GOOD CONNECTIONS,
Trout Lake
July 5-8, 2000
Dave Hughes, PI, made this trip alone to the Trout Lake
LTER, to accomplish several tasks.
1. Complete the tower connections for both 915Mhz and
2.4Ghz omni antennas atop the new 120 foot tower erected under
contract by Northern Technical Services.
2. Deliver, install and test the Freewave DRG115W (outdoor)
and R (indoor) model 1 watt, 915mhz no license (Part 15 FCC
Rules) radios that will link Campbell Data Loggers on the
project Lakes, starting with Sparkling Lake.
3. Deliver, install, configure and test, a set of Aironet
11mbps Direct Sequence 2.4Ghz no-licence (Part 15 FCC Rules)
radios, with YDI half watt amplifiers, which use the 120 foot
tower for their antennas.
4. Evaluate the installation, special configuration,
and operation of another set of Aironet radios aquired by
Paul Hanson, LTER Data Manager. Being used with the Big Buoy,
Remote Underwater Sampling Station (RUSS) which will be launched
with the assistance of the vendor, Apprise Technologies, while
I am at Trout Lake.
5. Evaluate the method and approach being used by two
U of W-M students, Ryan Ludgwig and Nick Offerman, funded
half of their time by this Wireless NSF Project (other half
by LTER funds), to move - to Madison - over the Internet in
data base format as well as in Web form, the data produced
on the Big Buoy system and wirelessly collected at Trout Lake
Station.
6. Work out Sparkling Lake Raft radio, battery, solar
installation details while the Raft is being overhauled, out
of the water.
7. Run tests on the possible use of small World Wireless
Microhopper radios ( 100milliwatts, 19.2kbps, 915Mhz no licence
(Part 15 FCC Rule) serial port radios on a series of Small
Buoys launched for limited periods of time on a variety of
lakes by the LTER staff.
8. Investigate reports of data-stream interference by
a pair of Freewave DRG115 radios purchased by the LTER project
and installed by them on a Small Buoy equipped with a sophisticated
infrared gas analyzer sensory system linked to a Campbell
CR10X data logger.
9. Address the unresolved software issue of data logger
data being delivered wirelessly to the Trout Lake Center,
while being accessed routinely for data base collection, from
the Madison Campus via cell phone, and secondarily by the
Trout Lake LTER. The requirement to change that entire system.
10. Meet Margot Bergstrom, our student assistant hired
for the summer by our NSF Project, have her help in the above
jobs while learning fundementally how these wireless systems
are planned to work with biological science data collection,
and carry out other tasks for the LTER Field Site for the
remainder of the summer before entering she enters college.
Setting up Operational
Base
The first task on arrival was to set up an Operational
Base for the Wireless systems. In previous visits we had decided
that the best place in which to terminate the antenna cables
coming 120 feet down from the top of the tower, would be a
small environmentally controlled storage room in the Trout
Lake Site Garage. This was to keep the RF Cable run as short
as possible. If the RF cable also ran from the bottom of the
tower all the way to the main Center building - approximately
150 more feet - the total RF Cable length would be nearly
300 feet. Even with the heaviest gauge LMR-600 RF cable, there
would be a loss of up to 12dB - very high. By terminating
in the closer Garage Building, running the last 25 feet from
the base of the tower underground and through the wall into
the Storage Room, limiting RF line length to 150 feet and
loss to more like 6db we would have to, in the short run,
be obliged to use a relay radio to reach the Center, its computers,
local area network, and the Internet. In the long run we want
an Ethernet extension from the main Center building to the
Garage, and storage room that becomes our Wireless Oprational
Center. There is light and wall power in the Storage Room,
which can be locked.
Equipment Control
The first duty I assigned Margot was to be, for the remainder
of the summer, the Wireless Equipment Inventory Control officer.
As we unpacked and stored the hundreds of seperate items -
radios, power supplies, connectors, cables, adapters, antennas,
batteries, chargers, solar panels, software, and packing cases.
she recorded, noted serial and model numbers, and labeled
each item. Since the wireless systems - valued at least at
$10,000 - will be intermingled with equipment that is the
property of the University of Wisconsin, and the Trout Lake
Center is remote from sources of supply, it is extremely important
to track the identity and location of every item.
