TWP/TSI/C1 - obstruction in lens

The TSI at Manus has been able to remain dry and produces an image but the percentage of 
cloud cover is determined to be 100% even on cloud free days. Moisture in the camera lens 
and neutral density filter was corrupting the sky images. The lens and filter were 
cleaned, the moisure seals checked, the camera housing sealed, and desiccant added to both the 
camera housing and main electronics enclosure. On 3/23/2006, the lens was replaced and 
the sky images were significantly improved.

• AVHRR satellite images for Darwin from BOM (RAW)(raw)

• JPEG data stream - documentation not yet available

• MPEG data stream - documentation not yet available

• Cloud decision images, Total Sky Imager, PNG format(png)

• Sunshine Meter(sunny)
• region.horizon.count
• count.opaque
• count.box
• region.sun.count
• sun.strength
• Cloud amount, opaque (percent)(percent.opaque)
• count.mask
• region.zenith.count.opaque
• count.unknown
• count.sub.horz
• count.thin
• region.horizon.count.opaque
• region.sun.count.opaque
• region.zenith.count.thin
• count.sky
• count.sub.proczen
• region.zenith.count
• Cloud amount, thin (percent)(percent.thin)
• region.horizon.count.thin
• region.sun.count.thin

NIM/TSI/M1 - Reprocess: incorrect daily movies

Due to file management problems on the operating computer, the daily movie files are 
either missing, incomplete, or contain images from other days.

• MPEG data stream - documentation not yet available

NIM/TSI/M1 - Reprocess: Data processing problem

Due to the high aerosol conditions at Niamey, the TSI appeared to be beyond the limits of 
processing the color JPEG image into cloud fraction values. Several attempts to adjust 
the sky filter values used to process the images did not yielded the desired results. The 
processed images indicate an opaque sky cover, even in the absence of clouds.

The problem was corrected by removing the neutral density filter from the lens and 
re-adjusting the sky filter values.

The data can be reprocessed.

• Cloud decision images, Total Sky Imager, PNG format(png)

• count.opaque
• region.sun.count.opaque
• Sunshine Meter(sunny)
• region.horizon.count.thin
• region.zenith.count.opaque
• region.zenith.count.thin
• region.sun.count.thin
• region.horizon.count.opaque
• Cloud amount, opaque (percent)(percent.opaque)
• count.thin
• Cloud amount, thin (percent)(percent.thin)

SGP/TSI/C1 - Birds obstructing view

Birds were frequently in the the field-of-view and soiling the mirror, affecting the 
visible sky image and resulting processed cloud fraction image. Bird deterant spikes were 
installed on top of the camera housing and arm on 5/16 which prevented their perching on the 
camera so that the impact on the sky image was reduced. But birds were still present 
around the edges of the mirror until additional spikes were installed on top of the four 
corners of the enclosure on 6/14.

• Cloud decision images, Total Sky Imager, PNG format(png)

• JPEG data stream - documentation not yet available

• MPEG data stream - documentation not yet available

TWP/TSI/C3 - controller board failure

The mirror control board failed and was replaced with a spare. Several images were 
collected in the iterim that should be ignored. These images include periods when the sun was 
not masked by the shadowband, the camera lens was covered by red tape to protect the CCD 
from direct exposure to the sun and, the instrument was inside the shop for repair and 
testing.

• MPEG data stream - documentation not yet available

• count.sub.proczen
• region.zenith.count
• count.sub.horz
• count.unknown
• Cloud amount, opaque (percent)(percent.opaque)
• sun.strength
• Sunshine Meter(sunny)
• region.sun.count
• Cloud amount, thin (percent)(percent.thin)
• lat
• region.horizon.count
• region.sun.count.opaque
• count.thin
• region.zenith.count.thin
• region.horizon.count.opaque
• time_offset
• alt
• solar.altitude
• count.box
• lon
• count.opaque
• region.zenith.count.opaque
• count.sky
• region.horizon.count.thin
• solar.azimuth
• base_time
• count.mask
• region.sun.count.thin

• JPEG data stream - documentation not yet available

• Cloud decision images, Total Sky Imager, PNG format(png)

TWP/TSI/C1 - Shadowband continuosly rotating

The mirror was continuously spinning during different periods of the day, usually in the 
afternoon, causing the shadowband to not properly block the sun. This problem was 
corrected by removing the real-time-clock integrated circuit DSI306, or EEPROM U10, from the 
control board and reloading the position and time parameters.

