The list of PSK31 capable software shown below (in no particular order) is by no means a full representation of the available crop.  It's just what I've used, and have demonstrated to myself to "get the job done".

   You might be tempted to conclude from the recent advertisements by the Big Three (Kenwood, Yaesu and Icom) that the latest DSP filtration schemes are ideal for PSK31, but that has NOT yet been borne out by my observations. To understand what all the fuss is about, click here

First of all I'd like to say "Thank you" to those hams, manufacturers and others who have taken the time to re-think the design and use of ham gear in the context of PSK31. I am pleased to report that there has been some improvement in areas that I have previously been critical.  Here are a few examples of new (and some old) ways that you have made things better:

1. I have received various emails from hams who have found that by "lying" to their radio about the installed filters, have been able to use a narrow (cw) filter in the SSB modes.  If you've been frustrated, this is a very good thing to try.  Good examples include the Icom 706, 756 (not Pro or ProII) and any other rig where you can tell it via a menu what filter is what.

2. I have heard from folks who say that TenTec rigs, such as the Omni 6 Plus, Jupiter, etc. allow PBT and narrow filters in SSB mode, so they obviously have their act together. From what I've seen, it looks like the really cool K2 (Elecraft) also has this ability.

3. Users of older rigs, such as the Kenwood 440, 450, and basically any radio that has IF-shift (or the equivalent) in combination with selectable filters for SSB mode have already found that they are super for PSK31.

So...in a nutshell, there are several ways of getting proper receive filtration for the rapidly expanding PSK31 mode.  As the bands get more crowded, and the opportunity for QRMing is greater, you're going to need every advantage these select rigs can provide! If you don't understand what I'm referring to, please be sure to read the ramblings provided below.

A hopeful note: I had been reporting (as late as December 8 of 2001) that the new Icom Pro II was not well suited for the needs of the PSK operator, and that I had difficulty getting anyone at Icom to listen to my input. Thankfully, I was subsequently contacted by a senior individual of Icom America, and after hearing (and reading) my ramblings, assured me that Icom corporate was very aware of the issues I have been presenting.  Now, this is no guarantee that they will act on this, but at least I've been heard. Stay tuned and check back here for further developments.

"Old rigs for new..."

(2/16/2001) I recently visited the local HRO (Ham Radio Outlet) to have a close-in look at how well the most recent HF rigs would do for PSK31.  I went with a VERY specific goal in mind, to see whether the rigs could help in a situation that I'm seeing more and more frequently.

Here's the scenario: you've "parked" your rig at 14.070 (as we often do), with the intention of doing some broad-band reception. You begin a QSO at an audio tone of 2000 Hz. While you're operating a very strong Pactor signal appears at an audio tone of 1500 Hz, 500 Hz wide, causing your AGC to kick in big time, and the weak station that you've been working disappears. OK, now what...

Ideally, one would like to be able to shape the IF bandpass to somehow eliminate the interfering station, and that's what I went to HRO to see how to do. The results were quite surprising, not very good I'm afraid. (Yes, I looked long and hard at the "big guns": the Icom 756 Pro and the Yaesu MK5, as well as several other very new offerings, such as the Kenwood TS2000.)

1. The High Cut/Low Cut fallacy
The first thing I tried (naturally) was to use the low cut knob that's present on virtually all of the new rigs.  My thought was to raise the low cut threshold to be above the interfering station at 1500 hz. Imagine my surprise when I discovered that the low cut could not be set to be any higher than mid-IF, approximately 1200 hz. That didn't help.  This is true for ALL of the rigs that have this feature, which (sadly) includes all of the new high-priced radios. The high-cut stopped mid-range too, and thus I'd have the same problem trying to fix a problem that was below the center of the IF.

2. Filter, what filter?
Next I went to see if maybe I could use a narrow filter, and move its center up to the 2000 hz location to get rid of anything that's on both sides of the QSO. Once again, I ran into a couple of brick walls.  First of all, many of the newer radios do not permit you to use the narrow filters with the SSB modes.  Second, even those that do often either don't have any facility to do the IF shift, or limit the shift excursion so you can't put the center on that high (or low) station.

3. Notch filter? Fuhgettaboutit!
So, I thought maybe I could use the notch filter to notch out the interfering station. Well, that too was a bust, since I couldn't see any way to vary the width of the notch, and they were all too narrow to have much effect on the 500 hz wide Pactor station.

