Xantrex: Xantrex 3.8 shuts down short periods

clyde · **Posted:** Fri Jan 19, 2007 3:19 pm GMT EthGMT

I have 18 Sharp 208's connected 2 strings of 9 homerun to xantrex gt 3.8 about 18 feet of equal length. The inverter shuts down and takes too long to start up stating insufficient solar energy. It is now 10am and it is just now starting up and immediately jumped to 2312 AC Watts. The panels have sun access starting about 7am. The last four days have had no clouds all day is nothing but sunshine. The following is some code where the system shuts down in the prime operating hours, varying times from 5 minutes to hours, every day except Jan 8 which I had no stopping and just short of 9 hours online production:
Xantrex Performance Logger Version 1.8 ... ｩ 2006 www.solar-guppy.com
Inverter Infomation Model = GT3.8-NA-240 Xbus ID = 284469
Serial Number = B11011527
Firware Revisions Bootloader = 01.00.03 Main = 02.04.00
xbus = 01.02.04 Protect = 02.02.00
Inverter Logged Performance Data
Time VDC DC Amps Status
8:46:15 319.8 0.06 ST:4 STS:0000 12da 0010 92da FA:10 10:31:43 312.7 0.06 ST:4 STS:0000 12da 0010 92da FA:10
10:32:11 311.5 0.3 ST:4 STS:0000 12da 0010 92da FA:00
10:32:39 291.4 5.66 ST:3 STS:0000 0000 0010 92da FA:00 10:50:55 262.9 10.2 ST:3 STS:0000 0000 0010 92da FA:00 10:51:23 316.3 0.06 ST:3 STS:0000 0000 0010 92da FA:10 10:51:51 315.7 0.06 ST:4 STS:0000 0008 0008 92da FA:10
10:52:19 315.1 0.06 ST:4 STS:0000 0008 0008 92da FA:10
11:48:11 306.2 0.06 ST:4 STS:0000 0008 0008 92da FA:10
11:48:39 287.2 5.45 ST:3 STS:0000 0000 0008 92da FA:00
11:49:07 267.1 9.29 ST:3 STS:0000 0000 0008 92da FA:00
11:49:35 248.1 10.99 ST:3 STS:0000 0000 0008 92da FA:00
15:49:59 256.4 1.94 ST:3 STS:0000 0000 0008 92da FA:00
15:50:27 240.4 2.22 ST:3 STS:0000 0000 0008 92da FA:10
15:50:55 308 0.09 ST:4 STS:0000 0008 0008 92da FA:00
15:51:23 286 1.27 ST:3 STS:0000 0000 0008 92da FA:00
17:05:35 177.6 0.06 ST:3 STS:0000 0000 0008 92da FA:00
17:06:03 166.3 0.06 ST:3 STS:0000 0000 0008 92da FA:00
17:06:31 156.8 0.09 ST:4 STS:0000 0008 0008 92da FA:10
17:06:59 151.5 0.09 ST:4 STS:0000 0008 0008 92da FA:10
17:07:27 169.8 0.06 ST:0 STS:0002 000a 0000 92da FA:10
17:15:11 121.2 0.06 ST:0 STS:1012 12da 1052 92da FA:10
17:15:55 107 0.06 ST:0 STS:1082 12da 1092 92da FA:11
17:17:19 114.7 0.06 ST:0 STS:1082 12da 1092 92da FA:11
17:20:27 96.3 0.09 ST:0 STS:1082 12da 1092 92da FA:11

