Although I just installed my own PV system, I'm new to the PV systems and I think I have a few problems I hope someone can help me out with.

I live in Austin, TX (about 30.5 degrees longitude) with 16 Kyocera KC158 panels (2.5kW worth) installed on my 30 degree pitched roof. I use the inverter in sell mode and have the modules wired in series in 4 sets of four. I have 8 brand new MK sealed gel batteries (8G27).

It has been extremely pleasant here the past week the system has been running, i.e. 78-80 degrees light breeze and not a cloud in the sky. The most I've been able to get out of this system is 11 Amps AC according to the inverter. I checked the DC output with an Amprobe and the max DC amperage is 27.4 (which is in-line with what the inverter says).

I've played around with the float voltage and the sell voltage in menus 10 and 17 respectively and it does indeed make a difference in the max amperage output. I initially set the float (& sell) to 49 volts thinking more juice for the grid that way. At that time I was only able to get 7 amps DC out of my system. I was also told to do set the float voltage low by a friend that has an identical inverter, but different panels. On his system on the same days he was getting close to 21 amps DC.

Since I noticed the higher voltage yielded a higher power output, I've been thinking ways to get the voltage higher at the panels.

I have tested all of the panels individually, each array, and all together and on these nice days I'm getting 107 to 109 open circuit voltage. When under a load the panel voltage becomes exactly (really close) to what the inverter is set to. I get these same numbers at the J-box on the roof and at the DC disconnect.

I'm at a loss on what to try next to get more power out of this system. If you do the math I'm only getting about 38% of the rated capacity of the system - sad. I understand I'm not going to get nameplate and expect that. But I should be able to get at least 16 amps AC out of this system.

I have noticed one thing. These panels are listed as 12V panels, yet their specs say they produce max output (6.8A)at 23.2 Volts! That would be 92 Volts in series. I'm thinking that if I can get the voltage higher at each individual panel I may be able to push more power. In other words, I'm thinking about rewiring and trying 5 arrays of three panels. Using my float of 56.4V into 3 panels will yield 18.8V per panel rather than the 14.1V per panel I'm getting now under load.

I guess here in lies the question. I don't think it matters the number of panels I'm wiring in series, but just so I'm not exceeding the rated specs, right?. Any thoughts on this? Since I have only a basic knowledge of electricity and how these PV panels work, does anyone know if 3 in series on a 48V system will have any negative effect? Anything I'm overlooking?

One final comparison. I recently helped with a sunny boy install here locally and they used the same panels (16 of the same panels I have) on a 2500 Watt system and they are getting right at 2000 Watts out of their system. I know I can't compare apples to oranges, but I think it may be partly because those individual panels are running at roughly 17V.

Any help or direction will be most helpful.

Anthony

The KC-158 is NOT a 12 volt panel This is the cause of all your issues.

You are feeding a 48 volt inverter with a string that has a mpp point of 93 volts at standard conditions (for comparision , the KC-120's have an mpp of 69 volts standard condidtions , 55-60 volts in real world tempatures).

The best solution would be to get the MX60 outback charge controller and leave the array exactly as it is. Then set the float voltage to match what you set the MX60 output too. With 2500 watts STC and using a MX60 and SW , you should get about 1900 watts peak to the grid. (only assuming the batterys are fully charged)

Gridtie inverters are better than SW style for selling and you have the the additional loss of having the charge controller

You could rewire , but without mpp tracking , you never get anywhere near the best power output ... mppt is a must

You have one obvious first problem. You have a bad match with the modules (as you have apparently discovered) You could try wiring them in threes instead of fours as you noted but there is probably a better solution.

You should do a little more research, read about "Maximum Power Point Tracking" MPPT controllers (google for "MPPT charge controller"). Since you didn't mention what charge controllers you have I am guessing you don't have any or if you did you are not using an MPPT controller.

In a battery system the the voltage system will operate at the battery voltage (as you also have discovered). A MPPT controller will let the panels run at a higher voltage.

