Saturday, October 31, 2009

Pass the COM ports please

I recently built a new computer for my ham station and installed plain vanilla Windows XP 32 bit O/S. If you're interested I have a blog entry on the new machine, and best as I can tell it's completely compatible with the F5K and PSDR, fast and stable. This computer has been running swimmingly without a hick-up 24/7 for a few weeks now. It is an architecture that has eliminated many of the old ports like the PS-2 ports for mouse and keyboard and it has NO com ports and NO parallel port. There is a header on the MB to add some kind of PS2 ports but there is no cut out on the case to install such.

No problem I just switched to a USB keyboard. I had one that I use to trouble shoot computers but its a bit to small for comfortable typing so I went in search of a USB keyboard that made me happy. Here is what I bought:

All of my laptops are Thinkpads and I love the keyboards and this thing is nice sized. This is just a Thinkpad keyboard that is put in a little plastic case complete with mouse and hooked up to a USB feed. For years I used an IBM server keyboard because it had the IBM feel but was small. I don't need number pad sized keyboards sucking up my desk space. This keyboard has a nice palm rest and and that little eraser mouse which is OK with me, I also have a regular mini AOPEN O-35M mouse plugged in another USB port which is what I mostly use. That mouse is on the smaller side so it also saves a few cm^2 of desk space. I can highly recommend the lenovo keyboard if you are in the market for a USB keyboard

Now to the issue of COM ports: I don't got any!!!

My winkeyer USB uses a usb port. I upgraded from the serial port version just for this reason. PSDR allows you to key the program from an external keyer using a serial port, and I have used that as my first choice for CW keying for many years now. I have a paddle plugged into the F5K directly, but I almost always use my favorite paddle and my winkeyer. I have a special little pushbutton block of 4 buttons I had Pierro Begali make for me years ago and I have that wired into the winkeyer so it makes for a nice easy memory keyer setup. Now my first glitch: How do I get a physical serial input so I can input the output from my keyer? I made a little adapter using a PNP transistor and a resistor on a 9 pin plug that plugs into a serial port that allows me to plug in my keyer and key PSDR, but I have always used it with a regular serial port. I had a USB to serial converter laying around so I installed that and plugged in my adapter and set up PSDR to read that port and voila' it works!!! So that part is done. Next I have a bunch of other programs that need virtual serial ports to interconnect. I use DXlab suite and I use Skimmer and I sometimes use N1MM and sometimes fool around with the dreaded "digital" programs, so I need at least 4 virtual COM ports available.

Up to now I have used VCOM for this duty, but VCOM is kind of buggy. There were two guys sitting on a porch rocking. Every so often a hound dog next to one guy would raise up and go OOOWWWWWWW. After a while the other guy said "why is that dog doing that?" The first guy said "because he's laying on a nail". The second said "why don't he move" and the first said "it don't hurt enough yet".... that was my relationship with VCOM. I would get BSOD with VCOM even with XP 32 bit, but it wasn't enough of a hassle to change. There are a couple of other alternatives but each had its show stopper for me so I just plodded along with VCOM. VCOM is no longer supported by its author, the formidable N8VB who has gone on Quicksilver SDR fame. As far as I know it only supports XP 32 bit so if you are more exotic than that as far as VCOM is concerned you're out of luck

I read on the Flex list that Steve Nance K5FR (also quite formidable) has created a virtual port program using the Eltima V.S.P.D. 6.9 serial driver. This is probably the best windows based virtual port driver out there and supports all versions of windows from 2000 onward. Suddenly the nail hurt wayyy to much. I downloaded the driver from this website. I had previously un-installed VCOM and rebooted. I installed this driver and rebooted just like it tells you in the instructions.

I then opened the configuration program and here is what I saw:

I proceeded to add some ports (since that is what this is all about) and low and behold:

I have serial ports out the ying yang. I always number my pairs such that 10 is added to the lowest number eg 6,16, 7,17 etc. This program recognized my naming pattern and all I had to do to add ports was push the friggin button. It was so easy and that damn nail is not hurting me anymore.

