Issue #24: Cool Devices to Learn CW

The variety, creativity, and innovation I see here reinforces my belief that....

These are the good old days!

On my early CW learning path I would have loved to have a learning aid better than W1AW code practice sessions, which were the only game in town. Nobody I knew could afford an Instructograph.

 

The Only De-Coding Morse (CW)
machine is Organic.
Your Brain Cells .

Paul K8PG Morse Music

 

Small sample size delusion; where the sample set only contains a statistically small number of humans and types of CW signals.

The vast majority of humans can't decode Morse code at all (without additional training time, often many hours, days, and months worth). Many humans who can decode audio Morse code can only do so given a limited range of signaling rates and SNRs. Coherent synchronous encoder/decoder pairs can often copy signals at lower or higher speeds, or at lower SNRs than are audible to typical humans. The original Morse Code decoding machine (as Patented by S. Morse) involved a inked strip chart recorder, and was a mechanized and visual mode, not CW audio.

 

Good luck with your talk on this Dan! Looks interesting, Dave

 

Thanks for the great comment! I agree . . . I would have LOVED to have had something other than my ARRL material and a doorbell buzzer.

 

The big difference these days is the ubiquity of mobile devices, which outnumber PCs. A simple search of Apple's App store for Morse code apps listed over 100 apps in just the U.S App store, many free (and there are probably at least 3X more for Android, as well as many more in non-U.S. stores). Some good, maybe some not. But all of that is now available for helping learn Morse code in a vast number of different ways, and to the majority of the world's population, as more people now have mobile devices than radios.

I think these cool electronic devices in boxes are quite nice. One major advantage to decoder boxes that also contain a radio is that there is no need for an audio cable to eliminate acoustic multi-path echos, nulls, and fading. But apps can be an even cheaper way to give learning Morse code a try for the majority of the population who are already carrying a mobile device.

 

Dan
Went to the GITHUB site you posted. I would encourage others to do the same.

I was especially struck by your comment of:
"This design was started after a discuss on the Collapse OS mailing list about hardware is maintainable even without access to parts suppliers, for example Digikey/Mouser, etc. or even Amazon. Projects such as civboot seek to provide a means of resurrecting technology and infrastructure after a catastrophic event. This design is to provide a means of communication if other forms of communications become unavailable, especially in rural areas or urban environments with compromised infrastructure.:"

The DSB design will not win you any friends, but that is a side issue. Access to parts, even scrounging, will be difficult, at best.

I would love to see you build the unit solely from salvaged parts.

Also, now that you have released the design and software, how long will it be before we see a Chinese version show up?

Other questions -
Are you planning to sell a kit for this BitBanger? (Cool name BTW)

Are you planning on selling an encoder / decoder as a spin-off of the design? This might be of great interest to folks with existing gear....

Thank you for your effort, and thank you for the links to the other sites (civboot)

I'm looking forward to seeing your presentation!

 

Ron, I like the ones that CW Academy uses, like Morse Runner. Yes - helpful, for sure, and cheap! (Wish they had a Mac version -- I have to use some "in-between" software). But, that's receiving practice. The cool thing about the two devices mentioned here is the SENDING practice. It's like having a buddy there to monitor your learning on the key and point out then your R sounds more like an AE (the Morserino does this). At $100, I don't think Willi Kraml (the inventor) is going to get rich, and I also think that while it's a lot of money for some newcomers, it's also a device where you can see and feel every penny invested,) Dave

 

The DSB design is a compromise. At least the receive is DSB and the transmit is SSB (USB or LSB) so that the transmit bandwidth is not doubled. Unfortunately, there does not seem to be a SSB receiver architecture that both uses simple parts and is simple to build. A superhet receiver is going to require a crystal filter, and sourcing the crystals and aligning the filter is not going to be easy. A direct conversion receiver is possible, but it requires either two separate mixers, or a device like a FST3253 with all the switches present (and this is a specialized part that is being discontinued). Separate I and Q paths requires some calibration to compensate for gain and phase differences. If this is not done correctly, one might as well be using a DSB receiver.

