A couple of weeks ago I began a series of tests to determine if cooling of the
A10 board area of the Orion II would result in increased frequency stability.
My first test was to place a 40 mm 12 VDC brushless fan in the area, secured to
the bulkhead near the TCXO with a cable tie, and blowing air upward onto the
oscillator. A wire was routed out the back panel through opening for the #2
Band Data connector.
The results showed improved stability, but my mounting method resulted in a lot
of acoustic noise. Today, I put in a new fan, this one 50 mm, and mounted on
improvised foam shock mounts for better mechanical isolation and the result was
greatly improved. Noise is now barely noticeable, and certainly much lower
than both my Titan 425 linear and my Sony P-4 laptop, and the increased air
volume made a definitive improvement on the stability.
Based on the drift of the original Orion reported by Martin, AA6E on his blog
(http://blog.aa6e.net/2005/08/orion-frequency-calibration.html), here is my
comparison (all relative to 15 MHz WWV):
NO FAN: Maximum drift = 20 Hz from turn-on to stabilization. After 15 minutes
warm-up, 20 Hz drift. After 30 minutes, 15 Hz. Full stabilization appeared to
be achieved after about 3 hours.
WITH FAN: Maximum drift = 8 Hz from turn-on. After 15 minutes, 3 Hz. After
30 minutes: fully stabilized to within +- .5 Hz.
The "no-fan" drift pattern Martin reported (and I verified on my O2) quickly
went down 20 Hz (making the zero-beat frequency reading go up 20 Hz), then
slowly returned to the same frequency as start-up over a period of 2.5-3 hours.
With the fan, the frequency drifts in mostly one direction, downward by a
total of 8 Hz from a cold start-up, making the zero-beat reading go up 8 Hz.
In the next hour of operation, it drifts back 1 Hz, and there was no further
audible drift for the following 3 hours that I tested, resulting in the spec of
+- .5 Hz or .03 PPM after 30 minutes warm-up. This compares to 1 PPM after 30
minutes with no fan. Total error was .53 PPM from turn-on vs. 1.3 PPM with no
fan, and improvement factor of 2.5.
Of course, changes in shack temperature would most likely impact drift, and
long-term aging drift will still occur. Greater stability can be achieved
using more sophisticated temperature control, including keeping the area warm
even when the radio is off, or using a crystal oven, however, for most
purposes, this should give the kind of stability needed to prevent being 20 Hz
high or low on a net (I heard several stations off by at least that much on the
Ten-Tec net today, a couple of them using Orions), which is noticeable on SSB.
I would be interested in hearing if anyone else tries this simple modification
and if the results are similar.
Ron Castro, N6AHA
Chief Technical Officer
Results Radio, LLC
TenTec mailing list