ad: Alphaant-1

NorCal S9 signal Generator

Discussion in 'Homebrew and Kit Projects' started by K9FV, Oct 10, 2008.

Thread Status:
Not open for further replies.
ad: L-HROutlet
ad: l-rl
ad: L-Geochron
ad: L-MFJ
ad: abrind-2
ad: Left-2
ad: Left-3
  1. K9FV

    K9FV Ham Member QRZ Page

    Hello All, I am in the process of building the NorCal S9 Signal Generator kit - it has a 50µV and a 1µV output - that is S9 and S1 on the S meter. The basic output is 50 µV, and the 1 µV is obtained by inserting a 34db pad that is a Pi network used for the attenuator.

    I would like to have a 0.1 µV output - or somewhere there abouts to check the receivers senstivity. What value of resistors would be required to drop the output from 1.0 µV to about 0.1 µV?

    edit: 50µV is -73 dBm, 1µV = -107 dBm, what does 0.1µV equal? a minus "what" dBm?

    I'm sitting here looking at the problem feeling very foolish - I do voltage drop calcs quite often for motors, but there I am dealing with 480 volts and several hundred feet - I just have a blind spot for this problem, so be gentle with this old man.

    73 de Ken H>
    Last edited: Oct 10, 2008
  2. W6ECE

    W6ECE Ham Member QRZ Page

    My Wavetek Signal Generator says -130 dbm = 0.1µV

  3. K9FV

    K9FV Ham Member QRZ Page

    Hmmm, if 0.1µV = -130 dBm, it takes a pretty good receiver to detect that signal - maybe a 0.2µV signal is more reasonable? I think someone said their SoftRock MDS was -123 dBm (or a -127??).

    I think the ideal would be a multi-turn pot to slowly turn the signal down watching the trace on an O'scope for the MDS of the radio? Poor man's Wavetek Signal Generator?

    Thanks for the info - 73 de Ken H.
  4. AI3V

    AI3V Ham Member QRZ Page

    I usually listen for the MDS when I test a audio receiver...

    Others disable the agc/avc and watch a voltmeter on the speaker lead, when the voltmeter reading increases 3db (V x 1.414) you're s+n/n will be Zero db

    Most modern rigs will copy a .1uv signal.

    A modern rig that does not receive a .25uv signal is broken.


    P.S. think about a step attenuator to go with the s/g, The handbook has plans.
    Last edited: Oct 10, 2008
  5. AI3V

    AI3V Ham Member QRZ Page

  6. K9FV

    K9FV Ham Member QRZ Page

    Thanks for the info - reading your responses caused it to click, and to confirm I went to the Elecraft website and downloaded the step attunator manual and all the info is laid out nicely.

    Takes about 20 db to drop from 1µV down to .1µV. easy formula.

    dB = 20 LOG (A/B)

    Just talking (writing) sometimes will help the answer to click.

    73 de Ken H.
  7. ZL3GSL

    ZL3GSL Ham Member QRZ Page

    A "-127dB" output is not really going to be -127dB except in theory. :D

    Unless the "signal generator" is built like a "real" signal generator, there'll be much more than that floating around from the oscillator.

    The resistor "attenuator" will have so much leakage that the 1uV output will be a lie, too.

    Have a look at the construction of the output attenuator of a professional signal generator. Have a look at the shielding of the oscillator. The service manuals generally stress the need to replace every screw and every shield plate after servicing. One SG I used a lot had 15 or more screws on each panel.
    Last edited: Oct 11, 2008
  8. VK2TIL

    VK2TIL Ham Member QRZ Page

    I looked-up the specs on the Norcal S9; a nice useful little generator.

    But don't ask too much of it; just as you wouldn't drive your family car at Indy, don't expect to use the Norcal for critical applications like MDS testing.

    If you have or can get hold of (perhaps from a library) EMRFD you will find discussion there on weak-signal generators; W7ZOI solved the problem raised by ZL3GSL by sealing single-frequency crystal oscillators inside PCB enclosures which, fully-soldered, are as close to leak-proof as you can get.

    (An bonus with EMRFD is that, on the book CD, there is a lot of useful information including the very good QST article by Sabin on sensitivity & noise figure).

    Note also the harmonic output of the Norcal; for instance the second harmonic is specified as -4dBc, +/- 2dB.

    As a guide, I would expect a decent bench instrument to have all harmonics at least -30dBc; some do better.

    But instruments with high levels of performance will cost you a great deal more than the Norcal. :)

    Use it for its intended purposes and have fun!!
  9. K9FV

    K9FV Ham Member QRZ Page

    Oh I fullyl understand the lack of "lab grade" results from the NorCal S9 SG - it's specs are only ±3 dB at best - but in the real world, that's half a S unit - hard to even see on my 756Pro S meter. So what if it's 8.5 or 9.5 at 50 µV? Even S8 to S10 isn't all that bad for "real world" use.

    I plan to put the NorCal in a Altoids case which will provide some shielding.... it will provide some indication if the radio is receiving an S1 or S9 signal as the MDS.

    The NorCal has 4 crystals - one for each of the bands from 80 to 20 meters - no need to use a harmoic. and a LOT better price than the Elecraft signal generator.

    That is a neat calculator - actually pretty close to the Elecraft values. Thanks for the link.

    73 de Ken H.
  10. AI3V

    AI3V Ham Member QRZ Page


    Use the harmonic to your advantage.

    14.200 x 2 = 28.400

    See what I mean ? :)

    I like your altoids idea, I have a 50 lb milsurp URM25/d, kinda inconvient for toting to a hamfest to test out that rig "that worked when I put it away"

Thread Status:
Not open for further replies.

Share This Page