|
New to Ham Radio?
My Profile
Community
Articles
Forums
News
Reviews
Friends Remembered
Speak Out
Strays
Survey Question
Operating
Contesting
DX Cluster Spots
Propagation
Resources
Calendar
Classifieds
Ham Exams
Ham Links
List Archives
News Articles
Product Reviews
QSL Managers
Site Info
eHam Help (FAQ)
Support the site
The eHam Team
Advertising Info
Vision Statement
About eHam.net
|
The Butterfly Effect
Ulrich H. Steinberg (DJ8GO)
on
July 10, 2008
View comments about this article!
In three previous articles on this site I have traced the development story of an advanced memory keyer, based on the AVR Butterfly microprocessor board, each time implying that there was essentially nothing more to be done - I'm wiser now and will not make that insinuation again. ( see http://www.eham.net/articles/11749 and http://www.eham.net/articles/14970 and http://www.eham.net/articles/16608 ). When I set out to do this, I had no idea that it would keep me busy for such a long time, let alone that I would spend literally thousands of hours making it all work. It turned out to be a perfect project to bridge the gap between the peaks of two solar cycles ;-)
Meanwhile the keyer has become a commercial product, the Begali CW Machine, with a small but dedicated group of users, and it is packaged in a nice enclosure that is made by TenTec. There is enough documentation of the hardware available online at http://www.i2rtf.com/docs/CWMachineHardware.pdf to let you build your own version of it if you are so inclined (you would still have to get the firmware and software from me, unless you want to spend a couple of years developing it yourself ;-)

Countless suggestions have been made, and I myself as a power user have had a lot of ideas derived from daily use, and many of them made it into the product in some shape or form. Having spent so much time on the development, I pretty much felt that I had covered all bases and done everything that is doable in a memory keyer. That is, until I was hit with a harmless sounding user question that I had never thought about because I don't own the latest and greatest radios: would it be possible to use your paddle to send CW from a radio at another site that is connected to you over an Internet link?
A little background is required to understand why this is a tricky issue, and nobody to the best of my knowledge has so far been able to do this in a convincing fashion.
Some modern radios can be completely controlled and operated by software that runs on a PC that is attached to the radio. Then there are packages on the market, and even built directly into Windows XP and Vista, that let you “see” and operate a PC over the Internet from another PC or laptop at a remote location. You can even hear the sound of that remotely controlled PC. Put these two together, and you have the building blocks to operate your radio from anywhere in the world. You hear the sound from your rig via the sound card connection, and talking into a microphone will send your voice via the sound card into your rig and out on the air. The TenTec Omni VII takes this one step further and eliminates the need for the remote control software by putting the Internet connectivity right into the radio.
So, how do you send CW with this kind of remote setup? The only answer so far has been to use a keyboard and to type text into a program that converts it to CW at the radio site. You can't use paddles because the information about closing and opening a contact, if you could send it over the communication link, would be distorted by the minute delays and irregularities in the connection, and it would be impossible to send clean CW this way.
I am a reasonably good typist, but I have never quite grown accustomed to using a CW keyboard for anything but short contest exchanges, where all you practically type in is the other stations call. For “real” CW QSOs I prefer my paddle by a mile. So how can a guy like me use his beloved paddle with a remotely controlled radio?
After some thought I realized that the CW Machine firmware / software combination has all the basic building blocks to make remote “paddling” a reality, and after some more days of programming the first prototype actually worked! At the core of the idea is the fact that the CW Machine firmware translates paddle movements into readable characters which scroll across the LCD screen of the device, and which are also sent to a companion Windows program, the CW Machine Manager, where they show up on a monitor screen.
The CW Machine now supports several techniques to make remote CW a reality, depending on whether you have CW Machines on both ends of the communication link or just one CW Machine, either at the radio location or at the remote site.
The simplest approach, assuming that you have only a CW Machine connected to your radio, is to just use a remote desktop program that lets you start the CW Machine Manager and the radio control program. The CW Machine Manager has a sophisticated CW keyboard function, far more advanced than a typical CW keyboard program - so this appears like a pretty good solution. In practice, however, typing over a remote desktop link is often a pretty frustrating experience, with characters being randomly delayed, and the end result is a choppy CW signal that is hardly acceptable. (There are functions in the CW Machine Manager to mitigate that issue, but it is far from the experience that you have with a paddle.)
