[3830] RTTY Roundup VE7FO Single Op LP

[jimsmith@shaw.ca]

                    ARRL RTTY Roundup

Call: VE7FO
Operator(s): VE7FO
Station: VE7FO

Class: Single Op LP
QTH: BC
Operating Time (hrs): 20

Summary:
 Band  QSOs
------------
   80:   90
   40:   45
   20:  142
   15:  144
   10:  204
------------
Total:  625  State/Prov = 55  Countries = 27  Total Score = 51,250

Club: British Columbia DX Club

Comments:

Wow, that was fun (mostly).

I've been working on the station layout, particularly on all the audio lines, of
which there are an astonishing number for such a simple layout.  Have spent some
time lately replacing all audio lines with 100% shield Beldfoil.  They all
terminate at a patch panel with 48 1/4" mono jacks.  Very convenient.  I found a
ground loop problem so now have 4 600:600 ohm line transformers connected to
their own jacks on the patch panel so I can connect them as I please.  My
standard configuration has them on the sound card lines, 2 on the Line Ins and 2
on the Line Outs.  I've even got an AC VTVM with it's own jack on the panel so I
can patch in here and there for level checking.  


PREPARATION

The last RTTY test I entered with any degree of seriousness was the WAE in Nov.
02.  If you read my story you might recall I had some ground loop or RF problems
which put a strange whooshing noise on my outgoing signal and caused the RF
output power to go up and down for no apparent reason.  Well, this time, for a
change, I started getting ready a couple of days early so I could deal with
problems like this before the contest rather than during it.  

OK, got WriteLog functioning so the MkV Main Rx output goes to RITTY (K6STI) and
the Sub Rx output goes to the PK-232.  Both print on the screen just fine.  (If
you want to see how to do this check out AA5AU's website.  Thanks Don, works
great.)  Input levels to RITTY (soundcard) and PK-232 are separately
controllable through a Radio Shack mixer.  (Six more jacks on the patch panel.) 
Sure beats messing with the Windows sound controls during a contest.  No sign of
ground loop/RF problems in Tx output.

My allowance of 2 days for getting ready made it possible for me to start the
contest only a few minutes late.


WHAT HAPPENED

Except as noted, all times are PST.  Add 8 hours for Z.  

1008  Started on 15, made 1 S&P Q, tuned around a bit and switched to 10. 
Quickly found a spot on 10 and started running. Wow, I can't believe this.  WL
short term rate meter showed 134/hr at 1037.  I didn't think you could do that
on RTTY, especially with 100W and an old tribander.

1108  59 Qs, 34 mults -  Rate started dropping so decided to S&P for a while,
still on 10.  I have always had difficulty with rapidly tuning in stations in
S&P.  This time it just seemed to fall into place.  Start at the low end of the
band and watch the vertical lines which represent M & S signals.  When you have
a pair of them in the 2 boxes which represent the M & S filters, you're there. 
Well, you're there if the 2 signals belong to the same guy. (AFC off for S&P of
course.)  I guess my ears are getting more used to the sound of a correctly
tuned signal because it now seems very easy to do.

1141  76 Qs, 40 mults - Rate and mult production not very good.  Found a hole at
28.102 and started running.

1215  128 Qs, 45 mults - Short term rate meter briefly hit 148.  Not good enough
I guess so back to S&P.  I have no idea how to defend this decision except that
the rate had dropped to 1Q/2 min.  However, that was over only a 4 minute
period.  Maybe I should have hung in there a while longer to see if it came up
again.  Just not a disciplined player, I guess.

1250  139 Qs, 48 mults - Things seem to be going down hill here.  Switched to 15
where I'll be fresh meat.

1400  175 Qs, 53 mults - I have a problem on 15.  I did something bad to the MkV
some time ago and now it only puts out half power on 15.  Too busy with other
stuff to worry about 3 dB on 15.  Hope to get around to fixing it when contest
season ends.  Started at the bottom and S&P'd my way up the band.  Found a hole
at 21.121 and started running.

1430  204 Qs, 53 mults - Switched back to 10 to wring as much out of it as
possible before it dies.

1600  259 Qs, 62 mults - Bye-bye 10.  Switched to 15, mostly running.

