CW Contest for data operators

I'm all for data operators having a go at cw contests using a computer, but please read this before the contest to save annoying every cw operator on the planet. I've had a go at a few data mode contests and it seems usual in RTTY or PSK31 to give your call sign and report twice. The logic (i think) behind this is that sometimes the computer gets it wrong and at least if given twice your attention is drawn to try and correct mistakes. However, in cw contests this practice is frowned upon and unnecessary.

Here is the expected behaviour for CW contests:
1) CW contests are all about efficiency
2) Always give your call sign and report exactly once
3) If the station gets it wrong the first time, simply repeat the last exchange exactly once
4) If no response leave a gap then repeat exactly once then leave a gap

CW is really cool and I like operators that are really efficient. If you get no response then simply repeat the last exchange exactly once. Leave a short gap for the operator to come back, then if not repeat again, he will come back when he gets it or gives up...simple!

There is nothing more annoying than someone who repeats things twice without a gap when you make a mistake. Sometimes when a pile up is large you only get a partial call. The idea is that when you give this partial call, everyone else stands by and allows the caller to complete the QSO. If this caller decides to repeat information more than once without a break, the pile up gets impatient and the QSO takes even longer to complete.

If you are one of these annoying people who believes "give my call twice without a break if you don't get it the first time" please remember by doing this in a contest, it singles you out as a "data operator" or a "LID" to everyone waiting, and the run station at the other end will be shouting "I only need it once!", whilst 3-4  stations waiting in the pile-up may tune away due to the length of time they are having to wait per Q. 

Worse than this, you may double with the run stations response on your second unnecessary callsign repeat and a stalemate occurs which is normally always broken by the run operator having to repeat your call to let you know the previous exchange was for you. Of course you probably don't remember the exchange as you are just playing and the operator has to give you the whole exchange again. By this point most of his Q rate will have vanished, together with the pile-up. At this point your Q will have been most unwelcome as it was at everyone else's expense....

Why am I such an angry man for such a little thing? Out of an entire contest of say 2000Qs over 24hrs, roughly 1 in 5 S&P stations do this during a pile-up, that's 400Qs where I lose the will to live.

REMEMBER... what is right for a normal cw qso is not correct for a cw contest!

CW for beginners, lesson 3

Unfortunately most cw conversations don't use plain English as this would be laborious and tedious to send in full. If a word can be shortened it will be and if there is a Q code for the sentence it will be used. The results can sometimes be confusing for the first time listener. Here is an example of a two way standard (rubber stamp) cw conversation:

*************
CQ CQ CQ CQ (general call) DE (this is) GM0ELP GM0ELP K (over)

GM0ELP DE M0GBK KN (prosign, back to you)

M0GBK DE GM0ELP GM (good morning) DR OM (dear old man) UR (you are) RST (Radio Signal Tone) 599 599 BT (prosign, break) OP (operator) Doug Doug ES (and) QTH (location) is Hamilton Hamilton BT So hw (how copy)? AR (prosign, end of transmission) M0GBK DE GM0ELP KN

GM0ELP DE M0GBK GM (good morning) DR OM Doug UR (you are) RST (Radio Signal Tone) 599 599 BT (prosign, break) OP (operator) Simon Simon ES (and) QTH (location) is Meltham Meltham BT Condx (working conditions) hr (here) rig Flex 6600 es pwr (power) 1KW es ant (antenna) OFCD 10m up BT Wx (weather) is wet es cloudy temp is 9 degrees BT so BTU (back to you) Doug AR GM0ELP DE M0GBK KN

BK (Break-in, now that calls have been established no need to repeat) OK Simon TNX (thanks) fer (for) condx (you conditions) hr rig is ftdx101d es pwr 10w es ant dipole 10m up BT Wx hr is sunny es temp is 15 degrees BT TNX fer FB (fine business) QSO (chat) es HPE (hope) CUAGN SN (see you again soon) 73 (good wishes) AR M0GBK de GM0ELP KN

BK es tnx fer INFO (information) dr om Doug es all CPY (copy) BT Hpe cuagn sn 73 TU (thank you) AR GM0ELP DE M0GBK KN

tu . .