The second scenario is that you have a CW Machine with you at the remote location and just a CW keyboard program at the radio site. You bring up the CW keyboard program on your remote desktop, similar to the CW Machine Manager program in the first scenario, and the CW Machine that you have with you is now able to “trick” that program into reading the characters that you create with a paddle just like you had typed them in on the keyboard. Now you're using your paddle for CW! That's a big step, but some issues have to be solved. The CW keyboard program will in all probability use different characters to represent the operational prosigns. The CW Machine, e.g., uses the # character for the [sk] prosign, while the TenTec Omni VII uses a > for that purpose. Also, the CW Machine supports many foreign characters, like the Spanish Ñ or German Umlaut characters or the Icelantic Thorn character Þ. You probably didn't know that there was a Morse encoding for those, and neither do most CW keyboard programs, which either do nothing or send garbage when they see these strange beasts. Fortunately these issues have an easy solution. What remains is the issue that the remote desktop connection may introduce fluctuating delays that result in a choppy CW signal.
Another tricky issue is, that the CW keyboard program is operating at a particular speed which can hopefully be controlled from your remote location, but the CW Machine that you use with your paddle is running at the pace that you set with the speed potentiometer. As long as these two speeds are about the same, there is no major problem, but whenever you readjust your speed potentiometer, you have to make the corresponding change in your radio control program to keep both sides in sync. It would be nice if the speed of both sides could be synchronized by just turning the speed potentiometer on your CW Machine.
And that is where the third technique comes to the rescue, which requires that you have CW Machines in both locations. The CW Machine Manager can now operate as a server at the radio location which “talks” directly over the Internet to another CW Machine Manager at your remote location. Turning the speed potentiometer at the remote location automatically adjust the speed on both sides, and, since they both support the same character set, the compatibility issues mentioned above don't exist. Since the communication technique does not involve simulated or real typing, it is practically immune against minor fluctuations in the link.
It feels almost magical when you turn the speed potentiometer of a CW Machine in a hotel room somewhere, and you know that the CW Machine at the radio will faithfully track that adjustment. And when you use your paddle to create characters they will come out as perfectly spaced CW at the radio site.
This type of remotely controlled operation may not be legal everywhere, and you have to consult your licensing rules and regulations. But then, it has to be legal only in the country where the transmitter is located, and your license must authorize you to operate in that country. This is “real” radio, using the Internet only as a fancy long cable that lets you adjust your radio and the attached CW Machine.
My intrepid pilot user, who put that harmless sounding question to me in the first place, is a ham in a country where this type of operation is still a matter of dispute - so let's give her/him the benefit of anonymity but a big applause. We had to resolve a few issues, but eventually I was able to operate an incredible station that I could never afford myself, from my crummy hotel room in NJ, using my favorite mode of operation - good old CW created with a paddle.
This new function of the CW Machine allows you to operate a Big Gun CW station from an apartment in the city, where you wouldn't even have the permission to string up a wet noodle for an antenna. Do I sense a business opportunity here, offering telecommuting to a super station? … ;-)
This works so well that I was seriously tempted to get myself a modern rig to operate remotely from the hotel rooms that I spend most of my nights in - but then, why not keep the green to develop more creative applications for the CW Machine, and find a few generous CW Machine owners who let me use their dream stations remotely every now and then? ;-)
Like with the famous “Butterfly Effect”, a small disturbance at your paddle can key up a kilowatt amplifier at a location far, far away, creating a storm of perfect Morse to be heard around the world.
This article has expired. No more comments may be added.
|
Creeping featuritis
|
|
|
by N3QE on July 10, 2008
|
Mail this to a friend!
|
My gut feeling, having looked over the Begali CW Machine, is that it has too many features for me.
I'm the sort of guy who reads a review of a rig in QST, and if it starts out contrasting option 13 of menu 24 and its setting vs option 7 or menu 12, I realize that it's not the radio for me.
In the same way I was overwhelmed with the CW Machine's configurability. I've seen it in use and I was impressed. At the same time I looked at the firmware manual and was completely overwhelmed.
There was an aspect that I potentially liked about the CW Machine: its self-logging ability. The ability to have the keyer do the logging, independent of a computer, was very appealing.
