On Sun, 12 Jan 2014 11:48:13 -0500, Jerry Stuckle
wrote:
I'm not talking about fiber, and I'm not talking about some half-assed
gadget cooked up in your basement. I'm talking about a real,
professional-grade TDR.
Make and model please?
I used to drag around a Tek 1503 but rarely used it. It was most
useful for find broken cables, usually where the rats chewed through
the wires. Also, for measuring the length of cables on a roll that
didn't have ft markers. One day, I needed it, and it was dead,
probably from bouncing around in my truck. Rather than fix it, I sold
it as-is.
http://www.ebay.com/itm/131091152872
I was already carrying a scope, so I saw no reason to buy another
scope just to get a TDR. So, I designed and built a TDR, mostly
checking CAT5 and telco wires, not coax. Something like this:
http://www.epanorama.net/circuits/tdr.html
We use expensive stuff because it gives accurate results.
I use cheap stuff because I know its limitations and can work within
those limitations. Much as I value my money, old test equipment is
good enough for most things.
http://www.learnbydestroying.com/jeffl/pics/home/slides/test-equip-mess.html
You stated that the losses come primarily from the "impedance bump"
presumably produced by the non-50 ohm connector or adapter in the
line. Well, time to put that to the test.
I didn't say anything about a non-50 ohm connector. The connector can
be 50 ohm and still get an impedance bump. With a TDR you can see every
connector on the cable (as long as they are far enough apart - typically
3-4 feet).
If the connectors were a prefect 50 ohms, you wouldn't see any
reflection. Therefore, either the connectors are not quite 50 ohms,
your coax is not quite 50 ohms, you have a high resistance shield
connection, or this is one of your cases where theory fails to
accommodate your reality.
That's theoretical. Reality is much the same.
First of all, we don't crush cables to try to emulate something we can
easily see in real life.
I crush, chop, dissect, and set fire to cable to see how they're
built. Amazing what you find if you look inside of things. However,
since you have an aversion to crushing coax able, you can get the same
effect with a short length of different impedance coax cable and some
adapters. Just insert where I suggested you crush the coax, and you
should see it clearly on the TDR. Then replace it with an identical
length of the correct impedance cable and compare the losses. At
worst, the loss difference will be fractions of a dB. That suggests
that your "impedance bump" isn't really responsible for much loss.
Of course, you have set fire to plenum, non-plenum, and riser CAT5
cables to see how they burn. I think you might be surprised. Learn
by Destroying(tm).
Second of all, you can gimmick up something all you want with signal
generators, spectrum analyzers, crushed cables, all you want.
For those that believe theory, proof by bench testing is possible. For
those who do not believe or understand theory, no amount of test
equipment will ever prove anything. Theory and testing go hand in
hand. If there's a discrepancy, then either one of them is wrong, or
the procedure is flawed.
We SEE the results in real time in real installations with the
appropriate test equipment.
You see the results, but for some odd reason, I haven't seen your
results or list of appropriate professional test equipment. Why
always so vague and non-specific? FoN (Fear of Numbers)?
So? Dipoles aren't 50 ohm antennas. They're typically closer to 75 ohm.
I guess you missed my point. If you can tolerate the 1.5:1 VSWR, the
reduced attenuation and cost of 75 ohm coax makes the effort
worthwhile.
No, it's worthwhile because your coax will more closely match the
antenna. And you can easily match the coax to the transmitter with a tuner.
Tuners can be lossy. Try this T-tuner Java app on the lower bands.
http://fermi.la.asu.edu/w9cf/tuner/tuner.html
The starting values (hit the AutoTune button) produce a 2dB loss at
160 meters. The losses are much less on the higher bands. Q=100 for
the inductor seems a bit high.
Please re-read the article:
www.belden.com/blog/broadcastav/50-Ohms-The-Forgotten-Impedance.cfm
http://www.microwaves101.com/encyclopedia/why50ohms.cfm
I read the article. Unlike you, I understand what it is saying.
What part of the following do you disagree with and why? (It's all in
the above articles). Please try to be specific:
1. Loss for approximately equal size is less for 75 ohm cable.
2. 50 ohm cable will tolerate a higher power due to a higher voltage
breakdown.
