Post by thread:[EE] Antenna recomendations for EMF detector

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 As part of a larger project I've been developing ways to detect mobile devices that emit radio waves, cellphones, bluetooth, wifi etc. I've got one design that seems to work, based on a RF log detector chip. More info, as well as test results detecting GSM cellphones and other wireless devices, is on my website: http://petertodd.org/tech/emf-detector What I don't have is an antenna. I'd like it to be a small omni directional pcb antenna, that'll pick up frequencies above ~800mhz. (GSM in Canada uses a 850mhz band and a 1900mhz band) My RF log detector chip is rated from 50mhz to 3000mhz, although I'm assuming that detecting lower frequencies would be a bad thing as stuff like CB radio's probably put out relatively large amounts of power. With no antenna the system seems to work decently well, at least for GSM phones, but I'm assuming better design would let me detect those phones from farther away. Any suggestions on antenna design? - -- http://petertodd.org -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.6 (GNU/Linux) iD8DBQFHk7O63bMhDbI9xWQRAj0+AKCf222nHodarcPFn3bCNkcbOK4hTwCgiqF7 mol2FQSox55+dz4L8nyCQGA= =MHXH -----END PGP SIGNATURE-----

Hi Peter, I'd suggest looking up the following: log spiral log periodic broadband antennas discone antenna heillical antenna coil antenna There are a general class of antennas which have no "characteristic dimension" like a dipole would (length). Instead, they have a "characteristic ratio", which, from a fundamental point of view, means that you can argue that they cannot possibly prefer one frequency to another (within some range). Some of these fit into that class (log spiral and log periodic at least - not sure about discone). In reality, they work by having a current distribution which expands and contracts to the right length to resonate at whatever frequency you are using. I think the disadvantage is that they are larger than a single-band antenna for any of the bands in their frequency range would be. If you don't need very good sensitivity, then a simple ground plane antenna set up for the lower end of your frequency range might work very well. It will be less efficient, attenuate to varying degrees, and exhibit more complex antenna patterns as you go up in frequency, but is probably OK for your application. A small coil antenna could also work well. I think you might be able to get that to maintain less reactance change over the frequency range, at the expense of lower radiation resistance (hence, lower efficiency). Sean On Jan 20, 2008 3:48 PM, Peter Todd <spam_OUTpeteTakeThisOuTpetertodd.ca> wrote: {Quote hidden}> -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA1 > > As part of a larger project I've been developing ways to detect mobile > devices that emit radio waves, cellphones, bluetooth, wifi etc. I've got > one design that seems to work, based on a RF log detector chip. More > info, as well as test results detecting GSM cellphones and other > wireless devices, is on my website: > > http://petertodd.org/tech/emf-detector > > > What I don't have is an antenna. I'd like it to be a small omni > directional pcb antenna, that'll pick up frequencies above ~800mhz. (GSM > in Canada uses a 850mhz band and a 1900mhz band) My RF log detector chip > is rated from 50mhz to 3000mhz, although I'm assuming that detecting > lower frequencies would be a bad thing as stuff like CB radio's probably > put out relatively large amounts of power. With no antenna the system > seems to work decently well, at least for GSM phones, but I'm assuming > better design would let me detect those phones from farther away. > > Any suggestions on antenna design? > > - -- > http://petertodd.org > -----BEGIN PGP SIGNATURE----- > Version: GnuPG v1.4.6 (GNU/Linux) > > iD8DBQFHk7O63bMhDbI9xWQRAj0+AKCf222nHodarcPFn3bCNkcbOK4hTwCgiqF7 > mol2FQSox55+dz4L8nyCQGA= > =MHXH > -----END PGP SIGNATURE----- > -

> On Jan 20, 2008 3:48 PM, Peter Todd <.....peteKILLspam@spam@petertodd.ca> wrote: > > As part of a larger project I've been developing ways to detect mobile > > devices that emit radio waves, cellphones, bluetooth, wifi etc..... > > > > What I don't have is an ... > > > > Any suggestions on antenna design? For PCB-etched antennae, maybe do a Google for fractal aerials (or antennae, in US English). Small size, broadband and quite elegant looking. eg. http://en.wikipedia.org/wiki/Fractal_antenna There's a Spanish site around - can't find it right now - that had some PDFs on fractal aerial design. best - Debbie. Make the switch to the world's best email. Get the new Yahoo!7 Mail now. http://www.yahoo7.com.au/worldsbestemail