4. Yeah, but they say you can set the "width" and "center"...
Nice try.  Too bad that you can't set the width and center in SSB mode (such as in the Kenwood 2000), it's only for CW. You're forced into the High Cut/Low Cut mode in SSB mode, see above.

There's some hope...
Believe it or not, I did find ONE newer radio that appeared to be just the ticket. The funny part of it is that it's also one of the least expensive radios on the market.  It's the ICOM 718, into which you can install a nice narrow 250 hz filter. Upon reading the manual, it seems you are permitted (with a single one-time menu setting) to use this filter in the SSB mode. In addition, the radio has a very nice "old-fashioned" IF-Shift control.  "Voila!"  Now, you can select the narrow filter and slide it to whereever you need to for your QSO.  I plan to bring one of these radios home soon to confirm that it actually is as effective as it promises to be.

Users of other modern radios, don't give up...you may find that by LYING to your radio, you can fake it into using a narrow CW filter in SSB mode. More about this here...

Why is this all true?
From my cursory examination, it seems that the HF rig manufacturers are still designing radios that pay little if any serious attention to PSK31 and other broad-banded AFSK-style modes.  They assume (wrongly) that any operating we want to do is CENTERED IN THE PASSBAND, which is clearly not the case for PSK31.  In fact it's just the opposite; we park the rig and tune with the PC. Sadly, it's probably about two lines of code in the new radio's firmware that stops the Low and High cut limits at the center point.  If they just opened it up to allow the low-cut to be moved up to the high end, and the high-cut similarly to be moved down to the low end, the problem would be solved. I just hope they're listening...

Don't throw out those old radios!!!
Lastly, don't assume that since this is a new operating mode that your old radio just won't cut it. The irony to all of this is that your old Kenwood 440, 450, etc. may turn out to be the best PSK31 rig you can find. They have IF shift, and you can install a nice narrow filter, a winning combination.  I personally use an OLD Drake TR-7 with an external synthesized VFO (for stability) and it's the best PSK31 rig I've ever had. So, why is it so good? Imagine a weak PSK in the middle of the strong boys.  Click in a narrow filter, say 300hz, and use the IF shift to isolate the desired signal.  The effect is truly spectacular.  The station comes up dramatically, the AGC relaxes, and copy improves; it's as if the nearbly stations just disappear. You've got to see it on the waterfall to appreciate how cool it looks and works.

(March, 2001) I received this interesting email from Steve Jackson, KZ1X/4, in response to the above commentary:

"The rig that bypasses all these problems is the TenTec Omni VI+. I have one, with all the factory filters. You can use any filter you want to, at any time. No Japanese microprocessor trying to make the radio idiot proof!  Just push the button to select one or more real, multipole, hi-Q filters in each IF.   (InRad has filters that are supposedly better.  Haven't tried 'em.) The PBT then puts the filter 'center' wherever in the passband you want it, after you've selected the filter; or, you can bypass all the filtering and just use the PBT.

All audio and control I/O are on the back panel, on individual RCA jacks.  Don't need to interface with the mic plug.

Full duty cycle at 100% power, no fan, and at any SWR and phase angle. No auto power cutback. Manual notch is narrow and deep enough to allow me to have a PSK QSO 1 kHz away from an automated PACTOR station, even without the filter! RX specs for close-in IMD (less than standard 20 kHz), etc, are second only to the Elecraft K2, according to the ARRL lab.

Yes, it's a weird duck, and lots of people just don't "get" TenTec.  I'm no cultist. I have JA rigs, too. But when I want to get serious, the power switch on the Omni is what's on."

Thanks Steve!

(March, 2002) I got this nice message from Wiley Sanders,KF6IIU:

"I'm succesfully using a Yaesu FT-101 with PSK31. There are three "unique" features using this rig in this mode:
- The rig has separate tuned circuits for generating the L.O. transmit and receive frequencies.
- The SSB bandwidth is fixed at 2.5 kHz.
- The AGC cannot be disabled and has a very slow time constant.
- The PTT current is 110ma at 13.5V, beyond most interface's current-sinking capabilities.

