Boltwood Cloud Sensor
I have had more than one person ask me "what does a cloud sensor do?". I usually
get a blank stare or a smirk when I answer "It senses clouds". Most amateur
astronomers at first don't see the point of a cloud sensor. Their feeling is
they can walk outside and look up to see if it's cloudy or not. My response
to them is "Have you ever fallen asleep during a long imaging session? What
if it had rained? Your equipment would have been soaked.
I have had the above scenario happen to me, which is why I ended up purchasing
a Boltwood Cloud Sensor. The Boltwood Cloud Sensor was designed by Paul Boltwood
of Boltwood Systems Corporation. I ordered my sensor back in January of 2005
from the exclusive distributor, Diffraction Limited (the MaxIm DL folks). At
the time of this review the sensor has been in use now for about 9 months.
Like any adult kid waiting on a new astronomy item I was inpatient and wanted
my new "toy". A week had gone by and I have not received any info
about my order, so I sent an email inquiry to Diffraction Limited. I promptly
received a response from Doug George (the Chief Cook and Bottle Washer) that
there was a problem with the heating circuit. This was allowing snow to collect
on the sensor of some of the first production units. Doug explained that this
had not occurred with the prototypes. He apologized, and then informed me that
there would be a delay in shipping while they fixed the problem. The sensor
showed up about 2 weeks later. It was about a month after it was originally
ordered. As it turns out, it was worth the wait as my system had no problems
with snow build up.
It should be noted that currently Diffraction Limited
states units are shipped within 2 days of the customer placing the order and
my unit was one of the very early units.
The sensor's software (known as Clarify) runs on Windows operating systems ranging
from Windows 98 to XP. The software installation went off without a hitch. I
ordered my sensor with a 100 foot cable and mounted the sensor on the lower
roof of my house (Fig 1.). The cable was hidden underneath the siding and snaked
along the outside of the house until it found its way into the family room.
This is where the observatory's weather computer resides. I could have gotten
away with only 50ft of cable, so the excess was coiled up and placed behind
the rack the holds the weather PC. The sensor was attached to the house using
2 right angle brackets purchased at the local hardware store. The 2 brackets
were screwed together and then screwed to the side of the house. The sensor
was then attached to the brackets via nylon wire ties. No installation brackets
are included with the unit because there is no way for the manufacturer to know
how the customer will mount the unit. Some may mount it to a house, or a observatory,
or on a pole. The unit should be mounted with about a 10 degree tilt facing
the direction of the prevailing winds. According to the manufacturers instructions
this serves two purposes; it helps prevent rain water from building up on the
sensor plate, and it allows the sensor to see the clouds a bit earlier as they
roll in.
Fig 1. Boltwood Cloud Sensor mounted (JAT Observatory)
The sensor is powered by an AC to
DC adapter. The power cord and sensor connect to an adapter box that has a
standard 9 pin serial port. A DB-9 pin cable is used to connect the adapter box
to the PC. My PC didn't have an
open serial port, so I used a Belkin USB to serial converter and it worked fine
on my system. When I ran the cable I ended removing one of the connectors so I
didn't have to cut a bigger hole in my wall. Fortunately the manual covers this
and gives you the pin-outs and wire colors to assist you in reassembling the
connector and cable. The components
contained in the system can be seen in Fig 2.
Fig 2. Boltwood Cloud Sensor components
The Boltwood Cloud Sensor works by
measuring the relative sky temperature within an 80 degree field of view
(according to the manual there is some sensitively out to 120 degrees). The
software looks at the difference between the sky, and the local ambient
temperature. It determines the sky temperature by sensing the infrared radiation in the 8 to 14 micron
range. The larger the difference between the sky temperature and the local
ambient temperature means the clearer the sky. The smaller the difference, the
cloudier the sky is. The sensor's software has 3 levels at which it will report
the cloudiness level. Depending on the delta temperature the Clarify software
will report "Clear", "Cloudy" or "Very Cloudy" (Fig 3.) The trigger level that determines what
the sky's cloud status is can be set by the user. This is useful for tweaking
the system, the is required due to the temperature change that occurs when the seasons
change. The software also displays the sky's temperature as well as the local
ambient temperature.
Fig 3 Clarify software displays
The Boltwood Cloud Sensor's
name only tells half the story. It also detects precipitation. Which means it functions
as a rain detector. It does this by means of a wetness sensor plate. The plate
senses a change in resistance when moisture contacts the plate. The software
reports this as "Rain". When rain is present on the plate a heater is
automatically turned on in order to warm up the plate and evaporate the
moisture. The heater will continue to warm the plate until no more moisture is
present on the sensor. (Notice the 100% heater setting on the rain display in
Fig 3). This is how the sensor's software determines when it has stopped
raining. Another feature of the
software is it is able to detect snow. It does this by keeping the plate warm
in cold temperatures. When snow falls on the plate, it melts. As the snow melts
the plate senses the change in resistance due to the melted snow and reports
it. But since the ambient temperature is below 32 degrees the software reports
the precipitation as "Snow".
