Anritsu Sitemaster general questions

[spiderbite]

Anyone with Experience using the S332d Sitemaster, (which is the one that does Spectrum analysis as well as Return loss distance to fault etc)

please advise on a couple of points.

I am making my own low-loss cables and need to baseline the cables for documentation and troubleshooting purposes later.

When performing the calibration on the unit after the open, short, load on the actual unit, you then attach the phase-stable cable and calibrate that.

At this point, do I attach the cable I am creating to the phase-stable cable and test? The user guide is a little vague.

Also, I want to do return loss and DTF-SWR tests on just the cable and the cable and antenna to have for documentation.

Can I also do the jumper-lightning arrestor-cable-antenna
whole shebang as well?

I would like to perform these tests with the cables and antennas separate, together, and as part of the whole unit before startup.


Is this overkill? I am not doing a 1400 ft radio tower, but maybe 6 runs of 50 -75 ft LMR400.

I know I need the DTF -SWR test, I think I want the return loss... but am I missing something

I have downloaded many manuals, pdf's etc...nothing quite as good as having somebody who has done it before explain it in English.

Thanks and Merry Christmas-

Spider

[Specialist927]

When my company tests LMR400 cable for loss we use a network analyzer.

[M/Q]

Hey Spider,

Site master has their own forum. I bet you would be able to find your answers there for sure.

http://www.us.anritsu.com/SMIU/

Sorry, but I got to ask 50-75 ft runs??? Ouch.

One more dumb question, do they say they can run a SWR test without the antenna?

[spiderbite]

Glad you asked MQ....

The project will offer VoIP phones in a relatively open outdoor area of the type of facility I work at. There are many areas under porticos, etc.

The units will be mounted indoors where I can reach them easily. The 50 foot of low loss cable will go out to the 8.5 dBi whisper patch antenna.

The CAT 5 will make up the difference to the closet.

The length of Low loss cable will help attenuate my signal being as all my antennas will be articulated at some degree toward the ground. This way, I can have 6 low powered AP's providing enough aggregate throughput to my users in a relatively small outdoor area (less than 8,000 sq feet probably)

without overlapping cells and providing physical load balancing to the network.

And I can steer the signal away from the street and anybody happening along with bad intent. Actually, I don't think it makes it to the sidewalk.


pretty cool, really

[spiderbite]

actually, it doesn't say. but in the calibration example they give you, there is the 50ohm load thing at the other end.

I suppose you could...but the results would be pretty bad.

You can do a return loss and a cable loss without the antenna though...I think.

[M/Q]

I knew you had great reasons. I just wanted to learn, being a bit nosey. Also being bit jealous, you always seem to get to play with the really cool stuff.

I have used what we call a dummy load in 2.4 gHz amatuer satellite work, but I was not even remotely impressed. It did not simulate anything worthwhile. A light bulb with a connector on it works as well load-wise. If you have to purchase the dummy load, I would forget it and just use the antenna. I for one would be much more interested in that. Just an opinion as usual.

I really would appreciate hearing more about the setup, if you have the time and inclination to talk about it.

[spiderbite]

I will definitely get back with you about.
As far as cool stuff to play with....

This project has has been discussed for three years, started last year, and finally we will get to put it in place.

Granted, if it works well, it will probably roll out nationwide.

But there has been a lot of wailing and eye rolling just to get here.

Actually, I have several questions to ask you about the build offline.....

[azkidd69]

I know this is late but I just foud this site.
About the S332D.

For the calibration you go ahead and hook up your phase stable cable and perform the open/short/load (OSL) calibration though your phase cable. That way the cable itself is removed from the measurements you make. Try to perform the calibration with any adapters needed already in place

For your baseline tests.

You want to perform the tests in about a 100 MHz band if below 1 GHz and about a 150MHz band if above 1 GHz.

Then I would connect the cable itself onto the phase cable and put the load (same one as used for calibration) @ the other end of the cable being tested.

You can do this test in either return loss or VSWR mode. The measurements are interchangeable. Personlly I am more comfortable with return loss myself.

Then I would change to distance to fault-return loss or distance to fault-swr and perform the same test.

Next you can place the short on the line and measure the actual cable loss (AKA Insertion Loss ) in db. If you want using cable loss mode or return loss mode. Remember cable loss mode does the calculation for you. In return loss you have to divide the peak result by 2 since it is actually measuring a round trip of the signal instead of just one way.

You can also perform a DTF-return loss measurement with the short on in order to get an accurate cable length measurement.

Fianlly you can start puttting the system together piece by piece and see how it affects the signal. You can put the whole system together and test through all components including the frequency sensitive devices. (i.e. lightning arrestors, antennas, etc)

The full system tests are going to be the most important tests for future diagnostics due to not needing to take the whole system apart in order to determine if the system is degrading over time.

Also, is this installation indoors or outdoors? If outdoors I would recommend using Mastic (Butyl Tape) to seal all connectors against the weather.

The preferred method for that is to first wrap the connection with regular electrical tape STICKY side out. (Makes it ALOT easier to take off down the road) Then wrap the connection with a layer of butyl and squeeze it together. Then finally you want about 3 -4 layers of tape on top of the butyl to keep it in place and to help it seal, and keep i from blowing through the tape.

If you have any questions feel free to post back.

[spiderbite]

That's exactly what I'm looking for....Thanks,

I am pretty sure I am following you all the way up to the full system. Which tests would you recommend for the full system?

As far as the tape trick - I had a tower guy out of Lake Fork, Texas show me that one...Thanks for the info...

More setbacks keep me from implementing this stuff anytime soon..probably at least three weeks out before I can begin the install...I'll keep you posted.