The workaround for the TX/RX offset is to call CQ and then tune in a replying station with the Clarifier (RIT). You can get within a couple dozen Hz this way. The FT-101 is still not that stable, though, so you have to lock the transmitting freq or you will drift around. (This is easy in Digipan, don't know about other software.) Plus, you must depend on the patience of the other op! Fortunately most ops are willing to tune a few Hz to accomodate my not-quite-steady signal.

For bandwidth/ACG issues, there is sometimes no alternative but to put the audio frequency at 900 Hz in the middle of the FT-101's CW passband and manually switch back and forth between receive-CW and transmit-USB. When I forget to switch, that explains why sometimes I don't come back to you right away!

The PTT current workaround is to use the FT-101's VOX. It works fine for me. I used straight cables with ferrite chokes for several weeks until I decided to splurge and order a professionally made interface.

Other observations - a lot of ops are still overdriving their rigs. Probably 20% of the signals I "see" have a 2nd, 3rd, or more set of "railroad tracks". I doubt most people monitor their transmitted signal. If you can rig a 2nd receiver to hear your own signal, tune in and zero beat your transmitted signal. If you hear *any* clicks or bumps, your signal probably isn't clean (assuming you are not overloading the receiver too much.)

Last weekend, one op was cranking out a HORRIBLE 60-Hz-harmonic signal that splattered over 3 khz of the 10 meter band. Obviously he wasn't monitoring his signal. I am sure the cause of his problem was the same problem I had when I first connected my Tigertronics SL-1 interface - my laptop computer was not grounded and it introduced a horrible ground loop. It is a Compaq 1700XL and it was floating at 18VAC above ground! Laptop power supplies are TOTAL CRAP; beware. Grounding the computer to the rig solved the problem.

PSK31 rocks! See you on the air."

Thank YOU Wiley!

1. Your soundcard is also used by Windows and other applications. Here's the scenario: you're in the middle of transmitting, and your soundcard is generating the tones for your rig. Then, you click on a menu or some other application, and Windows decides that you need to hear a sound (a click, whir, buzz, tuba, whatever). Guess where that sound goes? Yep, it's "on the air". Just try to explain to your QSO buddy why his rig just suddenly growled like a lion, or said "You've got mail".  The best solution would be to use multiple cards, and dedicate one to your Windows sounds, and use the other for your radio. The current crop of software, though, makes this really hard, and I've even heard of one case where Windows itself makes it impossible.
2. Your soundcard uses real-time wave generation. Thus, it's dependent on the cpu to make it do its thing.  So, if you suddenly have a burst of disk activity, your network card comes alive with email, or any other processor intensive action takes place, guess what stops working momentarily? That's right, your sound card. It will show up as a click or other weird glitch in the audio stream, and, depending on how severe or frequent the processor overload is, it could happen with maddening regularity, and last for seconds.  That's a great QSO buster too.
3. Your soundcard was never designed for super-clean output. At this point I've tried about a half dozen soundcards, and they ALL show some kind of digital artifacts in the output, ranging from a high-frequency (from an audio point of view) ripple on the surface of the wave, to strange in-signal period variations.  Every card is different, and you'll never see information about this in the specs. There doesn't even seem to be much correlation between price and quality from this aspect.
4. Your soundcard is a toy. Please don't be offended, but really now. What did you spend for the soundcard? $20, $30 maybe? You really can't expect such an inexpensive device to give you good linearity, clean hum-free signals, rf-shielding, stable DC levels, freedom from transients, wide dynamic range, etc. It's amazing they work as well as they do, considering how really cost-reduced they are.
5. Your soundcard has the wrong levels. I've yet to see a soundcard that doesn't need external level-setting hardware. For input you can try the microphone input, but that's too sensitive. Then you try the line input, but your speaker's got to be at ear-splitting levels to get useful display and accurate copy.  I've also found that the "Volume" controls on the typical PC are not even used by most of the software that is available, for reasons I don't really understand. The same applies to output for your rig's mic input, you've almost certainly got to reduce the levels and isolate the lines.  The good news here is that there are some good products now that can help, like RigBlaster and others.  Well worth the money!
6. Your soundcard uses real-time detection. Thus, just like with sound output, it's dependent on the cpu to make it do its thing.  So, if you suddenly have a burst of cpu usage, your soundcard goes "deaf" for a moment. This shows up as lost characters and, depending on how severe or frequent the processor overload is, it could really mess up your incoming copy.

