Help required: How to generate rf sweep

Hi all,


In my project, I need to generate sweep signals for the frequency range for GSM bands ie. 890-960Mhz and 1710-1860Mhz.

1. How to generate sweep signals for the above frequency range ?

2. Can it be generated for the above frequency or should i generate sweep signal for 2ghz and select the above two frequencies while sending?

3. What components should be used ?

Please help me with some ideas

You could buy a VCO (or 2) Micronetics | RF Components: VCO's

Keith.

You could use test equipment, eg a sweep generator or a signal generator with a sweep function fitted.

Or you could use a VCO as suggested above (either free running or with a PLL)
Or you could upcovert a DDS (this would give a very fast and precise sweep)

Note that the bands are harmonically related so you could make the sweeper for the lower band and then simply go through a doubler + filter to get the second band.


How quickly does it need to sweep?
How good does the phase noise have to be and what level of spurious signals can be tolerated?

---------- Post added at 16:14 ---------- Previous post was at 16:01 ----------

The most cheap and cheerful (accurate) design would be based on an ADF4350 from Analog Devices. You could either sweep under PLL control, or presumably you could sweep the internal VCO from an external ramp generator.

If you set a wide PLL loop bandwidth in the ADF4350 you should be able to tune across the band with a step retune time of 200us. So if you had 400 points in each band that would be 80 milliseconds to do an accurate sweep on one band.
This would equate to 250kHz step sizes on the 900MHz band.

The ADF4350 has everything included for the PLL eg dividers, phase detector and VCO and it can cover 137.5MHz to 4300MHz.

You just need to add a crystal oscillator for the reference clock plus a fast MCU chip (with a simple assy program) to control the ADF4350 over the SPI interface.

wow.. thank you very much...

saw the ic ADF4350, it provides the whole transmission for my project . But i got a doubt in tat ic, hope u can clear it

1. there is an auxiliary port in that ic RF out b+andb- .... does it give the same output freq as in RF a+and a- or can it be used to generate seperate freq range. ie. in my case 890-960mhz in RF OUT a and 1710-1860mhz in RF OUT b

I see, that you're moving ahead your project in little slices... Does the project also involve building and verifying the hardware? If so, you should consider, that it also implies theoretical and practical engineering skills and e.g. the ability, to identify suitable parts and to extract the necessary information from the datasheets. Reviewing your "virtual project folder" (the various contributions related to the project) up to now, I have some doubts. That won't be a problem, if you can get support and instructions for the project at your place, but you won't hardly be able acquire it completely through the internet.

I apologize for the possibly uninvited comment.

My concern about a DIY spectrum analyser is 'how good will it be?'. If you haven't the means to test it then it could be pretty bad. That's ok if it is a hobby RF project but if you are hoping to make real measurements with it you will have problems.

I have a couple of commercial spectrum analysers, one with tracking generator, and designing and building broadband amplifiers with a flat response is not trivial. Without such test equipment you are likely to have a very uneven response in your home built one.

That is why I would suggest looking at the projects on the internet which use pre-built modules.

Keith