Hi Peter: The simplest antenna would be a quarter wave whip above a ground plane. These are great for frequency ranges that are much less than an octave. The next simplest broad band antennas for this frequency range might be a disk cone or inverted disk cone. These cover a number of octaves of frequency. One element, the disk, is just some flat metal that doesn't need to be a disk. The other element is a cone with it's pointed end toward the ground plane. The low frequency limit is set by the cone point to base length and the high frequency by the geometry where the cone point meets the ground plane. A standard disk cone has the cone as the ground pointing up with a disk sitting on top. These are commonly used with police scanner radios or WiFi hacking. www.ramseyelectronics.com/downloads/manuals/DA25.pdf http://www.ramseyelectronics.com/site/default.asp?search=antenna&page=search The inverted design often uses the metal on a PC board as the disk ground and adds an inverted cone above it. The Biconical is just a discone where the disk is removed and replaced by a second cone (tip to tip), See: http://www.prc68.com/I/OE254.shtml This one is driven from a balun transformer contained in the hub and so all the elements are at DC ground. A very good thing to limit noise and signals that are below the frequency of operation. You may want to add a transformer or balun at the input to your RF log amp to high pass filter the input. I.e. eliminate lower frequencies like FM broadcast, taxi, police, pager base stations (probably the strongest signals) etc. from getting into the amp. Better a filter structure. Similar to the biconical is the bow tie. Commonly used in corner reflector TV antennas. More on various antennas at: http://www.prc68.com/I/Ant.shtml -- Have Fun, Brooke Clarke http://www.prc68.com/Alpha.shtml All my web pages listed based on html name http://www.PRC68.com http://www.precisionclock.com http://www.prc68.com/I/WebCam2.shtml 24/7 Sky-Weather-Astronomy Web Cam

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On Sun, Jan 20, 2008 at 09:39:36PM -0500, Sean Breheny wrote: {Quote hidden}> Hi Peter, > > I'd suggest looking up the following: > > log spiral > log periodic > broadband antennas > discone antenna > heillical antenna > coil antenna > > There are a general class of antennas which have no "characteristic > dimension" like a dipole would (length). Instead, they have a > "characteristic ratio", which, from a fundamental point of view, means > that you can argue that they cannot possibly prefer one frequency to > another (within some range). Some of these fit into that class (log > spiral and log periodic at least - not sure about discone). Yup, I think discone fits in there as well, at least, if I understand wikipedia's entry on it. > In reality, they work by having a current distribution which expands > and contracts to the right length to resonate at whatever frequency > you are using. ! Sounds like black magic to me. :) > I think the disadvantage is that they are larger than a single-band > antenna for any of the bands in their frequency range would be. More difficult fabrication too. I think I can rule out the discone for instance due to the cone shape, same goes for the log spiral antennas. Log periodic would be doable in a pcb, but it's very directional from what I read on it. I couldn't find much info on coil antennas, at least for higher frequencies. Looks like the ham world uses lots of loop antennas, for say the 10mhz band, but I didn't find any info on using them for higher frquencies. > If you don't need very good sensitivity, then a simple ground plane > antenna set up for the lower end of your frequency range might work > very well. It will be less efficient, attenuate to varying degrees, > and exhibit more complex antenna patterns as you go up in frequency, > but is probably OK for your application. A small coil antenna could > also work well. I think you might be able to get that to maintain less > reactance change over the frequency range, at the expense of lower > radiation resistance (hence, lower efficiency). My current pcb isn't an ideal ground plane by any means due to it's small size, though I did try simply attaching a length of wire to the pcb vertically. It seemed to work as well, once I trimmed the length a bit to about 1.5", longer and the noise floor gets higher, shorter and there is less signal. This seems to be a very simple solution, so I might just use it. I'll be doing some more experimentation, as well as making up a larger board with a bigger ground plane to test that out. - -- http://petertodd.org -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.6 (GNU/Linux) iD8DBQFHmQjm3bMhDbI9xWQRAvcMAJ4kCZd3WTRXW5ipGcqBfO41eZj6/ACfUmqr 3mz9rGKjggyeLvfdzwYm3Es= =wiQk -----END PGP SIGNATURE-----