2kw 48v off grid system with problems

Craig4321

Hello to all.
as you can see I am new to the forum but not new to solar power.
I have been living off grid for 14 years so I have gone through the good and bad of self sufficiency.
I live on Cyprus a small island in the Mediterranean,so lots of Sun...
my 48v system consists of..
24 x 12V 80w panels=2kw (6 strings of 4)
morning star. Tristar ts-45
wind turbine and controller= 2kw max (never seen more than 12amps in summer,not much wind)
5kw studer inverter/charger (6.5kw genny)
24 x 2v sunlight batteries 480ah 9 years old in bad shape.
and a brand new set of 8 rolls s600 6v batteries .not yet connected
the system has been running well for the last 14 years but my 2kw solar array has only been charging at 21.6 amps maybe the last 5 year's...
my questions are .
1. Would this be due to a faulty panel. Or the bad batteries.
2. Would an Mppt controller give me a better charging amperage with the system as is?
any help would be greatly appreciated
as I am planning to add another 1kw to keep my new batteries in good shape

vtmaps

Craig4321 wrote: »

1. Would this be due to a faulty panel. Or the bad batteries.
2. Would an Mppt controller give me a better charging amperage with the system as is?
any help would be greatly appreciated

Welcome to the forum, Craig.

The problem certainly could be your batteries... if the batteries are old and sulfated they would have a higher resistance and that would limit current flow. Since you purchased new batteries, try hooking them up and see what happens. If the problem persists, I can walk you through the necessary troubleshooting.

By the way, your new batteries will be ruined if you don't connect them and get them charged up.

--vtMaps

zoneblue

Im assuming your panels are 36cells, thus 6P4S is a workable enough arrangement for a 48V bank, providing your array cabling is sufficient.

Thats quite a bit of pv for a ts45 though. Youd expect about 40A from that array on a good day, which is pretty close to the working limit of the controller.

With your presumably warm and sunny weather you will see some thermal derating, however the only way to check the panels is to individually test their Voc and Isc with a multimeter. Even if your bank is tired, i wouldnt have thought that would limit bulk charge currents that much. Check for corosion on connections, after 14 years, and pre mc4, might be issues there maybe.

Craig4321

Thanks for replies
Do I have to disconnect each panel or can I clamp meter them...my panels are connected in connection boxes (old style.)as to cable,I am using very heavy duty armoured cable, not sure the gauge but VERY thick .
I was thinking maybe the controller might be lowering the charge because the batteries reach 50 to 51 volts very quickly ,long before the sun is at peak, that's why I got new ones..I have only had the rolls batteries 2 days, the company forgot to deliver connectors planning to replace them today.(if connector cables arrive)
I just wanted to make sure I would be charging the new ones enough with the 21amps...I use between 6 and 8 kwh with sunshine from 7am up to 7pm.
what about an mppt controller, looking at a morning star mppt,would that increase charging?
thanks
craig

BB.

Craig,

Your solar panels seem to be outputting a bit less than they should. Guessing:

• 80 Watts / 17.5 Volts Vmp = 4.57 amps
• 4.75 amps * 6 = 27.4 amps

So--That could be dirty panels, or one or more bad wiring connections. Or a bad panel or even simply panels getting old.

It could be one of the array strings is not producing any current and the rest are OK (5x4.75 amps = 22.9 amps)

If you can get your DC Clamp Meter (AC clamp meter will not work) around once cable from each string (if you put + and - cables from same string through the DC clamp, the meter will read ~0 amps--the + and - current flow canceling each other out).

If you do not have a DC current clamp meter--You can simply disconnect one array terminal at a time and read the charge controller meter for the current reduction from each disconnected string.

You leave everything connected normally and isolate one wire from a string & measure the current from each string. Ideally, all strings should be producing roughly the same amount of current (with the same amount of sun--Try to measure each cable quickly--so that sun/clouds/etc. do not vary the actual production).

If all strings are producing the same amount of current but less than rated current--The panels may be getting old.

There could be other problems too (if the solar panels are hot and the battery charging voltage is high--It is possible that the Vmp-array is being reduced and not high enough to pump full current to the battery bank).

I am guessing you used to see higher current from the array years ago?