Wireless installations are the worst example of the old
saw 'For want of a Nail the Shoe is Lost, for want of a shoe,
the Horse is lost...' for the tiniest missing connector, given
the FCC insistance on non-standard connectors on the antennas
of complex Part 15 radios, can prevent an installation from
being completed. In fact, I had to bring a simple $5.00 male-male
N Connector Adapter from Colorado to be carried after a special
trip out from Woodruff, WI, and up the 120 foot Tower at 10AM
by Joe San Fillippo, installer from Northern Telecom, to link
the female N connector of the Young Designs (YDI) amplifier,
to the female connector of the 2.4Ghz omni antenna. We overlooking
the fact that the seperately shipped to Trout Lake from YDI
in Virginia, and the Omni antenna from another city, both
had female N connectors, so would not mate. So Fillippo had
to call us and say it would take another trip to the tower
once he had the correct connector. What should have taken
one antenna/cable installation trip took two.
Even though Northern Telecom does many tower and antenna
installations, they simply do not install Part 15 radio antennas,
and so do not have in stock their connectors. Thus it is far
more important than it at first seems, for those who seek
to buy and install radios for field science, to check, double
check, and triple check the exact requirement for connectors
before undertaking the task - especially in distant locations
(like Alaska and Puerto Rico, where missing connectors discovered
at the work site can take days to arrive.)
One good thing we learned about Trout Lake is that both
FedEx and United Parcel Service deliver very rapidly. Two
items, one from Dallas, Texas, shipped early the day before
and another one ordered less than 12 hours before from Illinois,
were delivered before 1PM to the Trout Lake field station.
Missing connectors in both case.
So the Inventory and Property Control system was set in
place, and Margot knows it is her reponsibility. After our
wireless NSF Project is over, those radios and systems of
value to the LTER site will be formally transferred in ownership
from Old Colorado City Communications - which is currently
accountable for everything purchased - to the University of
Wisconsin, at the LTER Site.
Final Tower Set Up
We ordered the 120 foot guyed tower for antennas put up
close to the Research Center buildings, reaching about 40
feet above the trees, to either extend the - as yet unknown
- range of the 915Mhz, 1 watt Freewaves as far as possible
across the lake research area through the forest of trees
and/or making it possible to get a good link to the 250 foot
tall Wisconsin Electric Company Tower on Muskie Mountain,
3 miles east of the Center. There was some concern expressed
by the Trout Lake staff, that a metal tower might give the
mildly rustic field research center an over-industrialized
look. Everyone was pleased after the tower was erected that
is was a lot less intrusive than they had feared. Its carefully
selected siting, the placement of the guy wires, and the thin
metal structure made it even hard to see from a distance.
The highest antenna on the mast is an Omni single white
rod from Cushcraft, tuned to 902-928Mhz, with 10db of gain.
Below it is a 2.4Ghz similar Omni. Attached to the 2.4Ghz
antenna is a half watt YDI amplifier - a small gray device.
Both are hard to see without binoculars.
Then down the mast are two black LMR600, low loss, thick,
RF cables, fastened to the mast, and entering the ground at
the base. They then run underground for about 18 feet under
the garage storage yard and go through the walls into the
storage room, terminating after about 6 feet of loose cable
with male N connectors. When Mike Willett ordered the cables,
he also ordered end connectors attached by the cable vendor,
rather than leaving it to us to install. It saved much time.
Joe San Fillipo arrived about 10AM with an assistant,
and as soon as I gave him a missing male-male N connector
adapter, he quickly went up the tower, installed it between
the standard short cable from the 2.4Ghz antenna to the amplifier,
and sealed everything well from weather using a silicon sealer,
and electrical tape. A normal charge for such a climb by a
professional, and about an hour's work would be $250.
The tower erection, antenna and cable installation were
very professionally done.
