[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Instrument questions



Original poster: "by way of Terry Fritz <twftesla-at-qwest-dot-net>" <Mddeming-at-aol-dot-com>

Hi Paul,
         Many thanks for the insights. I was looking at my $170 sweep
generator, my ancient  $150 O-scope and ads for spectrum analyzers at $2K-$8K,
or service monitors at $10K-$30K. I knew there had to be a better way, but in
the "river of time and technical progress"  I'm further up the creek than most
people on the list, and I'm trying to play catch-up. It comes from spending 33
years in the pipeline industry being  an overpaid imitation engineer, instead
of the starving physics teacher I could have been back in '66. ;-)

Thanks again,
Matt D.

In a message dated 3/27/02 9:47:35 AM Eastern Standard Time, tesla-at-pupman-dot-com
writes:


>
> > Is there any relatively easy way to Kludge together a sweep
> > generator and a dual trace analog scope to make a poor man's
> > spectrum analyzer?
>
> You could try this for frequencies above audio:
>
>                      input
>                        o
>                        |
>                rf      |
>   |sweep gen|--------[mixer]------[audio amp]----[scope Y]
>        |
>        |      ramp 
>         -----------------------------------------[scope X]  
>
> The mixer to consist of just a diode. As you can see, this
> is just a crude, swept-frequency, direct-conversion receiver.
>
> The audio amp makes a nice narrow band lowpass filter, providing
> it's a cheap and cheerful one!  Use only low sweep rates.
>
> You can make a nice simple logarithmic amp out of an op-amp and
> a another diode or two.  Well worth the effort.
>
> Ought to work for low frequencies, eg those used by TCs.
> For UHF, yank a vco tuner module from an old VCR or TV and take the
> video output to the scope Y.  Apply ramp to vco and scope X.
>
> Just take a diode, a long wire antenna, your sweep gen on 'manual'
> put them together and see if you can get any audio across the
> diode as you tune around - doesn't matter if it's horribly
> distorted. If necessary use an RC or two between mixer and amp
> if the resulting spectrum looks too busy.  Adjust osc level for
> best results, then connect scope and turn on sweep.
>
> For better performance, replace the single diode with a ring of
> four and drive with a little center-tapped toroidal xformer so
> as to make a balanced detector - much cleaner.
>
> Or even better, use one of those analogue multiplier chips for
> the mixer.
>
> The nice thing about this direct conversion approach is you can
> start off simple and improve the performance in little steps as
> and when you need: RC filter, log amp, balanced mixer, lower
> noise amp, etc.  Great project for the electronics novice, and
> an interesting sunday afternoon for the expert.
>
> Maybe it could be combined with a little circuit to ping a TC
> resonator (pri, sec, or pri+sec) to give a tobacco-tin style
> TC analyser?  Forget the scope, use a manual frequency sweep,
> and an earphone or speaker to hear the audio frequency beats.
> Ping the coil at a rate of say 200Hz, and you should hear this as
> rough tone as you tune past each of the coil's resonant modes.
>
> Would make a good way to measure the effective k of the coupled
> resonators - if you have a freq counter hung off the oscillator:
> just tune for the upper and lower beats and note the frequencies.
> Potentially very accurate indeed, especially at low k.  For anyone
> wishing to determine the difference (if any) between k measured at
> 60Hz, and the effective k at resonance, this might be one way to
> go (hint!).
>
> I guess it could also double as a bat detector (replace pinger
> with microphone) or as a VLF receiver (replace pinger with long-
> wire). 
> --
> Paul Nicholson
> --