[Elecraft] K2 drift observations
N0SA
[email protected]
Sat Dec 21 19:45:01 2002
I have K2 #3045 and I did some testing of the drift. I have the 100 watt
module installed.I did most of the testing on 15 meters so this is the worse
case test.
First I did a cold start to what I term full warm up. Meaning until the
drift settled down. I guess the rig got as hot as it was going to get.
What I did was to key the rig into a dummy load and run the keyer with power
set at 100 watts. I did 2 minutes of transmitting code and two minutes in
receive. This kind of simulated a CW QSO. My rig drifted down in frequency
on 15 meters.
TIME in minutes DRIFT Hz TOTAL DRIFT
00 00 00
04 30 30
08 30 60
12 30 90
16 30 120
20 20 140
24 10 150
26 00 150
30 00 150
34 00 150
40 00 150 Hz TOTAL
As you can see it settled down after about 25 minutes. This will vary
depending on how hard you are driving the radio, meaning how much heat is
getting generated.
The funny part was that after it stopped drifting the actual transmit freq.
matched the the radios readout. Due to the way I have it calibrated.
Next I tried adding some heat sheilding inside the rig.I fabricated a little
piece of plastic to divert the airflow at the front end of the amplifier
heat shield. This diverted the air flow directly out the heat fin slots and
stopped it from hitting the control board. I also added a little fiberglass
insulation to seal around the speaker and the side slot for the amp wiring
to the main board. I could feel the increased air flow at the heat sink
fins. Then I tried a test again. This time I increased the heat build up
time to 20 minutes. With alternate transmit and cool down periods.
I ended up with 130 Hz drift in 20 minutes. I little bit of a gain probably
from the increased heat dissipaion and sheilding. Still not a big gain.
Again, this was from a cold start.
Next I let the rig warm up first. Meaning I turned it on and just let it set
for 10 minutes. It drifted about 30 Hz. This is at 21Mhz. Then i simulated
QRP (5 watt) QSO,s for about 1 hour. After that time it settled down with a
total of about 100Hz drift. I then kicked it up to 100 watts for another 30
minutes and it drifted another 50Hz. for a total of 150 Hz. This is without
the full sheilding that I had previously insatalled. This matches the first
test.
Finally I did a test of drift at 10 Mhz but I let the radio idle in receive
for 1 hour before using it at 100 watts.
TIME DRIFT TOTAL
0 0 0
8min 6 6
15 4 10
38 9 19
48 0 19
1hour 6 25
then I transmitted off and on for 20 minutes at 100 watts
and it drifted another 35 Hz till it settled down for a total of 60 HZ. But
only 35Hz after warm up of 1 hour.
Conclusions are: extra heat sheilding and airflow control inside the rig
helps a little. The drift is slow over a fairly long period of time. My
normal QSO's do not generate the heat that I simulated in these tests
therefore the drift time would be even longer or not as much because the rig
cools off more in betwenn transmissions. This is probably the reason no one
gets complaints on the air. The drift is either spread out between QSO's or
much less because total heat build-up is less. Also this is the worse case.
The other bands are not nearly as bad.
I think I will leave the little airflow diverter in my rig as it seems to
help but I have removed the fiberglass seals I put around the speaker. I am
also in the process of making a little fan device for the top of the rig. I
have two nice small muffin fans that I am mounting to a piece of plexi to
set on he heat sink. They are really quiet and move a lot of air.
I hope this hasn't bored anyone and may be of some interest to some. Now
enough testing. I'm going to be using my rig for actual QSO's from now on.
P.S.
I never noticed the drift thing until it was mentioned here.
I know this K2 has better drift figures than the earlier model K2's that I
have owned as I could hear the warm up drift in my narrow CW filter
passband. That is not noticable in this K2.
Larry
N0SA