bh | Harris RF-505A Receiver |
The Harris RF-505A is a synthesized-tuning full-coverage longwave-thru-shortwave receiver from the early 1970s.
Tuning to 100 Hz resolution is accomplished via six 10-position rotary switches, one switch per decade from Tens-MHz down to Hundreds-Hz.
The switch-based tuning tends to make this receiver unappealing for SWL'ing and band-scanning activities.
There is a continuous-tune VFO mode that can be switched in, but it only spans 10KHz so is only useful for tweaking up an individual station.
With the rather bland appearance, no colorful dial or warm NIXIE glow, one could easily pass by these units - dismissed as an obscure piece of lab or industrial equipment - until one looks more closely at the front panel to discern what it actually is. They don't seem to receive much love in the receiver community. Contrast with the Drake MSR-2, a synthesized receiver from the same period but with NIXIES and a combination of switch tuning and continuous VFO tuning, the VFO spanning 100KHz.
On the other hand, this was a very high-end receiver in its day and interesting as an early synthesised PLL-tuned receiver. A marketing brochure for the model has a copyright date from 1971. The two units I've worked on have component date codes in the 1977-1982 period. The physical design and IC and semiconductor complement are indeed indicative of the late-1960s / early-1970s.
CONTENTS (this page): |
SUB-PAGES:
|
RELATED: | EXTERNAL LINKS & REFERENCES: |
The RF-505A is a dual conversion design with full digital PLL frequency synthesis for tuning.
The first IF is well up in the VHF: 156 MHz.
The second IF is a more typical 500KHz.
After the 2nd-conversion mixer, dual IF filters, IF amplifiers, detectors and audio paths provide for simultaneous reception of both upper and lower sidebands, as well as AM and CW modes.
Immediate attenuation of the RF input is provided but incorporated with manual gain control of the IF amplifiers. The single RF-Gain knob controls both of these factors. AGC is always present, it cannot be switched off.
A preselector provides independant tuning of the RF input for frequencies above 2MHz, below 2MHz a fixed filter is present. The preselector also provides a single bipolar-transistor stage of gain. With the preselector switched out there is no input tuning or RF gain stage, the first active stage from the antenna is the balanced-diode-ring 1st mixer.
RF/IF signal path switching is done electronically, keeping the routing of such signals localised on-board and avoiding longer paths out and through mechanical switches.
The 500KHz IF filters are Rockwell/Collins mechanical filters.
The semiconductor complement to accomplish all this includes:
The analog ICs are mostly simple early RCA and Motorola differential/cascode amplifiers.
The digital ICs are TTL, mostly 7400 series with a few Signetics 8200 series and couple of Motorola 2000 series where higher speed is necessary.
Numerous fixed frequencies are required for mixing purposes and tuning frequency synthesis.
These are generated by digital division, harmonic selection and multiplication, and PLL techniques,
all tied back to a single 5 MHz master reference.
Three option modules for the internal master reference were available, varying in stability from 1-in-10^6 to 1-in-10^8. In any case an external master reference can be used, selected by an internal switch. The internal master reference can be fed out on the rear panel and thus a bank of RF-505As can all be fed by a single reference from one of them.
The 5MHz master reference is fed to a pulse generator which serves both as a source for digital dividers and a harmonic-rich source for harmonic multipliers.
The 2nd-conversion Local Oscillator frequency of 156.5 MHz for converting the 1st IF of 156 MHz to the 2nd IF of 500 KHz is generated by an offset-PLL. Although the VCO in this loop utilizes another crystal, it is tweaked by a varactor fed by the loop phase comparator, and both the offset frequency and the phase reference trace back to the 5 MHz master reference.
It is not clear why the designers used a PLL for the 2nd L.O., rather than just mixing the 160 MHz reference with the 3.5 MHz reference, and filtering and amplifying the difference. Spectral purity wouldn't seem to be the reason as in the PLL form the final 156.5 MHz is derived from filtering a pulse generator output.
The 2nd L.O. generation of the RF-505A is a simple example of an offset-PLL. With 160MHz injected into the loop feedback path via the mixer (see 2nd L.O. PLL diagram), to get the input into the phase comparator from the mixer to match the 3.5 MHz at the reference input to the phase comparator the loop must adjust the VCXO/multiplier combination to a final output of 156.5 MHz (160-156.5=3.5).
The two PLLs in the tuning frequency synthesis, discussed below, are also offset-PLLs.
The down-side to the offset-PLL is the overall loop stability is now dependant on the stability of the injected frequency as well the phase-reference frequency.