Weak Link #2: The PC itself

1. Your PC can't keep up. Sure, in general, your machine does fine for typing, paint programs, and lots of other non-time-critical things, but they were not made to take care of events that are "here and then gone".  This is particularly true at the application layer, where (unless you plan to totally hog the machine) you're at the mercy of the Windows task swapper. I don't care if you've got a SuperWhizBangPentium1000 machine, you'll run into this sooner or later.  The net effect is either lost characters, or glitches in the output stream. I even had a problem where the video driver itself was too slow (on a 400 Mhz K6 machine) and made a glitch in the audio output whenever the screen was refreshed.
2. Your PC wasn't made for controlling external devices So, you've got to play games with using the RTS/DTS lines in your RS232 port to simply key your rig.  What a pain.  Particularly if you, like I, have all these beautiful molded connectors everywhere.  You'll either have to cut into one, or make some kind of jury rigging to get to the lines.  Then, just to be safe, you better drive a relay or opto-isolator, or you're bound to have some nasty ground loops or rf feedback.
3. Your PC display runs out of room. Imagine this: you're running Logger, the volume control applet, and your Kachina all on the same screen. Problem is, you can't see all of them at the same time. So, you're switching back and forth, and give up in disgust. This is so typical of the PC it's laughable. And, if you're anything like me, you're no longer running your screen at 1600 X 1200, because, frankly, your eyes can't handle it..

Weak Link #3: Your rig

1. Wideband reception is a myth. There's some really great reasons to want to use PSK31 in a broad banded way.  The problem is that there are some even more compelling reasons NOT to.  Like when that KW packet stations fires up 1kc away from your QSO and makes your AGC go crazy. I defy you to copy clean PSK31 or anything else for that matter.  The problem is that the current design for radios don't let us effectively "notch" out one or more strong stations at the IF stage (most notch filters are MUCH too narrow), therefore the AGC kicks in.  So, we cut back the bandwidth, but it's getting to the point where even my 250 hertz crystal filter isn't narrow enough to get rid of the inter-modal interference.
    
2. Your rig wasn't designed to accommodate flexible digital modes. OK, you've got several crystal filters, but does your rig let you click them in when you're in SSB mode?  The answer for many users of older ICOM gear, and maybe others, is "NO". So you put up with the problems mentioned in #1 above, and try to find ways to work around the strong interfering stations. Or here's another scenario that trips up even the newest of radios, such as the Icom Pro and Pro II, and the Kenwood TS2000. You've got a desired PSK station at an audio frequency of 1700 hz, and an interfering station at 1300 hz that's 500 hz wide. OK, so you try to bring up the low-cut DSP filter, but it (incredibly) tops out at 1000 hz, leaving the interfering station intact and strong as ever. You're stuck, and the only way to get this guy is to turn the tuning knob on your rig (ackk!!!).
   
   One possible fix (believe it or not) is that several of the rigs that lack DSP are able to be lied-to, thus making them much more suitable for PSK31.  What you do is go into the appropriate menu, and tell them that that slick 300hz CW filter you installed is really a 1.9kHz SSB filter. THEN it will let you select that narrow filter in sideband mode, and with any luck you can slide the IF shift up or down to make it do its magic without having to touch the tuning knob. Rigs that I've successfully fooled this way include the Icom 706 (all current models), 746 (non-pro) and 756 (non-pro and non-pro-II). "What a world, what a world..."
    
3. Your rig doesn't really support AFSK input. Sure, you've been running packet for years, and that uses AFSK, so "where's the beef"?  The problem is now that you're using the rear connector on your rig to bring in the audio, and you don't have that handy PTT output from your TNC to key the rig, you're finding all kinds of difficulties. Things like that fact that your VOX doesn't detect rear panel audio, so you try to use PTT via software commands from the PC, but the rig doesn't mute the microphone (so you've got to unplug it). You then decide that's no good, so then you bring the audio into the mic connector, but then you find that you've got to set the level of your audio high enough to trip the VOX, but not so high that you drive your rig into the ALC area, then sadly you find that there's no overlap of the two ranges. These may not SOUND like big problems, but I've heard plenty of complaints from fellow hams about these very issues.  (A good solution to this, the one I use, is to get yourself a RigBlaster and call it a day.  By the way, I get NO commission, it's just a fine product.)