Signalink USB and WINMOR as a mode

Started by KC9TNH, October 18, 2012, 09:50:27 am

Previous topic - Next topic


DISCLAIMER: Don't know enough about the modern digital modes to address one of this sub-forum's premises - discussion of standardization - but that topic has legs for future threads at some point. So this is just one more thing - maybe - in the tool bag.
Over the past week or so I've been dabbling in this. The RMSExpress s/w was loaded and an initial WINLINK account created by sending a test message to myself using Telnet (this one routed off to Perth & back), just wanted to validate the s/w load in advance of the Signalink arriving.

Quote from: WA4STO on October 03, 2012, 05:27:53 pm...I went with a Signalink USB interface and will never look back.
As street jargon seems to go:


Having already read through the manual before it arrived, the whole thing could've been done in about 20 minutes, uninterrupted. However, 2 factors added to that install time. First, given that things are part of a working radio room the cables needed to be routed behind/through etc. because when I got it done I didn't want to get behind anything, pulling stuff out to troubleshoot & tend to keep a neat table. Also I'm of the measure-twice-cut-once generation.

The other is that I also took a bit of time to reflect on what interacts with what, as I'm pretty draconian about not having RF in the shack. With outboard speaker going to be run from the audio out of the Signalink and a USB cable running from the laptop, thence in proximity to the amp, and on to the little box there were some ferrites to be applied. (As a friend told me there might be a 12-step program for those whose drug of choice is compressed iron powder.)

The Signalink itself was a joy to configure and install. EVERY direction (and some possible screw-ups) are in the manual & address what needs to be done. RTFM completely FIRST, then proceed IN ORDER. The manual is in real honest-to-God ENGLISH (angels sing) and written by people who've seen it all and write the online FAQs. The box itself is extremely well-made.

I'd originally wanted to route an email to a very distant relay as I have a piece of wire that plays better than its theory on 20m. However, I forgot that when configuring the propagation prediction software add-on that RMSExpress uses to give best candidates I'd plugged in some info about my multi-band OCF dipole. So I used a relay in Canada that's literally over the bridge from Detroit on 40m, and it worked great. I initially used the slower speed to give the benefit of doubt to the atmosphere gods (500 vs. 1600) but will try 1600 tonight maybe.

In the base shack I have a pretty robust manual tuner, not as fast an auto-tuner but it solves alot of problems. With the temporary brain-freeze of suggested list of relays (ranked on best path by the software), listening to new sounds, and watching a waterfall of activity on the bands competing for my attention (I didn't hear any pings from the Red October) something I'd done before paid dividends. When I'd first gotten that manual tuner I'd gone through and made a little cheat-sheet of the dial-settings for the CW and SSB mid-points for all the common bands. This allows tuning almost as fast as an auto. Comes in handy when your first freq of choice is in the tank or busy and you need to make a rudder correction. After awhile with this method and a known antenna some settings are muscle-memory.

Finally pressed the Start button and watched the simultaneous logging activity, the 2 stations linked up, and the traffic was being passed.
Total time using that slower rate for a simple few-lines ASCII test was a bit over 2 minutes. (Power was set to 50w by the way, certainly achievable with the 817 and its little THP kicker amp and QRP will be an interesting test.)

Exited out of that* & popped up the browser to check email and, voila'. Test in my commercial account.  In sum, ready to start some further testing, faster rate, more text, balancing the band conditions with who might be a better relay based on the antenna available.  1 successful test isn't a result. More to follow.

Edit: Back to the Signalink, you don't have to have that one but it really does simplify alot of stuff and the little CD-ette that comes with it also contains a bunch of commonly-used client software for various modes.  In fact, in reading a couple of the real good primers on using WINMOR that are available, one used a $5 outboard sound interface, and one provides the schematic for what the thing simply has to do for those who like to roll their own. For me time isn't necessarily money but, more importantly, time is simply time.

* The manual is serious about not having any other communication programs running. Unless you tell it to go out and check the bands for candidate relays LIVE, via RF, updating the list will be done via the internet. If you have your browser, email, whatever, up & running at that time you'll get an error because the commo software wants your TCP/IP port.  You can do without it - that's the whole point right? - but those are changes to be made BEFORE running your session. See above about RTFM.

