Project Manager: Stefan Hock, Electrical Engineering
Technician at TEC-Institut für technische Innovationen
Author: Eberhard Zentgraf, Degreed Electronic Engineer
at TEC-Institut für technische Innovationen
Science-based team involved in planning, designing,
measuring procedures and evaluation:
Table of contents:
2. Procedural method
3. Experimental setup and execution
3.1. Initial monitoring after 4 weeks 1
3.2. Intermediate Inspection after 6 months
4. Test results
4.1. Visual inspection after cleaning
4.2. Measuring results of proximity parameters
4.3. Electrical module values
Roofs of farm buildings are often ideally suited for installation of PV systems. This also applies to roofs of barns, i.e. buildings for livestock. However, there have long been concerns that ammoniacal fumes of animals (especially poultry and pigs) could damage the PV modules in the course of time. Generally, the ammonia (NH3) attains into the ambient air through ventilation systems and thus also to the PV modules in the vicinity of such stables.
In appropriate concentration, ammonia may damage surfaces of objects and bring them to corrosion. This principle also applies to all parts of PV modules, which are exposed to the ambient air.
Durability of PV modules against ammonia is currently being tested in multiple research procedures. Very likely, the most common test methods are as listed below in chart 1.
Chart 1: Common methods, ammonia resistance of PV modules
As you can see on chart 1, the various test methods in some cases differ a great deal in NH3 concentration, the ambient temperature and the ambient humidity. They all have in common that test durations were less than one month and all tests were performed in artificial laboratory conditions. Tests sometimes took place in a sulfur dioxide (SO2) atmosphere instead of an ammonia (NH3) atmosphere.
2. Procedural method
Therefore we determined the following test methods:
a) Tests should take place in a realistic pig sty atmosphere (with all components contained in such air) and
b) as a long term test of a minimum of 6 months. Succeeding those 6 months a review was to be generated.
An acquainted farmer, who runs a pig-breeding farm with up to approximately 1000 pigs, offered the opportunity of long-term researches.
3. Experimental setup and execution
Images 1 thru 8 show details of the two ANTARIS ASM 180 and ASM 185 PV test modules we used, before they were brought to the pig sty. With these photos, the initial visible conditions were recorded enabling a comparison after 6 months.
Img 1: initial condition
Img 2: initial condition
Img 3: initial condition
Img 4: initial condition
Img 5: initial condition
Img 6: initial condition
Img 7: initial condition
Img 8: initial condition
In addition, both modules were ‘flashed’ on a solar simulator at standard test conditions (STC) in order to document electrical output data. Subsequently, the two modules were transported to the pigpen.
The sty is divided into several large bays, in which the pigs stay.
We mounted the two test modules (type ANTARIS ASM 180 and ASM 185) beneath the sty ceiling, one module facing upward and one module facing downward. Also see img 9 and img10.
Img 9: modules beneath the pigpen ceiling
Img 10: modules beneath the pigpen ceiling
For continuous monitoring of ammonia concentration a device called ‘GasAlert Extreme’ with NH3 sensor ‘SR-A04’ from BW Technologies/Honeywell was used which was also mounted beneath the ceiling of the pigpen close to the two modules, see img 11. This device has a data logger so data can be stored and retrieved for a long period of time.
Img 11: GasAlert Extreme for valuation of ammonia concentration
To verify the readings of the GasAlert Extreme, spot-tests were performed on a regular basis using so-called RAE detector tubes and a gas detection pump, model LP-1200 from ACE Gasmesstechnik, shown on img 12.
Img 12: Foreground: Test device from RAE with detector tubes and gas detection pump. Background: testo 174H, GasAlert Extreme.
Humidity and temperature are logged with a testo 174H. This device also includes a data logger. The testo 174H is not only displayed on img 12, but also on img 13. It is the small black box which lies on the right hand module and is partially covered by the yellow GasAlert Extreme.
Img 13: GasAlert Extreme and testo 174H
3.1. First examination after 4 weeks
About 4 weeks succeeding start of recordings both data loggers were read, a verifying measurement with test tubes was examined and photos were taken. There we found a layer of dust of approximately 1 mm, see img 14 and 15. The bottom of the modules were only slightly polluted, see img 16.
The inhabitants of the sty observed the control sequences (data reading, etc.) quite interested, see img 17.
Img 14: layer of dust after 4 weeks
Img 15: layer of dust after 4 weeks
Img 16: bottom side is only slightly polluted after 4 weeks
Img 17: The inhabitants of the sty observe the control sequences with interest.
3.2. Examination after 6 months
Succeeding the specified time frame, both modules were taken from the pigpen and transported to the TEC institute for examinations. The dust layer on top of the modules had grown up to about 1 cm in the mean time, see img 18 and img 19. In contrast the bottom sides were hardly more polluted than they were on img 16 (after 4 weeks).
Img 18: Layer of dust after 6 months
Img 19: Layer of dust after 6 months
4. Test results
4.1. Visual examination after cleaning
After both modules were cleaned thoroughly, the visual examination showed no damages even after 6 months in a pigpen. No aluminum frames, or backside foils, front glass, junction boxes, wires, plugs or silicon gasket seals showed signs of corrosion, see imgs 20 to 28.
Img 20: Condition after 6 months
Img 21: Condition after 6 months
Img 22: Condition after 6 months
Img 23: Condition after 6 months
Img 24: Condition after 6 months
Img 25: Condition after 6 months
Img 26: Condition after 6 months
Img 27: Condition after 6 months
Img 28: Condition after 6 months
Interesting in this context is a comparison of img 5 and img 25. On img 25 there are 3 brownish spots distinctly visible which are located in the area between the green letters of a ‘roman II’ and the junction box? These spots were caused by adhesive tapes which held a letter-sized sheet of paper long before the beginning of researching of this module. Taking a closer look, these three spots are also visible in img 5, although much lighter. Those adhesive tapes which were removed quite some time ago, left a residue of adhesion on the rear side of the modules. This residue reacted with the ambient environment in the pigpen and became a brownish tinting, whereas all components of the modules showed no reactions and remained the same.
4.2. Measurements results of proximity parameters
Chart 2 shows maximum-, minimum- and average values of ammonia concentrations [ppm] (parts per million), humidity [%] and temperature [°C] within the period of beginning of April 2011 until end of September 2011.
Period: April 2011 thru September 2011 (6 months)
Chart 2: Maximum-, minimum- and average values from 04-2011 thru 09-2011
The graphs on img 29 thru img 32 exemplarily show the characteristics of ammonia concentration, temperature and humidity over periods of several hours.
Img 29: Example: progress of ammonia concentration over a period of 14 hours
Img 30: Progress of temperature and humidity: same time period as on img 29
Img 31: Example: Progress of ammonia concentration over a period of 24 hours
Img 32: Progress of temperature and humidity: same period of time as on img 31
4.3. Electrical data:
The electrical data of both modules were not affected by disposition in a pigpen for 6 months. All values of performed tests with a solar simulator (under STC) before and after a 6 months period in a pigpen revealed that all parameters remained within permissible measurement tolerances. A difference of 0.11% was detected on the ASM 180 PV module and 0.39% on the ASM 185 module.
The sustained disposition in a permanently utilized pigpen took no visible toll or measurable damages nor any negative influence on the PV modules used.
We would like to give special thanks to the farmer Mr. Konrad Staab, who enabled usto carry out the tests in the pigpen of his farm.
Waldaschaff, October 14th 2011
Degreed engineer of electro technology