Users that search through noise all of the time probably will not know if their receiver is deficient in the performance characteristics necessary for this sort of search. The easiest way to check is to listen to the 'Note' of a stable signal in the SSB mode (an Amateur beacon, a crystal oscillator harmonic etc.). A clear 'sweet' tone indicates good performance, but a noisy/raucous tone indicates poor performance.
The narrow band performance of my R7000 has been progressively worsening over the years, and I have been in contact repeatedly with the local agent over those years in an attempt to get a fix for the problem. The receiver has been in to the local agent twice, but with little improvement. I have also checked on a range of internet sites, and tried the modifications relevant to poor narrow-band performance, but again found little improvement.
The modifications detailed below caused a dramatic improvement to the usability and performance of the receiver, particularly in narrow band modes.
I have also fitted an external heat sink on the power supply rectifier to move the excess heat generated outside of the case.
My R7000 performed well on SSB initially, but progressively worsened over the months and years until SSB became almost unintelligible, particularly on the higher frequency VCO.
VCO PHASE NOISE
Investigation of this fault revealed that the problem was caused by very high phase noise on the PLL sub-assembly VCOs. The bandwidth of the noise on the lower frequency VCO was 5KHz, and on the higher frequency VCO was 8KHz.
I found very high noise levels on the supply rails feeding the VCOs. The voltage out of the regulator IC1 was clean, but the switching transistors Q8 and Q14 had excessive voltage drop across them, allowing noise modulation of the supply voltages to the VCOs and other stages on the PLL board. The voltage drop across Q8 was 300mV, and across Q14 was over 400mV.
CAUTION, The modifications listed below involve the replacement of small and fragile components. Do not attempt these modifications unless you have experience working with this type of componentry and have the correct tools and a temperature controlled soldering iron.
I replaced the 2.2K switching transistor bias resistors R37 and R57 with 1K resistors. This reduced the voltage drop across the switching transistors to well under 200mV, and achieved a significant reduction in VCO phase noise. As an added bonus I also noted an improvement in general receiver performance.
To further reduce the noise on the VCO supply voltages, I made the following modifications:
The result of this modification is a dramatic improvement in SSB intelligibility and quality. The CW note changes from wideband noise that sounds a bit like a tone, to a tone with some residual low frequency noise modulation, again a dramatic improvement.
entire website copyright © The SETI League, Inc.
this page last updated 15 February 2003
Top of Page