Your internet starting point is:
From there, read, learn, search more & go forth.


Wes, great write-up.  Digital and HF-email are on my winter-spring list of things to add to my operating repertoire. 

It seems like the software uses the web for a few things.  How does it all work with no internet connection?  My main interest in HF email is for a reigonal SHTF situation where the area I'm in has no phone & web, but much of the rest of the country does.


Quote from: cockpitbob on October 18, 2012, 10:06:35 amIt seems like the software uses the web for a few things.  How does it all work with no internet connection?  My main interest in HF email is for a reigonal SHTF situation where the area I'm in has no phone & web, but much of the rest of the country does.
With my limited use of it to this point, the internet is convenient for a handful of things INITIALLY. From my perspective, here's what I used the net for:

1. The companion propagation prediction program (ITSHF, not to be confused with 'If The SHTF') is used by RMSExpress, the WINMOR client software, to produce for you an updated list of radio messaging relays and their ranked path reliability, frequency info and some other info. This is the source of the data that comes up when one presses "Select Channel" - so you can pick from one of the relays based on its perceived reliability given its freq, distance & azimuth so you can select proper antenna, etc.  The software DOES have a mechanism for updating this using RF, over the air, but presumably takes longer.

Besides, the propagation models are done on overall Space Wx prediction models using SMOOTHED Sunspot Numbers rather than the little daily gadget people watch where one minute the sky is falling & 10 min later the radio gods are smiling. The initial hand-hamming of these values for the coming year into the on-board prop software is time well spent.  Sources for where to get this are in the WINMOR primers. Winlink's website has links to some of them. This lets you...

2.  Use the internet, from within RMS Express (which uses the solar data you plugged in), to get the initial table of ranked best HF mail relays.  When you use this the top-ranked one isn't necessarily the best, you might have other considerations. So getting that INITIAL table planted into the software directory structure is a good thing.  Momentary glitches based on solar or geo-mag events are gonna happen no matter what. But you'll have something close when without the net.

3.  Before the Signalink arrived, I was able to validate the basic install & config of the s/w by doing a test message using a Telnet session to send / receive to a relay. Zip, took a few seconds. Done, life is good to that point. (Some things in the s/w won't come alive to setup until you actually have a device installed anyway.)

However, believe me, on the test regime in my head after checking for different throughput questions is to update that info mentioned in #2 simply via RF to benchmark what the payload is. After all, that's the intent and, absent SATCOM, how would a vessel do that? I would imagine that it will take longer, but I want to see it done. All the above is what I'd consider basic setup; the meat is gonna be how it performs without a TCP/IP connection at ANY point.

Thanks for asking, great question.


Quote from: KC9TNH on October 18, 2012, 09:50:27 am

From there, read, learn, search more & go forth.

If there's but one more thing I wanna do with ham radio before I become one with the great antenna farm of the cosmos, it's get going with Winmor.

The whole idea of a network that is designed to scan all HF bands, just looking for your puny little prepper signal to connect with, on a 24/7 basis, is too much to even contemplate.  It's just fabulous. 

But it gets better. 

Unlike voice, CW and some of the digital modes, with the Winlink system, your data either gets to its ultimate destination devoid of errors or it doesn't get there at all.  Still, not to worry, the system will keep scanning and keep trying, long after you're cuttin' firewood for your BOL in your dreams.

But it gets better.

Not sure what band will best propagate your signal?  Well, quit wondering.  The software tells you which station is most likely to work for you at any given time.  Sheesh.  They thought of everything.

73 de WA4STO


October 18, 2012, 11:22:35 am #4 Last Edit: October 18, 2012, 03:16:39 pm by KC9TNH
Quote from: WA4STO on October 18, 2012, 10:40:44 amBut it gets better.

Not sure what band will best propagate your signal?  Well, quit wondering.  The software tells you which station is most likely to work for you at any given time.  Sheesh.  They thought of everything.

