++++++++++++++++++++++++++++ Date: Mon, 9 Jul 2001 16:39:49 -0700 From: "Ron D' Eau Claire" Subject: RE: [Elecraft] Weight Jim asked: > I'm puzzled about weighting CW. Does it have to do with spacing? In other > words does it mean that you can change the spacing by "weighting" things > differently? I thought the point of CW was to maintain an even spacing for > the benefit of the receiving station and for clarity in sending. You are quite right, Jim, provided by "even" you mean "consistent". "Normal" CW has, as its basic interval, the space between a dot or a dash and the next dot or dash in that character. For example, the space between the dot and dash that make up an "A" is a 'standard interval' for whatever speed is being sent. Using that space as one standard interval, a correct dot is equal to one standard interval and a dash is three standard intervals. Longer intervals denote the spaces between letters and words. These spaces also have defined lengths but I must confess I'd have to go hunting in my reference books to find them. The most important thing is NOT to use the same spacing oryourcwwillstarttosoundlikethis. It IS normal for "prosigns" (SK, AR, KN, CL and even the venerable SOS) to be sent as ONE character - no letter spacing. These prosigns are assigned 'letters' (e.g. AR) only as memory aids. They are individual, single, characters when sent. Weighting refers to the length of a dot or dash relative to the standard 'space' interval. Many ops have found that they can get through heavy QRN better if they use a somewhat heavier weighting. That is, the dots and dashes are longer than the standard one and three interval durations you'd expect. Back in the days when everyone used a manual key, I think many ops just instinctively started making the weighting heavier when the other station reported weak signals and poor reception conditions in general. Stations using bugs might just slow down, and their dots and dashes got proportionally longer as a natural consequence. Personally, I think that is true; slightly heavier weighting helps copy in poor conditions. All one has to do is try to copy one of those "bug" ops who leave their dot speed set at 30 wpm while sending at 15 wpm so see how hard it is to copy such CW under less-than-perfect conditions. Variable weighting in 'automatic' keyers like the K2 has also became important back in the days when rigs were commonly keyed with relays (before solid state keying circuits appeared). The relay takes a bit of time to 'pull in' - often quite a bit longer than it takes to 'release'. At moderate and fast CW speeds this can have the effect of making the dots and dashes too short. Adjusting the weighting of the keyer compensated for this shift. Generally, if you set your weighing too "light" (dots and dashes too short), OT CW ops will complain that your CW sounds "choppy". Clearly, the weighting option was added to the K2 so ops could 'tweak' their CW to sound 'better' to their ears. As you pointed out "consistency" is very important. More important, I believe, that exact machine-perfect spacing. Ron AC7AC K2 # 1289 +++++++++++++++++++++++++ ate: Mon, 09 Jul 2001 21:45:11 -0400 From: "Mark J. Dulcey" Subject: Re: [Elecraft] Weight Jim and Kate Belt wrote: > > Hi All: > I'm puzzled about weighting CW. Does it have to do with spacing? In other > words does it mean that you can change the spacing by "weighting" things > differently? I thought the point of CW was to maintain an even spacing for > the benefit of the receiving station and for clarity in sending. The new > firmware 2.0 manual says on page 6, "The range is 0.90-1.40, roughly 90% to > 140% of 'normal' weighting." CW "weight" is the ratio between the length of a dot and the length of a space. Normal weighting has the two lengths exactly the same. Keyers allow the weight to be adjusted, for a variety of reasons. Many rigs don't accurately reproduce the symbol length on their input signal; the weight adjustment allows the operator to send correct CW even with such an incorrect rig. (This reason doesn't apply to a rig like the K2, where the keyer is part of the radio.) And people often prefer the sound of CW sent at a weight other than the officially "correct" one, to compensate for the imperfect keying waveforms of transmitters and the time-domain response of receivers. (Typical crystal filters are terrible performers in that regard, to the extent that some rigs are useless for high-speed data modes.) ++++++++++++++++++++++++ On: Monday, July 09, 2001 11:41 AM "Ferguson, Kevin" wrote > I have read the article you refer to. It seems the author's intent is to > make a super CW-contester out of you... > > > A sloppy straight key fist is MUCH less irritating to me than a sloppy hand > on the keyer. Keyers DO NOT insure perfect code.'' > > If you can send clean code on one straight key, you can send code on almost > any other key without much trouble. The same can't be said of keyers, > paddles, or bugs... > > IMO, it is much easier to learn to send CW on a straight key, than learning > to copy. I agree that straight key sending helps your copying...I think it > helps you to focus on the rythem of the letters...I also think it helps when > the