1700  287 Qs, 63 mults - Bye-bye 15.  Getting very worried about close-in mults
as I'm missing a number of W7s so bypassed 20 and switched to 40.  Oh, oh. 
Maybe got here too late.  I hear 0s, 9s and even 4s.  There's N7DB.  Call him. 
What the hell is that whooshing noise in the monitor and why is the output power
fluctuating so erratically and why isn't he answering me?  He's plenty strong
here and I don't hear anyone else calling.  Aaarrgghhh.  RF feedback problem
isn't really solved.  Well, my 40 dipole doesn't have a balun so I do get RF in
the shack.  Find that if I reduce the AFSK audio input to the MkV enough I get a
clean signal that N7DB can read.  Signal is clean but it is also only 20 Watts
now.  Damn, he isn't one of my needed mults.  Next Q, Bingo.  W0SD in, you
guessed it, SD.  S&P around 40 for a while.  So much for the idea of picking up
close in mults.  The only other mult besides SD was MS.  Even dipped into 80 for
a half hour but no mults there either.  During this time I found that wrapping
certain cables a number of times through a couple of snap-together ferrite cores
enabled me to advance the output power to 30w or so before the dreaded feedback
problem hits.

2000  332 Qs, 65 mults - Switched to 80 at this point.  Same RF problem as on
40.  Not too concerned about this in the long run as it only happens on RTTY and
the planned antenna changes should cool down the currents on the coax shields. 
Hmmm.... I bet a couple of simple low pass RC filters on the MkV AFSK connector
would do a lot more than any number of RF chokes.  

A couple of times during the 80m session I noticed the MkV output power was
suddenly 100W without the whooshing sound and erratic behaviour.  After a while
I discovered how to reliably reproduce this.  Like most everyone, I have a
number of Function keys programmed to transmit canned messages.  Loyal TR guys
will cringe when they read this but, when using WL, I simply copy AA5AU's
suggestions which makes F2 CQ, F3 the exchange, F4 the QSL message and F5 dumps
my call (3X VE7FO).  Here's what I found.  I have no idea why it worked and I
don't even want to think about it because it contradicts everything I think I
know with the consequent vertigo and nausea which occurs when the entire
universe suddenly moves out from under one's feet.  If I hit F2 I transmit a CQ
and, if I turn up the AFSK audio control on the MkV (mic) too much, I get the RF
feedback problem.  If (better find something to hang onto or else sit or lie
down before you read this) I rapidly hit F2 twice in succession the MkV puts out
100 clean Watts of RTTY with no adjustment of the mic level control.  How
rapidly?  I don't know, this isn't a science experiment it's a contest.  Hit the
key then, while it's transmitting, hit it again.  How do I know the signal is
clean?  Sounds fine in the monitor and people come back to me when I call them. 
Same for F3, F4 and F5 keys.  Well, this led to a somewhat different operating
style.  In S&P, if the guy is loud I hit F5 once to dump my call at 30W or so. 
If he isn't I hit F5 twice which dumps my call at 100W but sends it 6 times
instead of 3.  Same for the exchange but I usually didn't hit F3 twice figuring
that if the guy at the other end heard anything at all once the contact was
established he could probably figure out that it was 599 BC.  (For RTTY
shouldn't it be 599C?)  For running I always hit the CQ button twice.  

80 is slow but it is yielding Qs.  Did pick up NV and UT  Made 7 Qs on 40 also.


2200  379 Qs, 66 mults - Not too bad for me for 12 hours.  Half way.  Start
thinking I might break that 1,000Q barrier if I'm lucky. 