I do have a set of Begali paddles and simply simply love them. I use them with a much more simplistic programmable keyer (a K1EL) that I jury-rigged with some foot-pedals to let me work contests with my feet while logging and turning my rigs knobs with my hands :-).
Tim.
|
|   |
|
RE: Creeping featuritis
|
|
|
by DJ8GO on July 10, 2008
|
Mail this to a friend!
|
N3QE:
As a developer I am very aware of that danger, trust me ;-) But if you want to use the CW Machine as a simple memory keyer, like a K1EL, just attach it to your rig, load a couple of messages into it, and start sending. All the advanced stuff, which results in such a sizable user manual, is frosting on the cake and is there when the need arises. Even complex functions, like the automatic creation of a log, can be done without special configuration and without doing anything that you wouldn't do when using a simple memory keyer.
|
|   |
|
RE: Creeping featuritis
|
|
|
by K0BG on July 10, 2008
|
Mail this to a friend!
|
I think Tim is basically correct, but all one has to do is look around at the other amateur related devices. Just about every transceiver on the market has more knobs and features than any one single user will utilize. Heck, I've had an IC-7000 in my vehicle for two years, and I've never found the need to program any of the memory slots. It even does RTTY, but it will never see that mode either!
Ulrich, you mentioned the possible manual complications when you get too many features, and perhaps that's true. However, as I alluded to above, one user's need is not necessarily another's.
Where the fine line has to drawn, is somewhere between serving 80 to 90 percentile of the marketplace, and trying to be everything to everybody. I don't think anyone has ever accomplished the latter.
Alan, KØBG
www.k0bg.com
|
|   |
|
CW Machine
|
|
|
by DJ8GO on July 10, 2008
|
Mail this to a friend!
|
Alan, Tim,
One natural limitation on creeping featuritis for me is that I'm a single guy who has to maintain all these features ;-) And another one is that I am personally using the device on a daily basis and implement those features that I would potentially use myself. The list of features that have materialized over the last two years is still amazing even to me, but it is the result of carefully considered requests by actual users, and the fact that you can do things with the underlying microprocessor board that you couldn't dream of doing with a "normal" memory keyer. That 80%-90% coverage is an elusive goal, but I think I have come as close as I ever will.
|
|   |
|
RE: Creeping featuritis
|
|
|
by N3QE on July 10, 2008
|
Mail this to a friend!
|
About being "Everything to everybody", I don't think the CW Machine will ever be about that. I think it will appeal to certain subsets, and I count myself in the first subset:
1. Those who want some kind of electronic logging but don't want to sit in front of a computer while working CW.
2. Those who want a very highly configurable keyer.
A lot of old-fogeys like me will be perfectly happy with much more simplistic keyers.
In the meantime there's already a lot of contesters out there who don't mind sitting in front of a computer in a contest and let the computer do nearly all the keying. Some of these guys ragchew from a computer tooo. I'm hoping that at least some of them defect from this method as the CW Machine and maybe other tools become available that get us back to radio and away from sitting in front of a computer.
|
|   |
|
RE: CW Machine
|
|
|
by K6AER on July 10, 2008
|
Mail this to a friend!
|
Ulrich
Very nice article. It is nice to see you in print. I wish I had more time to play with the machine at the Begali booth at Dayton but I was too busy selling their beautiful keys. You look like you were in heaven explaining the operation of the box.
Not everyone is equipped to explore the intricacies complicated equipment but I appreciate the engineering that went into you device. You did a wonderful job on the design.
We’ll see you next year.
Mike – K6AER
|
|   |
|
RE: CW Machine
|
|
|
by W9OY on July 10, 2008
|
Mail this to a friend!
|
Seems to me you could have "CW machine" emulators at each end of the circuit and just plug the paddle into a serial port on the remote computer. Certainly not as elegant but parsimonious. Like you I have not been satisfied using the keyboard as a remote keying device. There are some programs that will allow TCP/IP serial ports to connect across the internet.
My Flex 5000's inboard keyer (which is a piece of software) can be keyed from a serial port. In fact it can be keyed from a serial port and also keyed from the internal KEY jack on the radio at the same time. If you were able to connect a virtual serial port on the host computer to a real serial port on the remote computer you could easily key the radio. Here is an example of such software
http://www.hw-group.com/products/hw_vsp/index_en.html
I don't know if other radios allow this means of keying but its an interesting idea.
