FM modulation of a CW signal

Yes of course, you are right, in my excitement i didnt remember:!: 

This problem looking for a solution has been going through my mind for about 18 months, ever since i bought this analyzer, This solution idea has actually got me quite excited, its so simple!

Hi Jomac,
I hope Harry, who proposed you the double conversion technique, will find a bit of time and answer your questions. To be honest, I am pretty unsure about the best frequency plan.

BR from Ivan

Hello Ivan,
              May i bother you for a little advice, to make sure im on the right track?

Im gathering bits together now for the first experiments in the first block, the 10Mhz to 50Mhz unit. Im trying to decide what would be the ideal crystal frequencies for the 2 conversions, could i safely go below the 10 Mhz to convert down and then back up, or would i be better going much higher in frequency by possibly using a separate oscillator?

I have several 'exact' frequency computer crystals, ie 4,6,8,10,12,16,20 Mhz and several odd frequencies ie 22.325Mhz etc.

If i have to buy a pair of higher 3rd or 5th harmonic frequency crystals to take me way above 50Mhz, then im happy to do so.

If you could point me in the right direction as regards to frequency for the down and up conversions, i would be really greatful.

John

OK Ivan, i can see the logic now, and i think the added advantage would be getting used to the principles and the construction of each module.

After some deep thinking, I asked myself would i need coverage for the frequency range 500 to 1050 Mhz, and the answer is no, with the exception of the telemetry and data band that falls within 860 to 870 Mhz and this could be a separate, stand alone unit to worry about later.

My question now is what would the ideal frequency be for the local oscillator on the 10 to 50 Mhz? I have many crystals which are exact frequencies from about 4 Mhz to 27 Mhz and if i used the 3rd overtone of say the 27 Mhz crystal, that would give me a LO of 81 Mhz well away from the operating frequency. I also have many salvaged crystals from computer boards which have odd frequencies like 14.350 Mhz etc.

J

Hello Jomac,
I tried to make the bands so that the ratio of lower and higher limit is cca 1:5 with some overlap. The 10 - 50 MHz would employ HF parts and construction, 40 - 220 MHz VHF and 220 - 1050 MHz UHF ones. Design other frequency division as you need, but keep the frequency limits ratio as small as possible.

BR from Ivan

Hello Ivan,
              Ive always liked the NE602/612, practically everything ive built with these has worked, and worked well, they are as common in my toolbox as a BC107.

Is there any technical reason why you have suggested starting at 10Mhz and stopping at 50mhz? I only ask, because of 2 things, 455khz is still useful, and to cover the 2mtr amateur band and PMR would be convenient, but i will take your advice.

Really when i think about it, in reality i wouldnt want to go much above 600 Mhz, and the 612, looking at the datasheet seems to be fine up to 500mhz.

J

Hi Jomac,
you are right, it would be reasonable to start FM at 10 MHz to cover the most common FM IF 10,7 MHz. The bands may be e.g. 10-50, 40-220 and 200-1050 MHz. Well, the mixer concept would probably still require switched filters to cover each one of them. The PLL concept seems to be more wideband.
The ICs based on Gilbert cell (NE612 etc.) are quite good mixers. Keep on them up to the frequency limit.

BR from Ivan

I could maybe have occasion if i was working on an IF subsection that had an IF of 10.7 or 455Hhz but that would be pretty rare. OK so if i concentrated on dividing the band into 3 subsections and concentrated on the lower frequencies first. What would be the ideal approach? My understanding of mixers is pretty basic and limited to devices such as the NE602/612 devices.

Hi Jomac,
I expect you have no need to frequency modulate the whole frequency range in one device. It has little sense to FM carriers below 20 MHz. The range from 20 MHz to 1,1 GHz can be covered by, say, 3 separate adapters.

BR from Ivan

Yes i can understand the theory behind this now, the idea seems so simple! But if i wanted to be able to FM modulate the full frequency range of the spectrum analyzers tracking generator range, which is 150Khz to 1050Mhz i assume i would have to mix the output of the tracking generator at a frequency higher then 1050Mhz?

Interesting that at my work in the laboratories we have a lovely piece of kit for simulating RF transmisstion to/from mobile telephones. One item can do exactly what you want (but costs US$200,000). The object is to simulate dopler shift with speed, for example when the mobile is in a train or aeroplane.

It works by feeding the signal into a mixer to convert to an Intermediate Frequency (IF). The IF is then converted back up to the operating frequency in a similar mixer. If both mixer oscillators are the same frequency then you get out exactly what you put in.

If you were then to modulate one of the mixer oscillators then the modulation will be applied to the output signal. It is an extremely simple method but it works. This same system is used to make voice changes that can make you sound like Darth Vader or Jimmy Clitheroe. If you are intending to use simple modulation to test receivers then this could be built up in an evening:

Mixer => LPF => Mixer => BPF => Amplifier

If you are not particularly fussy about the amplitude then you can skip the ampifier.

If you need a wide range, then as Ivan pointed out the PLL would be the way to go.

BR Harry - SM0VPO

Hi all,
this is really strange. When I am logged in, the link in my post is where I have put it and points where I want. When I am not logged in, my original link is hidden and another link, said being inserted by VigLink and pointing to a commercial, appears !   In any case, the links are barely visible...

BR from Ivan

Hi Jomac,
the links are badly visible here. They are blue like the other text, slightly darker only. In my reply they are the words "like this one". When you point to them, they turn red and get visible. Maybe Harry can change the color scheme ?

In any case, the link points here : http://www.ka9q.net/crackpots/arrl-pll.png

BR from Ivan

Hi,
you can probably use a PLL type FM modulator like this one. The tracking generator output would be the reference. 

BR from Ivan

I have a budget spectrum analyzer, its the Atten 6011 with tracking generator. It covers 0.15Mhz to 1050Mhz.

Its a wonderful piece of kit and i would like to extend its usefulness. Ive taken the output from the tracking generator and AM modulated the signal, but what i really want to do is to FM modulate the signal.

Is there any way i can do this, without opening up and going inside the equipment?

Having an AM/FM/CW signal generator would really extend its usefulness.