73 de WA4STO[/color]
I'll give an example of how the s/w helps out with that table. Right off the bat, several of the best candidates were very distant stations MMMMM  :) on 30m. Now you can set the threshold of what the s/w uses to rank Green (best), Yellow, Red. In this case it comes as a ranking of 75 or better means a Green path; it shows the raw ranking too. But as it turns out, after actually tuning up at that moment & looking at the receive levels in the waterfall and the little gauge that functions like a modern S-meter, those stations on 30m weren't gonna work out. Either the prop wasn't right, or they were ranked high and, as a result were damn popular ergo busy.

OK, well, 73 might be yellow, but it's hand-grenade range of 75, so a 40m station was chosen. Bang, done deal.

To segue away from the zombie apocalypse for a moment, remember that here we're talking about a protocol where you can watch live (in your logger there on the screen) the attempt at connection, the distant station's response, the hand-shake, and all the:

Got your little bunch of data packets, looks good, gimme some more.
Here's some more packets.
Got your little bunch of data packets, looks good, gimme some more.
That's all I have for you; do you have anything for me?*
tks de kc9tnh <connection terminated>
... or machine words to that effect.  This isn't your DS3 circuit down the street.

There's one other use for also having an internet connection brought up by one of the primer authors whose focus is MARS.  If you are in support of an EMCOMM situation, where there may be loss of net IN THE AFFECTED AREA, you need to use your precious RF bandwidth to support information going to/from affected area. Alot of EMCOMM SOPs say "All traffic must go via RF."

Horsehockey. Routine comms between other supporting stations, cells, zombie-fighting bands, that can go via a less demanding method should go that way.  So in an EMCOMM situation from outside an affected area where you have both, use them. That's my story & I'm stickin' to it.

* When you make a connection to an RMS HF gateway your connection also picks up any mail waiting for you.


Quote from: WA4STO on October 18, 2012, 10:40:44 amBut it gets better. 
A properly implemented piece of wire, in the air, in the clear, and your laptop and your (pick interface device) and your 10w transceiver, and you're good to go. Great for field environments and much of the premise of this forum.

For those of you uber-geek members of the Auxiliary, or who will work an EMCOMM event and/or sheltering-in-place (and you're a map junkie like me), you can download the list of the RMS relays for WINMOR to a .KML file. Pull that into your GoogleEarth and it will overlay all those locations into your view of the Big Blue Marble, whatever that is. Becomes a switchably viewable layer.

Just make sure when you go HERE that you select the radio-button for WINMOR, then after it sorts you can select the KML file.


When changing something in your station, be sure to take all the care you can to rule out the new guy on the block causing you any issues.

Thought I'd post this reminder because, fixed or otherwise, one thing the little Signalink USB (SL) box is going to do is aggregate a bunch of little connections on its backside in close proximity to each other.

The base station speaker has opportunity for a line out but I elected to take receive for the SL straight out of the radio as designed. It's the first honest version of what's coming in, and switching a speaker in/out can change the loading the SL sees. But not to worry, there is a separate audio out jack on the SL to feed a speaker for the shack, present whether the SL is getting power from the computer or not. So...

Coming into the back of the SL, in close proximity, are the USB connection to/from the laptop (which carries POWER for the SL as well), speaker feed from the radio, and speaker lead feeding the shack speaker as well. The RF monster appeared because all of that was coincidentally near a run that takes care of the mic hand-switch for a headset. A real witch's brew, but easily rectified.

I generally route things in a couple of distinct directions to/from the radio depending upon whether they are signal-related, or power-related, trying to keep them from coming near except when it can't be avoided. So what I did was properly dress the wires away from the SL coming away from the box as soon as possible. The other thing done was to use some small clamp-on ferrites at each end of all cables, including both ends of the USB SL<--->laptop connection. Outbound audio was easily returned to normal.

Just a lesson-learned for when adding a little box with wires near each other, especially those carrying multiple signals. As an old instructor mentioned once, "if you've pulled the throttle back & the airplane starts to fall out of the sky, let's think about reversing the last thing you did, hmm?"

Having a combo pack handy of various decent-mix clamp-on ferrites in your kit can help make your day. Not an endorsement, but one source is DX-Engineering's PN: DXE-CSB-COMBO. You can buy packs of all smaller sizes too, although the big ones that come along are useful for things like appliance power cords or really larged fixed-base feedlines.