73 W9OY
|
|   |
|
RE: CW Machine
|
|
|
by KA4KOE on July 10, 2008
|
Mail this to a friend!
|
Too much for me, although it may be ok for others. Gimme a Vibroplex and I'm happy.
FEEELEEEEP
|
|   |
|
RE: CW Machine
|
|
|
by DJ8GO on July 10, 2008
|
Mail this to a friend!
|
W9OY:
If you can, indeed, hook up a paddle (not a keyer) to the serial port of your Flex 5000, that just means that two wires in the serial cable work as the dit and the dah line, like they would in a normal cable with just three wires. You would run into the problem that I am alluding to: you want to transmit the open/close information over a link with irregular delays, and that would fail miserably.
Having two software emulators at both ends would work - after all, the CW Machine situation is just software on both ends, and part of that software runs in the keyer, some other part in a Windows program. There are some very tricky timing issues to resolve (and I'm not disclosing how that can be done ;-), but let me give you an example: let's assume you have hooked your paddle into some software that translates your paddle movement into characters that you send over the link (pure open/close information won't work as explained above) , at the other end it's reversed into CW that keys your rig. Now try sending "W9OY DE N2DE": the "W" would be sent over the link as soon it has been recognized, i.e., after the character space following the last dah, and your transmitter would start sending it. Meanwhile you have started forming the "9" with your paddle, which takes a lot longer than the transmission of the "W" because its a longer character, but it can't be sent before it's completed and has been recognized (after all, your software can't be clairvoyant ;-). The effect would be an improperly long space after the "W" ... The result would be choppy CW with improper spacing. These kinds of issues are far better handled by a dedicated device like the CW Machine, and they are probably the reason why remote "paddling" has not had a convincing solution so far.
|
|   |
|
Creeping featuritis- comments
|
|
|
by AC7ZL on July 10, 2008
|
Mail this to a friend!
|
I can see both sides of this discussion.
It is entirely possible for a device to be so overladen with "features" that it becomes all but useless for it's basic purpose. This situation reminds me of an absurd limited-edition collectible "pocketknife" issued by Victorinox. The knife features what appears to be fifty or more tools and blades-- basically every tool that has ever appeared on a pocketknife--- all on one knife. It's a neat thing to look at, and will probably have value as a collector's item, but as a tool, it is completely useless.
On the other hand, when you put a microprocessor into a tool or device, what comes with that is the potential for unimaginable function, flexibility, and configurablity. The danger is this: As a device's complexity and sophistication increases, the tendency for it to become the "pocketknife" mentioned above also increases.
Two things can prevent that from happening. One, is a properly designed user interface. This means an interface that is easy to understand, easy to navigate, and intuitive. The other factor is well-written documentation. My experience is that the latter is frequently poor, even when the former is good.
Assuming, for the moment, that the "CW Machine's" user interface *is* well designed, maybe it should come with *two* manuals. One would be a full-length manual for power users and people who have need for obscure features. The other manual would be very brief, feature larger type and clear illustrations, and would address only those topics pertaining to the machines basic use.
Pete
AC7ZL
|
|   |
|
RE: Creeping featuritis
|
|
|
by W9AC on July 10, 2008
|
Mail this to a friend!
|
One useful feature for remote Internet operation is the addition (if not already incorporated) of a local CW side-tone generator into the keyer.
I have been using the K1EL remote Internet system for the past year and the results have been very satisfying when using a WinkeyUSB on each end of the Internet. What the WinkeyerUSB does not have is a side-tone output for insertion into a laptop PC which, can then be used to mix side-tone with receiver audio into the headphones. Although the K1EL chip has a side-tone output, I needed to modify its square-wave output and convert it into a sine-wave. The output feeds the Line-In jack on my remote laptop PC.
Why the need for local side-tone? Actually, it's not needed if one only uses remote keyboard CW over the Internet. The side-tone from the host radio can be sent over the Internet and that works well even with the Internet's latencies. But what happens when you are using a paddle or straight key at the remote end? The human ear requires a local side-tone in order to keep sync with our sending. And so, if remote paddles are used, plan on finding a way to insert a local side-tone into the receive audio mix at the remote end.
