Do the adl5380 demodulator work whit RF freq at 5400MHZ and the LO freq at 5430MHz?
are the I/Q outputs valid 30 MHz outputs in this case?
An RF input frequency of 5400 MHz and an LO frequency of 5430 are both within the operating range of the ADL5380. And since the baseband ouputs have a bandwidth of 390 MHz, the device will be able to handle an output frequency of 30 MHz. You do do need to put an image reject filter in front of the ADL5380 IQ Demodulator. Without and image reject filter, input signals at 5460 will also be mixed down to 30 MHz.
Is the amplitude of 30 Mhz I/Q channels dipendent of the Phase-change RF signal?
i'm a beginner I/Q demodulation!!
The I and Q outputs provide a cartesian representation of the instantaneous phase of the input signal. So the instantaneous phase is equal ato ArcTan(Q/I). So, yes, as the phase changes, the amplitudes of the I and Q signals will also change.
1)With the quadrature demodulation, can i measure phase variation from -180 to 180 or only from 0 to 180 degrees?
2)In the ADL5380 datasheet (page 26) i see the image rejection graph. At 5400 MHz the rejection is 40 db! I need analogic filter also?
An IQ Demodulator can measure phase in the +/-180 degree range. Regarding the image, I think that you are mixing up rejection of the image frequency and the image rejection of the IQ modulator. Take a look at the attached article for a futher explanation
so, only with a frequency-zero IF i have not image problem?
mmm... i do not understand.
I bought the ADL5380 evalutation board from analog devices site.
I connect the LO sma connettor to HP generator and i set the freq at 5430MHz and 0dbm power.
I connect the RF sma connector to AGILENT generator and i set the freq at 5400 MHz and
I connect the CH1 channel of digital oscilloscope to IF-I sma connector (load 50 ohm).
I connect the CH2 channel of digital oscilloscope to IF-Q sma connector (load 50 ohm).
The signals I/Q are sinusoid-signal at 30 MHz:
1) The amplitude of 2 channels is the same amplitude.
2)The phase-difference of 2 channels is 10 degrees (I/Q).
3)If i change the power level of RF channel(5400MHz), the levels of I/Q sinusoids changes alike.(the power difference difference is 0 db).
4)the phase-difference of I/Q channels never change.
Is it correct?
Your setup is correct and what you're observing is correct. I need to back up a little and ask what you're trying to do and how do you expect to use the ADL5380? Is this for educational purposes or is it part of a project you are working on. Here are a few comments: The phase accuracy of the ADL5380 is approx. +/- 1degrees and the amplitude imbalance is less than 1dB. Mesuring the phase imbalance on an oscilloscope is very crude and not very accurate. You will not get the resolution needed on the oscilloscope. The phase difference of 10 degrees you had measured on the oscilloscope is quite high and most likely not correct. Depending on what types of equipment you have available to you, you may want to put the I and Q signal into a spectrum analyzer and look at the spectrum in the frequency domain. This is where you will see the image in the negative spectrum and this is where Eamon's email will start making sense.
I need a circuit for do this:
1)the signal to analize is a non-coherent RADAR-PULSE of 0.5uSEC. (The amplitude of this signal is variable, depending from distance of target. The phase of signal is RANDOM, magnetron pulse).
2)The microwave pulse is NOT modulated in freq or phase.
3)the circuit must down-converter this signal to IF frequency (with use of stable local oscillator) in 2 signals (I/Q), and i must have to outputs the information of amplitude and phase (referenced to LOcal osc).
is it possible with a i/Q demodulator?
A quadrature demodulator's I & Q outputs can be combined, say, in an FFT analyzer, to deduce the sign of the (assuming doppler shifted) IF also, in addition to the frequency and amplitude. In order to measure phase, the demodulator's LO should be locked to the Transmitter.
A word of caution: the phase between I & Q will always be 90 degrees (plus or minus the < 1 deg. of quadrature error) for a CW RF input, so if you trigger one osc. channel with the (low-passed) I channel output, the Q channel should be shifted by 90 deg. all the time. If you only saw 10 deg. difference between I & Q, there may be path difference (cable length, scope channel filtering, or baseband LPF phase response, etc.) causing the extra phase difference.
The FFT analyzer's output should provide the absolute phase information of each channel, referencing to the sampling clock.
If you only saw 10 deg. difference between I & Q, there may be path difference (cable length, scope channel filtering, or baseband LPF phase response, etc.) causing the extra phase difference. The FFT analyzer's output should provide the absolute phase information of each channel, referencing to the sampling clock. Benjamin
I use ADL5380 evalutation board and the cable length from board from EVboard to oscilloscope is the same.No extra filter is present.
IF the pulse to analise have width 0.5uSEC, for good FFT (phase accurancy 1°) can i sample the IF at 120 MHz?
Not sure about phase accuracy vs. sampling frequency. I believe you'll have to integrate the phase over many pulses to get the accuracy, assuming jitter from pulse is low enough. Need help from someone with better system/radar knowledge here :-(
is not possible use many pulses to improve the measure, because the phase is random from pulse to pulse (magnetron).
The idea is to lock the receiver's LO to the Tx's (magnetron), so that phase variations from the source will be subtracted out, and the "only" variation left is that from the reflected signal, which would change only if the target is moving relative to the RADAR.
this is a classic tecnique for coherent radar. (it use coho) locked to tx.
But moderns radar measure TX phase pulse , referenced at reference-oscillator, then measure RX phase pulse , referenced at same reference oscillator, and if the distance of targhet not change the difference from 2 measures never change.
Retrieving data ...