds like something I need to add to my list of projects with microcontrollers.  😁

Link to comment
Share on other sites

1 hour ago, Hyarion said:

The inverter is an older Axpert 5Kva (Mecer Branded): SOL-I-AX-5M

I think that is the MKS, which is good news for you - @Coulomb and friends have developed pathed firmware for those that eliminates some of the worst bugs, and adds some nice features.  Might well assist in stopping the drop out you saw.

1 hour ago, Hyarion said:

8 x 250W Solar Panels

Something does not add up - you have voltages that are consistent with 3-in-series (3S) 72 cell panels, but you have only 8 panels, (perhaps even 60 cell ones) which means 4S2P or 2S4P.  Will be interesting to see.

 

Link to comment
Share on other sites

11 hours ago, Calvin said:

but you have only 8 panels, (perhaps even 60 cell ones) which means 4S2P or 2S4P.

My wild guess: 4S2P of 60-cell (perhaps even 54-cell) panels. So then 3S would be better, but you'd need an extra panel of very similar characteristics to make it 3S3P. 2S of 60-cell might well be too low in voltage.

I would not expect a 145 V absolute max SCC to disconnect at 136 V, unless it's doing so because it's too hot. @Hyarion, what is your main inverter firmware? Some of the older ones (not the oldest ones, that were too noisy by far) didn't run the fans enough. There is now good evidence that running higher (>100 V) voltage into these solar charge controllers causes them to heat up. Running 73.00e patched firmware and turning the fans around to blow air upwards might fix the problem. You would also be able to read the maximum temperature of 4 sensors from the LC Display.

Link to comment
Share on other sites

@Coulomb Thank you for your response.

Inverter firmware is 72.40.  I will look into upgrading it.

It's possible my graphing was not catching the voltage soon enough.  Existing libraries for monitoring the inverter didn't quite suit my needs so I've been building a custom one but havn't had time to work on it properly, so it's only been recording every 2min.  I managed to work on it this morning and is currently recording every 15 seconds (and could record more frequently if I wish).

Looks like it's a combination of voltage and temperature.  Going back a few days I can see that:

On days that it spends more time on eskom than solar, the voltages are sitting around 140~144V, due to my 2min intervals at the point, I'd say it was switch to Eskom due to overvoltage.

On days where it runs mostly on Solar (and voltages are around 136V), the temperature slowly rises until it hits 67deg and then it drops back to Eskom until the temperature is back to < 60deg.

I will update once I've had an installer take a look later this week.

Link to comment
Share on other sites

1 hour ago, Hyarion said:

Looks like it's a combination of voltage and temperature.  Going back a few days I can see that:

On days that it spends more time on eskom than solar, the voltages are sitting around 140~144V, due to my 2min intervals at the point, I'd say it was switch to Eskom due to overvoltage.

On days where it runs mostly on Solar (and voltages are around 136V), the temperature slowly rises until it hits 67deg and then it drops back to Eskom until the temperature is back to < 60deg.

I think it is probably only due to voltage - your temperatures are going down due to the SCC tripping because of the over-voltage.  My inverters (Kings, but similar SCCs) would throttle on SCC temperature once the SCC PWM got to 87C, but that was a gentle reduction to keep the temperature at 86-87.  It did not trip the SCC  (A key difference is that you have only about 2kW of panels on the inverter - I have 4.3kW - so I don't think that there is enough power to get the SCC up to 87C).  On the other hand, if the PV voltage got too high the SCC tripped, which is what yours appears to be doing.

The good news is that, once I rewired my panels from 3S to 2S both problems went away.  Voltages are now at about 75-90, and the SCC temperature never gets anywhere near 87C.

 

Link to comment
Share on other sites

From everything above, the first thing would be to them to check how the panels are wired. As Coulomb mentioned, you are most likely wired current 4S2P and if this is correct and you can get an additional similar spec'ed panel, the ideal would be (depending on the actual confirmed panel details) to wire this as 3S3P.

I personally have 12 Panels wired in 2S6P which equates to around 80v & with a max of 80A 

I am busy installing at my parents house 12 Panels wired in 3S4P will result in around 90v with a max of 42A


Both of the above configs, will work very well within the clipping parameters of the inverter as the panels are of completely different spec's. The key is to get the specs of your panels, how they are wired and go from there

Link to comment
Share on other sites

Thank you all for your replies and assistance.

I had a solar installer inspect the system last week and indeed it is wired as 4S2P.  VOC of the panels is 38.1V, so it's hitting overvoltage even easier than expected.  He picked up some other glaring issues with the system as well.  I'll be using him to sort out all the issues as soon as possible and then things should run smoothly.

 

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.

 Share

×
×
  • Create New...