Hi TTT

Ja my issue was with the PV cables. Max PV string voltage is 50VDC which is stupid but it is what it is. Max amps 50A so can safely go to 4 panels and if Etienne is running for 8 hours that should be enough power. If it is a 24/5 operation then he needs a much bigger system. The only hesitation about other kit over Voltronic is price. My system has been running for 9 months now and the interest on the saving amounts to a substantial amount of cash. If I had bought a Victron Quattro and a Outback I would have paid about R50k. I got the equivalent for R10k. That is a R4k per annum saving in interest alone (2 panels a year). 

Having bought myself into a few corners, I have learned to figure the next purchase in such a way that you can slowly build to the higher end stuff. 

7 minutes ago, The Terrible Triplett said:

See, pushing the Voltronics PWM to max power, with the PWM being inefficient to boot and it being uncomfortably high DC amps, it is being pushing very hard.

And with it being in a container office their are the temps to consider on top of the PWM operating flat out (read heating up).

For I vaguely recall reading that you guys have mentioned the Voltronics are not the best at cooling, and this one is 800w, therefor probably even less so?

So I would split the inverter from the charge controller, as the 800w is more than enough for the load, just not for the batts, and at 12v ... cheaper solution.

Use it, or not, Etienne's choice at the end of the day.

See the thing is this.

He does have the Inverter with PWM that is up to the job being able to do 50A already paid for. (Maybe not the best at it but it can)

Now he just needs to add more panels

Your suggestion would be Price of more panels +- 5k for the separate MPPT +(wiring to the bank)

Mine Price of more panels + cost of thicker cables <---- Win win like you always say ROI

 

PS no problem with the cooling on the inverters just a bit of a design that could be changed as with all cooling solutions

22 minutes ago, The Terrible Triplett said:

Would it?

Also Axpert 1kva 24v with MPPT R4700 on first hit, will still work on his setup
 

I know I am a cheapskate but if you look at the number of Axperts around I am in good company. 

My current install I would move over to my staff cottages if I could find a 7-8 kW Axpert.  I haven't really the space to parallel two Axperts (could do but it would not be very neat). When that day comes I will get Trojans but the rest of the kit is overpriced considering what you get when you buy an Axpert. Yes sure its software is dicky and it doesn't have features I would like to have but ultimately it is saving me R100/day  and the ROI is second to none.

 

Good solid point all over.

 

On the lighter side. Ask me? TTT, wil you buy a Voltronic unit?

I will answer: No.

Why, would ask, the ROI is very good?

I will answer: I already have a Victron. 

27 minutes ago, The Terrible Triplett said:

Good solid point all over.

 

On the lighter side. Ask me? TTT, wil you buy a Voltronic unit?

I will answer: No.

Why, would ask, the ROI is very good?

I will answer: I already have a Victron. 

And if you did not have the Victron

Just now, viper_za said:

And if you did not have the Victron

But I HAVE a Victron. 

3 hours ago, viper_za said:

If he wants to go this route it would be cheaper to do a different Voltronics model with MPPT

Cheapest inverter would be the MKS 1k, which is 1) a 24V inverter and 2) costs about 5k.

So the challenge is to get him an mppt for less than 5k. Which is easy really... the Bluesolar 100/30 does that easily and you have R1000 left to take the wife out with.

If you serie-connect those 250W panels, you should end up with an open circuit voltage of around 90V, and at max power around 72V, 14 amps. So it will leave you room to double the size of your solar array.

I must say though, it remains a troubling setup.

1. MPPT from 70V down to 12... not really recommended. It would have been better to use 150W 12V panels and more of them, cost would have been the same more or less, but you can make a 900W array with 2 strings of 3 panels (which incidentally is how mine is configured).

2. 500W peak power into 400Ah batteries. So rule of thumb says 15% of capacity, or 60 amps. At 500 watts, you're two thirds of the way there. Let's say you go to 20% capacity, that means about the max you should put on there is 900W of panels, which would have been best configured as 6 panels of 12V 150W each, that would have had a more sane open circuit voltage of 66V, Max power around 54V.

3. If you had a better configured solar array, you could get away with 2 x Victron BlueSolar MPPT 75/15. That would have cost 3k, which is a little over half of a new inverter :-)

So sadly it seems I agree with others here: Best is going to be thicker cables and more solar.

Also, your load is not matched with your production. That's why it runs flat :-)

2 minutes ago, plonkster said:

Cheapest inverter would be the MKS 1k, which is 1) a 24V inverter and 2) costs about 5k.

Scratch that. If you've got two 12V batteries in parallel... bite the bullet and go 24V. Get a 24V inverter. With a built-in MPPT. In other words, get the 24V Axpert. Smous the 12V one on gumtree. Call the loss school money. Then add more solar panels.

Guys, sorry for not monitoring, but you've all run away with a assumption that is incorrect. I stated in my first post that max draw is 40%. I only switch those lights on when testing. Normal daily draw is about 10%. Zero draw over weekends or at night. In this scenario, I get about 9 day use, and as reported earlier, batteries fail on a Monday or Tuesday. Illogical considering that they've had the whole weekend of summertime sunshine, with zero load.

Settings and current status attached.

