>From - Sat Mar 02 00:57:15 2024
Received: by 10.35.119.11 with SMTP id w11mr8181417pym.1171919132925;
Mon, 19 Feb 2007 13:05:32 -0800 (PST)
Return-Path: <jm..._at_tscm.com>
Received: from lvs00-fl-n11.ftl.affinity.com (lvs00-fl-n11.ftl.affinity.com [216.219.253.186])
by mx.google.com with ESMTP id a28si1337331pye.2007.02.19.13.05.32;
Mon, 19 Feb 2007 13:05:32 -0800 (PST)
Received-SPF: neutral (google.com: 216.219.253.186 is neither permitted nor denied by best guess record for domain of jm..._at_tscm.com)
Received: from [70.22.244.211] ([70.22.244.211]:32787 "EHLO Raphael.tscm.com")
by ams011.ftl.affinity.com with ESMTP id S365591AbXBSVFn convert rfc822-to-8bit
(ORCPT <rfc822;T..._at_googlegroups.com>);
Mon, 19 Feb 2007 16:05:43 -0500
Message-Id: <7.0.1.0.2.20070219142659.08256510_at_tscm.com>
X-Mailer: QUALCOMM Windows Eudora Version 7.0.1.0
Date: Mon, 19 Feb 2007 15:56:14 -0500
To: "TSCM-L Professionals List" <TSCM-..._at_googlegroups.com>
From: "James M. Atkinson" <jm..._at_tscm.com>
Subject: Re: [TSCM-L] Wideband (AM) detectors improvement using filters
?
In-Reply-To: <1171551154.380997.148490_at_s48g2000cws.googlegroups.com>
References: <1171551154.380997.148490_at_s48g2000cws.googlegroups.com>
Mime-Version: 1.0
Content-Type: text/plain; charset="iso-8859-1"; formatowed
Content-Transfer-Encoding: 8BIT
Check out:
http://www.tscm.com/TSCM101noise.html
Response as follows:
At 09:52 AM 2/15/2007, cont..._at_yahoo.co.uk wrote:
>.
>
>
>Wideband (AM) detectors are very popular...because they are easy to
>use
>and relatively cheap.
>
>...the problem is that they are not very sensitive.
>...another problem is that they cant separate signals at different
>frequencies.
>
>
>Question
>--------------
>
>
>-/////////////////////////-
>
>How many times more sensitive would a wideband detector become
>if it had additional filters in the frontend ?
>
>-/////////////////////////-
>
It can be hundreds of thousands, or even millions of times more sensitive.
Fow example, let say you want to build a simple
diode detector that monitors 398.5-400 MHz (1.5
MHz wide), you build a simple 8 pole bandpass
filter that contains 4 high pass, and then 4 low
pass filters, or better yet, 8 high pass
cavities, and 8 low pass cavities. This will
knock down even the most powerful FM and TV
broadcast signals so you can pass the signal to
a LNA and then into a diode, and then sum the
voltage that comes out of this diode with a
output of a matching diode (reference diode/amp),
and apply the output of the summing circuit to
another amplifier that drives a simple 30 dB or larger column of LED's.
The goal here is to take a hypothetical FM radio
station that is presenting -20 dBm on site, and
to knock it down below -174 dB. This means that
we want to kill everything outside of the band by
154 dB, but I know some engineers who use only 40
db, others use 120 dB, and some even use 400 dB
as their design goals for TSCM and SIGINT
bandpass filters. An example is the 60 cycle high
pass filters that I make, they knock down 60 Hz
by 400 dB, so you can plug one end into a 480
volt circuit, and the other end into the RF input
of your spectrum analyzer or oscilloscope. We als
comb out all the harmonics of the line voltage
and knock down any nastiness that might cause blue smoke to leak out of the SA.
>For example the detector does 1 Mhz to 3000 Mhz
>
>Now i will devide that into 5 bands...using 5 separate filters :
>
>a) 1-80 Mhz
>b) 80-200 Mhz
>c) 200-520 Mhz
>d) 520-1000 Mhz
>e) 1000-3000 Mhz
>
>
>Ofcourse when using one of above filters... it seems very logical
>that
>the maximal detection range becomes much better then without a filter.