Freewave Installation and
Testing
Concurrent with Joe's high tower work - I configured and
attached to the base RF cable connector a Freewave DRG115R
- indoor model - radio, designating it as the Master Radio
in the network. I also configured its Call Book to talk only
to a second, portable, Freewave DRG115R, which I configured
as Slave.
While later, after more than one Slave radio is put into
use, we will drop the use of the Call Book - which requires
all radios be manually told which other radios it is talking
to - and use the newer Freewave 'Network' configuration. But
for now through the testing period, we want to retain the
simplest 'Green Light When Connected' configuration at both
end. The proof of a link, and even to some extent the quality
of the link can be instantly determined by looking at the
CD connect light on each radio. Link when Green Light is on;
no link when it is off.
Immediatly when both radios were configured, and power
applied, we had Green link lights on both radios. Which, even
though the radios were close to each other in the storage
room, the Master was firmly connected to the RF cable, so
that the status light indicated we were getting signal from
atop the tower, and there were no signs of a broken cable
or bad connections at the bottom or top.
The Tower arrangement passed the first test.
Rough Site Survey
The next, and most critical test for the Freewave radios,
which we were counting on to link to the data loggers out
on lakes, was to see how far the tower radiated signal would
reach from the Trout Lake Station - and in particular, in
the direction of Sparkling Lake, which has the raft on it
- the first priority to connect up wirelessly.
The method we used to make the first rough Radio Reach
Survey was to connect up a DRG115R radio via its 110v transformer
to a Recreational Vehicle power converter. The converter works
off the 12 volts of the cigarette lighter in a car, but contains
standard 110v plugs. Then we attached the radio via a 6 12
foot cable to an 5 db Omni. With me driving my rental car,
Margot in the right seat hand holding the antenna out the
window, and Tim Kratz, Trout Lake LTER PI in the back seat
watching the status lights on the Freewave radio, we headed
for Sparkling Lake.
To everyone's delight, the green CD connect light stayed
solid green all the the way to the near shore of Sparkling
Lake, indicating a good connection to the radio back in the
storage room, via the tower antenna. Omni to omni communications,
which is the most desirable situation, because the raft is
expected to move somewhat on the water. If a highly directional
yagi antenna is used at the raft, movement can throw it off
line and break the connection. And an Omni at Trout Lake station
can send and receive equal energy in all directions, permitting
other data loggers to be connected wirelessly.
Then we drove on a back road south of Sparkling Lake,
and reached several other potential research sites, the link
failing only at over 2 miles range through continuous tree
stands.
We then drove in several directions, including toward
Crystal Lake, over 2.5 miles east of the Center. While the
link failed just at the entrance to the lake, as soon as we
got to the boat launch location, the link came back. Still
with the omni. We swapped the omni for a Yagi, and the link
was stronger.
Then we drove to a small road on the far side of Sparkling
Lake, and continued to have a good link, proving that we will
be able to connect up the Campbell Data Logger on Sparkling
Lake on the raft in the middle of the lake, with margin to
spare.
And all this was, according to Tim Kratz, at the densest
vegetation seasonal level of the forests. It will simply get
better in winter when the leaves are gone.
The only thing we were not able to test was the range
limits during, or after rains. Margot Bergstrom, as one of
her duties will test and plot the limits of the periphery
of range when leaves are wet from rain as well as dry.
My rough calculation is that we can expect than any research
site within a 20 square mile area - 2.5 miles in radius, to
be reachable using only omni antennas. Yagis should increase
this to at least 3.5 miles (over 115 square miles) and we
can increase the distance using the Repeater function of the
Freewave radios. Especially since the new function permits
any radio to be both slave and repeater at the same time -
meaning that the radio connected to a data logger out on the
periphery of the study area can both communicate the data
and pass traffic from and to a more distant radio and data
logger. This could extend the distance to as much as 10 miles
radius - or 3,000 square miles.
All that aside, this tower-915mhz Radios-range test trip
proved out, for the first time, that Freewave class radios
can provide useful links out to reasonable research area ranges,
in forested areas.