..

*************

Don't worry about this as most conversations are rubber stamp and vary little from the above. Most constructs can be guessed and you will always get some sort of understanding.

Some tips:
1) When decoding don't write everything down, just the important things like name etc.
2) Don't write down capitals, a lower case scrawl is faster (it's not a quality record!).
3) When starting out listen to beacons (28.2 to 28.3) and slowly decode the repetitive message they send.
4) Always tune to the higher end of the cw segment of the band as this is where QRS (slow speed) cw is always sent.
5) Don't listen on cw contest days (weekend) as high speed cw is the norm and can be off putting for the beginner.
6) Always wear headphones to cut out distractions and close your eyes to help concentrate.
7) Don't be frightened to ask for a repeat "SRI (sorry) PSE (please) AGN (again) UR (your) name QSB QSB (fading) BK".

Doug GM0ELP

CW for beginners, Lesson 2

You will hear a lot of nonsense from old timers who have forgotten that there is one thing you have to do before anything else and that is.... simply learn the morse code alphabet!

This preceeds everything else...bar none!

We have 26 letters to learn, ignore the old timers (prosigns, koch method, just recognise the tune blah blah blah) and lets just cut to the chase as fast as possible by using simple association and memory techniques to try and fast track the code into those "little grey cells" believe me this step is unavoidable.

Firstly lets do all the letters that only consist of di's or ( . )s, **remember the letter group "eish"**:

e ( . )
i ( . . )
s ( . . . )
h ( . . . . ) That's 4 out of 26!

"Every Indian Shoots High" imagine the arrows coming towards you.

Second, lets do all the letters that only consist of dah's or ( _ )s, **remember the group letter "tmo"**:

t ( _ )
m ( _ _ )
o ( _ _ _ ) That's 7 out of 26!

"Totally Massive Ohms" imagine a PCB track circuit with small breaks, the ohms become massive with more breaks.

Third, lets do all the letters that start with di's and end with a dah, **remember the group letter "auv"**:

a ( . _ )
u ( . . _ )
v ( . . . _ ) That's 10 out of 26!

"Autonomous Underwater Vehicle" imagine the di's are the rope with the dah being the vehicle slowing going up the alphabet.

Fourth, lets do all the letters that start with a di and end with dah's, **remember the group letter "awj"**:

a ( . _ )
w ( . _ _ )
j ( . _ _ _ ) That's 12 out of 26!

"Awful Woolly Jumper" picture wool being pulled from a single point on a badly knitted woolly jumper, the dit being the single point on the jumper and the dah's being the wool.

Fifth, lets do all the letters that start with dah and end with di's, **remember the group letter "ndb"**:

n ( _ . )
d ( _ . . )
b ( _ . . . ) That's 15 out of 26!

"No Dust Bins" picture a bin lying on it side and the litter spilling out.

Sixth, lets face it if you've listened to any cw, you'll definitely hear this pattern all the time, **remember the group letter "cqdx"**:

c ( _ . _ . )
q ( _ _ . _ )
d ( _ . . )
x ( _ . . _ ) That's 18 out of 26!

"CQ DX" just listen to some cw on air.

Seventh, dah or dah's encapsulated with di or di's

r  ( . _ . )
l ( . _ . . )
p ( . _ _ . ) That's 21 out of 26!

"Right Lazy Person"

Eighth, multiple dah's ending in di or di's

g ( _ _ . )
z ( _ _ . . ) That's 23 out of 26!

"Giant Zebras"

Ninth, lets face it if you've listened to any cw, you'll definitely hear this pattern all the time, every general call ends in a 'k' and also a repeater sends it after you unkey.

k ( _ . _ ) That's 24 out of 26!

"CQ DX DE GM0ELP K"

Tenth, think of this as forming the letter 'f' with the first di being the tail of the letter and the dah being the strike through of the letter.

f ( . . _ . ) That's 25 out of 26!

 as follows:
    .
   _
  .
.

This leaves us with y, no help here

y ( _ . _ _ ) That's 26 out of 26 and we are done!

Now we need to practice recalling the alphabet in cw, if you stumble, punish yourself by going back to the beginning. Enjoy!

