After long time I decided to take my IC-7700 a part and clean the dust inside.
As it is an HEAVY radio, this operation was always postponed.
Very nice piece of equipment!!
In the mean time I decided to give a look at the famous spur.
I find the “f0 -36 kHz spur” on the scope very annoying.
I use the radio 99% as IF on 144 MHz for tropo o EME operations and of course the low band noise let me ALWAYS see the spur...
Icom documented the "possibility" of some spurious signals.
The spur is readable on all bands with variable intensity, depending on the band and preamplifier settings (on 14 MHz band it is lower with the preamp in line!).
It is more a psychological problem rather than a real degradation of the receiver characteristics: I really hate seeing it on the waterfall all the time.
The SCOPE UNIT works well, the problem comes from the RF UNIT which introduces an unwanted signal into the SCP_FROM_RF_UNIT input. The SCOPE UNIT therefore “reads” correctly also the spur at about -103 dBm (value is depending of mode or bands).
There is not any kind of interaction with a supposed 36 kHz signal on the SCOPE UNIT. The spur is ABOUT 36 kHz from the fo, changing from mode to mode.
Unwanted signal at the SCOPE UNIT input:
64.4910 MHz in AM or FM
64.4895 MHz in LSB
64.4925 MHz in USB or PSK
64,4105 MHz in RTTY
I measured the level of LO arriving from the PLL UNIT and all was correct.
The signal RX 2nd Lo R2LO coming from the PLL UNIT enter into the RF UNIT and then goes to the mixer IC1501+1502 with the correct levels.
For some reasons a low portion of it is picked up also by the buffer Q1352 which send part of the 1st IF to the scope.
I have had some results on reducing the scope spur acting on C1656 (RF UNIT) which is on the 90° phase shifter circuit. But degrading the RX performance was definitely not the right way...
Did the RF UNIT IMAGE RESPONSE calibration procedure to re-set the correct value again.
I suspected that the unwanted signal was somewhere radiated (after Q1651 the LO level is quite high).
I have put a good number of ferrite cores on the coaxial cables coming in or going out the RF UNIT having only a marginal reduction of the spur only applying a ferrite to the SCOPE BOARD inputs cables SCP_FROM_RF_UNIT + SCPT_FROM_RF_UNIT. I kept that core on the radio.
Another test was to move the coaxial cables on the RF UNIT. As there are a lot of wires it was quite difficult and I never get enough good results (variation +-3 dB).
Only in one CASUAL position the spur level decreased consistently but I was not able to reproduce that again!... That led me thinking that there is a multiple interaction and sitting the cover will modify again the results.
I have finished to resitting every coaxial cable on the RF UNIT on a different way, keeping the 2 wires going to the SCOPE UNIT and the RX_2nd_LO_R2LO more separated possible.
I did not remove the RF UNIT board to make further measurements.
The final result is a spur on the scope lower about 4 dB (+-1 dB depending on the band).
Actually not a big improvement, but better than nothing.
Further, to cancel the spur on the waterfall, I have setted the scope input attenuator at 10 dB for every band. This reduces the sensitivity of the scope but seems to be still usable for general use.
I hope that somebody will use these notes as starting point for further improvements on this clear lack of design.