You may want to think about what you are going to do next with your system... A 600 AH @ 48 volt battery bank needs around 5% to 13% rate of charge, and 10% or more is good for a full time off grid home/usage.

You have lots of sun, so 5% rate of charge (30 amps) could work--But it would be best if you charge during the day and use the batteries at night. If you charge and use power during the day, then you are not really even going to see 5% rate of charge.

Ideally--You should have a 10% rate of charge to quickly and vigorously recharge the battery bank.

When the system was built 80Watt panels where probably cost effective for your region... Today, there are 140 to 250+ Watt panels that are pretty cost effective... However, these larger panels, many times, will not work well with a PWM controller (you need Vmp-array roughly 70-72 volts). The "typical" large format panels these day are around Vmp~30 volts -- Which does not "map well" to a PWM controller on 48 volt battery bank (30 and 60 volts is too low, 90 volts is too high to be very efficient with PWM controller).

A MPPT type charge controller works very well with these large format panels--But you need to either "mix" old array + old PWM controller and new array + MPPT controller on the same battery bank (will work fine). Or, get rid of your old array+PWM controller and get a new array+large MPPT controller. (possibly sell old panels+controller to somebody else).

You will probably have to do several paper designs and see what works out best for you.

Do you have any idea of how much power you use per day (Watt*Hours or AH @ 48 volts)? It looks like you get around 4 hours to almost 7 hours of sun per day (winter/summer)--How much do you use the genset these days (when all is working well)? Would a larger array be helpful for you (less generator run time)?

These days, with solar panels being "cheap" and battery banks being "expensive"--Putting up a large solar array can help extend your battery life (higher charging current, less deep discharge, less generator run time during bad weather).

-Bill

Craig4321

I definitely feel that I would need more charging power ...if my panels were charging as they should then I would see at least 27 to 31 amps.from the full 40amps .as they do get quite hot.. (Air temp rangers between 32degrees Celsius and 42in summer)
I will check the panels and connections,I agree that there is a fault somewhere....
that's why I thought to add another 4 x 250w panels with an mppt in parallel to my existing controller.
or maybe
5 x 200w panels with a pwm controller in parallel to my existing controller (cheaper option)
or even
swopping my old PWM controller with an MPPT one..Would that not increase the 21amp production without adding another parallel set?

I consume between 6 and 8kwh daily


in the early days when my panels were producing around 30 to 34amps I never used a generator all summer.
I used it about twice a week in winter with only 5hours Sun, even less when there was good wind for my wind turbine.
As an afternote edit...I have just spoken to the company that first installed my system and the tech guy tells me 21 or 22amp production is about right for the age and time of year(HIGH TEPERATURE)...don't know what to think...very confusing..
craig

vtmaps

Craig4321 wrote: »

I was thinking maybe the controller might be lowering the charge because the batteries reach 50 to 51 volts very quickly ,long before the sun is at peak

That is another sign that your batteries are the problem. I am not convinced that your panels or controller are at fault. Of course, they might be at fault, but the batteries alone could explain the charging problem.

that's why I thought to add another 4 x 250w panels with an mppt in parallel to my existing controller. or maybe 5 x 200w panels with a pwm controller in parallel to my existing controller (cheaper option) or even swopping my old PWM controller with an MPPT one..Would that not increase the 21amp production without adding another parallel set?

With your new batteries and a consumption of 6-8 kwh per day, you will probably benefit from more charging capacity. Your plan to add 5 x 200 watt panels is not viable with either a PWM or a MPPT controller. 200 watt panels typically have a Vmp of about 30 volts. You will need to add them in strings of 3, and use an MPPT controller.

Note: you could add them in strings of 2 with a PWM controller, but they would only contribute to bulk charge (which is when you need high current). After bulk charge, when the need for current declines, your existing array could take over for absorption charge.

By the way, in a hot climate such as yours, swapping out your PWM controller for an MPPT controller on your existing array could make your problems worse. Your existing array, when hot, barely has high enough voltage to charge your batteries. An MPPT controller needs more headroom (higher input voltage) than a PWM controller. If you want to use an MPPT controller with your existing array, you should reconfigure it as 5 panels in series. Of course, with only 24 panels you need to lose 4 of them or gain one more.