0000  412 Qs, 67 mults - Thinking of going to bed.  I'm missing WY, MT, GA, DE,
DC and VT as well as the usual VE suspects.  However, there's no mug for a clean
sweep and, if I get up early and there's any propagation at all, I should be
able to pick up a bunch of Eu mults.  Decide to have a nice glass of chardonnay,
work a few more 80m Qs and go to bed.  Little did I realize when I opened the
door to Bacchus that Murphy would slip in with him.  So, I'm sipping my wine and
making Qs when I reach for the F2 key to call CQ and knock the wine glass over. 
Yes, it's right in front of the keyboard and most of the wine vanishes therein. 
This is one of those Salvador Dali inspired Microsoft keyboards.  Well, it's
Microsoft so it should be OK shouldn't it?  I pop off a few key caps and see
below a plastic surface with pools of wine on it.  I get out the paper towel,
poke it down between the keys and mop up as best I can.  Everything's still
working and it is a contest so I go back to making Qs.  I do notice now that the
E key doesn't send E anymore, it sends IE.  After a while I notice that each of
the Function keys repeats.  i.e. hitting the CQ key once sends the CQ message
twice.  Unfortunately at 30W not the 100W I get by actually hitting the key
twice.  So, if I want a 100W CQ, I have to hit the key twice which means that
the CQ message is sent 4 times.  Oh well, it's 80m so you do have to make a
little noise.  It must have puzzled a number of people, though, when I hit the
exchange or QSL key to hear the message twice.  It got particularly confusing
when I hit the F8 key.  This one enables you to, if two people called and you
could get both calls to say, "W1ABC TU AND NOW W2DEF 599 BC W2DEF".  As if this
weren't bad enough, in a little while whenever I pressed the "W1ABC TU DE VE7FO
QRZ" key it would follow this with the exchange message.  i.e.  "W1ABC TU DE
VE7FO QRZ 599 BC BC BK".  at this point I was afraid to use the F8 key.  Also,
the CQs started repeating more than the 4 times and it took several presses of
the ESC key to get it to stop until, finally, much like the Sorceror's
Apprentice, I couldn't stop the sending of CQ.  What do I do now?  Talk about
your alligator, there's not even any pretence of listening.  Help, I wanna go to
bed.  What to do?  Well, the mouse still works so I closed WriteLog.  that
stopped the CQing.  However, the MkV is still in transmit mode and I didn't want
to shut it off as the fan is still running.  In retrospect, I should have shut
it off to get out of transmit mode then turned it on again until the fan stopped
running.  Perhaps you will understand that, given the confusing circumstances
and the lateness of the hour, I turned the RF Power and Mic controls all the way
down so that nothing would get transmitted, propped the keyboard up on end to
allow the wine to drain out and went to bed, leaving the computer and MkV to
fend for themselves.

0022  422 Qs, 67 mults.  Set the alarm for 0630 to catch the morning Eu opening.
 Perhaps they will appreciate wine soaked Qs.  

0700  I'm up and dressed with orange juice in hand and wiggle the mouse to bring
the computer back to life.  Hmmm... nothing.  Notice the MkV is, as expected,
still in Transmit mode but the fan, as not expected, is running.  Feel the
heatsinks.  Oh, oh, pretty hot.  even though it is in Transmit mode it shouldn't
be drawing much power with RF control all the way down and no drive.  Anyway,
did what I should have done last night, shut off the power to get it back to
receive mode then turned the power back on to let the fan cool things down. 
Now, what's with the computer?  Hey, it's smarter than the MkV.  It shut itself
completely off.  Wiped up the pool of wine which had drained out of the
keyboard, hit the Power On button and went upstairs for breakfast.  Came down
and found the computer stalled part way through the boot process.  Wouldn't
respond to the keyboard.  Hit Caps Lock to see if the light would go on.  Nope. 
Hit Reset button on the computer and watched it do stuff, including flashing
lights on the keyboard.  Got partway through loading startup apps (Win 98SE) and
stopped.  Mouse works.  KB doesn't.  Took KB off daughter's machine.  Tried to
plug it into mine.  Wrong connector.  Think of buying new KB but stores not open
yet and time's a-wasting.  Shut computer off and disassemble keyboard.  Hey,
this thing is well built.  There are three main sub-assemblies under the keys. 
One for the left hand batch, one for the right and one for the numeric and
navigation keys way over on the right.  The wine all went into the left batch of
keys.  Unscrew and remove the rather solid aluminum plate under the left batch
of keys.  I see a clear, flexible plastic sheet with a bunch of rubbery things
attached to one side and a bunch of printed wiring traces such as one would
expect to see on a circuit board.  The rubbery things transmit force from the
key caps to the printed wiring.  Not clear how this translates a keypress into a
command.  Anyway, wine all over this plastic stuff.  As I handle it it becomes
apparent that there are 3 layers of this plastic material and that the wine has
got inside.  Just what I need, a multi-layer PC Board with wine in between the
layers.  Well, it isn't a board, it's 3 sheets of plastic which can be
separated.  So, I separate them and wipe off all the wine.  As I do, it becomes
apparent that each rubbery thing is compressed by its associated key cap.  When
it is compressed it presses a printed wiring land on the bottom of the top sheet
of plastic through a hole in the middle sheet of plastic against a printed
wiring land on the top of the bottom sheet of plastic.  It became obvious that
the increasingly more annoying problems I experienced last night were due to
capillary action as the wine was sucked ever deeper into the heart of these
plastic sheets.