To convert a square wave into a sine wave on the K1EL keyer, I first used a 3-pole, 18 dB/octave high-pass filter, followed by an extremely poor quality audio transformer from an old transistor radio. In almost any other case, designers want the best audio transformer money can buy in order to maintain excellent complex wave transient response, low distortion, and a high saturation point. Not so on this application. The worse the transformer, the better it will aid in converting a square-wave into a sine wave.
The K1EL system also allows the user to control every single parameter from the remote location and best of all, the K1EL system will produce perfectly-timed CW.
Another thought -- when transmitting CW characters over the Internet, there are only two viable options, in my opinion. Either the CW must be sent in completed, formed characters, or by forming and timing each keyed element. It is not enough to send key closures over the internet. Digitally converting either the formed character or keyed element will ensure the best timing possible. My preference is to use timed keyed element spacing in order to avoid the potential timing issue with some character formations. Consider the following call sign:
KE0Z
When character formation transmission is used, the "0" after the letter "E" cannot be sent until each long keyed element is formed. This produces a bit of a lagged effect on a call sign like this. By contrast, when each keyed element is sent digitally over the Internet, no such stuttering will occur.
The Ten Tec Omni VII uses character formation, but as a practical matter, the CW timing is so good, that one only notices the effect when after transmitting long-duration characters like the number "0."
I will be interested in more information on Ulrich's excellent keyer. It looks like it is very well designed and knowing Ulrich's attention to detail, I am sure it works flawlessly.
|
|   |
|
The Butterfly Effect
|
|
|
by W3FM on July 10, 2008
|
Mail this to a friend!
|
|
Recently I worked PY7COU/F on 40m CW. Paulo was operating from Recife,Brazil and by internet keying the rig of an F4 station near Paris. We had a solid ragchew. He has the ability use other stations in Australia, Pennsylvania and Virgina. He was keying from his PC keyboard but expected to soon be able to use his paddels.
|
|   |
|
The Butterfly Effect
|
|
|
by N2ZXE on July 10, 2008
|
Mail this to a friend!
|
I have to admit that it is an interesting piece of hardware and to be able to CW from a computer connected to the Internet seems like a cool feature.
However, there are moments that I'm afraid that contests will be done by CW automatons in which will be able to identify a call sign, get the the contest information, reply back while the "operator" is watching the game while having a beer.
If you think about it, the technology is all there and available.
As time goes by CW looks less than CW...
I guess I'm getting old :).
|
|   |
|
The Butterfly Effect
|
|
|
by N0AH on July 10, 2008
|
Mail this to a friend!
|
|
Hey, I thought I saw this in a MFJ catalog under some code thingymajigger.................Martin!
|
|   |
|
RE: The Butterfly Effect
|
|
|
by KB1OOO on July 11, 2008
|
Mail this to a friend!
|
DJ8GO,
Regarding sending open/close information: instead of transmitting open and close "events" which suffer from delays and reliability of the communication, why can't you encode the precise open/close times on the local machine and send that information to the server? I think that you would only have to buffer by some very small amount on the server. This is how CWIRC http://myspace.voo.be/pcoupard/cwirc/ works to transmit cw over the internet to another internet user and it works quite well, in spite of inefficiencies do to the fact that its built on top of the IRC network with possibly quite a bit of load. In fact, CWIRC can interface received CW signals over the internet to your rig so I don't think that your statement about there being no way until your device to remotely send CW with a paddle is quite correct. You can also use a straight key or a bug.
Your approach would kill someone's bug swing that they are very proud of (although make me happy on the receiving end) :)
73,
Marc
|
|   |
|
RE: The Butterfly Effect
|
|
|
by DJ8GO on July 11, 2008
|
Mail this to a friend!
|
KB1OOO:
that approach sounds easier than it is when you actually try to implement it, unless I misunderstand your suggestion. At 30wpm the elementary timing interval (length of a dit and time between elements of a character) is 40 milliseconds. Let's assume you want to send CQ by transmitting the keydown and keyup times with perfect timing:
down 120, up 40, down 40, up 40, down 120, up 40, ...
