basic RF lab setup

[yanir]

Hi, I've been designing electronics for work for a few years and started as a hobbyist. Still relatively new to this world. I've recently gotten into RF and want to build a basic lab for this purpose. As far as electronics go, equipment isn't cheap but jeez, RF is another order of magnitude! Logic analyzers and network analyzers are ridiculously expensive.
Anyone know of good brands and used models that are good for tinkering? I'm interested in microwave bandwidth, I've been playing with some cellular modules at home and would like to do impedance match testing on antenna microstrip. Those frequencies are in the ghz range. That equipment can get pricey. I'd also settle for playing around in lower frequencies if the equipment is dirt cheap.

So 1. I need to perform an impedance match test on 900mhz - 1800mhz for some signal tracks on my board.
                How can I do this cheaply?
     2. I'd like to play with wifi and other 2.4ghz technologies
     3. I'll take anything that is good for learning.

Anyone done this before? Any recommendations? I have a rigol scope 100mhz (originally 50mhz, thanks the forum). I need some other basic equipment I can build, like a lab power supply and a signal generator.

I've heard that you can use a signal analyzer and tracking generator to substitute a VNA in some situations. This seems like a cost effective way to go. Thoughts?

Thanks

[Short Circuit]

There is some PC/USB based equipment on the market;
http://www.signalhound.eu/buy.htm
http://www.sdr-kits.net/ seems nice too, but limited to 1.3GHz

No experience with either one though.

[olsenn]

the RF Explorer WSUB1G + 2.4G may be useful to you. It's a handheld spectrum analyzer. You can buy one here: http://www.seeedstudio.com/depot/rf-explorer-model-wsub1g-p-922.html?cPath=174

Other than that, make sure you have all the basics: 5.5+ Digit DMM, scope (your Rigol should suffice), DC power... another thing to remember is to make sure your dicrete components (transistors, caps etc) can operate at RF frequencies. You may wish to start off with a 27MHz transmitter/receiver project, which you can fully analyze using the FFT feature of your Rigol oscilloscope. Once you get good at designing at (relatively) low frequencies, then you can go into the GSM bands.

Good luck!

[Zad]

It depends what exactly you want to do. Ideally you want a spectrum analyser which goes up to 5x the fundamental frequency you are operating at. There can be too much obsession with test gear though, most design is done in your head, and not playing around with $20k analysers. Once you are at the test gear stage, most of the hard work has (or should have) been done.

Generally speaking though, the amateur radio community is the best place to look. Magazines, local clubs, "radio rallies" and of course Ebay. Simple network analysis can be done with a signal generator, directional couplers and a detector. The ARRL has all sorts of books you may be interested in, for example:

http://www.arrl.org/shop/Microwave-Know-How/
http://www.arrl.org/shop/Microwave-Projects/
http://www.arrl.org/shop/Microwave-Projects-2

[yanir]

Zad, Your right, It's easy to get obsessed with all of that equipment and drool. I'll check out those books.
olsenn, Building some 'from scratch' 27mhz projects sounds like a fun way to learn. I think I'll do that.

I do have to build a board with a cellular module though. all of the real rf work was done my the module manufacturer, I just need to do some impedance match testing for the antenna microstrip line after I design the board. There is a lot of good info on how to design a microstrip for a particular impedance. I would like to test it after the pcb fab though. Does anyone know of a cheap test setup for just this one purpose?

[gregariz]

Zad, Your right, It's easy to get obsessed with all of that equipment and drool. I'll check out those books.
olsenn, Building some 'from scratch' 27mhz projects sounds like a fun way to learn. I think I'll do that.

For good or for worse, this board has turned into a cross between hobby zone and advertising space for whatever online shop is flogging. I find I still have a read out of a self destructive interest that hasnt subsided yet but I ignore most of the posts these days - particularly the almost constant obsession with test equipment.

I do have to build a board with a cellular module though. all of the real rf work was done my the module manufacturer, I just need to do some impedance match testing for the antenna microstrip line after I design the board. There is a lot of good info on how to design a microstrip for a particular impedance. I would like to test it after the pcb fab though. Does anyone know of a cheap test setup for just this one purpose?

IMO if you need to lay microstrip to an antenna, you should just follow the appropriate microstrip calculations and layout to the letter, and 'if it doesnt perform' as advertised then you should ask someone who has an RF shop to perform the measurements with you. Or worst case rent the equipment. It won't be worth getting set up for an occasional job unless its a real money spinner for you. The frequency is just a little high for most of the ham instruments you will run across.

[yanir]

IMO if you need to lay microstrip to an antenna, you should just follow the appropriate microstrip calculations and layout to the letter, and 'if it doesnt perform' as advertised then you should ask someone who has an RF shop to perform the measurements with you. Or worst case rent the equipment. It won't be worth getting set up for an occasional job unless its a real money spinner for you. The frequency is just a little high for most of the ham instruments you will run across.

That's the plan. I just want to do some testing before sending the board to a lab to make sure I'm not way off. Lab fees are expensive too. If there is a "quick and dirty" method of checking impedance I was hoping some "gray beards" my have some insight.

[ejeffrey]

I have a signalhound USB spectrum analyzer.  It performs acceptably for my needs, which is measuring the modulation spectrum of a laser.  It has a tracking generator option, but I am not sure if it can do VNA mode or only SNA.  I haven't really done any tests on the noise floor or timebase stability, my signals tend to be quite bright and narrow, and they drift more than the SA would anyway: something like a 3 GHz carrier with 2-3 sidebands at several MHz.  The biggest problem I have is that the software is a bit clunky.  Most importantly, it is a bit slow -- both the spectrum analyzer and the software.  The spectrum analyzer only has 240 kHz of realtime bandwidth except in the low resolution 5 MHz RBW mode.  Also the UI controls don't respond except at the end of a sweep.  If you are at a high resolution narrow span and the signal you are watching drifts off the screen, when you bump up the span you can easily get to 30 second sweeps and it doesn't appear to let you cancel or increase the RBW until the 30 seconds are up.  A 4.4 GHz SA for $1000 + $500 for the tracking generator it is not a bad deal, but on the other hand I am used to a bit more polish in a $1500 product.

If all you need is SWR you can do that with a coupler / circulator and a power meter.