• JPEG data stream - documentation not yet available

• AVHRR satellite images for Darwin from BOM (RAW)(raw)

• MPEG data stream - documentation not yet available

• Cloud decision images, Total Sky Imager, PNG format(png)

• count.sub.horz
• count.thin
• region.horizon.count
• count.box
• count.opaque
• region.sun.count.opaque
• region.sun.count
• region.horizon.count.opaque
• region.zenith.count.thin
• count.sky
• region.zenith.count
• count.sub.proczen
• Cloud amount, thin (percent)(percent.thin)
• sun.strength
• region.horizon.count.thin
• Cloud amount, opaque (percent)(percent.opaque)
• region.zenith.count.opaque
• region.sun.count.thin
• count.mask
• count.unknown

TWP/TSI/C1 - rust spots on mirror

During the evening of 7/8/2006, something caused several rust spots on one side of the 
hemispheric mirror which continued to grow in size and to be interpreted as clouds in the 
processed images. This problem was corrected during a site visit when the mirror was 
replaced.

• Cloud decision images, Total Sky Imager, PNG format(png)

• count.thin
• Cloud amount, thin (percent)(percent.thin)
• count.opaque
• Cloud amount, opaque (percent)(percent.opaque)

TWP/TSI/C1 - high control board temperature

The TT8 control board temperature was high, ~50C, so that the cooling fan was constantly 
on. It appears that the heat sinks were not transferred from the original board to the new 
one when it was replaced. The problem was corrected when heat sinks were installed.

• count.sub.horz
• Sunshine Meter(sunny)
• count.thin
• region.horizon.count
• count.opaque
• count.box
• region.horizon.count.opaque
• region.sun.count
• region.sun.count.opaque
• region.zenith.count.thin
• solar.azimuth
• count.sky
• count.sub.proczen
• region.zenith.count
• Cloud amount, thin (percent)(percent.thin)
• solar.altitude
• sun.strength
• Cloud amount, opaque (percent)(percent.opaque)
• region.horizon.count.thin
• count.mask
• region.sun.count.thin
• region.zenith.count.opaque
• count.unknown

TWP/TSI/C1 - solar shadowband inoperative

The mirror/shadowband was not turning and there was no communication with the control 
board. The shadowband was stuck in the position of 0 degrees from 8/29 to 9/20/2006.

On 10/12/2006, the mirror was replaced, the real-time clock chip was removed and heat 
sinks were added to the control board, and the Ethernet hub and serial adaptor were replaced. 
On 1/8/2007, the slip clutch assembly, timing pulley, and timing belt were replaced 
(which were damaged when the mirror was replaced). These actions did not correct the problem. 
The instrument was swapped with the unit from Darwin so that the control board could be 
repaired and tested in Darwin.

Several images were collected in the iterim that should be ignored. These images include 
periods when the sun was not masked by the shadowband or the camera lens was covered by 
tape to protect the CCD from direct
exposure to the sun. There are also extended periods of missing data when the instrument 
was shutdown, waiting for parts.

• JPEG data stream - documentation not yet available

• MPEG data stream - documentation not yet available

• Cloud decision images, Total Sky Imager, PNG format(png)

• Sunshine Meter(sunny)
• region.horizon.count
• count.opaque
• count.box
• region.sun.count
• solar.azimuth
• sun.strength
• lon
• Cloud amount, opaque (percent)(percent.opaque)
• time
• count.mask
• region.zenith.count.opaque
• count.unknown
• count.sub.horz
• count.thin
• alt
• lat
• base_time
• region.horizon.count.opaque
• region.sun.count.opaque
• region.zenith.count.thin
• time_offset
• count.sky
• count.sub.proczen
• region.zenith.count
• Cloud amount, thin (percent)(percent.thin)
• solar.altitude
• region.horizon.count.thin
• region.sun.count.thin

NIM/TSI/M1 - Shadowband Misalignment

From installation, the shadowband was positioned south at solar noon until the vernal 
equinox when it began to migrate counter-clockwise until reaching east 37 days later on 4/26 
(about 9 days before mid-spring). The shadowband at solar noon continued to migrate until 
reaching its most northerly position (~15 deg) on the summer solstice (if everything was 
perfectly aligned, it probably would have been pointed north on this day). It then began 
to migrate clockwise and reached east again 56 days later on 8/16 (about 14 days after 
mid-summer). As expected, the shadowband reached its most southerly position on the 
autumnal equinox but was about 10 deg off of being due south. 