I suspect you will want an Outback MX60 or RV Power Products SB3048 when you are done reading.

On another front, why did you choose gel batteries?

arnold

Thanks Arnold and Solar Guppy! I sincerely appreciate your responces. Today, I have read numerous sources and talked to several people about this and they all say the identical same thing you two did. I will order the Outback MX60 tomorrow. It must not be a coincidence that you both mention the MX60. I guess this was a tough lesson.

Thanks for your time on this.

Oh, the gel batteries. I was told they would hold a charge longer "on the shelf", and since I'm using them as backup they should be pretty much on the shelf. Also, since these are in a somewhat enclosed container in my garage I wanted minimal gasing. I also got a really good price on these NEW batteries.

I'm willing to listen to any arguments either way. I'm still learning.

Thanks again for you help!

Anthony

My main concern about gel batteries is they are sensitive to being overcharged. The default settings on the inverters and charge controllers tends to be bad for them so you have to make sure you get the settings correctly.

For this application (standby power with a PV system) I would have suggested AGM (Acid Gas Mat)/VRLA (valve regulated lead acid)/starved electrolyte batteries. They are nicely between conventional flooded acid batteries and gel batteries.

arnold

Actually , The KC-158 would be concidered a ~18 volt panel , as it has 48 cells in series vs 36 cells the KC-120 has

If you only put two in series vs the four now done , the mppt point will be BELOW the battery float voltage and you will get low power output since the panels will be sourced.

No harm , just allot less power than is available from the panels when operated at mpp

[ 10-22-2003: Message edited by: Solar Guppy ]

I think I understand. You simply can't "make" panels produce more voltage than their Vmp. I assume the battery voltage won't back feed to the panels in a case such as this?

Ok, I've done lots more reading and have another question about these KC158G panels. It's clear that I've bought the wrong panels and the MX60 should take vare of my problem - can't wait till it gets here.

Since I've posted my original message, I've found someone with a near identical situation as me - huge voltage mismatch between the panels and battery voltage. He read and showed me the links, that some people are calling the KC158's 24 volt panels and are wiring them up as such.

Again, I'm new at this and so I'm not sure would happen if you were to do this? I've not been able to get anything out of Kyocera on this... Anyone know what would happen if you took a 23 volt panel and ran it at say 26 to 27 volts? Overheating problems I would assume? Since reading this my "buddy" wants to try it rather than buying a MPPT. (Case of cheap B&$t*ard syndrone I think) I've asked him not to do for his sake until I can get some info. I'm not going to risk it with mine!

Thanks again,

Anthony

P.S. Thanks again Arnold for the info on the batteries. I did want the AGM's initially, but they were getting rid of these as an order cancelled and they were stuck with lots of these. They let go of them for a nice price like I mentioned.

There is a lot of personal opinions around about battery preferences, but I have a good bit of battery experience, and I would say that the gel batteries you bought are a high quality battery well suited to your use. AGM batteries are also fine. Flooded (liquid electrolyte) deep cycle batteries like the common L-16 are NOT good for grid tied systems. They take much more current to keep charged, and that gets much worse as they age. I would be satisfied with your purchase and not consider any change.

I believe that the SW4048 system will always hold the batteries at float voltage, so you must set that voltage at a level acceptable for the batteries. The factory gives that voltage as 13.8 to 14.1 per battery or 55.2 to 56.4 for your 48 volt system. I would set it at the lower part of that range since they are under constant charge.

Also important for any sealed battery, if the temperature of the batteries varies much from 70 degrees F (>5 deg F) you should use a battery temperature probe for both the MX60 and SW4048. It would be a good idea to contact Outback, or the dealer you purchased it from, and make sure to set the temperature compensation rate on the MX60 the same as for the SW4048.

Although it appears that you have already ordered the MX60, an alternate would have been to add two more panels for a total of 18 and then wire them into six sets of three series connected panels. Of course, this is still possible with the MX60, since it will easily handle 3000W

Robert [img]images/smiles/icon_smile.gif[/img]