Today is...
Monday, December 22, 2014
Welcome to Control.com, the global online
community of automation professionals.
Advertisement
Featured Video...
Featured Video
A quick introduction to EtherCAT for motion control and I/O...
Advertisement
Our Advertisers
Help keep our servers running...
Patronize our advertisers!
Visit our Post Archive
Radio Modems
I am looking for some experiences, either positive or negative, regarding the use of license-free ISM spread spectrum radio modems.
By Gee, Jeffrey on 29 June, 2001 - 9:45 am

I am looking for some experiences, either positive or negative, regarding
the use of license-free ISM spread spectrum radio modems. We have a PLC5
and we need to communicate to 6 remote sites, the farthest of which is 1/2
mile away. I thought we could use DF1 protocol to Micrologix controllers at
each location. There are a few houses and trees in the way out to the
farthest remote location, and we do not have 'line of sight'. A local
vendor/integrator, with whom I have no past relationships with, is telling
me to put up 20-30 ft towers for antennas. Does this sound reasonable? It
does not seem right to me, that I can go 25 miles with these if I have 'line
of sight', but I can't go 2500 ft because of a couple single story ranch
houses and trees? Any advice?

Jeffrey Gee

By Walt Boyes on 29 June, 2001 - 11:12 am

I suggest you speak to Jim Steffey at Data-Linc. He is an acknowledged expert in this field. If you can use SS radio under your installed conditions, he'll know. You can reach him at mailto:jsteffey@data-linc.com.

Walt Boyes

---------------------------------------------
Walt Boyes -- MarketingPractice Consultants
walt@waltboyes.com
21118 SE 278th Place - Maple Valley, WA 98038
253-709-5046 cell 425-432-8262 home office
fax:801-749-7142 ICQ: 59435534

"Strategic marketing, sales and electronic
business consulting for the small and medium-sized
enterprise: http://www.waltboyes.com"
---------------------------------------------

By Carlos Araya \(PLC INT'L\) on 29 June, 2001 - 11:14 am

Jeffrey:
Depending on the type of radio you select, I think you can use a repeter radio instead the towers. The tower are expencive and remenber you need power in the place (solar?) Many of the Ethernet radios for free frequency have the capacity to be radio and repeter at the same time, the speed change to half of capacity.
Try DataLink radios as an example. Good luck!

Carlos Araya Mackenney
Automation Manager - PLC International S.A.
Phone (562) 269.8727 Fax (562) 269.8728
Santiago de Chile

1/2 mile with no line of sight might be trivial or next to impossible
depending on what type of radio modems you are considering . The most common
license free ISM bands are 902-928MHz and 2.4GHz .
The 900MHz stuff propagates much better than the 2.4GHz , and it
sometimes works even without line of sight for shorter distances.
Then there's the issue of the radio modems over the air data rate,
independent of the frequency band . The radios designed specifically for
long range have a low to moderate data rate, ranging anywhere from 9600bps
to 115kbps or more . The other variety are fast data rate radios which tend
to adhere to some wireless network standard such as 802.11b,a etc . The
higher data rates result in much shorter ranges due to several reasons among
which :
-higher datarate = wider bandwidth , which in turn means higher noise
bandwidth for the receiver . This equates to a higher front end noise that
limits the maximum receiver sensitivity.
-higher datarate = wider bandwidth also means more in-band interference and
consequently requires higher system gain margin for a stable link .

In a nutshell , frequency hopping systems in the 900MHz band are
usually going to exhibit longer range than direct sequence systems in the
2.4GHz .

Usually people who sell 2.4GHz systems like to brag about the new FCC
relaxation of rules which allows for more antenna gain , however this is not
applicable if your configuration needs the basestation to talk to several
slaves randomly scattered around it . The higher antenna gain comes at a
price and that is more directivity , larger size and cost . It also is
highly unlikely to have all slaves grouped in the high gain antenna's main
lobe .
Also at 2.4GHz the antenna coax cable is going to represent a
significant loss for the system margin , and for your pocket .

There are simple range calculator tools on the web that allow you to
play what-if scenarios , including obstacles, antenna height , gain, etc .
You might try RFPROP , free download from
ftp://ftp.simtel.net/pub/simtelnet/win3/ham/rfp103.zip and then just plug in
your numbers .


Best Regards,
Matt Tudor , MSEE
Elmar Technologies
http://personal.lig.bellsouth.net/~mariusrf consulting
http://www.rfdatacorp.com spread spectrum radio telemetry

By Joe Manns on 29 June, 2001 - 2:51 pm

This topic is covered on Ian Wiese's SCADA gospel list, an old discussion is
here, http://members.iinet.net.au/~ianw/archive/book1.htm , More recent archives are at http://listmaster.iinet.net.au/ but they were not reachable
recently. http://www.tapr.org/tapr/html/ve3jf.dcc97/ve3jf.dcc97.html gives
some background on the topic.

1/2 mile can be a headache or slam-dunk depending on what's between the
sites and what surrounds it. You might have problems with noisy 50 Watt
paging transmitters if they are located nearby. Cordless phones can cause
problems too.

A good integrator will do a signal strength and interference survey of the
sites using various tower heights. This is always a good idea. You might
find you can do a single 20 ft tower at the base station and get by with
stubs at the other sites. Keep in mind if your antennas are up in the air,
they are less vulnerable to curious kids.

Both Dataradio and MDS have equipment that will transport DF1, as will
Data-linc, Freewave, Locus Inc, etc..