With the apparent alignment problem, I don't understand why the shadowband was mostly 
blocking the sun. When I started noticing this offset in May, I chose to leave it alone 
because I felt that adjusting the shadowband to where I thought it should be would cause it to 
not block the sun at either sunrise or sunset. 

The alignment was corrected on 9/29 (6 days after the day of the autumnal equinox).

• MPEG data stream - documentation not yet available

• Cloud decision images, Total Sky Imager, PNG format(png)

• JPEG data stream - documentation not yet available

NIM/TSI/M1 - Shadowband glare

When the neutral density filter was removed from the lens, the reflection of the sun from 
the smooth shadowband strip became much brighter causing a glare in the image on clear 
days. To minimize this reflection, a rough surface (the "fuzzy" side of an adhesive velco 
strip) was placed along the length of the shadowband.

• Sunshine Meter(sunny)
• sun.strength
• region.sun.count

TWP/TSI/C2 - hazy image

Condensation in the camera lens and/or filter caused the sky images to be "hazy". The 
desiccant in the main enclosure was changed and the neutral density filter was cleaned.

• Cloud decision images, Total Sky Imager, PNG format(png)

• sun.strength
• solar.altitude
• region.zenith.count
• region.zenith.count.opaque
• count.thin
• Cloud amount, opaque (percent)(percent.opaque)
• region.sun.count.thin
• lat
• region.sun.count.opaque
• count.sub.proczen
• Cloud amount, thin (percent)(percent.thin)
• alt
• count.box
• region.zenith.count.thin
• count.sub.horz
• count.sky
• lon
• region.horizon.count.thin
• region.sun.count
• time_offset
• region.horizon.count.opaque
• Sunshine Meter(sunny)
• region.horizon.count
• count.mask
• count.opaque
• count.unknown
• base_time
• solar.azimuth

• JPEG data stream - documentation not yet available

• MPEG data stream - documentation not yet available

TWP/TSI/C1 - hazy image

When the TSI from Darwin was installed (so the instrument from Manus could be returned to 
Darwin for repair) moisture in the camera lens and filter caused the sky image to be very 
"hazy". The U10 chip was removed from the control board, desiccant was installed inside 
the enclosure, additional small desiccant bags were installed inside the camera housing, 
the camera housing was sealed with silicon to prevent moisture ingress, and the camera 
lens filter was cleaned and the focus adjusted.

• JPEG data stream - documentation not yet available

• MPEG data stream - documentation not yet available

• Cloud decision images, Total Sky Imager, PNG format(png)

• count.sub.horz
• count.thin
• region.horizon.count
• count.box
• count.opaque
• region.sun.count.opaque
• region.sun.count
• region.horizon.count.opaque
• region.zenith.count.thin
• count.sky
• region.zenith.count
• count.sub.proczen
• Cloud amount, thin (percent)(percent.thin)
• sun.strength
• region.horizon.count.thin
• Cloud amount, opaque (percent)(percent.opaque)
• region.zenith.count.opaque
• region.sun.count.thin
• count.mask
• count.unknown

NSA/TSI/C1 - camera lens covered

The shadowband was not properly masking the sun so the camera lens was covered on 4/13 to 
prevent damage to the CCD. On 4/20, the mirror was blown off by a snow storm and the 
instrument was brought inside for repair. The mirror was replaced, the lens uncovered, and 
the instrument reinstalled.

• region.zenith.count.opaque
• count.sub.horz
• count.sub.proczen
• count.opaque
• Cloud amount, thin (percent)(percent.thin)
• count.thin
• region.horizon.count.opaque
• region.sun.count
• region.horizon.count.thin
• count.unknown
• region.horizon.count
• region.zenith.count
• region.sun.count.thin
• Sunshine Meter(sunny)
• region.zenith.count.thin
• sun.strength
• region.sun.count.opaque
• count.box
• count.sky
• count.mask
• Cloud amount, opaque (percent)(percent.opaque)

• Cloud decision images, Total Sky Imager, PNG format(png)

• JPEG data stream - documentation not yet available

NSA/TSI/C1 - shadowband misaligned

At several times during this period the shadowband did not mask the sun because the mirror 
was not properly secured. Attempts were made to align the mirror on 4/2, 4/3, and 4/10.

