Hi all

 

Im quite excited to share this with you. So I connected all my house lights to my inverter. Yesterday afternoon I started my logging and then ran this report this morning

 

Displays battery capacity and battery voltage overnight till this morning.

 

Lowest battery capacity = 71 % but started recovering after everybody went to bed. See orange line. It recovered to about 89 %

Battery voltage stayed pretty constant as per the blue line.

 

By 9 am this morning my solar panels had already charged the batteries back to 100% capacity.

 

Hi what monitoring software are you using.. Is it may be the watchpower that's on the CD that came with inverter?

Yes I am using watchpower but i am also monitoring it with software that I wrote to confirm the values. I am still in the progress of finishing the program. Why you asking?

Your graph looked very good. Watchpower appears to suffer when running 24/7/365 to handle all the data. Wish they had a option one can set when looking at stats the user can change default year to example default to "today"

Also another future which will really be nice if one can display more than one inverter on same graph since it is logging all inverter when having them running in parallel mode.

Hi Warren 

 

Man that is a good feeling and it just does not wear off. I have bought diesel for the generator today for the first time in 6 months.

 

You can ignore the Axpert SOC it is wildly inaccurate. Look at your charging current. If it drops to 2-3A and the sun is shining brightly, your batteries are charged. My 260 Ah the charging rate drops down to 1.2 - 1.3 A and the batteries accept 70 W or so. At that level my batteries are fully charged. Even my Victron battery monitor reckons the batteries are fully charged a bit early but I am making the necessary adjustments slowly. 

Went for a bit of Victron training yesterday. That voltage thing was stressed a few times, that you cannot determine whether a battery is full by just considering voltage. Apparently that's what the Axpert does :-)

The software I am writing will allow for calibration. As long as you have proven values tested with other devices like a multimeter and clamp amp meter on ac and dc you can calibrate the software to compensate. For example I noticed that the axpert watts currently used are about 30 watts less than the actual watts displayed on my efergy energy monitor. If you have 2 of the 3 values you can calculate the third value. For example with volts and amps you can calculate watts.

I also noticed that the watchpower software and the inverter display also display some of the values incorrectly. I actually decompiled the watch power software and found even more protocols. The watchpower software was written in Java by sun micro systems. The watchpower software also has logic bugs so some of the values are incorrectly displayed.

I also noticed that the watchpower displays the battery fully charged however the inverter displayed it was still charging so I need to see which one is correct.

Hope this info helps others.

Your efergy meter may not be entirely accurate. It cannot accurately measure any inductive load. Warren, you, scottday and jdp are all trying to improve on WatchPower (which is pretty crap) you should do so co-operatively. I have done a little programming all through my teen years and early adulthood but I am not about to tackle something like that (my time  and the dinosaur's time has come and gone!). 

Hi Warren

Man that is a good feeling and it just does not wear off. I have bought diesel for the generator today for the first time in 6 months.

You can ignore the Axpert SOC it is wildly inaccurate. Look at your charging current. If it drops to 2-3A and the sun is shining brightly, your batteries are charged. My 260 Ah the charging rate drops down to 1.2 - 1.3 A and the batteries accept 70 W or so. At that level my batteries are fully charged. Even my Victron battery monitor reckons the batteries are fully charged a bit early but I am making the necessary adjustments slowly.

I realized that the batteries were charged so I scanned through the decompiled watchpower and it looks like a bug in the software. I am trying to map out all the logic faults hardware and software wise. It just takes time which I don't really have. I only get to play with the software late at night or on weekends. Very tough life when you do programming for a living and then get home and carry on. lol

Program something worthwhile and sell it back to Voltronic.

Your efergy meter may not be entirely accurate. It cannot accurately measure any inductive load. Warren, you, scottday and jdp are all trying to improve on WatchPower (which is pretty crap) you should do so co-operatively. I have done a little programming all through my teen years and early adulthood but I am not about to tackle something like that (my time  and the dinosaur's time has come and gone!).

I must admit that the efergy monitor is quite accurate from my calculations between amps drawn and voltage which I tested a few years ago as I was doubting the accuracy myself at one point.

O for the love of Pete! I will pay you all to do this! I know at least one other non forumite that will pay for a decent app perhaps integrating victron bmv readings.

Must do aggregate calcs as well and graphing.

Like seriously man.

I with Edmund on that one - add readings from BMV and you have a winner.

I must admit that the efergy monitor is quite accurate from my calculations between amps drawn and voltage which I tested a few years ago as I was doubting the accuracy myself at one point.

 

The efergy meter, if it is a clamp type meter, will only measure apparent power (VA) and not real power (W). For pure resistive loads, e.g. the kettle or geyser element, the apparent power and real power is exactly the same, but as soon as you have inductive and/or capacitive loads you have to measure the relation between the voltage and current at any given time, known as the power factor (PF) and take that into account as well. For mixed loads in your home, you have to take the power factor into account, otherwise your measurements are pretty much useless.

 

For example, I've measured the standby power on one of my old TVs - the apparent power drawn was 18VA, but the real power drawn was 3W. Luckily we pay eskom for real power used, but in an inverter system, this TV will draw 18W from either the batteries or PV in standby mode. 

For example, I've measured the standby power on one of my old TVs - the apparent power drawn was 18VA, but the real power drawn was 3W. Luckily we pay eskom for real power used, but in an inverter system, this TV will draw 18W from either the batteries or PV in standby mode. 

 

Really? From previous discussions, I was under the impression that it will draw 3W from the batteries, BUT when working out the total load on the inverter (rated in VA) you have to use 18 and not 3.

Eg, I have the same thing with LED lamps with a poor power factor. 5W, but closer to 20VA. They will use 5W of power from the battery, but if I had a 100VA inverter (for arguments sake) I could only put 5 lamps on it and not 20... :-)

Am I correct?

To be honest and blunt I dont really care if it is a bit inaccurate as long as it works and I dont drain my batteries to fast and I have lights when loadshedding starts and I also have a ROI.