Which IC could be used to downconvert signals around ~100 to ~1000 MHz to audio band?
I found UPC1686g
datasheet: http://www.cel.com/pdf/datasheets/upc1686g.pdf
but it seems it is not suitable for my task.
Please suggest some IC, ideally controlled over I2C or similar digital interface, which is able to set internal VCO with small frequency steps and provide downconversion.
Almost all mixers can downconvert to audio, the issue is why? The reason for asking is the technology and device you need may be different depending on what the output is to be used for. You also need to give us some idea of tuning resolution you need, bearing in mind that most I2C devices with a built in VCO are based on PLL circuits which have fixed frequency steps. There may be ways to overcome that but without knowing the resolution you need, the I2C speed may not allow some methods to be used.
Brian.
Look at "Software Defined Radio". Analog Devices is nowadays very popular but there are others.
For instance..https://myriadrf.org/
They are nonprofit organization and there are lot of information including softwares,drivers,ICs etc.
downconvert WHAT to audio band? You will hear NOTHING.
IF the RF signal had some audio modulation on it, then you could use a demodulation circuit to turn the modulation back into audio output. The exact type of demodulation circuit would depend on what type of modulation the RF signal had, such as AM, FM, SSB, Phase Modulation, and on and on
My first step is to hear unmodulated carrier. For unmodulated carriers i will hear an audio tone i guess. I was searching for single package mixer with integrated VCO, without demodulation schematics or with complementary analog IF outputs. Simplest configuration is VCO + single ended diode mixer. I want frequency steps of less than 300kHz, it may be higher, but the lower is better, as it reduces requirements to the audio part. 20kHz steps would be perfect, then PC mic input could be used directly.
You won't hear anything from an unmodulated carrier, the mixer output would be DC.
The reason for being so cautious with answers is it sounds like you are expecting to make a receiver with some vital parts missing. You have to bear in mind that if you (example) receive a signal at 500MHz and mix it with a 500MHz LO, the output will not be audio, you get any deviation from one signal to the other, such as phase and minor frequency shift instead. Any modulation on the signal will also produce two products, LO+modulationl and LO-modulation which may be difficult to utilize. That is the primary purpose of using an IF, it keeps the products apart in the frequency domain so they are easier to filter.
However, for some signals it may be possible to directly demodulate into in-phase and quadrature (I & Q) signals using two mixers but to do so you have to consider how to produce the 90 degree phase shift over a wide frequency range.
I would suggest if you want to do some inexpensive experiments, you buy a DDS frequency generator board from a well known on-line source and first experiment with mixers over the frequency range it can produce. Afterwards, you can substitute it with a PLL to widen the tuning range.
Brian.
If we take 2.4GHz transceiver and inject 2.400001000GHz , would not it be any 1000Hz tone? I think it would. I am going to mix unmodulated signals. One from vco, another - random unmodulated carriers. I do not care about image, I do not need quadrature receiver. Only unmodulated signals. Asking about IC with integrated vco and single ended mixer. Tuning bandwidth around 1GHz. No high pass filters on mixing output, so I can obtain mixing result in audio band. Are you really sure there are no any mixing output if we tune VCO near some unmodulated carrier frequency and put it to mixer? Example of unmodulated carrier is test mode if nrf24l01. I do not need data, demodulation, etc. Just mixing result of incoming signal with vco in a tiny package. I do not receive data, do not receive voice or any other kind of modulated signals. Consider all signals are simple sinusoidal tones.
Almost - but not quite right.
If you took a 2.4GHz signal source and mixed it with 24.00001GHz it would beat to produce a 1KHz tone because that is their frequency difference. It isn't quite what you asked before. You need both signals to produce the difference, just injecting an LO will not produce any output unless there is a fixed and constant carrier to mix with.
Incidentally, if you use devices like the nrf24l01 to produce unmodulated carrier it will not give good results. Devices like that tend to produce ripple in their frequency due to the way their PLL works. It isn't significant when used for wideband data transmission but the 'noise' will be much more noticable when dropped into the audio range.
Brian.