--vtMaps

Craig4321

Thanks for the info..That sounds like what I thought
so to conclude, the best way to go if I understood it correctly,according to my budget is to get 4 x 250w panels 3 in series plus 1 parallel(Is that correct?) and hook them up to . Tristar mppt. t_45 parallel to my existing system.
I can always add another 2 panels in the near future.
Rgds
Craig

vtmaps

Craig4321 wrote: »

get 4 x 250w panels 3 in series plus 1 parallel(Is that correct?) and hook them up to . Tristar mppt. t_45 parallel to my existing system.

No!

When you put panels in series we call that a string. One panel, by itself, can also be a string. When you make an array by putting strings of panels in parallel, each string must have the same voltage. Therefore you cannot put a string of three panels in parallel with a string of one panel. If your string length is 3, you need either 3 or 6 or 9 or 12 panels, etc. to make up an array.

If you have more than one array (each array being composed of one or more equal voltage strings), each array needs its own controller.

by the way, the advice I gave in my previous post (where I suggested a string length of 3 with an MPPT controller) was based on Vmp of about 30 volts which is typical for panels of that wattage. I need to know the exact spec of the panels you are considering before you finalize your plans.

--vtMaps

Craig4321

The panels are 250w grid tie ...vmpp 30.75 and uoc 37.31
originally I thought uoc 37.31 was enough for 2 in series
I was thinking to join 2 panels in parallel then adding the other 2 panels in series
charge controller mppt Tristar.

vtmaps

Craig4321 wrote: »

The panels are 250w grid tie ...vmpp 30.75 and uoc 37.31
originally I thought uoc 37.31 was enough for 2 in series
I was thinking to join 2 panels in parallel then adding the other 2 panels in series
charge controller mppt Tristar.

You typed "uoc", but that should be "Voc". That is the open circuit voltage. As soon as you try to draw power from the panels, their voltage will drop closer to Vmp. Two panels in series will have a Vmp = 61.5 volts. That is at standard temperature. When the panels are in the hot sun, their Vmp will drop too low to charge a 48 volt battery bank. Therefore you need to have a string length of 3 panels in series with a Vmp = 92.25. With a Vmp = 92.25 you will need an MPPT controller. You will need to choose between either 3 or 6 of those panels.

As I mentioned previously, you could use 4 panels with a string length of 2 and a PWM controller, but the panels will only provide power during bulk charging. That's not necessarily a problem as long as your existing array is functional and can do the absorb charge.

--vtMaps

Craig4321

O.k thanks.....I will update as soon as there is something to update about.
kind regards
Craig

BB.

A quick shot at the math:

Nominally, a 450 AH @ 48 volt battery bank used for 2 days of storage (no sun) and 50% maximum discharge:

• 450 AH * 48 volts * 0.85 AC inverter eff * 1/2 days storage * 0.50 maximum discharge = 4,590 WH per day average load

Based on 5% to 13% rate of charge for your new battery bank:

• 450 AH * 58 volts charging * 1/0.77 panel+controller deraring * 0.05 rate of charge = 1,695 Watt array minimum
• 450 AH * 58 volts charging * 1/0.77 panel+controller deraring * 0.10 rate of charge = 3,390 Watt array nominal
• 450 AH * 58 volts charging * 1/0.77 panel+controller deraring * 0.13 rate of charge = 4,406 Watt array "typical cost effective" maximum

And then based on your usage... Worst case assumptions:

• 8,000 WH per day * 1/0.52 off grid system efficiency * 1/4.0 hours per day average winter sun = 3,846 Watt array minimum

Or, another way of looking at it, with your present (if it was working well) 1,920 Watt array:

• 8,000 Watt*Hour per day * 1/0.52 system eff * 1/1,920 Watt array = 8 hours of noon time equivalent sun per day

It appears that with a fixed array tilted about 55 degrees from vertical for Cypress that you get around 4 to 7 hours per day of average sun (winter to summer) depending on where you live (solarelectricityhandbook).

