[TenTec] FSK, PSK, let's call the whole thing off...

[Adam Wade]

----- Original Message ----
> From: Bob, K4TAX

> Yes, true, but what about the other 20 or so current digital modes that are 
> available these days?  These are efficient with highly reliable data 
> throughput modes as opposed to the slow antiquated RTTY mode.  Remember 
> RTTY is the ONLY one that uses the FSK method of transmission.

This is not completely accurate, at least technically.  While many rigs only allow binary FSK, all the AFSK modes can be modulated by non-baseband FSK means,  This means Olivia, Throb, MFSK, etc.  It would not be wonderfully easy to build a rig that would modulate all of these modes with data in and data out and no baseband, as most folks find it a lot easier and more convenient to simply switch to a sideband with a voice bandwidth and "modulate the baseband" (i.e., turn the binary data into audio tones, then let the rig modulate the audio as RF).  The added advantage, as noted, is that you can let software adjust the zero point as long as the whole signal you're trying to receive is within the audio passband of your receiver (and all of that audio is going to your computer for decoding -- not really an issue for anyone who has a computer that was made in the past 20-25 years or so).

The reality is that, as noted, it would require slightly different hardware for each FSK mode to allow direct modulation/demodulation in your rig from digital data (binary, ASCII, what have you).  The QRP portable standalone PSK box I've seen for sale does this with phase shift rather than frequency shift, but the idea is the same.  And unlike a computer with a sound card and an SSB rig, you would need to have the hardware set up specifically for the mode(s) you wanted to work, rather than just cutting yourself a wide slice of passband and letting software handle it all.

I can't really see a great advantage to using FSK for hams these days, unless you have a really lousy sound card or a very, very slow old computer.  While a hardware-based FSK should have some advantage in terms of SNR for weak signal stuff, most ham receivers can't do any better than a good sound card/chip and the computing power in your computer.  Remember, a typical 10 year old computer has about 10-20x the computing power available for DSP or what have you than the latest, greatest $10k contest radio.  IT's what is DONE with that power that makes all the difference, IME.

Interestingly, current SDRs take advantage of the significant improvement in IMx (most of us only think about IM3, but all harmonics can be important in locking out strong neabry signals -- as anyone who has tried to work low-power PSK on 20m while an RTTY contest is going on could testify, if they did the math!) that can be had with direct conversion to baseband using quadrature outputs -- the technology was developed to the current state of the art (and thus made available for incorporation into modern SDRs) for digital cellphone use, where multipath and especially very strong signals not far up and down the band can essentially ruin two-way comms.  Cellphones can fend off nearby signals into the kilowatt range while getting solid, reliable comms with only milliwatts, and most of the noise sources are within several miles of the base and mobile cellphone units.  I am very interested to see if ham radio SDRs will eventually take more advantage of this
 with direct hardware encoding and decoding for digital modes (old-school FSK would be one example of this) instead of simply converting to an audio baseband and handing it off to other software for decoding and encoding, like the software we currently use for digi-modes that transmits/receives via audio into/out of an SSB rig.  There should be significant improvements in ability to copy weak signals, get reliable sigs out at QRP power levels, and a vastly improved ability to fend off strong nearby sigs (nearby in frequency, that is) well beyond concerns about IM3 (how about being able to fend off IM11 or better from more than a kilowatt near your receiver?)...  The DSP power to handle that on a hardware level is available, and would not really cost a lot more to implement in these radios; further, you would no longer need a computer to connect to your SDR.  Whatever computing functions were necessary would be built into the radio, making the unit very
 portable.  Using electronics made specifically for low power consumption would mean portable digi-modes could become quite self-contained in a single unit with relatively small, lightweight batteries -- the military already has this technology and is using it for, among other things, controlling their UAVs and relaying data from UAVs and sats to the field, and it should be possible to get it affordably into radios soon.

As another aside, I have sourced a 192 khz sound card with balanced inputs (an E-Mu 0404 PCI) for under $50 shipped off of eBay, and combined with the $25 Clifton Labs Z10000 buffer amplifier and the $12 SoftRock II Lite IF "Special" quadrature decoder/receiver, I have made a complete SDR receiver that interacts with my Orion 565 directly and provides me with over 96 khz of real-time spectrum analysis and/or waterfall with decoding of digital modes as well as direct click-tuning of the Orion from what I see on the screen from the SoftRock.  Total cost for all the parts (which, granted, must be assembled by hand, and they are SMDs which are a bit of a pain to work with, and of course there was some minor surgery required on the first IF board of the Orion) including isolation xfmrs that can handle 100 khz bandwidth and a shielded case to keep the noise floor low and ground loops out, was under $100.  Yes, that's right...  I have a SDR receiver that is as
 good as pretty much anything you can get from Flex Radio (but is receive-only, at least for now) for $100, using the VFOs and first IF of the Orion as a front end.  If anyone here contests using digital modes, this is a truly amazing set-up to have, and it can be used just like the Flex Radios for CW contesting as well, using CW Skimmer and so on.  With a fast computer or two controlling the Orion via Ham Radio Deluxe along with running CW Skimmer and the logging software of your choice, you can do almost everything a $5k Flex Radio can with your existing Orion or Orion II.  Other radios that have serial or CAT control can have most or all of this capability as well, although I haven't built anything like this for any radios other than the 565 Orion so far.