Assume that there are slight irregularities on your link, which is absolutely normal, and you don't have an unbroken and uniform incoming flow of these "packets". If the first "up 40" is delayed by 5 msec it would act like a "up 45", messing up your "C". So you want to buffer these packets on the receiving end to smoothe out these effects. How many should arrive before you actually start your transmission? Once your buffer is empty you start filling it again, increasing the gap to your next element until the buffer tells you that it's ready again to supply packets. These things are not easily resolved, and any approach that does not take into account random delays on the link may work fine in a controlled lab environment without appreciable transmission delays, but hardly in real life.
|
|   |
|
RE: The Butterfly Effect
|
|
|
by K4JSR on July 11, 2008
|
Mail this to a friend!
|
KA4KOE said, "Too much for me, although it may be ok for others. Gimme a Vibroplex and I'm happy."
Being a good old Georgia boy, what Feeeelep meant to say was a quote from Butterfly McQueen, "But I don't
knows nuthin' about birthin' no CW, Miss Scarlett!"
(Ms. McQueen is not to be mistaken for Madame Butterfly of Grand Opera fame!)
That's my story and I'm sticking to it like burnt Possum meat in a skillett!
73, Cal K4JSR Loading up for a hamfest!! :-@
|
|   |
|
RE: The Butterfly Effect
|
|
|
by KB1OOO on July 11, 2008
|
Mail this to a friend!
|
DJ8GO,
Buffer underrun comes up all over the place in real computer science
applications (OpenGL 3D rendering, audio streaming, video streaming,
VOIP, etc). Given the lower required bitrate, CW is less complicated
than any of those situations.
Let's suppose for simplicity, that we needed to buffer by 1 second, and
let's suppose we have a 3 second circular buffer. We don't start
playing until we have 1 second of the buffer filled. Now suppose we have
a 200ms delay in the middle of your 30wpm keying of C. E.g.
+120-40+40-40 then (200ms) delay. Since we are not bit rate limited,
the +120-40+40 data fills in immediately after the delay and there is no
compound effect of the delay. By the time it gets played, you have an
intact signal of +120-40+40-40+120-40+40.
There are also plently of simple schemes, like double and triple
buffering, for dealing with required output device buffering.
I'm on CWIRC every day and happy to have a 30wpm QSO with you over the
internet to show you that it's not a "controlled lab environment". And
CWIRC has tons more overhead to deal with than would be necessary for a
simple client/server link. If you want to go faster than 30wpm, DJ1YFK
is usually there also so you can ramp it up to higher speeds with him.
Marc
|
|   |
|
RE: The Butterfly Effect
|
|
|
by DJ8GO on July 11, 2008
|
Mail this to a friend!
|
|
I am using similar logic, except that I don't transmit open/close information but characters, and I know of course that you can mitigate those underrun situations. The tradeoff is during the time you need to fill the pipeline. In higher speed conversational CW you often have short exchanges like "?" or "R" which do not take much time, and in your example with 1 second front-end delay they would not get sent as intended. I am glad that other folks have run into the same issues and found very similar solutions. It is an interesting exercise because the trained ear is so sensitive to altered timing and gaps - in streaming audio a 100msec gap hardly matters, but in CW it is truly annoying; in that sense, something as simple as CW has its own set of complications and is not so simple after all.
|
|   |
|
RE: The Butterfly Effect
|
|
|
by KB1OOO on July 12, 2008
|
Mail this to a friend!
|
DJ8GO,
The 1 sec was just an example. Unless you end up being bit rate limited, I don't think that you'll need more buffering than you need to achieve your character sending. I use voice over IP Skype for phone conversations every day. Wouldn't saying what?, yes?, and OK? during a conversation be similar to sending ? and R during a CW QSO?
73,
Marc
|
|   |
|
RE: The Butterfly Effect
|
|
|
by KB1OOO on July 12, 2008
|
Mail this to a friend!
|
I just set the recv buffering to 200ms on CWIRC. No blips at all. Tried 20wpm through 60wpm, all different types of characters. This is from Belgium to US. Again lots of overhead because the link is handled by multiple IRC servers serving thousands of users. 100ms produces a few blips. With a direct link, I'm sure that you can drop this considerably.
