Help with designing a Microstrip.

[EC MEISTER]

I'm planning on making a reflectometer to measure efficiency in antenna design and I plan on using a pcb microstrip design. Trouble is I'm unsure on how to calculate this. I've found this calculator but I can't make sense of it. My design will be 50ohm, but I can't see how this calculator would give me track widths and spacings required to obtain such a figure. Any help as to how or where to start would be appreciated.

[vpunk]

I know nothing about this but if you change the track width and length parameters in the calculator, the impedance Z0 changes and by trial and error changing the width to 102.04 (everything else at its default) gives an impedance of exactly 50Ohms but it all depends on other factors like your frequency etc.
You have to click on the arrows between the Width and Impedance boxes to get it to calculate

[2xCPU]

I'm very interested in how you go with this - if it's looking good, there will probably be enough interest here to justify a batch of well made boards.

When I looked briefly at striplines, I quickly gave up because Eagle does not seem to have the ability to do per-trace-clearance (only global). So I just laid out the board to keep the RF tracks as short as possible. Also, I had no data on the PCB I was using. Guessing the values for a simulation rather defeats the purpose.

The easiest way may be to lay the board out as best you can. Then import it into a CAD/SIM package to do the final track/clearances.

2.

[EC MEISTER]

The calculator fails for me, the height is set at 58 and is invalid, yet height is the only section i can't modify or click on. I'm working with very little knowledge here myself, should clarify I'm modeling this off an old swr meter. In the Meter it used a central through rod with two smaller precise sized and spaced rods either side of it.

In PCB terms, should turn out to be something like this:

Click to view full size!

[EC MEISTER]

I've found this which is of interest. I'm just wondering if I'll have to use the whole 50mm pcb length or if a slightly longer or shorter piece would affect the 50ohm rating.

I'm a bit stuck as to where a 1nF germanium cap would be connected to and where the pot would be connected to. Is there an alternate way of making a meter? Possibly using a micro-controller or the like.

[paulvk]

It will all depend on what frequency you want to deal with, the higher you go the more loss in a PCB you may even need teflon board. So what requency? I have built and have a number of these devices.

[2xCPU]

Originally Posted by EC MEISTER

I'm a bit stuck as to where a 1nF germanium cap would be connected to

germanium cap? I think you may have miss-read it.
Futurlec have BAT81 @0.40c each

and where the pot would be connected to.

Once the signal is rectified, you don't need to treat it as RF. So the pot just gets mounted to the front panel and connected with DC style wiring.
You'll need to shield the sample board of stray RF will drive the meter.

Is there an alternate way of making a meter? Possibly using a micro-controller or the like.

I had one bookmarked, but it's now 404. Fortunately I took a copy. It uses an 16F877 and 16x2 LCD, but is only intended for HF (I assume you're planning to use it somewhat higher?).
The beauty of SWR meters is that the sample deck and display are independent. So you could use the OH2NLT display on your deck.
* Just found it on the authors site: OH2NLT

In the Meter it used a central through rod with two smaller precise sized and spaced rods either side of it.

That sounds familiar, is it the project that uses lengths of semi-rigid coax soldered together with matching slits in the jackets for sampling?

2.

[EC MEISTER]

Originally Posted by 2xCPU

germanium cap? I think you may have miss-read it.
Futurlec have BAT81 @0.40c each
2.

Would a BAT81S-TR do or does it have to be BAT81. (going off farnell)
I've been advised to use a germanium cap somewhere in there.
Frequency is around 433Mhz.

That sounds familiar, is it the project that uses lengths of semi-rigid coax soldered together with matching slits in the jackets for sampling?

It'll be coax with F connectors on the board. Direct line from the antenna.

[2xCPU]

433Mhz - Amateur or LIPID?
You said in the first post that you are designing for 50 Ohms. Do you realise F connectors are 75 Ohms?

BAT81S-TR
The TR just indicates the packaging (Tape on Reel).
I think the S just indicates Vishay is the manufacture.

2.

[paulvk]

Ok 433MHz what power levels are we talking about?

[EC MEISTER]

Ok, maybe the F connector was just for the end of the coax cable to copper wire di-pole connection.

Ok let me share more of the details. Essentially there are a few "distress signal" radios lying around. Using an RX2 receiver I'm attempting to make a system to give a bearing of the direction of the distress signals. The transmitters will intermittently transmit a carrier and an identifier (in morse code) on 433.19 MHZ at approx 10 mW.

The purpose of the microstrip is to make an SWR meter to make sure the antenna has a SWR of no more than 1.2

[paulvk]

At that low power level you will need a different sampling system. At such low power levels you should be more concerned with resonance of the antenna, there are plenty of examples with great detail on how to build for this frequency range and a centimeter or so will not have any great affect. You need to look up "Fox Hunting" this is where a transmitter is hidden and people go out and find it you will find a lot of information on how to build the fox and devices to find it.

[EC MEISTER]

Fox hunting... ok thanks that gives me something to work with.

[2xCPU]

I assume this is just a 'conceptual' distress signal and your not planning to rely on it.
As it's a receiving only antenna, the VSWR really isn't important.

As paulvk says, fox hunting is the place to start.
Have search for Doppler RDF.

2.