[Lowfer] Frequency stability - was => Re: Icom R75 Frequency Jumps

[Graham]

Dex,

Rather interesting in fact.

I have been spending a lot of time using a program called SBSpectrum by 
Peter Martinez G3PLX and monitoring WWV and CHU.

The focus of this is to observe the ionospheric effects on the carriers 
of such transmissions, you can often see these effects as doppler on the 
carrier. All of this you may already be aware of and I apolgize if I am 
preaching to the choir as it were.

WWV is on the other side the continent from me or at least relatively 
speaking. CHU's transmitters on the other hand are only 20 miles from my 
location. I received CHU's 3330 and 7850 ground wave on a continuous 
basis and should should little of no doppler effects. However, I started 
to see this:

https://www.dropbox.com/sh/dcp8e4h95g93pkt/IATp0f9XA4/ve3gtc_exp_grabber_20130809_072001.bmp

The image is a dopplergram of WWV on 10Mhz. The darkest line is the WWV 
transmitter, the light line just above is a 10mhz OCXO that I have been 
testing recently. The undulations and the fuzziness in the WWV carrier 
is due to the changing ionosphere. The cause of the cyclic pattern with 
a period of an around an hour was the mystery.

At first I didn't know what was going on but it sure appeared to be some 
cyclic effect. Keep in mind the bandwidth of the "dopplergram" is 7.5 hz.

I spend much time tracking the source of this behaviour. I couldn't be 
CHU transmitter could it? I turned everything in the house off and back 
on over a period of days trying to identify the root cause. It wasn't 
until one evening I was sitting by the radio, the air conditioner was 
off as was the TV and other noise sources and then I noticed the trace 
starting to make it's characteristic upward swing. What was causing it? 
Then it hit me like a light bulb turning on! No, it wasn't a new fangled 
light but rather, if I listened very carefully, it was the fan in my 
FT-950 running. This fan should not normally running when the radio is 
just sitting in receive mode but yet there it was - the start of fan 
coincided exactly with the upward swing and when the fan stopped, the 
trace started it's long downward swing. Now, the weather had been hot 
and the air conditioner was cycling on and off but I had not seen this 
behaviour earlier in the summer. What was causing the problem was that I 
put some plastic bags containing parts for a project on the table right 
in front of and just under the front edge of the radio blocking air 
circulation. The master oscillator for the FT-950 is on the bottom near 
the front right side. I cleared away the stuff from in front of the 
radio and it settled back down again. After much testing I did discover 
that whenever the temperature in shack was above 25 degrees C and the 
relative humidity was high, the radio's fan would run and cause the 
drift, the little bags of parts and what not in front the radio just 
made it worse. Whenever the temperature is below that the fan would 
never run except after transmitting for some time period when it should. 
The shack is in the basement and the temperature is usually around 20C 
it just happened to an exceptional summer where the shack was warmer 
than normal and more humid.

The FT-950 has a TXCO as it's master oscillator and is described as 
"Built-in TCXO for incredible ±0.5 ppm (@ +25 ℃)". The frequency shifts 
seen in my dopplergram are entirely within that +/- 0.5 ppm spec.

I have a new R-75 making it's way to my QTH which will be put to a 
variety of uses including monitoring WWV and CHU. I will try it out 
first stock and see what I get - Dex's image of his test will give me a 
baseline against which to compare. I suspect that I will want to improve 
the radio's stability and I think I will try a crystal heater as a first 
step for comparison but will likely end up implementing one of the 
external reference schemes. It would be nice to try one of ICOM's CR-282 
high stability oscillator's as a comparison, it may be sufficient but it 
is a bit pricey as others have mentioned especially since it is stated 
as +/- 0.5 ppm as well.

We sure do get fussy when it is so easy to have in our shacks rubidium 
standards or GPSDO's - +/- 0.5 ppm is at least a couple of magnitudes 
above what my trusty old TS-520 is capable of.

cheers, Graham ve3gtc



On 13-08-10 11:52 PM, Dexter McIntyre W4DEX wrote:
> Stock R75 6 hour QRSS120 slow Argo screen receiving the 10 MHz output 
> of a Z3801 GPS receiver:
>
> https://dl.dropboxusercontent.com/u/15095569/r75_10mhz_z3801ref.jpg
>
> FWIW, not much I guess.
>
> Dex
> _________________________