[azkidd69]

as for your system tests. I would do 2 tests.

first one with all cables and antenna with NO lightning arrester or filters, duplexers, rx amps etc.

then the second one with all the xtraneous tidbits in-line

what brand of cable, brand and model of antennas, tx/rx frequencies,lightning arrestor make and model, etc are you using? I should be able to give you some baseline numbers to shoot for.

And before you fall into the DTF trap that alot of people do. Here is some good reading material on the 'proper' DTF usage. It is not a pass/fail thing.

From RF Dan (an Anritsu engineer)

DTF takes an average of the frequency swept VSWR, puts that into an
equation that also takes the default value for the attenuation of a
feeder cable, somehow the default velocity percent gets thrown in and
it redistributes this information over distance.

First, the attenuation values stored in the machine are based on a
manufacture’s catalog specs which are typically spec'd at an ambient
temperature of 68 degrees F. If it is not exactly that temperature
outside when you test the line, you have already introduced a
variable into the equation.

The velocity percent shown in catalogs is also plus or minus a
certain tolerance and is not exact. Another variable.

If you have a 70 foot length 7/8" feeder, and say for example, a 15
foot 1/2" superflex jumper at the bottom, and a 10 foot regular 1/2"
jumper at the top, you are actually applying the loss of the 7/8"
feeder cable to the 25 feet worth of jumpers which have a higher loss
than the feeder - but the DTF calculation can not account for this.
Another variable.

Real life examples:

I visited a site where the installer had a DTF spike in feeder cable
about 35 feet from the top jumper and could not find anything wrong.
The spike was above the operators "spec" so the line was deemed
faulty. At the end of the day, we switched the bottom jumper, 275
feet away from the so called fault in the cable, and the spike went
away. Why? The jumpers "electrical lengths" were different and they
were in a different phase. Changing the bottom jumper had no affect
on the overall system performance but the DTF plot looked better.
Basically, the installer had lost time chasing ghosts.

I received a DTF plot of system (with antenna)that showed a spike in
the middle of a feeder run. Site could not be passed. A DTF plot of
just the cable showed no spike in the feeder. The phase relationship
of the feeder and antenna together created a spurious spike in the
cable. Both the antenna and feeder were good. The system sweep with
both together was stellar. But, the installer had wasted time chasing
ghosts to satisfy his customer's "spec"

Had a guy in Canada saying his cable was bad because he was getting a
1.05 DTF on his connectors. He insisted the cable was making the
connectors look bad because his connector attachment visually looked
great. His system sweep of the transmission line was -31 db. He
replaced the cable, used the same connectors, got 1.02 DTF numbers
and called me to tell me he was right. His system sweep was now -26
dB. But hey, the DTF looked nice, guess it doesn't matter his first
lines actually performed better. Another example of wasted time.

I know for a fact that the following scenario happens. A line is
installed, system sweep exceeds customer's spec, but a DTF spike at 2
connectors is above the DTF "spec." Installer loosens the connector
until the DTF "looks good" and does a balancing act between the DTF
and system sweeps, making the DTF look good enough while having a
negative affect in the system sweep, but keeps the system sweep
barely passing. At the end of the day, the operator, by setting specs
DTF specs, now has a line that does not perform as good as it could
and has a line that is set up for premature failure because the
connectors are not tightened the manufacture's specs.

Actually, I am stern advocate for DTF's "intended" use, which is not
pass fail.

You sweep a line, it passes, then you run a DTF and save it. A year from now, the line is not performing as it did at commissioning for some reason. You run a new DTF plot, compare it to the old, and look for something that has changed so that you are not troubleshooting
blindly. This saves time on troubleshooting. Pass /Fail on DTF wastes
time.

You sweep a line, and it does not pass. You use DTF to look for
something that is grossly wrong. A 1.04 versus and 1.05 versus a 1.07
on a connection is meaningless. If you have a 1.35, then you may have
something to fix. Where is the line versus what is a good connection
versus a poor connection? There isn't. It is indefinable because of
all the factors / variables that go into it. Operators have set DTF
specs on what is "typically" seen in DTF. Something above "typical"
is not necessarily something bad.

I have had operators tell me they set DTF specs to ensure that the
installers are doing a good job regardless of what the system sweep
results are. Little do they know.

If the line passes the system return loss sweep, you have a quality installation with quality products, regardless of what the DTF plot looks like.

[spiderbite]

Thanks for taking the time to share your extensive knowledge.

I apologize for still being confused on a few points....

the following stuff you asked for:

Cisco 1231G AP's 2.412 - 2.437 - 2.462

PolyPhaser DSX-ME Filter EMP Prot 700MHz-2.7GHz
Andrews equivalant to Times microwave LMR 400 (100 ft =6.7dBi loss)
Centurion 8.5 dBi whisper patch antenna (part no. CAF94118)


My confusion is the testing sequence and what results I am looking for:

1st test - cable by itself with load - return loss for ?

2nd test - cable by itself with load DTF -return loss for ?

3rd test cable by itself with short - cable loss for insertion loss

4th test - cable by itself with short - dtf-return loss for accurate cable length measurement

5th test - cable with antenna - return loss for ?pass / fail? is this like VSWR 1:1.5 ratio..

6th test - cable with antenna - dtf return loss ?

7th test - cable with antenna and arrestor -?return loss?


Thanks for taking the time. I am saving all this info for future reference...

Spider

[RFGEOFF]

The act of calibration with the sitemaster series in effect nulls to the load, therefore, any act of calibration should null to the end thus removing any losses in the accoutrements and cables used to test the system. RF Geoff