I had a system installed the middle of last November, and overall I am pretty happy. But I think that I should be getting more out of my system. I am grid-tied with the following in Southern San Diego county.

60 Kyocera 120 Solar Panels - ground mounted facing 170 degrees. ( not sure of the angle )
3 Xantrex C-40 Charge Controllers
16 Trojan L16 Batteries( not sure if P or H)
2 Xantrex SW4048, each connected to a GTI

The most kwH I have sold back so far is 20. Monday and Tuesday of this week I sold back 17, but today only 13. ( I determine this by reading my meter religiously every morning when I leave for work, and every evening when I get home ).

When I am home during the day, I have noticed that often at almost 1:30 PM on the dot ( 2:30 after daylight savings ), that the inverters lose sync with the grid often (i.e. the grid-tie and inverting lights are yellow, AC1 in is blinking green ) over a one hour period or so.

During this hour, the inverters will re-sync after 5 minutes or so, but then lose sync again after a couple of minutes, and then repeat this cycle. The GTI's click incessantly during this period.

Since this is right in the middle of the day, I figure I am losing a good 4-5 Kwh that I could be selling back, and I think also explains the difference between the 17 Kwh days and the 13 Kwh days.

Does this sound familiar to anyone? Is there anything I can do about this?


Thanks, [img]null[/img]

Well if you were using Sunties , you would be putting 30-40 kWr into the grid easy ...

I've never used a SW , but if there going offline , you need to find out why. All inverters that sell to the grid have to say within 211-264 (105-132) VAC and the frequency must be within 59.3-60.5 hz

The most common reason for going offline is the grid voltage goes out of spec ... this can be caused by using undersize wiring on the AC mains .. inverter feed power , voltage rises beyond spec , inverter goes offline

Since both SW's are doing it I would say the issue is on the grid side

But , if there is somthing wrong on the DC side , like a charge controller not working , bad wiring , ect .. then you can get low voltage cut offs

who installed the system ? , have you told the SW's to sell maximum power using the batterys to verify you can pump 8kWr into the grid ? if you discharge the batterys , what charge rate do you get.

Your also going to lose power since your NOT doing mppt , this can be 20-30%

Hope this gives you some ideas , my 48 KC-120 with 2 sunties is doing 24-28 kWr on sunnydays right now ...

Henry [img]images/smiles/icon_razz.gif[/img] [img]images/smiles/icon_razz.gif[/img]

While MPPT helps, I do not think the gain is anything near 20 to 30 percent. When exposed to direct sunlight, the panels get to hot to allow this much gain.

Robert

Robert,

Checkout Mpp tcharge contoller.

and this by SandiaSandia mppt

Solar panels generate there peak power in a very narrow window. When charging batterys , or in this case looking for a voltage point concidered charged enough to sell power , it is almost for sure NOT at mppt point.

In very hot weather , say in the 90's and above, yes the mppt point get very close or even below the battery charging voltage.

Even durring the day , the mppt point will vary 6-8 volts , forcing the array to be at the battery voltage is a sure way to not harvest available power

I stand by my 20-30% statement , with documentation and design experiance , do you have something i could read to suggest otherwise ?

Henry [img]images/smiles/icon_razz.gif[/img] [img]images/smiles/icon_razz.gif[/img]

I agree that operating PV panels at battery voltage is not the way to get the maximum amount of energy from a PV array.

My comment was based on my own personal experience. I recently replaced 4 C40s with 4 OutBack MX60 MPPT charge controllers. To try and evaluate the actual real world benefit of MPPT, I replaced one C40 with a MX60 and periodically compared the output of the MX60 and the remaining C40s for nearly six weeks. Each charge controller was connected to 40 BP275 (75W) panels wired into ten 48V strings. I am at a latitude of 39.8N (Chico, CA.) and the panels face South with a tilt of 26.4 degrees. While I saw peak benefits of up to 20 percent, average benefit was closer to 10 to 12 percent.

Robert

Hi Robert,

Good information ! ... I like real world comparisons ..

What is the Battery side of the system look like , thoughts I have would be combining on the battery side may or may not be loading the array to extract full current from the array ?

Outback makes some impressive claims on the MX60 , your the first person I have heard from with actual results

Thanks

Henry

My first system has RV PowerProducts SB3048s for charge controllers.

Using data the SB itself reported (PV current, PV voltage, output current and battery voltage) I computed a boost percentage. I want to acknowledge once again the great help from Rick at RV power in creating a technical bulletin that allowed me to do the instrumentation.

Based on my experience I would be comfortable in predicting at leasts a 7% gain overall when the batteries are full (with the caveat that this is for climate where I measured this which is Sunnyvale, CA). Naturally the boost varied with conditions. Over 30% gains do happen at the edges (when it's cool for instance). Also if the batteries are depleted then the benefit MPPT increases.

Some additional background data. The system used BP-585s which have a slightly higher Vmp so the benefit of the MPPT is also increased. The SB had adjustable set points that in theory could improve (or worsen) its ability to find the approximate maximum powerpoint, I didn't adjust those.

Someday it would be interesting to have a "bake off" between the different MPPT controllers.

arnold

---------
The most common reason for going offline is the grid voltage goes out of spec ... this can be caused by using undersize wiring on the AC mains .. inverter feed power , voltage rises beyond spec , inverter goes offline
------------
Solar Guppy, are you saying the problem may lie between the electric meter, and the inverters?

I seem to remeber that one of the first things we did after going online was to sell from the batteries, but we only ran for a couple of minutes in this mode. This was at the end of the day, so all of the juice was coming from the batteries.

also have seen you mention washing the Solar Panels. Do you have a favorite soap?

Hi Paul,

I was giving some suggestions of things to look at. The SW should have codes explaining why it went off line ... since you paid to have this system installed , have the installed debug the problem.

When I suggested doing the sell test , it was to prove that the system could sell maximum rated wattage and still function.

This really is a straight forward thing to track down , you need to get your installer to come and help debug , that is his job.

Soap , same stuff I use to wash the dishes.

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

Concerning Arnold de Leon's previous post about the efficiency of his MPPT controllers measured at around 10-12%, my measurements reveal the same numbers - about 10-15% over the C40's performance. I had taken my spot measurements from 10/2001 to 1/2003 and live within 10 miles of Arnold. I used a pair of RVPower's 3048s to power a single SW4048. This info is a little off of the original thread title, but it's nice to see the numbers I got corroborated.

A point of argument for MPPT is that performance 'shines' when the sun doesn't, and offers minimal benefit when it does, leading one to surmise that a 30% increase to a few hundred watts on overcast days doesn't add much (cumulatively) to the couple thousand watts attained in bright sunshine at a 3% efficiency gain.