• region.zenith.count.opaque
• count.sub.horz
• count.sub.proczen
• count.opaque
• Cloud amount, thin (percent)(percent.thin)
• count.thin
• region.horizon.count.opaque
• region.sun.count
• region.horizon.count.thin
• count.unknown
• region.horizon.count
• region.zenith.count
• region.sun.count.thin
• region.zenith.count.thin
• sun.strength
• region.sun.count.opaque
• count.box
• count.sky
• count.mask
• Cloud amount, opaque (percent)(percent.opaque)

• Cloud decision images, Total Sky Imager, PNG format(png)

• JPEG data stream - documentation not yet available

TWP/TSI/C2 - Mirror continuously rotating

The mirror began to continuously spin. Diagnostics were performed on the encoder, control 
board, and Ethernet adaptor. The instrument was removed on 1/2/2008 and shipped to Darwin 
for repair. The encoder and control board were replaced and the instrument was 
reinstalled.

• Cloud decision images, Total Sky Imager, PNG format(png)

• region.horizon.count.thin
• region.sun.count
• region.horizon.count.opaque
• region.zenith.count
• region.zenith.count.opaque
• Cloud amount, opaque (percent)(percent.opaque)
• region.horizon.count
• region.sun.count.thin
• region.sun.count.opaque
• Cloud amount, thin (percent)(percent.thin)
• region.zenith.count.thin

• JPEG data stream - documentation not yet available

• MPEG data stream - documentation not yet available

FKB/TSI/M1 - Reprocess: Missing time annotation

The time annotation in the MPeg movie is missing.

• MPEG data stream - documentation not yet available

TWP/TSI/C2 - Shadowband Misalignment

The shadowband was not masking the sun beginning at sunrise because the control board had 
the wrong date & time. These were reset and the shadowband came into proper alignment.

• Cloud amount, opaque (percent)(percent.opaque)
• region.sun.count.thin
• region.sun.count
• region.sun.count.opaque
• count.opaque
• Cloud amount, thin (percent)(percent.thin)
• solar.altitude
• count.thin
• solar.azimuth

NSA/TSI/C1 - frost/condensation in lens

The sky image appeared hazy and was corrected by cleaning the neutral-density filter and 
lens of the camera. The determination of cloud fraction may be affected.

• Cloud amount, thin (percent)(percent.thin)
• Cloud amount, opaque (percent)(percent.opaque)

• JPEG data stream - documentation not yet available

SGP/TSI/C1 - Missing data

Data are missing and unrecoverable.

• Cloud decision images, Total Sky Imager, PNG format(png)

• JPEG data stream - documentation not yet available

• MPEG data stream - documentation not yet available

• solar.altitude
• region.horizon.count.opaque
• count.mask
• region.zenith.count.thin
• count.unknown
• region.zenith.count
• count.sub.proczen
• lat
• region.zenith.count.opaque
• count.sky
• count.opaque
• region.sun.count.opaque
• region.sun.count.thin
• lon
• solar.azimuth
• region.horizon.count
• count.sub.horz
• region.horizon.count.thin
• alt
• count.thin
• Cloud amount, thin (percent)(percent.thin)
• base_time
• Sunshine Meter(sunny)
• region.sun.count
• count.box
• Cloud amount, opaque (percent)(percent.opaque)
• sun.strength
• time_offset

SGP/TSI/C1 - Contamination of mirror

A persistent bird was landing on and polluting the mirror, multiple times throughout the 
day. Spikes were placed on the corners and top of the instrument (an affective deterrent 
in the past) but this bird had the agility to perch on the top-center of the mirror.

