[R-390] Resistor Noise

[bernie nicholson]

Thank you Barry Mat Roy & Sam For your thoughts on Noise,  Tongue In Cheek & otherwise, My thoughts ,  Boltzmanns constant is Just THAT we are stuck with it , Sam I read your Treatise on noise in resistors and am cognisant of the fact that , there is more thermal agitation in a 1 meg resistor than a 10k  and that wire wounds generate less noise again , all understood , But a  cracked  carbon as opposed to a metal Film , Perry talks about good science ect ect  ,BUT I would like to see references to support this view ,  So our Collins Engineers who designed  these circuits Have optimized all the resistor values for the tubes in use PRESUMABLY ,  Its easy to Arbitrarily pick out a more modern tube with a better equivalent noise resistance to replace the 6dc6  But  HOW Do we optimize the tuned circuits  that have been designed for the original tube ???
also allowing that it will have a different input /output capacitance ,HOW do we make the circuit track ??? I am a great believer in NOT  trying to reinvent the wheel , I have a lot of time & admiration for the likes of    SABIN , SHOENIKE ,PAPENFUS , and BRUENE  and  " WHO  " am   "I" to try and second guess their  designs 60 years On , and somebody mentioned Ray Osterwald ,Frankly I think he got well and truly OUT OF HIS DEPTH in the ELECTRIC RADIO ARTICLE , "  A COMPETITION GRADE 390A. "        This receiver is the ultimate compromise between sensitivity , selectivity, dynamic range act act ,I have three of them and with a bit of tweaking  they achieve  0.15 Uv for 10db signal to  noise+noise, in a 2KHZ bandwidth   and a Dynamic range of over 80 db  This even today outperforms lots of modern  high end radios !!!   GUYS this is about as good as it gets , The latest  HF  technology doesn't do anybetter , There is no point is there?       NOW about caps across filters , In A production environment  SELECT ON TEST  COMPONENT TWEAKING is time consuming and COSTLY , the reasons for the  substitution of variable caps  to me would be self evident,   Also  there appears to be NO difference in performance between the brands , They all conform to a Gov. Specification , BUT I can See that There is SNOB value in owning a COLLINS produced radio, But no practical advantage ,   Overall the 390A is a very quiet receiver after the ultimate in tweaking and the correct adjustment of the IF Gain pot. you can turn up the audio and RF gain and all you hear if correctly aligned is a hiss , connecting an antenna just about blows you out of the room , The final test I perform after realignment is to establish the input impedance on the frequency that you are interested in ,I use a GR916a across the ant input and then I substitute a carbon resistor of the same value whilst looking at an audio O/P meter , If you have done everything right you will see a  perceptible increase in output, YOU ARE LOOKING AT THE NOISE VOLTAGE IN THE RESISTOR !!! Very few radios will show this the first I experienced was with an Collins  ARC58, Which obviously evolved from the 390 series , I think the Genesis of the ganged Slugrack Idea first appears in the ART13 transmitter with the 20 turn vfo mechanically connected to the slug rack for the next multiplier stages  Circa 1937, Finally  please excuse the long post  ,I hope your collective attention spans have held up ok  ,regards From Oz