Did some initial testing. As the primers indicated (linked or downloadable from the site at just because the s/w ranks stations at the top doesn't mean one a little further down the list won't work better (within reason) since you know best (or should) how your antenna plays and the different constraints on a given band based on time of day, is the corn dryer going down the road at the Co-op, etc.

Have made a few tests forwarding test msgs to my com'l email account, as well as creating some messages from that account for pick-up when connecting. The shortest an overall session has been is about 1.8 minutes, for a simple 1-2 line test.  But only 2.2 min for a full 1/2-page of text from the USMC Antenna Handbook, WELL in excess of the ARRL "gram" limit and so I can see why this mode is appearing in alot of EMCOMM planners' ARES/RACES SOPs. That manual extract wasn't typed in, but was dumped into Notepad and sent as an attachment.

The longest session has been 8.8 minutes. During that one I was doing 2 things, only one of which was intended. I first wanted to see if I could pass a .JPG graphics file, so in honor of RadioRay I dumped a 43k thumb-nail of his old outfit's shoulder patch into the message. It went, and then also a SERVICE message came back into the Inbox saying the RMS relay was unable to deliver. However, I checked my private email a moment later and it was indeed there, so not sure that the SERVICE message was about. Maybe the final ACK dropped into the bit bucket.

By the way, as with many stations one gets familiar with, certain stations just seem to be always there, solid regardless of band conditions. I'm gonna continue to try to pick some others, but there are a few on 40m and one on 20m that aren't always busy, but seem to have pretty good stuff at the other end. Also, the information that indicates whether a particular relay is up 24/7 is also in the ranking.

All in all, this has some pretty nifty potential; within RMS-Express it functions pretty much like most familiar email clients - except for the terminal node controller window and waterfall (spectator privileges to the band activity, literally) which popup when you start an actual session. It will be interesting to see how throughput fares in a Peer-to-Peer session, literally radio to radio, with another user of WINMOR. It's feasibility there will make it a mission enhancer, rather than just something interesting. I'll have to see about that. All of this has been done on 50w.

LL:  When you feel you've done your best efforts and no station is answering, no matter how "idle" they are or their receive level, make sure you've actually reached over and pushed the ON button on the Signalink.


WINMOR is absolutely the best PC emcomm software going.  REAL internet e-mail over radio, and ham-2-ham radio e-mail. . . I am very impressed by it.  On the sailboat I used PACTOR III for years and from my little cabin in Idaho before that. The PIII modems cost as much as a new rig or more, but the WINMOR is FREEware.  If a station can work PSK31, then they have the hardware to work WINMOR.

Well - time for Zzz!

de RadioRay ..._ ._
Vigilant Always
"When we cannot do the good we would, we must be ready to do the good we can."  ~ Matthew Henry


Update: As a sometime-NCS on a voice net, and gnarling through the AM's kerfuffle while the rest of the world (seemingly) playing SSB contest, playing a bit with CW and digital. Some update notes, no particular order:

  • WINMOR: really an advantage to KNOW strengths & weaknesses of your antenna and especially if you run more than one. Just plugging in an updated SFI value when in session mode will trigger re-eval of the channel propagation estimates. HOWEVER, this is also based on whatever's in the software considers your antenna properties. Most leave that stuff at default, especially if running more than one. So...
    IF you really know what your antenna does/doesn't do you are much better able to evaluate the "ranked" list of channels produced for you. The list this afternoon produced the usual suspects floated to the top, nearly all on 40m but I wanted to play further away. The 3 40m choices highest in terms of likelihood of success were no-joy.  But there was a juicy 30m channel ranked way down but, given bearing/distance, is right in the wheel-house of my OCFD, so I gave it a try. BANG! success, 2 sec after hitting the start button. And 30m is notoriously less busy than the "usual" bands. Livin' large.

  • Am enjoying the companion fldigi software, and learning better how to read a waterfall. It is also kinda fun to watch how it reacts, even when its properties are tweaked for wpm, how it copies at least the gist of a CW connection. No question someone who can send quality code but still has a copying problem could be helped by this. There is a slight delay in the character appearing and, with the code at your favorite pitch in your ears, I find myself eventually getting the ear back to doing what it should.