73,
Marc
|
|   |
|
RE: The Butterfly Effect
|
|
|
by DJ8GO on July 12, 2008
|
Mail this to a friend!
|
KB1OOO:
The difficulties with CW arise because it contains so little information (compared to VoIP, say), and every small deviation from "perfect", which are masked in the volume of other information streams, becomes immediately obvious. Plenty of room for experimentation to achieve a good balance ...
|
|   |
|
The Butterfly Effect
|
|
|
by LX1LH on July 14, 2008
|
Mail this to a friend!
|
Dear Ulrich,
you have build a nice machine there with - from my subjective point of view only - one catch: the user interface. I read the manual (yes I did) and so many features with so many options to get lost. A bigger display with some more status LEDs maybe would have harmed the now pristine optics but would improve handling the otherwise convincing featuritis.
Best 73 es gb de Lutz, LX1LH
PS: Why a big DIN port for the keyboard ?
|
|   |
|
RE: The Butterfly Effect
|
|
|
by DJ8GO on July 14, 2008
|
Mail this to a friend!
|
LX1LH:
The options are there if and when you need them, but nothing forces you to even look at them ;-) Sure, a bigger display and blinking LEDs might be nice, but they would rule out the Butterfly board as a base component and increase the cost significantly. The "big DIN" connector is a PS/2 connector, and the only alternative would be to support USB keyboards / keypads, which again would increase the cost and hardware complexity significantly without adding value. (the device comes with a matching keypad, so you won't have to look for one) I think that looking at the documentation, which spells out every option in detail and might be intimidating, and actually using the keyer and taking a look at the options if and when you see a need, are two very different experiences ...
|
|   |
|
RE: The Butterfly Effect
|
|
|
by LX1LH on July 14, 2008
|
Mail this to a friend!
|
Dear Ulrich,
> The options are there if and when you need them,
> but nothing forces you to even look at them ;-)
I don't just want to look at them - I want to use them all. For much less features there are alternatives but the features are great.
> Sure, a bigger display and blinking LEDs might
> be nice, ....
Not nice, in my opinion essential. Think of the logbook function. What much more you could do and without the use of a PC - if you would have a proper GUI. More costs ? Hm Maybe but how about using the onboard RS-232 for more user interaction like the Tasco CWR620 or even the Elecraft K3 ? A little terminal program on the other side - just an idea. Its all there and just a matter of software ... you already interact via serial with certain logger programms, why not go the last mile and have a simple user interface ? Logging with a PC (ever), not just contest style ...
> The "big DIN" connector is a PS/2 connector,
On the Begali Homepages there is a model with a fat DIN one .... maybe I am blind - my apologies - but good news.
A word in general: I am counting my cents to buy one of your great creations. I just try to trigger a constructive dialogue .... ;)
Best 73 de Lutz, LX1LH
|
|   |
|
RE: The Butterfly Effect
|
|
|
by DJ8GO on July 14, 2008
|
Mail this to a friend!
|
The DIN must look big because the box is so small ;-) It always was a PS/2 connector, even on the Begali page.
In "real life" a larger display would not add that much value: the device is designed to do all logging operations without an attached PC, and all essential information is scrolling across the LCD (so it really is not limited to the 6 characters), and several indicators on the LCD provide status information. If you do connect to a PC, then you have a nice interface program, the CW Machine Manager, which shows much more than a simple terminal program could do.
|
|   |
|
The Butterfly Effect
|
|
|
by W2ST on July 14, 2008
|
Mail this to a friend!
|
|
As a user of this very compact and full-featured keyer, I am really impressed with what you have done, Ulrich. The fact that it is software-defined means that features can be altered or interchanged (as you do with the CW trainer software) and updated indefinitely (as you have done several times, in response to comments from me and others). Those who have commented on complexity -- or on features that are or are not included -- might enjoy spending a few hours with the keyer; I'm betting that most would find that it sets a new high standard.
|
|   |
|
Feature request
|
|
|
by KASSY on July 14, 2008
|
Mail this to a friend!
|
There seems to be a very basic feature that is completely absent from the CW machine. A feature that shocks me in its absence.
No history readout! When I send CW, I need a history readout- a way to know what I've sent to the other person, so that I don't repeat myself.
I do that function now by simply connecting a CW reader up to my sidetone jack in the rig. But it amazes me that a so-called do-all machine would be completely absent this feature.
