Sensitivity of various HF receivers...


 As HF receivers pass through the shop here, I make a crude attempt to measure their sensitivity -  this page is a record of what I've seen...

 I use an H-P 8640 RF Signal Generator, externally modulated by an H-P 4935A Transmission Test Set. The 4935A set is intended for telephone system repair & troubleshooting work -  it has an automated 'signal-to-noise' test function which I use in these receiver tests.

 The originally intended use of the 4935A signal-to-noise test is to loop an internally-generated 1004 Hz tone through some portion of a telephone system, and then examine the returned signal and determine how much noise was added to the audio passband by the telephone system.

 The 4935A calculates the audio S+N/N ratio by alternately measuring total audio passband power and then measuring audio passband power with the original 1004 Hz tone notched out. That alternation is plainly audible in the 4935A speaker and the withnotch/withoutnotch test cycles in roughly two seconds - this cycling continues until the user exits the 'signal-to-noise' test mode. The difference in 'unnotched' audio passband power vs. 'notched' audio passand power is the S+N/N ratio.

 Rather than passing the 1004 Hz tone through anything as pedestrian & uninteresting as a telephone system, I run it through an HF radio link...
 I amplitude modulate the 8640 RF signal generator with the 1004 Hz tone from the 4935A, send a microvolt or so of 8640 RF out to an HF receiver, and plug the 4935A's receive section into the audio output of the HF receiver. The receiver is set up to receive AM, since the 4935A must receive exactly 1004 Hz as a return signal.

 The 4935A is always set up with a "C-Message" audio filter & I always do the tests with a 60% modulated signal at 2010 KHz. The "uV" number shown in the table is the RMS voltage shown on the 8640 output meter.

 Other than having the receiver set up for AM detection, receiver settings are 'free' -  that is, any & all receiver adjustments can be made to optimize the S+N/N ratio number displayed on the 4935A front panel.

 The numbers I get are consistently better than what the receiver's manufacturers claim -  likely has to do with the goofy way I'm making the measurement... The numbers may be relative, but I do claim to be reasonably consistent from one receiver to the next.

 Got a hot receiver? Bring it by here & I'll test it & list the results...

 Any comments?
EMAIL me...        Dave Ross     N7EPI


Tests done with 2010 KHz 60% modulated @ 1004 CPS to produce a 10db S+N/N ratio
-------------------------------------------------------------------------------
3010C, ITT/Mackay Marine, s/n F0141                             06/09/99  .12uV
3.1 KC mechanical filter
-------------------------------------------------------------------------------
651S-1, Collins, MCN 571                                        06/10/99  .55uV
6 KC mechanical filter
-------------------------------------------------------------------------------
RT-1230(V)1/URC-94(V), Harris RF-280A, s/n 2600                 06/15/99  .45uV
_very_ wide crystal filter
-------------------------------------------------------------------------------
HF-8054A, Rockwell-Collins, MCN T-732                           09/03/99  .70uV
16 KC filter
-------------------------------------------------------------------------------
3010C, ITT/Mackay Marine, s/n F0119                             10/16/99  .11uV
3.1 KC mechanical filter
-------------------------------------------------------------------------------
R-392/URR, Western Electric, s/n 1266, 15283-PP-63              11/10/99  .65uV
~4 KC LC filter
-------------------------------------------------------------------------------
MVT-7100, Yupiteru, s/n 30101433                                12/16/99  .40uV

