I started my solar installation 12 months ago with Cheap Chinese Inveter/MPPT and have since upgrade some or the parts (The last outstanding part is the Inverter).  I want to make the jump from a 24V system to a 48V system when I get a new inverter.

The plan is as follows:

Stage 1: Add another 2 x 250W Solar Panels (Limit charge on MPPT to avoid overcharging)

Stage 2: Increase battery capacity by adding another 4 x Trojan T105 

Stage 3: Add a 48V Inverter

The Process has one minor issue - Stage 2: Adding 4 New batteries to +/- 12 month old batteries which is not ideal, so I need a way how to measure the current available Ah to make a call on whether it is viable or not.

Now here is where it gets tricky - During the last 12 months we moved and for a 4-5 month periods the batteries was not in use (I did charge it twice during the period); In the beginning I used a Chinese MPPT charge with limited data/settings and only got the BMV700 4 months into my new hobby (Which indicates a total discharge of 127kwH from the batteries (i.e Around 100 cycles at 25% discharge vs 2,400 according to the data sheet)

Any suggestions on testing the state of my batteries????

H Louw,

Jumping from 24v to 48v, why? Both have pro's and con's, 48v biggest pro being your inverter size.

Reducing your max inverter load requirement makes 24v more attractive in that you need 50% less batteries, and therefor everything become cheaper.

 

From everyone I have spoken to over the years, never match batteries that are not less than 6 months difference in age, depending on use.

Also, I would look at T105RE's.

An idea is to put you bank up for sale here ... and get all new.

That is what I would have done.

The 3 reasons for going 48V

- MPPT has double the capacity on 48V vs 24V

- Wire size/breakers/fuses on the battery side of the connections

- Ability to go 21.6kwH storage capacity using Trojan T105 (2 Strings of 8 batteries)

Not sure if it has changed but on my previous calculations the cost for T105 per KwH was actually cheaper than the T105RE

 

51 minutes ago, The Terrible Triplett said:

Also, I would look at T105RE's.

An idea is to put you bank up for sale here ... and get all new.

TTT is in the market for a second set of REs. 

Cool.

I just view solar as something you first reduce, keep it as small as possible, unless you can make money selling back to Eskom ... and that ain't gonna happen soon. 

T105RE's supposedly are better due to newer tech, and should therefor theoretically should last longer.

BUT, we all have different needs, views. 

I for one wish I had taken the 48V route back in the day. Don't get me wrong, a 24V system has plenty of advantages, and back in the day when I had to install the best system I could with the money I had (in the face of the looming load shedding and running a home office) it absolutely was the right decision... BUT... If you can afford the 48V route for heavens sake don't go with 24V! :-)

One possible option would be to mix the old and new batteries, add some balancers, and simply deal with the fact that your new additions are probably going to die an earlier death because of what you've done. It's probably not the most economical option, but it might be the best option in terms of cash flow, as your little unequal bunch might just last 18-24 months which might be long enough. As long as you know what you're doing. Of course we cannot in good conscience tell you that there is nothing wrong with such a move :-)

48 minutes ago, The Terrible Triplett said:

I just view solar as something you first reduce, keep it as small as possible, unless you can make money selling back to Eskom ... and that ain't gonna happen soon. 

Another reason on this is.
I don't really care how much I make back on my setup. What I want is not another freezer full of rotten meat because of power failures. Thus less dependent on Eskom
So no reason to keep it small, want to do basically all I did before the system was installed (Okay you need to adjust your usual ways)

Go 48V as @plonkster said, it just makes more sense.
The T105 RE's cost more but you get more cycles out of them

Edited June 20, 201610 yr by viper_za
It does not hurt not spending that much on Eskom anymore

Hi,

Agreed 48v is the way to go. But the battery cost was the concern, I found the Trojan TE35 6volt units to be the best priced and least cost per kWh at the time, and the cycle rating was higher than most other models.

Anyone else using the TE35's? So far been very happy with them, topped them up a bit once every three months or so, and try to keep the dod around 25% or less.

Our load typically around 5-600watts typically, at night going to around 900watts.

With the +/- 500 watt load they seem to hold up fairly well. I don't expect to get a full nights power from them though, it's more for standby - and with the split db would turn off non essentials at night and load shedding.

Have been tempted to add another set to up the amp hours or maybe two. But haven't gone there yet

Did add strings 4 and 5 to the Axpert and with the low winter sun it's definitely helping .

4 minutes ago, Arandoza said:

Have been tempted to add another set to up the amp hours or maybe two. But haven't gone there yet

Thinking about this also

4 minutes ago, Arandoza said:

Did add strings 4 and 5 to the Axpert and with the low winter sun it's definitely helping .

What is your total wattage on the PV now and your string config (panel brand watt etc) 

What is your total wattage on the PV now and your string config (panel brand watt etc) 

The total pv wattage is now 3750watts. 3 X 5 strings 250 watt panels - specs below

Pmax 250

Vmp 31.02

Imp 8.02

Voc 36.99

Isc 8.62

Model : ODA250-30-P

The Axpert has shown around 3600 watts pv at times.

Prior to adding stings 4 & 5, when only had 2250 watts pv it would occasionally go as high as 2600 watts being generated, and was often giving 2350 watts?

The only challenge i started to run into now in winter is very cloudy days like last week Monday and Tuesday here in pta, and also with the sun so low in the sky and longer shadows found higher generation numbers started much later in the morning, so I wanted to improve that by compensating with additional strings.

Also just started using AICC in the wee hours of today, and then Telkom had the main brakfontein exchange cable stolen at 4am, so the posting to pv output and emoncms stopped

Was looking forward to seeing how these systems plot the data and then comparing it to some

Of the others on pvoutput in za.

I guess the cable theft problem will take a few days to resolve ? I hope not much longer.

7 hours ago, Arandoza said:

The Axpert has shown around 3600 watts pv at times.

This is good to know, thank for the update Arandoza

2 hours ago, viper_za said:

This is good to know, thank for the update Arandoza

+1

I will be adding a 5th string to mine fairly soon. When I weld up the frame I will use solar power whereas last time we had a gennie clattering in the background.

Some other thoughts regarding the panels is that one has to have the entire or at least the biggest part of the array facing where you need it 90% of the time.

The reason I say this is that I tried pointing two of the strings in the east and west direction - hoping to get a noticeable increase in the early morning and late afternoon pv generation. However with only two strings one in each direction and the fact that the sun has to go thru more atmosphere early morning and late afternoon, I have started to think that this is not such a practical arrangement.

Also if one increased the strings, it may mean that you almost end up with three arrays, east , north, west. So it may make more sense to look at a 12 panel frame with tracking I am starting to think this would give the best price performance result? But how to do it ?

We have sections of flat'ish roofing, and the erf is not really big enough for ground mounting without shading, to overcome the shading I would need to raise the panels around 7m up, and then I would almost have a wind turbine

So I have been looking for plans on how to do this on a flat roof without increasing the weight loading in a big way. So far haven't found the ideal design.

Also on the strings some of the articles suggested not to over do by more than 25-30% which seemed to be a safe percentage, unless, one was running multiple Mppt or pwm chargers. As there seemed to be a heat buildup concern.