As PI, and with Mike Willett, sub-contracting
Senior Technician on this project with me, I visited one of the
company's whose scientific sensors and data collectors are most
used, it appears, by biological scientists in the field. Campbell
Scientific, Logan, Utah.
http://www.campbellsci.com
We flew into Salt Lake City, then
drove the 90 miles north to Logan to spend 6 hours with some of
its key staff. I had sent a letter in advance to its CEO introducing
our Project and asking for such a meeting so that I could determine:
1. The technical specifications
of the most popular (for scientists) data loggers, so we could devise
wireless interfaces for them in the field.
2. The ways they connected their
devices up now wirelessly (they generally use, so far, only licensed
VHF, and cell phone links, and those for only some products.)
3. What might be under development,
particularly new types of data loggers, integration into TCP/IP
communications, client software, and any plans for further communications
work.
Campbell was extremely cooperative,
committing three professionals to meet with us - Art Heers, President
of Engineering, Neal Israelsen, Marketing Administrator, and Doug
Neff, Applications Engineer in the Environmental Group.
Campbell has an extensive array
of devices they manufacture (their catalog runs to 200 pages) and
they sell internationally. And they do custom design of sensory
data gathering for special projects, such a monitoring nuclear waste
sites.
Their data loggers range upward
from limited (2 sensors only) CR510 Basic Datalogger in the $725
base price range, through what they say is the most purchased CRX10X
Measurement and Control Model with its own operating system, that
can handle 12 single ended analog input channels (or 6 differential).
Priced in the $1,300 range, less sensors. To the quite complex and
capable CR9000 system in the $8,500 range.
It was encouraging to learn that
their philosophy of design tries to make instruments that use as
little power as possible (they regard a 1 watt radio as a real power-drainer).
In fact have a business relationship with a company in Florida,
that is making, for them, an extremely low power spread spectrum
RF radio (with limited range) that is connectable to their specific
data loggers. We offered to be part of their Alpha or Beta test
of this equipment-in-progress. They accepted the offer and will
contact us when they are ready.
We also learned most all of their
equipment is rated at 12 volts input power supply, which makes it
convenient and compatible with many radios run from battery sources.
Because data logger data rates
from sensors have historically been low across the industry by my
assessment, it was no surprise to find that most of their equipment
had been designed to output via RS232 serial ports, with a maximum
data rate (hardware port) of 115kbps, with most data streams being
lower. This is ok for analog or digital data short of real time
audio and or video. We pointed out we have been asked to consider
connecting video and audio devices in some field situations. They
know that is the direction things are going, so appear to be developing
to accommodate it also, in the future.
Now there a number of radios I
have used, such as the Freewave DRG115, which is fitted for RS232
input/output. They are aware of that radio and have used it with
their equipment.
http://www.freewave.com
We pointed out, however, that almost
all current radios, because their rates go up to very high speeds
(E1 2Mbps, 11Mbps, and 100Mbps) terminate with standard Ethernet,
10BaseT ports. This was somewhat new information for them, since
they are not much into wireless connectivity. But they were keenly
interested - so in some ways the meeting resulted in our educating
them on the direction wireless is going, which they know will be
an increasing requirement. Their chief engineer, a very creative
designer started brainstorming outloud about possible new products.
They do have, however, not yet
reflected in their catalog (1999-2000) a device 'NL100' which interconnects
RS232 serial inputs to 10baseT Ethernet. We will certainly buy some
of those, since many radios we expect to use will have 10baseT ports.
It already includes a TCP/IP stack, so can be accept an IP number
when configured.
While they are not that far advanced
in incorporating TCP/IP protocols (Internet and IP addresses) into
their current equipment, their developmental direction definitely
is taking them toward Internet compatibility. As part of that effort,
they have already shipped one model of a Server-Client system called
'Loggernet' which can manage multiple data loggers, but can put
the data on a local LAN network, or go out to the Internet. The
Client software CR10X operates on Win98/NT platform. $285.
They also have an 'RTDM' software
that the users can use to design web display screens into which
captured data can flow and be presented tabularly, or by graphical
forms. $315
Finally, we got to see production
work on Java, that can send applets to the clients, containing data,
for web display.
We stressed that one BIG interest
we had detected among researchers, was the ability to have true
bi-directional command of the data loggers and sensors, in real
time. Not just the capture and one way delivery of data. And urged
them to take that into account as they do future designs, for which
TCP/IP protocols permit. That some scientists we talked to want
to be able to control the sensor, command it to take readings, and
even to send control commands to one which might involve motion
- such as a camera, or light sensor.
Additionally we learned of 'Media
Burst' company which does satellite data, and their own product
called 'Pantex' which wirelessly communicates just two readings
- RH and Temperature, short distances. And we watched demonstrations
of their VHF connected weather stations.
The trip and cursory survey of
Campbell's range of current, and planned project, was quite fruitful,
for both sides.
We made it clear we next will be
visiting two Field LTER Sites (Wisconsin and Puerto Rico) which
use Campbell Scientific data loggers and peripherals. When we (1)
determine, in conjunction with the Principal Investigators there
what field-deployed Campbell Data Loggers they would like us first
to connect up, wirelessly and (2) get the precise models of Campbell
systems they are using, so we can order like units from Campbell
and configure and test here in Colorado Springs, before carrying
them back to install in the field, for real scientific data gathering.
< Previous
| Next >
