Rev 2.0 | SC5309A & SC5310A
Hardware Manual
SignalCore, Inc.
16
SC5309A & SC5310A Hardware Manual
intercept point (IIP3). The IIP3 of the mixer is typically better than +18 dBm, and its P1dB is better
than 5 dBm. In applications that require better signal linearity, it is recommended that the level at
the mixer input is kept at -20 dBm or less. Typically, the IMD is better than -75 dBc for two -20 dBm
tones at the mixer. Increasing the level at the mixer improves SNR of the IF output signal at the
expense of linearity.
Another important characteristic of the mixer is the LO-IF isolation and LO-RF isolation. The higher
the isolation, the lower the leakage LO signal is at the RF and IF ports. The IF1 mixer is driven with
a
≈
+20 dBm LO1 signal and the LO-IF isolation is about 45 dB, resulting in a -25 dBm LO1 leakage in
the IF1 path. If the RF signal level at the mixer is -20 dBm and the mixer conversion loss is -8 dB,
the converted IF1 signal is -28 dBm. For this case, the LO leakage is greater than the converted IF
signal. For most of the RF input range, the LO1 leakage is not an issue because the first IF filter (IF1
BPF) will suppress it. Furthermore, when the leakage passes through the IF2 bandpass filters, it will
be further suppressed. However, when the downconverter turns to lower frequency, especially
when the RF is near the value of the bandwidth of these filters, the LO1 becomes an unwanted
spur.
The frequency relationship between the three ports of the mixer is given as:
𝐿𝑂 = 𝐼𝐹 + 𝑅𝐹
As RF approaches 0 Hz, LO -> IF, resulting in the LO leakage being inside the passband of the filter.
The filters are not able to optimally reject leakages inside their passband or within their filter slope.
This LO1 leakage will mix with the converted IF1 signals or with LO2 to produce higher order
unwanted in-band spurs that could be large enough to affect the recovery of signal data. These
intermodulation spurs are usually not so easy to determine and deal with. Having high linearity
components helps maintain these spurs at acceptable levels. The linearity in these modules keeps
these types of 3
rd
order effects at typically less than -60 dBc. In addition, the LO leakage is a known
signal that shows up as the DC value when the final IF spectrum is digitized and processed. For
lower RF frequencies it is recommended to switch the RF directly to a digitizer at the IF port instead
of putting it through the conversion stages. The user should determine whether the application
can tolerate LO leakages, what the effects will be, and decide on the most appropriate route to
take.
The first IF (IF1) frequency is centered typically at 3.675 GHz with a minimum bandwidth of 100
MHz. This center frequency may be adjusted programmatically from 3.675 GHz to ± 50 MHz offset.
The default center frequency is set at the factory at the time of calibration to compensate for shifts
in the IF1 and IF2 filter responses due to fabrication variations. Another advantage of setting the IF
frequency higher is to place it closer to the filter slope region, providing better LO1 leakage
suppression.
This stage of conversion is an up conversion rather than a down conversion. This broadband
architecture simplifies the input RF and IF1 sections to achieve:
1.
Superior image rejection without the use of multiple RF bandpass filters.
2.
Great RF suppression and isolation in the IF2 passband without multiple IF and LO frequency
bands.