other fellows fist is a little off, or when conditions are obliterating > part of the signal. > > At beginning speeds, certainly, a straight key is no handicap....and will > serve you well. > ... Kevin has a number of good points, some of which contradict my advice. I have no reason to argue against any of it. There has been advice on both sides posted that reinforces what I originally opened with in that some of this may turn out to be personal preference. There one statement above that deserves reinforcing: "If you can send clean code on a straight key..." I say you can wrap a wire around your finger in an emergency and send good code. This is probably the best reason to ensure that you can send on a straight key; you can always cobble something together to make that happen. Good luck to those with this decision in front of them. Fran, KA4FRH ++++++++++++++++++++++++ Date: Tue, 10 Jul 2001 06:02:22 -0500 From: Tom Hammond =?iso-8859-1?Q?N=D8SS?= Subject: Re: [Elecraft] Weight GM Jim: >I'm puzzled about weighting CW. Does it have to do with spacing? In other >words does it mean that you can change the spacing by "weighting" things >differently? I thought the point of CW was to maintain an even spacing for >the benefit of the receiving station and for clarity in sending. The new >firmware 2.0 manual says on page 6, "The range is 0.90-1.40, roughly 90% to >140% of 'normal' weighting." 'Weighting' refers to the duty-cycle of the characters, rather than their timing. Timing has to do with element spacing, not weighting. Some ops (such as myself) prefer to have out characters run just a bit 'light' in timing, especially at higher frequencies... others seem to prefer to have them a bit 'heavier' at the higher speeds. It all depends upon how YOU percieve the other guy's ability to best copy what you send. 50% weighting means that a DIT is ON for one BAUD and OFF for one baud. A 'light' DIT might be ON for .9 baud and off for 1.1 baud. This weighting 'lightens' the DIT a bit while retaining the same timing of the code at the current sending speed. I find that, in my K2, a weighting of 0.90 yields just the level of 'lightness' in weighting (about 47% weighting) that I prefer to have, especially when contesting. YMMV, so Wayne has given us the option to vary our 'personal' weighting to meet our needs. Constructive dissenting opinions are always welcome. 73 - Tom N0SS ++++++++++++++++++++++++++++ Date: Tue, 10 Jul 2001 06:42:58 -0500 From: Tom Hammond =?iso-8859-1?Q?N=D8SS?= Subject: Re: [Elecraft] Weight Jim: I wrote: >50% weighting means that a DIT is ON for one BAUD and OFF >for one baud. While this statement is correct, it's only correct for consecutive DITs, as in an I, S, H, or 5. Where an E is sent, there will be three bauds (timing units) following the DIT, ot create a character space. And where the E is the last character in a word, it will be followed by seven baud spaces, to form a word space. Tom N0SS +++++++++++++++++++++++++++++ Date: Tue, 10 Jul 2001 07:44:36 -0600 From: "Santa Fe" Subject: [Elecraft] Re: Weight (yikes) Yikes. "Weight" refers to the ratio of the dot to space. 50 - 50 is prescribed. It affects the dash to space ratio as well, but is stated as the dot to space ratio. Some people try to make up for a radio's keying delay by beefing up the weight, a mistake. "Ratio" refers to dot to dash (ratio). 3 - 1 is prescribed. Bugs almost invariably produce longer dashes than this. The variability of dashes with bugs is known as "swing" and can be used by the operator to apply emotion and inflection to the characters, something which is very difficult with machine sent code. A well adjusted bug will produce dits for some time, all very close to 50 - 50. But the keying delay in most modern radios will make these dits sound staccato or under weighted. To use a bug with such radios forces one to adjust the bug for heavy dits to make up for the keying delay, and this will make the dits come to an end after perhaps 8 or ten dits, and the end is a key closure. The CMOS Super-keyers (and some others) have a "constant" which allows one to adjust for the delay in many modern radio keying circuits. The constant will apply the same amount of time, 5 ms for example, to both a dot and dash character, regardless of speed. The keyer's weighting will not accomplish the task since it is speed related while the radio's keying delay is not. If you look on an oscilloscope at the keying output of a radio for a string of dots at 15 wpm, they are likely to look very close to 50 - 50. If you look at the dame radio and keyer at 60 wpm, you will that the dots look very much less than 50 - 50. Both had a keying delay of perhaps 5 ms, but proportionately, the 5 ms was not significant at 15 wpm, but is very significant at 60 wpm. Injection of the keying constant can make it perfect at both speeds. Do NOT use weighting to fix this problem. I have never heard the word "timing" used to describe any adjustment on a keyer. Different "modes" require different timing from the operator when used with an iambic paddle, however. An operator certainly becomes accustomed to the timing for a mode and will find changing modes difficult at best. 73, FD kt5x ++++++++++++++++++++++++