>
>I suppose that is caused by below factors :
When not use the wide band detector (3 GHz) to
gate the sampling side of the post filter
detectors. Simply put, when the 3 GHz diode
captures energy on the log scale that exceeds a
certain level (say -50 dB) that it uses to gate
all of the output of the diodes assigned to each of the five bands of interest.
Lets assume (grin) that I have a stack of
Astro-Med Everest analog chart recorders in a
rack. Each chart recorder is capable of 32 analog
channels each, and I gang or slave 4 of them
together so that I have 128 virtual channels. I
built a trigger circuit for each chart recorder
that monitors a wide swath of RF spectrum, so we
have the first recorder listening and reacting to
100 Hz to 1.8 GHz, the second 1 .5 to 18 GHz, the
third from 12 GHz to 30 GHz, and so on. When the
broadband energy at say 500 MHz goes over -50 dB
it triggers the chart recorder to start moving
the paper for 5 seconds, but looking at the chart
you can see which band of the 32 the energy
appears in. The tricky part is that each of the
32 channels have to have thier own filter and
LNA, then diode, and interface for the chart recorder.
Maximum range requires first maximum filtration,
folwed by maximim amplification, folwed by the
most sensitive detection diodes you can find,
then further amplification to drive whatever is
giong to be your indicator (LED column, speaker, chart recorder, etc)
>1) The environmental noisefloor is much lower so a signal would stick
>out much easier.
Yes, but you have to know the frequency on which
the bugs appear, and this is a tricky
proposition. You have to break the bands into
things like 135-150 MHz, then 150 to 174, 174 to 200 MHz, and so forth.
>2) Bandwidth in the receiver is directly related to noise.
>
Yes, it is all part of the KTB formula, with B
being that which we can control the most.
>Offcourse the adantage of filtering would also be that :
>
>1) The wanted signal is not masked (overpowered) by other out of band
>signals
Big rule of TSCM is to estimate that which you
should see in the area you will be sweeping, and
then to bring along filters to knock down things
that may be making your measurements "dirty".
Personally, I carry a fixed 8 pole band reject
filter for every TV station allocation, plus a FM
band reject filters (more then one), and filters
to knock out the paging bands.
You know what your going to need by looking up
FCC allocations for everything within five miles,
but 75 miles for any broadcast (AM, FM, TV)
channels. If I am performing a sweep in Needham,
MA there are huge antenna farms for FM and TV
broadcasting, so every inch of the office that I
am sweep in getting hammered with high levels of
RF. Not only do I have to know down the main
carrier, but even the 15th harmonic of the main
carrier is radiating off of structural steel and
causing massive headaches. This means that I have
to use a filter that kills not only the main
carrier, but also filters that kill every
harmonic out to the 20th or so carriers that is
going to tilt or saturate the sweep gear.
Once you get above 2 GHz you can use one of more
YIG filters to help you do this, but the low
frequencies (below 1 GHz) are going to be quite troublesome.
>2) You have an indication of the frequency-band of the found signal
>
>...etc
Simple sample and hold circuit that triggers a flip-flop circuit.
>The mentioned unit lets you either switch to each filter by hand...but
>it can also go into a so called "Scanning-Filter mode"
>
>Filter-scanning mode will switch between filters like this :
>
>Filter 1 >> Filter 2 >> Filter 3 >> Filter 4 >> Filter 5 >> NO-
>Filter ......and over again.
Simple six pole mechanical coax switch/relay with
a control circuit. It has been done before with
good results. Basically, you take a six pole
switch, and hand a dedicated filter, amplifier,
and antenna off each of the six input connectors.
One key is not to apply power to any of the six
amplifiers other then the one you are using at the time.
>That NO/filter seems to be a good idea in practice...to also receive
>out of band signals
>not covered by the filters (in this case below 1 mhz and above 3000
>mhz)
>
>The scanning-speed has still to be determined...it depends on the
>shortest duration
>of possible digital signals (pulses) that you want to detect :
>
>...if it is to slow you might miss a short on-time signal (non
>continious bursts)
>...if it is to fast to fast there is no time to validate the signal
>(continious but very-short pulses)
A very effective method of detecting tracking
devices in a vehicle is to use a gang of generic
diode detector such as the CPM-700, add your own
bandpass filters for each of the "phone to tower"
frequencies, and follow it by a small 20 dB
preamplier, or "convert" the RF probes.