Chris: I loaded that Dropbox link while at our head office this morning, and forgot that the proxy server stops Dropbox. Now correctly synchronised, so the link will work again: https://dl.dropboxusercontent.com/u/30819730/AxpertKS-1-5KVA_manual.pdf

 

24 minutes ago, Etienne said:

Normal daily draw is about 10%.

10% of what? 10% of inverter capacity? :-)

When your installer says you're "overloading" the system he doesn't mean you're overloading the inverter. He means your loads take out more than the solar panels put in.

I suspect he's wrong though. That would be a good explanation... but now that I read your post properly, you do say that they last a week once you charge them properly, indicating that your load is around 5kwh a week. You have enough solar panels to make that every 2 and a half days.

One other thing. As far as I know, 250W panels are 24V panels. Peak power production will be at around 36V, so peak current is around 7 amps. When you connect them to a 12V battery (this is what PWM does), you pull them down to 12V (from 36), so it operates at 7 amps/12V which is only 84 watt. Times two. So 160W solar array, 5 hours a day (on average), 0.8kWh.

So picture this, batteries are empty by Friday. Over the weekend you put back around 2kwh (so they are just below 50% SoC). And then they run out about mid-week? Sounds about right...

I reverse my position. Get the MPPT :-)

Edit: The more I think about it, the more I'm convinced. It's not that the charger in the inverter doesn't work. It's the wrong kind of charger for the panels outside. A Victron BlueSolar 75/15 will cost you 1.5k. If you put the panels in parallel, they will make 7A each, 14A total, and you're well within the max open circuit voltage. This is the cheapest good quality MPPT, can't go wrong with that.

Edit2: Oh crap, I need some sleep. The 15A rating is for the battery side. Scratch that... fact remains, PWM is the wrong kind of controller when the panels are that high above the battery voltage.

Ok. That is making more sense. Not what I want to hear. It actually p!sses me off a bit. What I have in essence been supplied with is a 160W system. On a good day. For interest's sake, I attach the quote that the guys gave me. It was sized to be almost 2.2kWh per day. What you're telling me is I can get about 0.8kWh per day.

The trick comes with the 250W panel providing 7 amps at 36 volts but stepping down to 12 volts, but still only giving 7 amps. That is a three times reduction in power output. I clearly need an MPPT.

The more I research this stuff, the more I realise that these installers don't know jack...

Ettiene Du Plessis.pdf

Problem is everybody is jumping on the solar bandwagon and nobody checks what and how they do it. I had 3 installers at my place and when I asked them technical questions they could not answer me. So I did everything myself except putting the pannels up on the roof as I am only one person and that is a job for a team.  

Hi,

Just a thought what about the "Shading" resistors in the panels? Could they not be causing a problem? Where the Volts look ok, but no amps.

That quote includes a 30A Microcare MPPT. If you have that installed... then my PWM argument should be ignored. If you're using the PWM controller in that inverter (and once again, I don't know the axpert, but I know the lower-end models are PWM), then my argument holds.

I see what they did. They are right, you are overusing the system, from THEIR perspective, but they are wrong due to practical application of the maths. 

Estimated load = 2,2kWh
2 x 250w panels = +-400w incoming (due to losses)
5.5 peak hours = 2,2kWh coming in.

So in their minds, they are 'right'. If batts are flat, the only reason can be, you are overusing.

But they are wrong, prolly due to inexperience for they are assuming a LOT:
- that every day is 5.5 hours of perfect sunlight;
- you are NOT going using one watt over 2.2kWh per day;.

To solve the problem quick, swap the PC for a laptop. Pc 150w becomes a 40w, you save 880wh per day, then your are at 1.3kWh load with 2.2kWh panels.

Or get more panels onto that Microcare controller as has been suggested.

5 hours ago, Etienne said:

 

Hi Etienne 

Is the charging happening via the Microcare MPPT? If so then you need to be careful using the AC charging capability of the Axpert. You may inadvertently over charge using solar from the Microcare and AC via the Axpert.

If you are using the Microcare MPPT I would not expect to see any PV Voltage in Watchpower. Something is seriously wrong. You either have been charged for a piece of kit that has not been included or somehow PV panels that should be connected to the Microcare MPPT are connected to the Inverter. 

Out of interest. If the MPPT and Eskom are both connected to charge the batteries, either or both will scale down or not charge at all. 

Seen that with my system, where one controller laat loop, the other controller does nothing, zero to the batts.

It looks like the power from the panels is being fed to the SCC built into the Axpert. Yet the  Axpert is not setup to charge using solar, charge source priority is set to Utility. Set Program 16 to CSO and then see what happens.  I put together a 4 page guide to settings for Axpert which you can download from this forum. It looks like you have paid for a Microcare MPPT and they have not installed it, or some clown wired the PV panels to the inverter and not the Microcare and it is sitting there doing nothing.

It would appear that the solar panels carry the load and any deficiency is made up from battery power but there is no charging of batteries with excess solar not used to cover the load so the batteries slowly run down.

Since the MPPT solar charge controller is more efficient I would move the panels over to the MPPT. JUst make sure you cannot over charge the batteries via solar and AC.

Watch out for the lower-end MPPTs. Given the price on the quote, this is not the high-end LCD model (with a 150V Voc), it's the LED model with Voc of 50V. It's already somewhat close to the max, those panels have a Voc of probably around 45V. A bit close for my own liking, but each to his own :-)

VOC of panels 38V so panels in parallel.

Ensure Program 1 is set to SBU.