Doug GM0ELP

CW for beginners, lesson 1

The following is an explanation of sending the word "paris" in cw to help the beginner understand the sending cadence required to allow someone to decode it properly at the other end. The word "paris" is used as the standard for determining the speed of sending in words per minute (wpm) i.e. if you send the word "paris" 12 times in a minute, you are sending at 12 WPM.

Definitions:
1) Element : A length of time equivalent to the length of a "." (di or dit) in cw. The actual time used in sending a "." changes depending on words per minute being sent, so we just refer to this time duration as an element for this explanation.
2) Dah Symbol: ( _ ) equivalent to 3 time elements.
3) Di Symbol: ( . ) pronounced "di"(mid-letter) or "dit"(located at end of a letter), equivalent to 1 time element.
4) Symbol space : (') duration equivalent to 1 element.
5) Letter space : (,) duration equivalent to 3 elements.
6) Word space : (/) duration equivalent to 7 elements.

The word "paris" consists of the following letters (given with cw equivalents in symbols with time duration in elements):
P (. _ _ .) = 2 Di symbols (2 elements), 2 Dah symbols (6 elements).
A (. _) = 1 Di symbol (1 element), 1 Dah symbol (3 elements).
R (. _ .) = 2 Di symbols (2 element), 1 Dah symbol (3 elements).
I (. .) = 2 Di symbols (2 elements).
S (. . .) = 3 Di symbols (3 elements).

A total of 22 symbol elements, however we need to include symbol space elements:

P (.'_'_'.) = symbols (8 elements) + symbol spaces (3 elements).
A (.'_) = symbols (4 elements) + symbol space (1 element).
R (.'_'.) = symbols (5 elements) + symbol spaces (2 elements).
I (.'.) = symbols (2 elements) + symbol space (1 element).
S (.'.'.) = symbols (3 elements) + symbol spaces (2 elements).

A total of 31 time elements,however we must also include 4 letter spaces and 1 word space to complete the understanding that we are sending a 5 letter word:

P (.'_'_'.), = symbols (8 elements) + symbol spaces (3 elements) + letter space (3 elements).
A (.'_), = symbols (4 elements) + symbol space (1 element) + letter space (3 elements).
R (.'_'.), = symbols (5 elements) + symbol spaces (2 elements) + letter space (3 elements).
I (.'.), = symbols (2 elements) + symbol space (1 element) + letter space (3 elements).
S (.'.'.)/ = symbols (3 elements) + symbol spaces (2 elements) + word space (7 elements).

A total of exactly 50 time elements. So to send the word "paris" takes exactly 50 time elements. 

Example: To define the time duration of 1 element when sending at 12WPM:
12 words x 50 elements = 60,000 ms.
element = 60,000ms/600 elements
element = 100ms.

The purpose of this exercise is that now you will understand why sending cadence is everything when sending cw successfully. You will also now appreciate that a computer will send perfect cw whereas a human will send cw with imperfections. This is why the majority of serious cw contesters use the computer to send cw 90% of the time.

Doug (GM0ELP)

Yaesu FTDX101D performance in CQWW CW 2019

Before the contest started I played with the setup of the rig display. I wanted 2 bands (one left and one right side) with minimal space reserved for meters (just to show PO and SWR for transmit) to allow maximum waterfall/spectrumscope space. The waterfall/Spectrumscope ratio isn't adjustable when setup this way, but the default is good with more screen realty given to the waterfall. The left hand (VFO A) half of the display was configured to:
(CURSOR)/FIX/CENTER,
SLOW1/SLOW2/SLOW3/(FAST1)/FAST2/FAST3 and 1k/2k/5k/(10k)/20k/50k/100k/200k/500k/1000k. This allowed me to call CQ and see (in good detail) the stations around me. It also allowed me to search and pounce the whole band without the signals moving as I turned the VFO (until the screen needed pushed along a bit). The right hand (VFO B) half of the display was configured to:
CURSOR/(FIX)/CENTER,
SLOW1/SLOW2/SLOW3/(FAST1)/FAST2/FAST3 and 1k/2k/5k/10k/20k/(50k)/100k/200k/500k/1000k.
This allowed me to see the whole cw portion of the same band as VFO A or check a different band for possible mults (in one gulp, starting from XX.000KHz). I found the waterfall speed of FAST1 as being sufficient to see morse element detail without them disappearing off the screen during short gaps in rx signal. This setup was used till the end of the contest. I found search and pounce with VFO A a bit tedious as after tuning to the edge of the display, I had to push the displayed band segment along +10KHz with the VFO, then tune back to search and pounce the new segment. This allowed me to tune without the waterfall trails 'bending in frequency', which I find distracting.