The accuracy of both the
cloud and rain detector are very good. After spending some time with the sensor
the user will be able to look at the displayed sky temperature and without
looking outside know if it is completely overcast or just light wispy clouds
(which may provide some opportunity to get in some observing). The moisture
detection system is able to detect very light sprinkles and very light snow flurries.
The software also allows the user to set an alarm based on if it is "Clear",
Cloudy", "Very Cloudy" or "Wet". The "Clear" alarm is very useful if you are
waiting for "sucker holes" so you can observe. That way when a clear area of
sky is over the sensor the alarm will sound.
The software can also log the
data to the PC's hard drive. In addition if the user it has some programming
skills they can write their own software to capture the data and display it via
the COM or Active X interface, or via the single line data file. An example of
what can be done is shown in Fig 4. The example represents a real-time plot of
the relative sky temperature and the local ambient temperature. The time span
is user selectable. This particular software was written by Russell Croman and is
offered free of charge and is available as a download on Russell's website. This
is the software I use in order to tweak the settings on my system. The graph is
also uploaded it to the internet so I can remotely monitor the Sky's status.
This is useful when I operate my system remotely, as I will know if it is safe
to open the dome.
Fig 4. Russell Croman's Cloud Sensor Graphing
software
But the best feature I have saved
for last. It is the whole reason I purchased the Boltwood sensor. In addition
to the software having the ability to sound alarms, the hardware has the
ability to send a signal that can be used to close a dome or roll-off roof when
it gets "Very Cloudy" or when "Rain or "Snow" is detected. The nice thing about
this feature is it can do this without the need for a computer to be connected.
It sends the closure signal by shorting the connection across a cable connected
to the 3.5mm audio jack on the adapter box. From this jack you can run a line
to control the closure circuit on your dome or roof to your observatory. The
only restriction here is that your circuit that closes the observatory only put
between 0 and +30 volts on the inner conductor of the jack and not more than
150ma of current.
(It is recommended that a TIP127 (PNP Darlington)
transistor be used instead of connecting the Boltwood directly to the relay
(Excessive current draw will damage the Boltwood unit. The transistor should be
connected as follows: The Emitter to power supply, collector to relay and a 10K
resistor from the base of the transistor to the Boltwood output. "When the
Boltwood pulls low it will turn on the transistor and that will power the
relay".)
If the user does not have a motorized
roof or shutter on their observatory the sensors output can be used to trigger
an audible alarm or light. This is the way that that imagers that don't have
permanent setups can utilize the Boltwood sensor. In my case since the sensor
is mounted on the house and not the observatory the sensor's closure output is
routed to a custom built wireless transmitter (Fig 5). In addition to saving me
the trouble of having to dig another trench to the observatory, it solved the
problem of connecting the closure line to the shutter circuit with is located
in the rotational part of the dome.
Fig 5. Custom
Built Wireless Transmitter (JAT Observatory)
When in become "Very Cloudy"
or rains the transmitter receives a closure signal from the Boltwood Cloud Sensor.
The wireless transmitter relays that signal to a wireless receiver mounted in
the dome (Fig 6.).
Fig 6. Custom Built Wireless Receiver / cover
removed (JAT Observatory)
The receiver then triggers the
shutters motor circuit and closes the dome. The automatic closure feature of
the Boltwood Clod Sensor is a godsend to imagers who do automated imaging while
they sleep. No longer do they have to worry about their equipment getting
soaked by a passing surprise storm. Having said that, if an observatory owner
installs this system they should be sure their roof or shutters are in good
working, reliable condition and capable of automated closure with no
obstructions before they place the safety of their equipment in the hands of
the Boltwood system. Also the user
should install a manual bypass inside the observatory that will allow the auto-closure
system to be disabled. This is mainly for safety reasons so the roof does not
unexpectedly close during maintenance if it becomes cloudy.
Now the pros and cons to the
system. The main con to the system is the price. But that is not to say the
system is not worth the cost, they are just not cheap to buy. A system with 100
ft of cable is currently $779. Subtract $25 dollars for a system with 50ft of
cable or add $25 for systems with 150ft of cable. Systems used in areas with
220 volt service are $40 more. Another drawback of the system is since it was
designed for astronomy use it was assumed it would be used at night. The system
is not shielded so it suffers from solar heating during the warm daylight
hours. This affects the local ambient temperature reading (raising it) and can
causing the system to report the sky as being cloudier that it actually is. The
good news is as the user gets used to the system this is not much of the issue.
Now
the pros; The system works and it is reliable, this translates to sleep for automated
imagers. The system is also versatile and adaptable. I personally believe the
system is a must for anyone that operates a remote or unattended automated observatory
or does automated long duration imaging.
You can now know if it is truly safe to open the observatory. You can
now sleep and not worry about if it is going to rain. My system has functioned perfectly and
constantly for the past 9 months
On the rating scale I give
this device a big 2 thumbs up (Hubble Telescope rating). It is the only
complete, commercially available cloud sensor that I know of on the market with
all the features and functions listed above.
Marcus Thompson
JAT Observatory
(http://jatobservatory.org)