I had to hide some columns and rows to display what is here but I will gladly forward the log files to anyone that can help with this issue.
Date AC Wh AC Peak WTime Online Lifetime AC Wh
1/4/2007 5763 3432 6:54:28 245873
1/5/2007 9786 2760 4:41:05 255884
1/6/2007 3898 0 4:36:39 259782
1/7/2007 12368 2517 7:53:55 272150
1/8/2007 12883 2510 8:47:35 285034
1/11/2007 0 0 0:00:00 301610
1/14/2007 13362 2657 7:32:11 340410
1/15/2007 13477 2014 7:43:24 353887
1/16/2007 13168 2849 8:08:20 367056
1/17/2007 8687 2677 5:37:06 375743
1/18/2007 11503 2697 7:49:29 387247

ps. I turn the computer off at night and leave it run during the day except when I was out of town.
Thanks
Clyde

mike90045 · **Posted:** Fri Jan 19, 2007 4:58 pm GMT EthGMT

I don't know if this has any thing to do with it, but you should download the latest patch for the GT-View, with a tweak for the GT3.8's
Scroll down the page to find it:

GT-View 1.9 Beta (English Only)
http://solar-guppy.com/downloads/GT-View_1.9_Beta.exe
GT-View Change History:
1.9 (Beta)
AC Peak Watts limit corrected, was limiting to 3500 watts AC reported

I see my west facing panels wake up the inverter about 7am ish, with about 300VDC, and the the power curve climbs as I'd expect thruout the day, till sunset.

Have you paid your installer yet ? I held off final payment till it passed final inspection approval, and the electric company paperwork was well underway.

Solar Guppy · **Posted:** Fri Jan 19, 2007 8:36 pm GMT EthGMT

if this was a previously good performing system, just call xantrex and request a replacement inverter

If this is a new install, I would try running on one string at a time to make sure something isn't funky with the wiring ..

If it still has issues's, Xantrex will send a replacement, they have been really good at keeping GT customers happy

clyde · **Posted:** Fri Jan 19, 2007 11:20 pm GMT EthGMT

Yes it is a new install by myself which was completed and signed of by the city the first week of December. I didn't get the cat5 adapters until Jan 4 so that is the first logging in the computer. I had to shut the system down today because PG&E came to install the new E-7 rate meter and do their inspection which was basically just talking to me, took about 10 minutes total. I knew the 1.9 beta was there but chose to stick with the non beta program. I'll upgrade to the 1.9 beta now and see what happens tomorrow. If that doesn't change I'll try unhooking one string and see what happens.
Date AC Wh AC Peak Time Online Lifetime AC Wh
1/4/2007 5763 3432 6:54:28 245873
1/5/2007 9786 2760 4:41:05 255884
1/6/2007 3898 0 4:36:39 259782
1/7/2007 12368 2517 7:53:55 272150
1/8/2007 12883 2510 8:47:35 285034
1/11/2007 0 0 0:00:00 301610
1/14/2007 13362 2657 7:32:11 340410
1/15/2007 13477 2014 7:43:24 353887
1/16/2007 13168 2849 8:08:20 367056
1/17/2007 8687 2677 5:37:06 375743
1/18/2007 11503 2697 7:49:29 387247
1/19/2007 421 1659 0:41:43 387668
1/19/2007 10284 3497 6:40:57 397934

BB · **Posted:** Sat Jan 20, 2007 1:55 pm GMT EthGMT

Clyde,

I am in San Mateo too (south of 92, near El Camino) and have been running my Xantrex 3.0 for since September 2005 without any problems (20x BP 4175 panels or 3,500 STC rating).

By the way, how did you get the E-7 power meter (I have that too)--The program has been closed for since middle of last year for new customers.

The E-7 rate is much better than the E-6 (IMHO) for solar.

