Ceramic 2.4GHz antenna

If you are a vendor and you spam my email address you will definitely _not_ be getting my business. You are free to state your parts here, but I'll expect you to back it up with samples and information.

Cheers

PeteS

I'm in the business of calculating such stuff. What kind of characteristic in terms of directionality, gain and bandwidth do you need ? Then a bit closer to pricing : if you have a cable in between, the cable is 50 ohms, but if you have it close to the electronics, you might have a different impedance in mind.

Rene

In this application, I will be feeding the antenna on an impedance controlled track of no more than 2 inches length, 50 ohm. If you are interested, this is for a 802.11b/g unit I am integrating into the latest and greatest.

I can deal with the usual issues. One thing is there will be a huge ground plane (LCD display) about 12 mm away. I have the time to do experiments with tuning components.

Bandwidth: I would prefer the antenna not to splatter beyond the ISM band (let's say 15dB attenuation out of band minimum)

Gain: I can live with 0-3dBi

Directionality: I would prefer a dual polarised device. This is a handheld unit and it can be (obviously) held at a number of angles.

Cheers

PeteS

Hi Rene

I might have an interest in that cable tester you host on your site. Can you send me (or have them send me) information on pricing / availability for just a couple of units?

Cheers

PeteS

My experience with 2.45GHz ceramic antennas is that once they are in a real environment the performance bears little resemblance to anything found in the datasheet. Very good results can be obtained with pcb antennas, but they do need a little more space. (They don't need to be PIFAs - 1/4 wave monopoles adjacent to a ground plane can work very nicely.)

In a small device you won't get much directionality, whether you want it or not.

John

Dumb question: why 50 ohms and not 75 ohms?

Rene

The impedance of the chipset I am using is 50 ohms (bog standard RF), and the simplest method is a direct connection (well, through a small cap anyway). I will no doubt end up with a couple of tuning components, but that's simple compared to a full impedance transform.

:)

Cheers

PeteS

If you have 2 inches to spare, you are very close to a quarter wave anyway !

Isn't that backwards?

I thought higher impedance meant smaller dia center conductors and hence higher losses. (I use the tiny center conductors in scope probe cables to remember the sign bit.)

I'm pretty sure the old coax Ethernet used 50 ohms rather than 75 because it had lower losses.

snipped-for-privacy@ip-64-139-1-69.sjc.megapath.net (Hal Murray) wrote in news:Jd2dnSFc3diBBcLYnZ2dnUVZ snipped-for-privacy@megapath.net:

Impedance is defined by the spacing made by the dielectric though, not by the conductor size. In TV coax the solid core is usually thicker than the total of the stranded cores in thinner 50 ohm lines. I don't know how much the 'skin effect' matters at sub-1GHz frequencies though. Anyway, I don't think that the relation between loss and impedance is fixed, it will depend on capacitance, and also at really high frequencies, on the surface area of a given length of conductor, because that's where the signals is, with this 'skin effect' that no doubt I know not nearly enough about.

characteristic impedane is not solely due to the size of the conductors.

overall dimensions of the cable and the electrical properties of the dielectric play the largest part.

Bye. Jasen

...and has nothing to do with free-space impedance of (about) 300 ohms?

For a fixed power, the current scales as sqrt(P/R), meaning with a higher impedance you indeed need less copper. The dielectric is specified for almost a kilovolt, perhaps even more. For the usual RF guy, the cost of a cable is not an issue, everything else costs far more. Not so the TV companies. Their investment are the cables.

Rene

Pete,

which chipset are you using? I am in the process of NDA-ing (Marvell) so I can get some data to do probably the same which you are doing, so I wonder (if you would/could discolse that, that is, privately or publically :-).

All, I also wonder what the impact of a metal case would be on such an antenna - one of the options I have is a metal (0.5mm thick brass) case, largish handheld size (with an opening for the TFT etc.).

Dimiter

------------------------------------------------------ Dimiter Popoff Transgalactic Instruments

------------------------------------------------------

PeteS wrote:

Telly antennas generally use a folded dipole / semi log period array , being frequency unrelated a single folded dipole presents an impedance of 300 OHMS at resonance and the 4:1 balun reduces this to 75 OHMS which is why RG59 au is normally used ( or RG6) The reason for 50 OHM cable being used is most transistor circuits are built with a 50 OHM impedance because of several factor some of which are convenience and available baluns Hence the antenna being built for 50 OHM at resonance .

IBM used RG-62, 93 ohm coax for their terminals for decades.

On a sunny day (Sun, 19 Nov 2006 17:24:08 GMT) it happened "Michael A. Terrell" wrote in :

They also invented OS/2, and MCA (microchannel bus) ;-)

Jan Panteltje wrote in news:ejq4mc$qf1$ snipped-for-privacy@news.datemas.de:

And the so-called Deathstar drives. In which case, the obituaries are fortunately premature. They never let me down. I never had the known bad ones though.