-------------------------------------------------------------------------------
EN-R2, Intermarine, s/n 1087                                    12/17/99  .19uV
 (anyone know any history on Intermarine MF/HF receivers?)
-------------------------------------------------------------------------------
R-1184/ARC-58(V), Collins 618C-3, s/n 298, AF09(603)45754       02/06/00  .34uV
6 KC mechanical filter  (pressurized ARC-58 variant)
-------------------------------------------------------------------------------
R-761/ARC-58, Collins 618C-3, s/n 840, NOm-71001                02/25/00  .32uV
6 KC mechanical filter  (TSC-15 simplex receiver/exciter)
-------------------------------------------------------------------------------
R-761/ARC-58, Collins 618C-3, s/n D073, NOm-73568               02/27/00  .30uV
6 KC mechanical filter  (TSC-15 duplex receiver)
-------------------------------------------------------------------------------
PRC-515, Collins 719D-2, 671V-2 s/n 088, with coupler tuned     03/23/00  .48uV
6 KC crystal filter                      with coupler untuned            1.00uV
-------------------------------------------------------------------------------
E6862, Telefunken manpack receiver, s/n 2350125                 05/19/00  .30uV
6 KC crystal filter
-------------------------------------------------------------------------------
RT-671/PRC-47 Collins transceiver, s/n 169 of '68 contract      08/27/00  .85uV
AM conversion in process, 6 KC mechanical filter
-------------------------------------------------------------------------------
R-174/URR (GRR-5) Emerson Radio & Phonograph Corp.              11/04/00  .52uV
~10 KC LC filter, s/n 3757 of 3105-PH-51 contract
-------------------------------------------------------------------------------
R-761/ARC-58, Collins 618C-3, s/n C301, NOm-73371               02/06/01  .27uV
6 KC mechanical filter  (TRC-75 receiver/exciter)
-------------------------------------------------------------------------------
R-390A/URR, Capehart Corp, Order No.21582-PC-61, s/n 418        05/15/01  .30uV
2 KC mechanical filter
-------------------------------------------------------------------------------
651S-1, Rockwell-Collins, MCN 3181                              10/11/01  .30uV
3 KC crystal filter
-------------------------------------------------------------------------------
R-1051C/URR, Radionics Technologies (CAGE 50097), s/n 001       01/09/02  .54uV
Contract F04606-78-C-0390, 6 KC mechanical filter
-------------------------------------------------------------------------------
R-105A/ARR-15, NOas 52-961, s/n 98:CQW  (Bendix??)              06/19/02  .48uV
LC filters in tunable IF, NAS JAX rebuild in early 1960's 
-------------------------------------------------------------------------------
R-105A/ARR-15, Collins, NOas 53-983, s/n 1099:COL               06/19/02  .46uV
LC filters in tunable IF, NAS JAX rebuild in early 1960's
-------------------------------------------------------------------------------
R-22 ARC Type 12, Aircraft Radio Corporation, s/n 304           08/24/03  .32uV*
LC filters                                                  *tested at 1500 KHz
-------------------------------------------------------------------------------
671U-4A Rockwell/Collins receiver/exciter (-007 unit), s/n 3774 12/16/03  .25uV
6 KC mechanical filter
-------------------------------------------------------------------------------
671U-4A Rockwell/Collins receiver/exciter (-011 unit), s/n 5284 12/16/03  .23uV
6 KC mechanical filter
-------------------------------------------------------------------------------
671U-4A Rockwell/Collins receiver/exciter (-011 unit), s/n 6301 12/16/03  .24uV
6 KC mechanical filter
-------------------------------------------------------------------------------
ARC-174 transceiver, Rockwell/Collins 718U-5, modified to operate below 2 MHz
671U-4A receiver/exciter s/n 3774, 6 KC mechanical filter
548S-3 PA/coupler s/n                                           04/25/05
                                                            2010 KHz  |   .34uV
                                                            1990 KHz  |   .34uV
                                                            1600 KHz  |   .36uV
                                                            1300 KHz  |   .36uV
                                                            1000 KHz  |   .41uV
                                                             700 KHz  |   .51uV
                                                             500 KHz  |   .70uV
                                                             400 KHz  |   .91uV
                                                             300 KHz  |  1.40uV
-------------------------------------------------------------------------------
R-1444/URR, Cincinnati Electronics, s/n 329 (1972 contract)     03/17/09  .15uV
4 KC mechanical filter
-------------------------------------------------------------------------------
95S-1A, Rockwell/Collins, s/n 123                               10/27/10  .34uV
2.16 KC DSP filter  (through "2-2000 ANT" connector)
-------------------------------------------------------------------------------
95V-1, Rockwell/Collins, s/n 129                                11/26/10  .34uV
2.16 KC DSP filter  (through unknown antenna connection)
-------------------------------------------------------------------------------


      And now for something entirely retrograde...


 I'm trying something similar with HF receiver & teletype modem combinations, but it's a somewhat more subjective process than the simple receiver test. I use a Stanford Research DS345 programmable signal generator to create the AFSK tone sequence corresponding to a Baudot radioteletype signal, then use this AFSK signal to amplitude modulate the H-P 8640.

 The subjectivity comes in when I have to say what is a 'good' and what is a 'bad' output from the receiver & modem combination. The modem's output is of course text, either on a screen or on a page printer. For now, a 'good' output is one that prints without any errors, and a 'bad' output is one that prints any errors at all. Sorry, no BERT available here yet...

 The output of the 8640 is plainjane AM AFSK, so the receiver & modem combination has to throw away the opposite sideband & the carrier. My guess is that this scheme may not work well with a 'real' RTTY receiver like an R-390 & CV-116 combo or an R-392 & CV-278 combo.

 Note that I'm using 100% modulation for this test -  it makes the math simpler. According to the 1950 "Radio Amateur's Handbook", page 262...   "With a sinewave modulating signal, the average power in a 100 per cent modulated wave is one and one-half times the value of unmodulated carrier power;   that is, the power output of the transmitter increases 50 per cent with 100 per cent modulation."
 Armed with this bit of info, I'm going to make a couple of assumptions. The output meter on the 8640 indicates RMS output voltage, and does not change when I switch in or out the modulating signal. The assumption is that when I switch in the modulating AFSK signal, I'm adding 50% more power to the 8640's unmodulated carrier. If the 8640 is putting out 1 microvolt (2x10^-14 Watt @ 50 ohms) of unmodulated carrier, then 100% modulating the carrier must add 1x10^-14 Watt of power to the carrier. Each sideband must add 5x10^-15 Watt of power to the signal, and therefore each sideband must have an RMS value of .5 microvolt.

 In short, the 'uV' reading in this chart is one half of the indicated reading on the 8640 meter, and hopefully reflects the equivalent amplitude of a single MARK-SPACE tone pair as seen by the receiver under test.   Again, the numbers may be somewhat relative, but I do claim to be reasonably consistent from one receiver/TU to the next.


Tests done with 2010 KHz 100% modulated AM AFSK tones producing error-free copy
-------------------------------------------------------------------------------
PRC-47 & CV-2455 combo, driving Model 28 KSR                    06/10/00  .50uV
2.4 KC mechanical filter
-------------------------------------------------------------------------------

 Any comments?
EMAIL me...        Dave Ross     N7EPI

Back to main page...