So, it would look like you probably use closer to 4-6 kWH per day or less on average... Especially if you were doing fine with a 1,920 Watt array:

• 1,920 Watt array * 0.77 panel+controller deratings * 4 hours average winter sun = 5,913 Watt*hours per day if using most power during day (drawing most power directly from solar array)
• 1,920 Watt array * 0.52 panel+controller deratings * 4 hours average winter sun = 3,994 Watt*hours per day if using most power overnight/bad weather (drawing power from battery bank)

Of course, your power usage may be variable by season--use more power during summer (fans/refrigerator/water pumping/etc.). Less power in winter.

-Bill

vtmaps

BB. wrote: »

A quick shot at the math:

Nominally, a 600 AH @ 48 volt battery bank used for 2 days of storage (no sun) and 50% maximum discharge:

The Rolls S600 batteries are 600 ah at the 100 hour rate. But they are also 450 ah at the 20 hour rate. I think Bill's rule-of-thumb calculations (correct me if I am wrong) are usually based on the 20 hour rate, so the array sizes he calculated should be multiplied by 0.75.

That probably won't change his conclusion, based upon your consumption, that you could benefit from some more panels. That was also my conclusion (as mentioned in post #7 in this thread).

--vtMaps

BB.

Oops... Missed that... Thank you vtmaps.

I have fixed the above post with 450 AH @ 20 hour discharge rate battery specs.

-Bill

Craig4321

Thanks Bill...
I am in agreement with your calculations,including the fact that the batteries are 450ah at c20
so I have opted to use correctly sized panels to match another pwmTristar ts-45 making a new array of 1kw.
I cannot afford a mppt +panels right now so I will see how it all goes and maybe upgrade a bit more later.
I can always start the generator up a couple of mornings a week just to pump things up a bit.

Craig4321

Thanks Bill.....I agree with you on your calculations and yes 450ah at c20.....
I am going to use a Tristar ts-45 and match the correct panels to it creating a second 1kw array.
I cannot afford a mppt+panels at the moment .
if the new array+existing system is still not enough then I will have to use the generator a couple of mornings a week to pump things up a bit. I can always upgrade a little more when the cash is available.
kind regards
Craig

zoneblue

At 15 years, those panels deserve one of two things:
- per panel Voc, Isc check. So itll take you a couple hours, but then youll know exactly whats going on. Exactly.
- flog them on ebay, and buy a new array, much much cheaper these days, no matching problems, and lots less conections to worry about.

Craig4321

That's probably the best for them,they've served me well. I am installing the new battery bank today so I will See what kind of charge is going in and give the panels a check-up....then take it from there.
The new array of panels won't be arriving for another week .so I will probably be using the generator for a few days to help charge if there is no improvement.

Photowhit

I think, you are seeing the results of bad batteries/batteries on their way out. VT Maps has caught all the signs.

I would configure your system with the new batteries and see what kind of charging. checking the strings with a clamp meter would be nice, but if you don't have one, and you have proper DC breakers, you could disconnect each string one by one. This is NOT normal procedure, but the breakers are designed to handle it. In this manner you could check the difference read at the meter.

Also if you have not purchased the 250 watt panels yet, you might check around, it is not uncommon to find 35-37 Volt VMP panels in the 250 watt range (at least in the USA) You might just squeeze 2 strings of 2 - 250 watt panels on the 45 amp charge controller you have, particularly in your hot climate. I would rather see you switch the a 60 amp charge controller if you need to add capacity and the 45 amp won't work. This would give a a good backup if problems arise.

Craig4321

No I haven't actually purchased the panels yet but the company that first installed my system 15years ago is going to find me 35-37v 250w panels to match up to the pwm controller...I have tried myself but nearly all the solar companies here are trying to sell with very little info given,even saying it's not possible to connect 2 tristar ts-45s parallel amongst other nonsense....I have received more info and knowledge here on this single post than talking to over 20 companies in Cyprus..
my old system installer is the only one who truly wants to help,who coincidentally is a New Zealander,Zoneblues fellow county man.

Craig4321

Update....finished installing batteries....charging current has gone from 21.6amp to 24.1amp so there has been a slight increase..the sky is clear but there is a lot of humidity so light intensity is not so great plus it's 43degrees out there....that's pretty hot..

Photowhit

Sounds like the old panels are in the ball park!

Glad someone is helping you... and not themselves. A TS-60 might be easier to read off production and is only 20% or so higher cost than the TS-45, at least here.