Neat stuff, and most of it really doesn't use any concepts that experienced, knowledgable hams probably don't already know, they are just arranged in new ways that at first are non-intuitive.  And just like with the advent of solid-state, we get to play around and see what advantages the old stuff has (traditional superhet receivers with roofing filters vs. direct to baseband quadrature receivers) when compared against the "latest and greatest" (which with solid state was tubes vs. transistors -- and tubes have some very handy advantages if you care to use them!).  It's not really that complicated or difficult to learn, it's just a little different...  And aren't hams above all else adaptable and knowledgable about basic principles, or at least isn't that what ham radio was supposed to have been about, once upon a time?

> All of the 
> digital modes are available via current free software and no interface is 
> needed.

It can be hazardous to your equipment, and can introduce a heck of a lot of nasty noise and potential problems, to NOT isolate your computer from your rig, especially at audio frequencies.  I STRONGLY recommend that at the very least we isolate the audio path from rig to computer with quality 600 ohm 1:1 xfmrs.  When dealing with SDRs like the SoftRock, where the sound card is being asked to work with sigs well beyond the hearable range of audio frequencies, finding appropriate xfmrs that are flat across the required bandwidth becomes a much larger challenge.

Also, many radios do not have a serial port or USB port interface, and those ones will require an interface box of some kind to allow rig control.  While not specifically required for working digi-modes (I have successfully worked a lot of digi-mode contacts using a Kenwood TS-520S, tube finals and all), if your rig CAN be controlled via computer interface it can work greatly to your advantage to have the rig connected to the PC.  For instance, the Orion can have nearly all of its adjustable features controlled through software off a computer, in mny cases to a far higher level of precision than the faceplate controls allow (how about not having logarithmic jumps in the microvolt threshold for programmable AGC?  With Ham Radio Deluxe, I can control the threshold in 1 uv increments, which can be very handy!), and rather than having to manually key the rig for xmit/rcv with digi-modes, or rely on VOX, the rig can be keyed by software via the serial
 interface.

> My motto:  more boxes + more cables + more connectors = more 
> problems.

That's true to an extent, but if everyone adhered to that principle to the extreme, we'd each have a single transmitter, single receiver, single antenna, single feedline, and a pad of paper for a log.  In reality, the more reliable you make an automated or even just fairly complex piece of hardware, in any field, the longer you can go without a problem, but the more difficult or costly it will likely be to diagnose and repair the problem.  This is why we end up with "appliance operators" today -- it's easier to throw away your broken VHF handheld and get a new one if it goes wrong, and it's cheaper, too.

> Of course what do I know, I still run Class C high level modulated AM using 
> a 1956 transmitter and not a 2000 technology transmitter.  I got rid of my 
> Model 25 RTTY floor shaker at least 40 years ago.  Haven't been back on FSK 
> RTTY since.  And oh by the way, my AM transmitter doesn't use any more 
> sideband bandwidth than current technology SSB rigs and in many cases it 
> uses less.

Well, most "modern" rigs default to 2.7 khz b/w for xmit, and some to 2.4 khz.  Are you saying you are really running an audio passband of only 1.35 khz?  I doubt I would be able to copy you if your xmtr was not producing more than -6 db of your peak above 1.35 khz audio...

Horses for courses and all that, and I would assume that you probably know a lot more about a lot of things radio-related than I do...  And certainly I (and perhaps most of the list) have a lot to learn from you that can be of great use to us.  I find most of what I learn that is most useful was known in radio circles over 50 years ago, and a chunk of it over 75 years ago.  Of course, a lot of those things can be applied to the new and latest, but HOW is not always readily apparent.  Me, I like to learn PRINCIPLES so I can apply them as needed, and I love anything that can teach me where principles apply in whatever I am facing.  I continue to experiment, with the new AND the old (I'm as interested in tube PA design and engineering as I am in quadrature-type direct-to-baseband receivers), with antennas and circuits and software all, because I want to understand.  I appreciate all contributions to that, and I also appreciate that not everyone wants to
 learn on that level -- I even appreciate that there are times when I want to turn on my rig and have a good rag-chew without having to think, know or understand how it all works.  But it's awfully nice to know that I have the knowledge and the tools to at least diagnose and often correct problems as they crop up, and to design and build some of my own stuff, for fun as well as to save money and just plain be able to say "I made this, and not from a kit".  And building kits has been a huge help in not only assembly technique and general good practices for RF circuit design, but also to help learn some basic concepts through physical motions rather than just reading and thinking.

YMMV, and for many it does, I think.  But that's my $0.02 worth, and you can keep the change.  :D
 
! Adam Wade -==- AF6ME -==- Ten-Tec Orion & SteppIR BiggIR Mk III !
! "I write down everything I want to remember. That way, instead  !
! of spending a lot of time trying to remember what it is I wrote !
! down, I spend z time looking for the paper I wrote it down on." !
!       - Beryl Pfizer    -=-    http://www.allthings550.com      !