-Bill

clyde · **Posted:** Sat Jan 20, 2007 4:33 pm GMT EthGMT

I was so disappointed when PG&E discontinued E-7 as I was in planning stages for my system. I went ahead and ordered all my stuff and got the R1 paperwork in before the end of May. I was told the panels would take about 4-6 weeks. They arrived Thanksgiving week. I got wind that PG&E was removing the requirement for AC disconnect that same week. I found it on their web site the following week. I installed the system and was producing Dec 5. I always think it is better to ask for forgiveness so I started to get the paperwork together for the city permit and received the permit on the 11th and scheduled for final inspection the next day. Inspector did not know anything about PV and net metering so it was a session of me explaining benefits to him and 15 minutes later he was gone and I had the signature on final inspection. Another gust of wind told me PG&E was bringing back E-7 due to public pressure. I could not find it on their website until after Dec 22 but it is open as of Jan 1 for the first 5000 customers. I confirmed it with PG&E NEMS manager and I asked if should wait to send in my paperwork application until after 1st. He said send it now but specifically ask for E-7. Needless to say I received a call from the office Dec 27 saying E-7 not available, I told her to look on her web site and she said she would hold application until after 1st of year. It took a several more phone calls to PG&E but yesterday the meter installer inspector came and he was shocked to see E-7 rate meter. He thought it was closed also and mentioned how the customer has to inform him of his own company policy and programs. I just said it is on your web site. I am elated the way it looks and how lucky the timing turned out to be.

The DC disconnect is in the Xantrex and I don't have an AC disconnect box... very clean looking wall with the Xantrex right next to a new E-7 Meter and one conduit going up to the panels. For the AC side I was able to drill a hole in back of Xantrex and 3" conduit directly into back of sub panel.
Bill, I am on the water close to Fish Market. If you don't mind i would like to drive by and look at your system, send me an email with your address clydeminter@yahoo.com
Back to my original problem, the 1.9 Beta did not change the slow online start up this morning. I'll look at the logs tonight to see if there is still a mid day shut down, and if there was I will unhook one string and see what happens. The 1.9 beta did not start up cleanly this morning when the computer started. XP asked for confirmation to run the program where 1.8 did not ask. I have always just put a shortcut to the program in my startup folder.
Clyde

clyde · **Posted:** Sat Jan 20, 2007 5:25 pm GMT EthGMT

Well tonight I will unhook one of the strings

It is now almost 1pm and the system shut down in bright sunshine for last 15 minutes saying "insufficient solar power" Then it started up and is now producing 1400w
Clyde

(alt+p)[/img]

Solar Guppy · **Posted:** Sat Jan 20, 2007 7:24 pm GMT EthGMT

Re-check all of your wire connections ... an intermittent connection could cause what is happening .. a un-torqued screw say at the GT or array would cause intermittent operation underload. the GT would try and start-up, wire connection fails, rinse and repeat

It can also be a the DC connections from the BOS box to the inverter inputs, check the screws on those wires as well

The XP thing is the windows/defender firewall, it asks for permission when a new program is run. GT-View 1.9 with its different revision looks like a new program to XP even though is the same program name. You only have to do that the first time a program is run

mike90045 · **Posted:** Sat Jan 20, 2007 10:24 pm GMT EthGMT

Log follows, system comes up, and down frequently, as sky brightens up, but then settles in.
This was on one of the coldest AM's here in LA, nice and frosty. My startip voltages are nowhere near the expected voltages in the sizer calculator.
At first it's dithering around, barely turning on. Actual sunrise is about 6:55 that day. Real power starts up about 8:30 AM, with ambient light only on west faceing panels. I'm pretty sure at that time, they are just barely above freezing, as they are still in shade.

At the end is the shutdown at sunset. about 5:06pm

And I can't seem to force the formatting to use a fixed font - sorry

Solar Guppy · **Posted:** Sun Jan 21, 2007 12:36 am GMT EstGMT

If you really want to know the VOC, turn the GT's disconnect to OFF and at sunrise ( or when panels get direct light ) measure with a DVM ...

In your case, panels not facing east, the idle tar load of the unit will be keeping the panel voltages down

The numbers in the calculators are very real, the variances are panel oriantation, sun angle, tempature and inverter idle load, which for NEC purposes must all be calculated for worst case, which would be east facing, on a tracker so panels are perpendicular to the sun on a cold day with no inverter load. A Charge Controller for a battery based system would be more prone to having issues as it's tar loss and number of converter stages are much less and out west in the mountains, this is a very real and likely problem

Xantrex: Xantrex 3.8 shuts down short periods

Who is online