To convert a probe you unscrew the probe body,
unscrew the antenna, unsolder the screw that the
antenna attached to and install an SMA female
connector. Next visit your local hardware store
and purchase some one inch tubing and fittings.
Mount the SMA/probe assembly into the end cap,
and solder the ground of the probe to the copper
as well. Then slide the whole thing into a length
of copper tubing that it roughly half the
wavelength of the frequency of interest. The
other end of the copper tube is a Male BNC
connector so that you can merely plug the tube
into the BNC connector of the CPM-700
Next fabricate an antenna out of a length of
copper clad welding wire and terminate it on a
female SMA connector so that you have a high performance dipole antenna.
Inside the copper tubing you will want to add a
small 8 pole bandpass filter and kill off the
tower-to-base and trunking frequencies by at
least 80 dB befoe you go into the probes amplifer.
The simplest this is to make a standalone box the
size of a carton of cigarettes into which the
phone-to-tower antenna where mounted, then feed
into separate filters for each band (8 poles
each), and then take the output and apply it to
dedicated amps all built into the same case.
The box would best be a brass block that you
drill out with cavities for each of the filter
poles, and then a large cavity for the
amplifiers. You could go with an interdigital
filters for each of the bands, but you really do
not need to displace much heat, and the physical
size of the 800 MHz band will be unpleasantly
large and heavy. I would (and have) leaned toward
a simple LC filters for a project like this and
simply add a lot of pole to keep the skirts of
the filter as sharp as possible.
Companies that make filters for the cell phones
themselves also make very small duplexer circuits
that you can use in reverse, but they tend not to
knock down the signals by more then 20 ot 30 dB,
but it may be a good place for you to start experimenting.
>Forget the scanning speed...lets talk just about sensitivity-
>improvement when using filters.
I am a *** BIG *** fan of carefully designed and
built filters, and their use in TSCM.
I am also an even bigger fan of using big-ass
(excuse my French) wide band antennas to suck in as much energy as possible.
>I know that its a difficult question...but lets say i have to write an
>article or propaganda
>for a unit with those 5 switchable filters...and i have to say how
>many times more sensitive
>a Wideband(AM)-detector with 5 filters like above is...compared to the
>same one without filters
>
>I suppose that the smaller the filter-bandwidth the more "sensitive"
>so lets simplefy the question
>and lets assume that the 5 used filters all have the same bandwidth so
>the whole range is equally
>devided
>
>------------------------------------------------------------------------------------
>How many times more sensitivity or range would you get ?
>------------------------------------------------------------------------------------
>
>
>Thanks !
>
>
>Contranl
>
>
>
>Remark...i dont understand why not one single manufacturer of (AM)
>wideband-detectors
> has not incorperated some filters...they are not
>difficult to produce
> and not expensive...but greatly improves them.
>
>
>Please do not respond (in private) to my (old) Yahoo-emailadress...i
>´ve lost the password
>and after weeks of correspondence they dont want to give it...Yahoo
>SUCKS !
----------------------------------------------------------------------------------------------------
World Class, Professional, Ethical, and Competent Bug Sweeps, and
Wiretap Detection using Sophisticated Laboratory Grade Test Equipment.
----------------------------------------------------------------------------------------------------
James M. Atkinson Phone: (978) 546-3803
Granite Island Group Fax: (978) 546-9467
127 Eastern Avenue #291 Web:
http://www.tscm.com/
Gloucester, MA 01931-8008 E-mail: mailto:jm..._at_tscm.com
----------------------------------------------------------------------------------------------------
We perform bug sweeps like it's a full contact sport, we take no prisoners,
and we give no quarter. Our goal is to simply, and completely stop the spy.
----------------------------------------------------------------------------------------------------
Received on Sat Mar 02 2024 - 00:57:15 CST