So, with the wine all gone, I screwed the bottom plate back on and, with the KB
not yet reassembled, restarted the computer.  Hey, the Caps Lock, Num Lock and
Scroll Lock keys all turn their associated LEDs on and off.  Computer boots
successfully.  Great!  Type a few characters.  No response.  Not great!  Mouse
works fine so I check out Device Manager (Win98SE).  It tells me that the
keyboard isn't working.  I already know that.  However, it also tells me that
the required drivers aren't installed.  Well, I thought Chardonnay was a wussy
drink but this sure opened my eyes.  Why, just by spilling some into the
keyboard it was able to reach into the hard drive and dissolve some driver
files.  Totally astonishing.  Well, Device Manager is offering to install the
files so I tell it to go ahead.  In it's smarmiest tone it asked me to insert
the Windows CD.  Oh, oh.  Where the hell is that.  Looked in the filing cabinet
where I keep all the program CDs.  Looked upstairs in the old hamshack.  Looked
on my daughter's desk and in her CD drive.  Came back downstairs and found it 2
feet away in the bookshelf immediately to my left.  Sigh.  Put it in the CD
drive and let it do its thing.  BINGO, KB works.  Now try the MkV to see if it
survived being in transmit mode all night.  Hurray, it's fine.  But what a
terrible time loss.  3-1/2 hours

1027  422 Qs, 67 mults.  Started on 20 hoping to catch the tail end of whatever
Eu opening there might have been.  Didn't figure 15 and certainly not 10 would
support any prop into Eu with the SSN being what it is.  Seemed to work.  By
1055 I had 7 Eu mults including a rather nice LX.  After that they tapered off. 
Bounced around between 20, 15 and 10 for the rest of the day.  Couldn't seem to
get any runs going for more than 15 min or so, no matter what band.  Thoughts of
making the 1000Qs went away.  Still, I'm pretty happy with the result.

1559  627 Qs, 83 mults.  

I tried simulating SO2R by tuning around with the B VFO which fed the PK-232. 
As it's one radio, you can't listen while transmitting.  I found that the PK-232
tuning indicator doesn't seem to be well adapted for rapid tuning.  I did make a
few Qs with it but not many.  Next time I'll try to have a scope connected to it
which I think will make the process faster.

Having a convenient level control for AF into RITTY was rather handy.  If a
signal wasn't quite printing, adjusting the level seemed to help.  I guess I
found the right level after a while as I didn't touch it for the last half of
the test.  I was amazed to find that signals that didn't move the S meter and
were barely audible printed just fine.  Seems to bear out what others say about
RITTY.  I tried MMTTY somewhat over a year ago but couldn't get that level of
performance out of it.  It has, no doubt, improved during that time but I think
I'll stick with RITTY.  I did notice, however, that some S5 signals with no QRM
didn't print well at all.  I did notice that some S5 signals didn't print well
at all.  Don't understand this.

I noticed that a number of signals seemed to be wider than 170 Hz.  i.e. When
their Mark tone was lined up with RITTY's Mark filter, the Space tone wasn't
lined up with the Space filter.  Any thoughts anyone?

The keyboard still has a problem.  The Shift key under the Enter key doesn't
work.

Thanks for the Qs, everyone, particularly those who so patiently endured the
keyboard problems.


Posted using 3830 Score Submittal Forms at: http://www.hornucopia.com/3830score/