Hopefully, it's not very expensive.
- k
|
|   |
|
RE: Feature request
|
|
|
by W2ST on July 15, 2008
|
Mail this to a friend!
|
Actually, that function IS present in the CW machine. If the keyer is connected to a computer, there is a real-time readout of everything that you send, on screen.
73,
Stan, W2ST
|
|   |
|
RE: Feature request
|
|
|
by DJ8GO on July 15, 2008
|
Mail this to a friend!
|
|
... and I am shocked that someone would make factual statements without even having taken a cursory look at the documentation. Of course the CW Machine displays your history, if you want to for hours or days back and far more than a code reader would show. There is even a print button that allows you to have a hardcopy of your history.
|
|   |
|
RE: Feature request
|
|
|
by W9AC on July 15, 2008
|
Mail this to a friend!
|
"...Of course the CW Machine displays your history, if you want to for hours or days back and far more than a code reader would show."
Excellent feature. I use the echo and history feature on the K1EL remote system to try my best at sending perfect word-spaced CW as completed characters are formed and transmitted through the Internet.
Until I began monitoring my own sending in real-time, I was sending words spaced much too closely together. Under normal circumstances, a slight tightening between words can still be interpreted by the operator on the other end. But when producing machine-generated CW from a paddle, the keyer/computer allows for no special dispensation unless ample time is left between words. Otherwise, words will run continuously together.
Because of this, running CW over the Internet has actually improved my sending as an operator.
Paul, W9AC
|
|   |
|
RE: Feature request
|
|
|
by DJ8GO on July 15, 2008
|
Mail this to a friend!
|
|
The CW Machine allows slight deviations from perfect spacing, but still, you will often catch yourself running words together or breaking them up in some situations, and you will see that right away because your transmitted text scrolls across the LCD. This instant feedback is particularly important for the CW Machine since some of its functions, like e.g. automatic logging, depend on analyzing "words" in real time.
|
|   |
|
RE: Feature request
|
|
|
by KASSY on July 15, 2008
|
Mail this to a friend!
|
"RE: Feature request
by W2ST on July 15, 2008
Actually, that function IS present in the CW machine. If the keyer is connected to a computer, there is a real-time readout of everything that you send, on screen.
73,
Stan, W2ST
Yes, that's the problem - it is NOT built into the machine. You must have a separate computer connected. If I have to have a computer involved, then there are dozens of feature-rich free packages that can be used to send CW.
I admit I did not read the docs far enough to find that feature. I only went through about 20 pages of the PDFs on the website. Far enough to find out that you turn it off by removing the battery. No power switch on the world's most expensive keyer.
I confess that I'm a tad impatient. If I have to read the full manual in order to learn the feature set, I'm not too interested in paying top dollar.
- k
|
|   |
|
RE: Feature request
|
|
|
by DJ8GO on July 15, 2008
|
Mail this to a friend!
|
"I'm not too interested in paying top dollar" is probably the key issue (although you said you didn't look at the price), I would suspect, and everything else is a pretext to find some fault. Built into the keyer is a scrolling display of your transmission, but it is of course limited by the size of that display. If you need to hook up a CW Reader to remember your last 50 or so characters that a stand-alone reader is able to show (for anything longer you would need a computer), then your short term memory is probably a medical issue that no device can cure ;-)
|
|   |
|
The Butterfly Effect
|
|
|
by G0UUT on July 28, 2008
|
Mail this to a friend!
|
Uli
you are doing a fantastic job not only on this Butterfly, as well on the cw key keep on the good job .
i am looking to order it soon from Piaro
thank you Uli
Elan g0uut/Dl9fcc
|
|   |
|
Email Subscription
You are not subscribed to discussions on this article.
Subscribe!
My Subscriptions
Subscriptions Help
Related News & Articles
So-Called 'Code Practice'
Finally, Enough Nerve to Get on the Air!
The Hijacking of 40-Meter CW
Morse Code for KB3HQH
Is CW (Morse Code) a Digital Mode?
Other Products Articles
10-Pound 700-Watt Automatic Amplifer
Today's Amateur Radio Equipment is a Bargain!
Audio Games
IC-7000 Woes
Simple Power Injector for Preamps
|
|
|