There is too much gain in many LNB's (many are +60dB) so you need to tame them and remove microwave filters if they do not cover the correct frequency. Microwave block down converters are great to use as 'off the shelf' and often cheap ways to 'see' a section of the microwave spectrum. Whether they are able to work at the required frequencies is another matter as most are dedicated to a task and filtered as such. The Local Oscillators are also normally at some odd frequency rather than a simple 1,2,3,4 GHz etc. A bit of simple maths is then needed to work out the displayed frequency on the SA. You also need to know whether they have a forward or reverse IF output (high or low side LO injection).
The cheap and dirty method to up and down convert is to buy a passive mixer that covers the frequencies that you want, both IF and RF, you then inject a Local oscillator at around +7dBm at the desired frequency.
In your case I would choose a mini circuits mixer that can handle the 1 to 2 GHz RF side whilst providing 0 to 1GHz Intermediate Frequency. This is all Superheterodyne (Superhet) receiver theory.
http://en.wikipedia.org/wiki/Superheterodyne_receiverWithout filtering there are issues with mixer products but if you know the frequency you are wishing to observe you should be OK . You can also upconvert the output of a Tracking Generator in the same way. In my example you could upconvert a 0 to 1 GHz sweep to 1 to 2 GHz using a 1GHz LO.
The local oscillator can be anything decent with enough drive for the mixer or use aMMIC to amplify the drive level. Mini-circuits produce suitable VCO's that will work well for the task and they can be set to the desired frequency.
Calibration is a challenge. If you are using a VCO, you need to either Phase Lock it to a crystal or let it free run and monitor the output frequency, adjusting the frequency manually as required. A fixed, crystal controlled 1GHz oscillator would bee ideal but these are not easy to find cheaply. Amplitude calibration is best done by comparison to a signal generator producing a near-by frequency against which the level calibration may be set with an attenuator.
The simplest converter that I built was formatted as follows:
A manually trimmed 1GHz Voltage Controlled Oscillator was fed to a Minicircuits ERA MMIC RF amplifier, and then split between two mixer LO inputs using a Minicircuits splitter/combiner. The two Mixers were microwave types from minicircuits that provided an IF spec of 0-1GHz and and RF/LO of 1 to 2GHz.the LO feed requirement was +7dBm.
One mixer was fed with the output of the Tracking Generator into its IF input, producing 1 to 2 GHz from its RF output. The other mixer was fed with the RF into its RF input and produced an IF of 0 to 1GHz out to feed the SA input.
All this was done using convenient Minicircuits modules and SMA interconnects. This was the fast way to build such a unit but a cheaper method is to use any commonly available VCO and mixers that meet the desired frequency coverage. Simple SMD MMIC amplifiers may be used in combination with SMD attenuators to get the levels right.
www.minicircuits.comWith a combined up and down converter sharing the same VCO frequency, return loss and transmission tests may be carried out using a decent directional coupler. Minicircuits sell this as well !
Fraser