VFO B was a mess on 20m with 500+ stations packed across the 50KHz, but I found using the rig mouse brilliant for search and pounce until RF caused it to stop working. I changed the mouse to one with a screened cable and a ferrite added for good measure and all was well to the end of the contest. I found using the mouse across a 50KHz span to be faster than just turning the VFO and more so, I could see when I had skipped a station during gaps in their transmission and back click to bag them.

I used the rig's RS232 port for OMNIRIG (used by my Gemini HF-1K for band/antenna selection) and the rig's USB port for rig control with N1MM+. I also used the rig's linear port (15w d-type) to control the KAT500 antenna tuner by sending PTT and band selection detail to allow pre-tune and antenna port selection on band change. This produces a fully automated setup and leaves very little to get wrong during the heat of battle.

I left the roofing filter on 600Hz (I don't have the 300Hz optional filter) and used the width control down to 150Hz for search and pounce or really bad adjacent station QRM. I did try the "VC tune" auto preselector (utune equivalent) on 40m but it made no discernible difference. The pre-amp was set to IPO on 40m and below, but I found AMP1 necessary for 20m and AMP2 for 15/10m good for weak signal when conditions allowed. The receiver never once gave me cause for concern with these setting during the whole contest, except for the usual eastern european QRO +2KHz wide brigade that dominate the results (Italy/Hungary/Bulgaria/Croatia being the worst offenders closely followed by Russia). A real downer for the sport having what can only be regarded as blatant cheats allowed to 'reserve' +2KHz to themselves and cause misery to the rest of the EU, there really is no contest possible with this situation.

Hopefully these eastern EU super stations will eventually be named and shamed with the proliferation of SDR recording stations. Being able to see the station means you can identify who is causing the clicks no matter how far away they are from you. If you have a home brew linear amplifier that you tune for maximum power and your receiver isn't the top spec available and you still have hair at the end of CQWW CW, you should know that your TX IMD is appalling and everyone hates you and your super station no matter what result you get.... stop cheating and clean up your act! Rant over ;-).

The FTDX101D worked brilliantly and so did the LinearAmp Gemini HF-1K/KAT500 combo for the whole contest. Looking forward to the next!

Yaesu FT-991A spectrum-scope / waterfall display noise through headphones work-around

After the work-around for the "no PTT in cw mode" bug. I started to think I might be able to use the FT-991a in a cw contest. On plugging in my headphones to the phones socket and listening through them, I noticed a background noise that sounded like a whine with some data like noise inter-mixed. I changed headphones but the noise was always present (more noticeable with low volume). For some reason I pressed the function key and noticed the noise disappeared (same effect with pressing the menu key). I soon realised that the noise was only present whilst the spectrum-scope / waterfall was visible and actively updating. I had a think about making up an audio low pass filter to go in series with the headphones but then thought better of it in case it let loose the magic smoke on such an expensive radio. I then started to play with all the scope settings like dimmer and colour to see if the whine would go away. Eventually I hit the solution, the noise would vanish completely when the "moni" (monitor, in the function menu) value was reduced below 20. Hopefully the time I spent finding this work-around will help someone out there!

Flexradio SSDR multi-client v3

Multi-client is an interesting feature developed by flex to allow 2 connections to one radio. I suppose the need to create useful application for their 2 SCU transceivers was great but I can’t help thinking that they have missed a trick. Why not a feature to allow SSDR connect to multiple radios? This way one could achieve the functionality of 2 SCUs in physically separate locations with once instance of SSDR. Surely this would be of more use to everyone?, but I expect this would have been too technically challenging for them?