This is accounted for in the RF-505A by all the offset-PLL injection frequencies tracing back to the master reference (with the exception of the manually-tuneable VCXO in VFO mode).
The objective of the frequency synthesizer is to produce the 1st-conversion Local Oscillator frequency for injection into the 1st mixer,
the frequency being determined by the setting of the 6 decade tuning switches.
This frequency must be adjustable from 156.0000 MHz to 185.9999 MHz in 100 Hz increments.
The synthesizer is based around 2 offset-PLLs. The first PLL (upper portion of synthesizer block diagram) produces a variable frequency in accordance with the setting of the 3 least-significant-digit tuning switches. The second PLL (lower portion of synthesizer block diagram) produces a variable frequency in accordance with the setting of the 3 most-significant-digit tuning switches while also being offset-adjusted by the output of the first, to produce the complete 1st Local Oscillator output.
The VCO of the LSD PLL varies from 10.000 to 10.999 MHz. This is down-mixed with a 12 MHz reference prior to digital division for frequency selection and comparison to a 1 KHz reference in the loop phase-comparator. The VCO output is also divided by 10 prior to mixing with another frequency. In all-switch tuning mode this latter frequency is a 15 MHz reference. The loop equations for the LSD loop in this mode are:
LSD3 : tuning switch setting of 3 least significant digits [000,999] NL : LSD loop division factor F(LSD) : frequency output from LSD loop and mixers NL = 2000 - LSD3 F(LSD) = 15,000KHz - (12,000KHz - NL*1KHz)/10
In VFO tuning mode the 2 LSD switches are disabled, their division-factor contribution fixed at 10*10, and the 15 MHz reference switched out in favor of output from the VFO oscillator. This is a crystal oscillator but tweaked across approximately 11 KHz by a varactor controlled by the front-panel VFO potentiometer. The exact range of the VFO can be adjusted via two trimmer pots. In VFO mode the LSD loop equations are:
LSD3 : Tens-KHz tuning switch setting [0,9] * 100 NL = 2000 - LSD3 VFO = 14,995 ± 5.5 KHz F(LSD) = VFO - (12,000KHz - NL*1KHz)/10
The loop equations for the MSD loop are:
MSD3 : tuning switch setting of 3 most significant digits [000,299] NM : MSD loop division factor F(LO1) : 1st Local Oscillator mixing frequency NM = 100 + MSD3 F(LO1) = 4*NM*25KHz + 160,000KHz - F(LSD)To determine the final received frequency, the receiver IF injection frequencies must be accounted for:
F(rcv) : received frequency F(rcv) = F(LO1) - (156,500KHz - 500KHz)
There are two versions of the front-panel preselector tuning scales.
On one version (shown in photo), 5 and 10 MHz appear twice on the scales, with the appearance of overlap between the upper end of one scale and lower end of the subsequent scale.
In fact, the upper end of the scales with such overlap is not operative.
The preselector tuning scales apply as follows:
00,nnn.n - 01,nnn.n MHz preselector tuning inoperative 02,nnn.n - 04,nnn.n MHz use inner-most scale 05,nnn.n - 09,nnn.n MHz use middle scale 10,nnn.n - 29,nnn.n MHz use outer-most scale
There are also two versions of the preselector module.
The circuit arrangement around the attenuator, gain control, and preselector in/out switch differs.
The Units-KHz knob pulls out to double in function as the VFO-mode tuning control.
A small drive belt links the knob to the VFO potentiometer.
The belt is a metal cable but with plastic teeth.
The plastic deteriorates with age and teeth may eventually strip from the cable, as seen in the photo.
Neoprene O-rings make a less-than-perfect but workable replacement. Two types from the common AS568B series O-rings have been observed to fit:
The photo shows a #127 O-ring fit in place.
Note the hole for the potentiometer is slotted, permitting adjustment for belt tension (best minimised).
Replacement requires removing the receiver module to access the Units-KHz switch/pot drive assembly, unscrewing of the switch plate (to avoid unsoldering all the wires to the plate), unsoldering 3-4 remaining wires to the assembly, and removal of the assembly.
The belt should be placed such that mid-positions of the potentiometer and knob align.
After reassembly the VFO range may be set by adjusting the two trimmers on the assembly.
Monitor VFO board IC A4.6 with a frequency counter, trim so frequencies match when knob is pulled out at 0 and 9.
Oddly, the lamp type for the meters does not seem to be listed in the manual. There are two types which may be valid:
Unit Log
Harris RF-505A |
bhilpert 2019 Aug |