By L. Sue Scott-Thomas on 2 July, 2001 - 2:23 pm

Mr. Gee:

There are a number of issues that will determine if the spread spectrum will
work for you in this particular application. They all revolve around
several basic issue:

1) Every radio has a certain threshhold (called receiver sensitivity)
that is a measurement of the minimum signal required for the radio to
work. You MUST have a field strength greater than the threshhold.
2) For a RELIABLE signal you also need a FADE MARGIN -- in other
words, you must have a enough signal between the threshhold and the
actual signal strength. Most agree that for a telemetry system 20 dB is
adequate.
3) You also need to consider the NOISE FLOOR where the system is
operating. If you are in a congested area you may need a stronger signal
to overcome the ambient noise from other RF sources.

What can affect the signal levels? - There are a number of things that
can cause attenuation (or a decrease) in your signal level including Rain,
Interference, Intermod, and obstructions, connectors, losses in feedline,
etc.

At 900 MHz or above most RF engineers will want you to have a line of site
path. This is because radio transmissions above UHF tend to be very
badly affected by obstructions (i.e. - houses, vegetation, land masses,
etc., etc.) You can lose anywhere from 10 to 40 dB of signal from an
obstruction. Although we have modeling that approximates these losses -
the best way is to determine their magnitude is to measure them. Your
vendor is certainly right in recommending that you have the antenna
mounted high enough to clear the obstructions. If, howevever, you
don't want to do that you need to measure the signal strength.

Most of the better spread spectrum radios have a receiver sensitivity in
the -105 dB to 110 dB range. Some of the cheaper ones may have a much
poor sensitivity - (-85 to -90 dB). For a 20 dB fade margin you need to
measure -85 to -90 dBm if you radio has a -110 dB receiver sensitivity.
Any reputable service shop will have a service monitor that can make these
measurements.

What you DO NOT want to do is to just test to see if you can get a signal
through. This will not show you if you have adequate fade margins to
give you a reliable path. You also need to be sure to test in the SUMMER
when the trees have leaves (leaves can cause substantial attenuation).

I would encourage you to be cautious working with unlicensed links -- you
are limited by LAW to a 3.85 dBi gain antenna -- if you use a larger
antenna you MUST TURN YOUR POWER DOWN so that you ERP (effective radiated
power) is compliant with the part 15 regulations of the FCC. This means
that you will have less than 1 watt of power and a small antenna. With
licensed UHF you can use large antennas and significant gain.

There are formulas that we use to calculate the gains and losses -- Your
last question asked why you could go 25 miles with line of site but not 2500
feet without line of site:

Free space path loss at 900 MHz for a 25 mile path is approximately 124 dB,
for a 2500 ft path is about 90 dB. If you sustain 34 dB loss because of
the houses and trees (and you can sustain those kinds of losses depending
upon construction) - you are close to approximating the losses from the
long path.

The math that is used to calculate the losses is very accurate and in
practice I have, on a number of occasions, calculated the paths - then
measured them. I am usually within 1 to 2 dBof the calculated losses.
The things that are not easily calculated are obstruction losses (leaves,
vegetation, houses, or topography). Those really need to be measureed and
then applied to your sitaution.

I am sure a lot of people will tell you it will work -- and it will probably
work. The issue is whether you will have adequate field strength to provide
your client with a reliable system. I am a fairly good living fixing
systems where people have ignored the propagation characteristics and have
just installed systems regardless of the path.

If you have additional questions, feel free to contact me. I hope this
answers some of your questions.

L. Sue Scott-Thomas, President
KNS Communications Consultants
10625 West Evans Avenue
Denver CO 80227-2089

I am doing this right now with LOCUS radio modems. It is really very easy.
Contact Dan Blome at LOCUS radio or any app engineer there. www.locusinc.com
Ed Schmidt
Associated Technologies, Inc.
Midlothian VA 23236

By Al Pawlowski on 2 July, 2001 - 3:08 pm

Unlicensed radios operate at such low power and high frequency that they do not have much ability to go through or around objects in the signal pathway.

If you have line-of-sight, my experience is that 25mi is possible under perfect conditions. Under average conditions, 5mi is more normal. However, I have seen problems under 3mi line-of-sight conditions.

If you don't have line-of-sight, you can not tell if you will get reliable
communications without some trial and signal stength measurements.

In either case, you need enough signal to overcome your path losses due to
things like rain, fog, wet trees/foliage and uncertain reflections. Towers
are a way to get above obstructions and achieve line-of-sight. Lower frequency and/or higher power (licensed) operation can also increase signal strength.

Your supplier should know what it takes, but a path margin (signal strength
over the minimum required to get through) of ~10db would probably cover most situations.

If you just try two radios out,

Hope this helps.

By Mark Weiss on 31 July, 2001 - 10:47 am

Jeff:

License free spread spectrum radios are limited in output power and therefore require 'line of site' in order to obtain solid communications. 'Line of Site' is a general term as it really depends on the physical characteristics of the site. Other ways of getting around obstructions besides using towers is through the use of store/forward repeaters.

The only real way to determine whether or not you will obtain solid, robust communication is to perform a site survey. Some radios have built in capabilities to do this. Locus, Inc. (www.locusinc.com) manufactures a 2.4 GHz radio that has built in diagnostics along with signal to noise and signal strength meters. Locus's distributors also all have site survey kits with various antennas and would be more than happy to assist you in doing this.

You can also contact Locus directly at 608-270-0500 and ask for either Dan Blome or Mark Weiss for more assistance.

Hope this helps.

Mark Weiss