I admit that building from salvaged parts will not be the easiest task, especially if all of the parts must be salvaged from existing equipment. At the very least, however, most of these parts are common jellybean hobbyists parts and would be much easier to obtain than the kinds of parts used to manufacture almost every other transceiver. I may try to show how this would be done but my first priority is getting the radio into as many hands as possible.

I am working with the Open Research Institute ( https://www.openresearch.institute ) to have a kit for this hopefully by next month. I fully expect there to be Chinese versions and I welcome this, as having more emergency radios available can only be better if they are needed. I am not looking to make money off of this, just to make radios available when they are needed.

One of the main purposes of the radio is to make the SCAMP digital mode available and encourage its widespread implementation. I don't have plans at present to make a standalone encoder/decoder because my current efforts are getting the transceiver released, but a standalone encoder/decoder would include SCAMP and therefore help encourage its adoption. A digital mode like SCAMP could communicate at lower SNR conditions than CW and not require an extensively trained operator which would be better for emergency communications.

73,
Dan
KW4TI

 

I'm going to disagree with the reviewer and say the morserino kit is easy to build! I have barely any soldering experience. I know which end to hold. So I went to jeffbezos.com and spent a bundle to soldering "stuff." I had three friends over, one had significant experience, one had some, and then, well, there was me. :- )

Two kit worked perfectly the first time. Mine and the medium experienced builder. (He had to unsolder something during the build and resolder something the next day.) The guy that didn't follow the instructions, well, his worked but he could't put the case on since he didn't read the instructio

We all thought this was an easy build. So... find your friend who can build, and have them WATCH you as they build it, with their soldering iron stuff of course. You can do this. We're all here for you!

 

A bit more about SCAMP
DCC2004 - SCAMP (Sound Card Amateur Message Protocol) (tapr.org)
The paper is undated.

Not to be confused with Scamp
Scamp is a self-contained Forth computer that you can use as the computing engine for your projects. It's easy to interface, and easy and quick to program using Forth, the world's best embedded programming language.

For some background - a 'historical' thread with chat about the (then very new) SCAMP.
SCAMP, Win Link,  and Black Helos? | QRZ Forums
(and yes, the usual snark and cat fighting are included)

 

I found this a while back, and it doesn't look like the original SCAMP proposal went forward and was eventually subsumed into WINMOR, which is now being phased out in favor of ARDOP.

73,
Dan
KW4TI

 

I would only note that PSK-31 is keyboard to keyboard and does well at low signal levels. These is no error correction and since the seeming demise of AMTOR (SITOR) there is a bit of a gap.

 

PSK-31 can indeed work quite well with low signal levels. The main disadvantage is that it is actually differential PSK. This causes problems when there are ionospheric conditions that cause Doppler shifts which modulates the phase. This can cause an "error floor" where the signal is still received, but the phase is somewhat scrambled and is not properly decoded. Most of the low bitrate modes intended for weak signal work (FT8/WSPR) use MFSK rather than PSK because of this. For a given bit rate, one can communicate at a lower SNR with more tones, though with diminishing returns as more tones are added, which is why most use 4 or 8 tones (as in FT4 or FT8). More tones, however, requires a more stable receiver LO, so something like a TCXO is necessary to achieve on the order of 0.5 ppm stability. If one doesn't want to require great frequency stability, or a more sophisticated demodulator for receiving (such as a FFT-based demodulator), then BFSK is a compromise which reduces the demands on the receiver and the signal processing hardware with perhaps a 2-3 dB increase in SNR for a given frame error rate.

The main difference between SCAMP and its predecessors is that it has been defined with different goals and priorities, namely to be usable with very simple cheap hardware so that a PC, soundcard, and stable SSB transceiver are not required for weak signal communication. It would have been possible to implement SCAMP even back in the late 70s or early 80s with 8-bit microprocessors, but I suppose weak signal propagation was not as popular as it is currently, with FT8 being the dominating digital mode.

73,
Dan
KW4TI

 

"This causes problems when there are ionospheric conditions that cause Doppler shifts which modulates the phase. This can cause an "error floor" where the signal is still received, but the phase is somewhat scrambled and is not properly decoded. "

Indeed, in Alaska, this is a constant issue - aroura 'wobble' is a feature, not a bug, eh?