[RE-wrenches] [Following up on] Low array voltage stumper on legacy system

[Allan Sindelar]

Wrenches,
What a delight it is when a client is so fully able to understand and 
follow directions! Albert, the homeowner of the 1997 system with 16 BP 
75W modules on a tracker, sent the results below. In short, he isolated 
each string and measured input and output voltage and current as shown 
by the display on the MX60.

The results suggest that while two of the four strings are worse than 
the other two, all are severely degraded, and none put out close to 
original MPP voltage or current.

Given that the tracker has recently stopped operating (due to an 
obviously failed internal electronic component, most likely from 
lightning damage) and replacement controllers are no longer available, 
SunXtenders that have reached the end of their life after nine years of 
operation, and these modules, I'm not recommending rebuilding this 
system with new batteries and modules. I have suggested a batteryless 
grid-tied system, and either keeping this as a modest backup resource 
for occasional outages (with new batteries, of course) or using a home 
standby generator to meet backup needs.

And both the client and I would be interested in any observations and 
further conclusions that can be taken from the numbers below and in the 
attached spreadsheet. It appears that string three pulls down the 
overall output beyond what it contributes. We have one spare module, if 
the client chooses to keep the existing array in operation. The other 
three strings put out between 50% and 61% of original nameplate current 
rating. All have decidedly low Vmp, which at STC would be around 68V.

Thank you,
Allan

*Allan Sindelar*
allan at sindelarsolar.com <mailto:allan at sindelarsolar.com>
NABCEP Certified PV Installation Professional
NABCEP Certified Technical Sales Professional
New Mexico EE98J Journeyman Electrician
Founder (Retired), Positive Energy, Inc.
*505 780-2738 cell*

**


-------- Forwarded Message --------
Date: 	Sun, 28 Aug 2016 09:11:31 -0600


Allan,
I completed the test yesterday during late morning sun.  Results are 
shown in the attached spreadsheet.  It appears that both strings 3 and 4 
gave lower Vmp, although 3 was worse.
Albert

PV Array Output 	
	
	
	
Observation date: 	8/27/16 	
	
	
Observation time: 	11:10 - 11:45 	
	
Values displayed on MX-60 charge controller @ MPPT 	
	

	
	
	
	
	
String* 	Input 	Output

	V 	A 	W 	V 	A

	
	
	
	
	
1 	52 	3.7 	184 	26 	7.1
2 	46 	3.4 	153 	26 	6
3 	38 	3.3 	122 	26.2 	4.7
4 	37 	4.2 	152 	26.2 	5.9

	
	
	
	
	
1+2+3+4 	41 	16 	610 	26.2 	23.3
1+2+4 	47 	12.3 	544 	26.2 	20.9

	
	
	
	
	
*selected using breakers on combiner box at array 	
	



Hi Allan,
I would try turning off string #3 and see what happens to the Vmp.
I'm guessing it'll go up to a more normal voltage.
Replacement is my thought.
Jay
Peltz power.


On Aug 25, 2016, at 7:04 AM, Allan Sindelar <allan at sindelarsolar.com 
<mailto:allan at sindelarsolar.com>> wrote:

Wrenches,
Here's one I haven't seen before. 1997 grid-tie battery backup system 
that was one of my first, although actual installation was by another 
before I was licensed. 16 BP 275, 36-cell 12V modules, the standard of 
that time. Early Outback MX60 replaced C40 around 2005; array rewired to 
four 48V strings of four modules at that time, serving a 24V battery 
bank. Tested with an end-of-life 9-year-old AGM battery bank in the 
system, but in Sell mode with a 26.4V Sell voltage setpoint.

At 11 am, 68ºF, 900W/m^2 on the Daystar, with the array under charge, I 
measured 3.5A, 3.4A, 2.4A, and 3.6A. Short-circuiting each string, I 
measured 4.0A, 3.7A, 3.8A, and 3.7A. Original rating was 4.45A, so 
except for the third string, current is what I'd reasonably expect from 
20-year-old module degradation. However, watching the periodic MPPT 
sweep from the MX60 at the array, the maximum power point voltage for 
the entire 48V array settled at 14.6A at 37.2V, or about 540 watts from 
a 1,200W (originally rated) array.

I have commonly seen the current reduce substantially with age-related 
degradation, while the MPP voltage remains fairly close to original. I 
have never seen the voltage drop this far - an average of 9.3V MPP per 
each 12V module. All of the cells are seriously browned. I have other 
systems out there with the same modules of similar age that don't 
exhibit this weird behavior. Can anyone help me understand why the MPP 
voltage would drop so far below what is typical?

There is a slight possibility of corroded buried input conductors, 
because of a serious but corrected issue of galvanic corrosion from a 
ground fault (a Romex connector pinching a conductor and allowing 
seepage to ground). This was corrected years ago, but the input 
conductors were direct burial for ~100' and there was never a bonding 
conductor between the array and the house, so there is a potential for 
corrosion damage. However, the voltage measured at both ends while 
charging was identical, so I think that would indicate that high 
conductor resistance isn't the issue.

As usual, thank you for the collective and individual wisdom shared here.
Allan
--
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.re-wrenches.org/pipermail/re-wrenches-re-wrenches.org/attachments/20160829/93efe485/attachment-0001.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: PV-array-output_Aug2016.xlsx
Type: application/vnd.openxmlformats-officedocument.spreadsheetml.sheet
Size: 45709 bytes
Desc: not available
URL: <http://lists.re-wrenches.org/pipermail/re-wrenches-re-wrenches.org/attachments/20160829/93efe485/attachment-0001.xlsx>