Craig4321

Yes the batteries made a difference..they are rolls s605 not s600 after close inspection..I doubt there's much difference between them other than the s600 has been discontinued...the Tristar ts-60 costs about 110Euro more here and the 45 will do for now...I can sell it easily here if I need to upgrade..
one thing I noticed was when connection my negative terminal ,I tried the spark test before connecting and strangely there was a spark. All batteries were connected correctly and the system had been shut off from the main switch and all other circuit breakers.pannels etc...I discovered by disconnecting the positive feed to the batteries on the wind turbine controller I was able to connect the negative terminal to the batteries without sparks..
reconnecting the positive back on the turbine controller was o.k but with a small spark. The wind turbine was not turning at the time.....now it is and producing 256w.
Is there a problem with my connections...other than the fact that I should have put a breaker on that feed.

Craig4321

If my wind generator is 48v 40amp, must I use a 60amp non fuse type.?

BB.

In general, you never want a wood turbine to blow a fuse/breaker. If a breaker pops in high wind, the turbine can over speed and self destruct

-Bill

Craig4321

Oh...o.k
at the moment it's connected directly from its controller to the batteries, without any sort of breaker...do you think I should leave it like that?
it's been that way since the system was installed 15 years ago.

BB.

More or less--You are looking to prevent fire (breaker protects wiring shorts from being fed by the battery bank). And to prevent the turbine from over speeding (shedding blades, causing a tower collapse).

So--You don't want either to happen.... But it something does, you want it to fail as safely as possible. I.e., breaker opens and prevents a wiring fire, and turbine is mounted far enough from house that falling blades/nacelle/tower does not hurt anyone.

And that is why a wind turbine would usually have several methods to "safely" shut down. Besides the load on the turbine (battery bank/dump load) keeping the turbine from over speeding, the turbine may have:

• furling (turn from high winds)
• feathering (blades change pitch)
• mechanical brake in turbine
• Aerodynamic braking (blades "flutter" in high winds/rpm)

As long as wiring is in good shape, then the braker/fuse should never trip.

However, remember that fuses/breakers are designed to be "unreliable" (open if there is excessive current before the wiring fails). Breakers and fuses have been known to false trip as they age.

You can do other things (like an over voltage shunt--If the turbine voltage/rpm goes too high, a relay/switch shorts the output of the turbine to, hopefully, bring it to a stop).

That is the problem with engineering solutions--Simple problems/questions can become "overly" complex. In the end, you have to look at what can happen in A, B, and/or C happens, and what the results would be. For example, is a fire worse than a turbine throwing a blade?

You can mitigate some of the issues through design--For example, use heavier gauge wiring/larger rated fuse/breaker so they will never trip except in the most dire of circumstances. Do not mount a large turbine right next to a living structure/play ground, etc.

-Bill

Craig4321

Thanks Bill....I need to think ahead....
for a clearer picture, the cables running from the wind generator. Were originally for 5kw unit but it was obliterated by a tornado type thing which is very rare here...
the present generator is a cheap ebay purchase called istabreeze 2000.
to be honest, it's not very good, it produces but I don't think it will manage 2kw.
for anyone intending to buy one,DONT.
going back to over spin issue they say it's protected but even if it did then it's own internal wiring would probably melt...it looked too thin to cope with anything higher than 40amps..
I think your recommendation of an overvoltage shunt is probably a good way to go.
Thanks for the advice.
Rgds
Craig

Craig4321

I have found luxor 200w 24v(37.16 vmpp 44.27voc)panels New at a great price..
would these go well with my Tristar ts-45?
they are monocrystalline I think.
2 strings of 3 parallel for 48v

vtmaps

Craig4321 wrote: »

I have found luxor 200w 24v(37.16 vmpp 44.27voc)panels New at a great price..
would these go well with my Tristar ts-45?
they are monocrystalline I think.
2 strings of 3 parallel for 48v

You want to put three of these panels in series to make a string, and then put two of these strings in parallel? If so, you should have an MPPT controller.

Or are you saying that your string length is two panels in series and that you will have three of these strings in parallel? If so, you would probably be better off with a PWM controller (because the PWM needs less headroom and two of those panels in series will just barely have high enough voltage for a 48 volt battery bank).

--vtMaps