• Cloud decision images, Total Sky Imager, PNG format(png)

• JPEG data stream - documentation not yet available

• MPEG data stream - documentation not yet available

• count.sky
• count.opaque
• Cloud amount, opaque (percent)(percent.opaque)
• region.zenith.count.thin
• region.zenith.count
• Cloud amount, thin (percent)(percent.thin)
• count.thin
• region.zenith.count.opaque

TWP/TSI/C3 - mirror continously rotating

The mirror/shadowband was rotating continously because of a serial communication problem 
with the control board. The problem was corrected when the Ethernet-to-serial adaptor was 
replaced.

• Cloud amount, thin (percent)(percent.thin)
• region.sun.count.opaque
• Cloud amount, opaque (percent)(percent.opaque)
• count.sky
• count.thin
• region.sun.count.thin
• region.sun.count
• count.opaque
• count.mask

• Cloud decision images, Total Sky Imager, PNG format(png)

TWP/TSI/C1 - shadowband misalignment

The TSI mirror controller was configured with a north latitude instead of south causing 
the shadowband to not mask the sun beginning at the vernal equinox. The problem was 
corrected by changing the sign of the latitude in the configuration file.

• JPEG data stream - documentation not yet available

• Cloud decision images, Total Sky Imager, PNG format(png)

• count.thin
• Cloud amount, thin (percent)(percent.thin)
• count.opaque
• region.sun.count.opaque
• region.sun.count
• Cloud amount, opaque (percent)(percent.opaque)
• region.sun.count.thin

TWP/TSI/C1 - Missing data

Data are missing and unrecoverable.

• JPEG data stream - documentation not yet available

• MPEG data stream - documentation not yet available

• Cloud decision images, Total Sky Imager, PNG format(png)

• Sunshine Meter(sunny)
• region.horizon.count
• count.opaque
• count.box
• region.sun.count
• solar.azimuth
• lon
• sun.strength
• Cloud amount, opaque (percent)(percent.opaque)
• count.mask
• region.zenith.count.opaque
• time
• count.unknown
• count.sub.horz
• count.thin
• alt
• lat
• base_time
• region.sun.count.opaque
• region.horizon.count.opaque
• region.zenith.count.thin
• time_offset
• count.sky
• count.sub.proczen
• region.zenith.count
• Cloud amount, thin (percent)(percent.thin)
• solar.altitude
• region.horizon.count.thin
• region.sun.count.thin

NSA/TSI/C1 - Mirror frozen

The mirror was not rotating so the shadowband was not masking the sun. This appears to 
have been caused by snow that was melted by the mirror heater and then re-frozen by high, 
cold winds.

• region.zenith.count.opaque
• count.opaque
• Cloud amount, thin (percent)(percent.thin)
• count.thin
• region.horizon.count.opaque
• region.sun.count
• region.horizon.count.thin
• region.horizon.count
• region.sun.count.thin
• region.zenith.count
• region.zenith.count.thin
• region.sun.count.opaque
• count.sky
• Cloud amount, opaque (percent)(percent.opaque)

• Cloud decision images, Total Sky Imager, PNG format(png)

• JPEG data stream - documentation not yet available

TWP/TSI/C1 - Shadowband Misalignment

When the TSI was reinstalled on 4/22/2009, the shadowband was aligned relative to the 
enclosure but either the camera arm & housing was not plumb or the enclosure was not properly 
aligned with true north. This was corrected on 6/30 at solar noon.

• count.thin
• Cloud amount, thin (percent)(percent.thin)
• count.opaque
• sun.strength
• region.sun.count.opaque
• Cloud amount, opaque (percent)(percent.opaque)
• region.sun.count.thin

GRW/TSI/M1 - Mirror reflection

A deformity in the mirror produced an anomolous refraction within the mirror surface that 
appeared in the images as a white streak near the top of the mirror and shadowband. The 
mirror was replaced on 6/19.

• region_sun_count
• region_sun_count_thin
• region_sun_count_opaque

TWP/TSI/C3 - Reprocess: Missing daily movies

The computer began "locking-up" every day at 11:15 UTC when the program that makes the 
daily movies is automatically run. Site Operations had to reboot the computer everyday upon 
arrival at about 21:30.

On 7/28, basic disk maintenance was performed that included error-checking, file purging, 
and defragmentation.

On 7/29, file management was performed, the programs to create the missing daily movies 
was manually run, and the times of scheduled tasks were optimized.

• MPEG data stream - documentation not yet available