  • Another thing you can do with your sound interface and fldigi is practice your own keying and see how well it is copied by the software. I've listened to some really pretty code (done by humans) but always remember that fldigi is a machine looking for perfect intervals. The only xmitting station that got close was W1AW with its evening lessons on 'nn47.50'. Listening to pretty human code, I see that nearly everyone puts a bit more time between characters, even at a given character rate. I don't care, I'm not counting dit intervals, and I like well done HUMAN code. So fldigi is just a tool-only in that regard.


Ray, if you could have only one band on a boat, which one would use take?



One band?  Offshore 20m   Coastal 40m.  I can find more local'ish information and etc. on 40m, but may not have the reach is seriously off-shore, except at night.  My #1 frequency is always 14300USB MMSN.

"When we cannot do the good we would, we must be ready to do the good we can."  ~ Matthew Henry


Finally!  After six months -- maybe more -- of piddling with Winmor (I'm a slow learner...), I got it to establish error-free contact AND pass traffic, with one station.

As it happens, I had gotten into a long-winded Olivia QSO with a fellow that I used to work with on NTS matters.  He was in Georgia.  It was interesting to note that his signal pretty much died off on 20 meters, but Olivia kept chugging, even with no real error-correction involved.  I may have to modify my mindset about FEC.

Anyway, he and I got yappin' about Winmor and I was whining about how I couldn't get it to work.  He suggested that I craft a message, addressed to him and have it in my outbox.  I was to make sure that I marked the message "peer to peer" as that was the key to getting stuff to transfer.

I had one serious problem, and that was that the port to or from the Signalink USB wouldn't initialize.  A reboot fixed that, but I'm suspicious of FLDIGI and other Windows sounds causing difficulties down the road.

He connected successfully, the familiar sound of ARQ occurred and the message automatically started transferring.  I'm hooked.

SO many things to experiment with now!  I'm particularly interested in seeing if other preppers will sign on to this idea AND if the auto band scanning will work.

I would like to envision:

BOL A only has a ten meter dipole.  And it needs to communicate on a reliable basis with BOL B which is 15 miles away.

BOL A has a fabulous path to BOL C on ten meters.    BOL B and BOL C both fully implement the band-scanning features of WL2K so the A > C > B path should work nicely.

At present, I'm noticing that signals are lousy and that I haven't yet discovered any one station that is consistently strong here in Nebraska.  Gotta work on that.

73 de WA4STO


Hey guys -- can one of you hold my virtual hand so as to get it thru my thick noggin how to use telnet from the RMS Express GUI to get the latest stations list?

I can't spot the appropriate menus to get there from here.

Tks es 73 de WA4STO


November 18, 2012, 06:22:38 pm #14 Last Edit: November 18, 2012, 09:22:57 pm by KC9TNH
Quote from: WA4STO on November 18, 2012, 12:36:51 pm
Hey guys -- can one of you hold my virtual hand so as to get it thru my thick noggin how to use telnet from the RMS Express GUI to get the latest stations list?

I can't spot the appropriate menus to get there from here.

Tks es 73 de WA4STO
Assuming you have an internet connection present once you're in the basic screen just click on Start Session (where the mode says WINMOR).
Once the next pair of windows appears on the session window click on Select Channel.
NOTE:  If you would like to update the actual SFI number at this time from some live propagation info you have you can modify that first.
Once you click on Select Channel, it will start building in the list of stations, their rating, bearing, distance, etc.

I'll go back later & read your other post above; I'm smoked, just got back from getting the 8-pointer home, bless his freezer-fillin' heart.

My feeling about WINMOR is to test it regularly, both via Telnet which I'd use if available for routine stuff, and via HF if an emergency or other situation. Regardless, I still test both periodically.

Assuming you've already done the initial stuff & have your WINLINK acct established, try going out to your com'l email and send yourself a message so you can test both the send & receive aspect. Don't forget to put //W2LK as the first part of the subject line.
Good luck!!!

* (arggh, that's what I get for posting when tired)