[R-390] Interesting Finding... R-390A

[John Kolb]

On Fri, 29 Nov 2002, Michael Melland wrote:

> This is interesting.... here is an unusual finding on an R-390A.
>
> Radio works fine.... a quick check for operational condition "sensitivity"
> (per manual... cal on, antenna peaked on each band and carrier meter to read
> no less than 40 dB at each 100 KC cal point) reveals excellent sensitivity
> but an anomoly on one band.  3 MC reads strangely.... carrier meter readings
> vary exactly 20 dB each 100 KC calibration point.  The odd frequencies read
> 20 dB lower then the even freqs. Interestingly, 2 MC and 20 MC (share same
> cans/crystal I think) read normally up and down the scale (65 dB+/- all 100
> KC points).
>
> See below..... any thoughts ?

If it were not only one 1 MHz wide band, I would suspect that the
xtal calibrator output were not level. Any sign at all of this effect
at the top of the 2-3 MHz band or the bottom of the 4-5 MHz band?

Have a calibrated output sig gen available to chart sensitivity with?

If I remember correctly, doesn't a perfect square wave have only
odd hormonics, no even hormonics at all? A very narrow pulse
will have pretty equal odd and even hormonics. I can see a calibrator
output waveshape causing this being more likely than ripples in the
response of the RF amps having peaks and dips at 100 kHz intervals.

Many moons ago, I built a xtal calibrator box out of 7400 and 74S
logic which could output a narrow output puls 15-20 nS wide as I
recall. Spectrum analyzer showed it as being flat out past 100 MHz.
With the fast logic families available now, a much better one
could be built.

My design has a chain of divide by 2's and a chain of divide by
10's switch selected starting with a 4 MHz crystal producing
outputs of 4, 2, 1, 0.5, 0.25, and 0.125 MHz, and base 10 submultiples
of those freqs. With this arrangement, a varity of calibration
marks can be generated, 10 kHz by selecting 1 MHz and 2 divide
by 10's or 125 kHz by selecting 0.125 and one divide by 10, for
example. Very useful for checking dial linearity.

I built this when I bought a HRO-500 a previous owner had adjusted
the dial linearity by bending the split plates on the main tuning
cap to look like the teeth of a saw /\/\/\. On perfectly on
the xtal cal freqs but of to 10 kHz out in between. Horrible.
Took about a month to get it right again.

John



>
> The 3 MC band exhibits the following readings:
>
> 3100  - 50
> 3200  - 60
> 3300  - 40
> 3400  - 60
> 3500  - 40
> 3600  - 60
> 3700  - 40
> 3800  - 60
> 3900  - 40
> +0000 - 60
> --
> Michael Melland, W9WIS
> Winneconne, Wisconsin USA EN54pc
> qrp-l #1656 - qrparci # 9875 - iparc #252
> ars #1075  -  http://webpages.charter.net/w9wis/
>
>
> _______________________________________________
> R-390 mailing list
> R-390@mailman.qth.net
> http://mailman.qth.net/mailman/listinfo/r-390
>