Diary
#32
SUCCESS
WITH THE MICROHOPPERS
In
attempting to better scale smaller, cheaper, lower power spread
spectrum radios to data collection tasks that don't require full
1 watt, and 115kbps radios like Freewaves which cost in the $2,000
range when fully equipped, we continue to test the small World
Wireless Microhoppers which could be purchased and configured
for under $250.
The
Microhoppers are spread spectrum radios which operate in the 902-928mhz
bands, but are only 100mw in power, and top out at 19,200bps data
transfer rate.
World
Wireless Microhopper
Paul
Hanson, who operates the Technical Lab supporting Trout Lake at
Madison, annually deploys several 'Small Buoys' which measure
limited data - usually a single sensor out to 1 mile from shore,
keeping them there only for a few weeks, and then moves them to
different lakes, some of which are 20 miles from Trout Lake. This
is done all summer long.
Gathering
data from the CRX10 Campbell Data Loggers in these small Buoys'
requires launching boats to fetch the modules, unless radios are
used. Time consuming.
Paul
had already tested Freewave pairs for this task, which work fine.
But they represent radio overkill - cost ($1,700 or more each)
and power (they can handle 115kbps, while the Data Loggers only
require 9,600 baud radios.)
We
supplied Paul on our last trip with a pair of Microhoppers, encased
in small plastic boxes, and interfaced with 9 Pin RS232 board
and plugs, 6 inch, 0 db rubber duck antennas and 12 volt DC power
connectors. The idea was for him to mount one on the buoy, powered
by the same battery used for the sensors and datalogger, and then
fetch the data from shore - possibly just driving up to the nearest
shore line, use the cigarette lighter for radio power, and a laptop
running PCW208 Campbell software to collect the data from the
logger on the buoy
Paul reported at the end of last summer that these radios, as
he had them installed didn't work well. He had to get - on a boat
- within a few feet of the buoy before the radios connected. And
then he had trouble uploading a program to the data logger by
radio, from the software.
We
quickly determined the reason for the extremely poor connections
was that Hanson used only the small 6 inch 0 db gain antennas
at both ends, and left the radio on the buoy - including the antenna
inside the waterproof box. Of course the Microhoppers, with only
100mw of transmit power just can't handle being boxed in.
The
solution was to equip the Buoy Microhopper with the special pigtail
connector (reverse SNA at the radio - male N connector at the
end of a 3 foot RF cable.) Then to connect it to a 6 foot, thin-rod
Antennex omni antenna, which has 6 db of gain. The antenna then
must be mounted outside the waterproof box, while the radio stays
inside.
Then
we then suggested he use a similar attached antenna - a 10db yagi
- at the radio attached to the laptop on shore. This is not always
necessary for short ranges, but will be for longer distances between
the two radios.
We
knew this should work well - the only question being over what
expanse of water, from shore, would this particular arrangement
work. Water can cause abnormal behavior of radio waves propagated
close to its' surface. The radio on the raft - its' antenna -
would be within 1 foot of the water's surface.
Before
testing the changed antennas the third day, we tackled the thornier
issue of not being able to upload a compiled (test) program from
a laptop to a CRX21 data logger via two Microhoppers even on the
same table. Paul had tried that, but it failed every time once
3% of the upload had occurred.
It
became almost humorous, as 6 sets of qualified brains tried over
4 hours every combination of settings. The indication was that
the Microhopper was not able to pass the data and pause for buffer
overruns properly. That it was a serial data flow control problem
endemic to the Microhopper.
Bench Testing
We
called Dan Withers in Seattle over a speaker phone. He was the
Microhopper expert with whom I had furnished a Campbell Scientific
CRX10 data logger and associated software, so he could test both
with the radios and interfaces he was working on for me. We tried
several things he suggested. By the time we broke for supper he
had not solved it either - even though he had the same radios,
PCW208 software, and CRX10 data logger.
Just as we reached our hotel at about 6:30 PM, Dan reached me
by cell phone. He got it to work by matching the radio packet
byte count, with the 208 software. The microhopper can handle
packets up to 142 bytes in length. The PC208W is set, by default,
to 2,048 bytes. He set the Microhopper packet size to 130 bytes,
and then set the PC208W to the same length. The next morning when
we tried that, it worked fine. The Microhopper should be able
to handle flow control in a more standard way (Xon/Xoff or Hardware
flow control)- but this fix solves the problem.
So
we rigged one data logger with small battery, radio, and the 8
db Antenna and cables with adapters, and went outside to test
the range, both across water, and through trees.
Test
Kit in Boat - radio, battery, omni
Pressed
for time before their departure, Mike and Tom took out a power
boat on Trout Lake, and stopped at about one third of a mile,
and then traveled to the anchor position of the Big Buoy - 1 mile
out on the lake. We then used the newly fabricated - by Dan Withers
'Range Tester' device to see whether we had 'link.' After insuring
that, then Paul Hanson, using the hand held yagi, laptop, other
Microhopper connected to the cigarette lighter for 12 volts, connected
with the logger out on the lake, up and downloaded just fine.
Good
Link from Vehicle
This
arrangement has general application to many circumstances. The
steps to be followed are:
-
A Microhopper Radio is placed with the Data Logger at the
data collection site, and one person attaches the Range Tester.
In this case out on the lake. It helps to have either handheld
two way radios at each end, or, where they work, cell phones
to coordinate the exact configuration at both ends.
-
The remote Microhopper needs to be set to an ID, or Group
Code, or both, to differentiate it from any other Microhoppers
which will be deployed nearby.
-
The other person on the team - in this case on shore - attaches
a Range Tester to the other Microhopper, and both ends confirm
that they have a data link or not. If not, a combination of
placing larger antennas on the lake end, or using a stronger
Yagi antenna on shore, or moving around to a better 'line
of sight' location, can fix the workable set up configuration
at both ends.
-
Then real data should be passed, without the Range Testers
attached, to insure there is integrity of the data and not
drop offs.
-
After this testing, all that will be required is for one person
to drive a vehicle (or carry a battery for powering the Microhopper
along with a laptop) to the predetermined - by testing - Read
Out site, connect by radio across the water, and capture the
data.
This
is much less labor and time intensive than having to tow and launch
a boat to retrieve the same data. Which usually would require
two persons. And the Microhopper radio, together with associated
peripherals can cost less than $250, which is 8 times less than
using the more powerful - and costly - Freewaves, or similar radios.
This
idea can be used in other situations around Trout Lake. One such,
now requiring manual efforts, is 'reading' the depth of water
in 40 small, PCV-Pipe Wells. This now requires someone to reach
each well, remove covers, read the depth with a gauge of some
type, cover it back up, record, and move on. In winter this also
requires a snowmobile.
With
Microhopper radios at each Well site, along with a depth sensor
(and no data logger), it would be possible for someone equipped
with a Microhopper, laptop, yagi, and power attachments, to merely
drive close to the site (perhaps half a mile through trees), and
read the data from each well.
We
are going to use the same technique at the Bonanza Creek LTER
down the Tanana River near Fairbanks, Alaska. In this case Stephanie
Pike, Data Manager, will be able to navigate her boat near the
shore of the river, near the fixed data loggers in the woods nearby,
and 'read' the data from her boat with laptop and Microhopper,
using boat battery power.
Previous
Next