3. Maximum power is with about 30 ohm coax cable.
http://www.microwaves101.com/encyclopedia/coax_power.cfm
4. Mismatch loss from 75 ohm cable in a 50 ohm system is 0.18dB.
5. If the mismatch loss from using 75 ohm cable is less than the
difference in attenuation for a comparable 50 ohm cable, and there are
no high power levels involved, then it is better to use 75 ohm cable
and tolerate the mismatch.
You don't know our installations. We've been in business for several
years, with lots of satisfied customers. Trying to tell a professional
what their job is only makes you look stupid.
I only debate a persons ideas and never attack them personally. If
you are unable to do that, you lack "professionalism" which presumably
goes with your professional title.
No MoCA installations?
http://www.mocalliance.org
Nope. It's used by few professionals. Coax ends up with too many
problems. Category cable for runs under 100M or so and fiber for longer
runs.
Fiber is generally better and much more expensive. I also do fiber
where possible. However, when faced with a large apartment complex,
with no possible way to economically rewire the building, I was forced
to consider what could be done with the existing wiring. The building
had CAT3 going to all the apartments, but CAT3 is useless (no twists).
Even DSL had crosstalk problems. I tried to use the CAT3 as a pull
line to drag in CAT5, but that also failed (no smurf tubes in the
walls). I also tried 2.4/5.7 wireless, which proved erratic because
of aluminum foil backed insulation, untouchable firewalls, and
interference from existing Wi-Fi systems.
So, I tired MoCA and it worked.
http://www.netgear.com/service-providers/products/in-home-connectivity/moca-bridges/
http://www.netgear.com/service-providers/products/in-home-connectivity/moca-bridges/wm2500rp.aspx
They're now bringing in Comcast TV, internet, and phone via coax,
using MoCA to distribute internet around each apartment. Depending on
wiring, Jperf shows about 100-300 Mbits/sec between clients (16
maximum). Speedtest shows whatever was ordered from Comcast from/to
the internet, typically 22/2 Mbit/sec.
To be honest, things did not go smoothly. MoCA is expensive. The
isolation between apartments on the main Comcast splitter was not
enough to prevent leakage between bridges. Using couplers instead of
splitters were a big help. I also needed to select a different
channel, which is not allowed by every MoCA unit. I started with
Actiontek, which was cheaper, but does NOT offer channel selection and
was therefore useless. Some of the apartments had satellite TV, which
had to be accommodated with diplexers. However, the worst problem was
that I rapidly became the single point of contact for any problem with
the system. Fortunately, I was conveniently underbid for the
maintenance contract by another service company, which has taken over
the system. To the best of my knowledge, it's still running nicely
after about 18 months.
Next time you run into a situation that does not economically allow a
retrofit of fiber or CAT5/6, and there's existing coax, think about
MoCA.
Find a different distributor or ask for it by part number.
Thomas and Betts SNS1P6U.
We use several distributors - both U.S. and international companies.
Did you any of them if they specifically carry T&B or stock T&B
SNS1P6U connectors? Where we left off, you were having difficulties
finding a stocking distributor. Are either of your unspecified
distributors on the T&T distributor search page?
http://www.tnb.com/ps/dls/dls?ca=corp
Within 25 miles of your location, it shows 5 stocking distributors
with 13 locations.
Quick check: Stocked by Allied, Newark, Tessco, Farnel(UK), MCM.
Not stocked by Digikey, Mouser, Arrow.
T&B distributor search:
http://www.tnb.com/ps/dls/dls?ca=corp
These are consumer (retail) outlets. They are not used by most
professionals.
None of those listed above have a retail (brick and mortar) store.
Many do sell online and from catalogs, but even the largest
distributors do that. Legally, if they charge sales tax, they are a
"retail" operation. Offhand, I can't think of a single large
distributor that does not accept taxable sales or requires a resale
permit in order to process an order. There used to be those as B2B
(business to business) operations, but not any more, because the
growth of internet commerce has made retail processing a necessity.
--
Jeff Liebermann
150 Felker St #D http://www.LearnByDestroying.com
Santa Cruz CA 95060
http://802.11junk.com
Skype: JeffLiebermann AE6KS 831-336-2558