Steve of course is the author of DDUTIL a program without which my ham radio experience would be far less rich. If you're in need of ports get off the nail and give this program a spin. Slicker 'n mucopolysachrides (aka snot)



This morning I'm sitting at the radio. It's 8:30 and I've been here for about an hour or an hour and a half. I have the head phones on and I'm reading the mail on a couple of guys down on 40CW while I'm checking email and reading the Flex list, checking Drudge etc. I pretty much just get my news from Drudge anymore since the rest of the media is basically out to lunch and at least if the world blows up Drudge will put a couple of spinning bubble gum machines and give you the head line


Fortunately I did not read that headline rather it was about all the movie stars that are sucking up to Obama. Guess that means I'll live through the day. I glance over at the panadapter on PSDR and here is what I see

Pile ups are very distinctive in their appearance. You can see all the stations in the middle of the screen piled up on each other. When this started it was just three or four really loud stations calling the P29 all basically on the same frequency. I'm thinking what the heck is this???? It almost looked like some kind of weird new multi channel digital signal like Olivia or something because it was a bunch of signals in a tiny little bandwidth. It's 8:30, the DX is usually well down into the sun noise by now, the gray line has moved on to New Orleans and the best I'm gonna do as far as DX goes is call Chicago. Now I like Chicago. I'm from Chicago, but as far as DX goes Chicago ain't it.

I flip down to this "pile of signals" and try to find the DX and loud as heck is P29VLR working the pile. With Skimmer I quickly determine where he is listening. He is moving up the pile working the next station in line he can hear, and then moving on up the pile to the next station 100hz up or what ever. I check the bandswitch on my AL80B and its set to 40M. I flip on the amp. I use a AL80B on most of my contacts because I can just flip it on and be ready to transmit if the amp is all tuned up. I put myself in his line of movement and BANG he's in the log. I look over at the SpotCollector (my DX cluster program from DXlab) and he is just now showing up on the cluster and I already have him in the log. In a couple of minutes I see that station spotted on 80M but I am way too far past gray line for that QSO. The guys in Texas have a shot at that one, but FL is dead meat. I then see a DU9 and a KH0 spotted on 40, but by now its 9 am and the sun owns my receiver as far as propagation goes and its time to go call CQ Chicago. Pretty soon I won't be able to even hear Chicago. HMMM if I had a remote base Flex station in California I could still get in on the fun.....

I check for some dope on P29LVR. P29VLR is operational on New Guinea for only a few days from Oct 27-31. It is now their evening on the 31st, probably just a couple hours from pull the plug time. Were it not for the F5K and its panadapter I doubt I would have even noticed


Friday, October 23, 2009


Heard him peaking at -107dBm. He heard me not at all. This receiver is amazing!


Wednesday, October 21, 2009

Fiji at 4:30

I went to bed early last night since I was beat from a long day. That meant I awoke early. Usually I just try to roll over and get another hour or so but this morning I was awake so I got up stopped by the fridge to scarf a bite of chicken from some left over panang curry. I hit the mouse to wake up my computer, and when the screen lit up I was on 160, and there was 3D2KJ calling CQ.

He was about S4 (-100dBm) and my band noise this morning was a pleasant -115dBm, with the usual deep long cycling QSB of 160. I tuned up the AL80B on 160 and gave him a call. I have other more powerful amps in my shack, but 99% of my work is done with this AL80B I bought 20 years ago. There are 2 reasons. First it puts out a solid KW and second it's instant on and the tuning is absolutely reliable. I fitted a clear plastic page protector to the front panel and that has little arrows pasted on it that gives me a place to tune to for each band. I just set the dials, do a quick dip of the plate and check the grid current and I am ready to go. Ten seconds tops. With my other amps it takes 3 minutes to condition the cathode and all of that, then you gotta tune and then make sure you don't draw too much grid current bla bla bla. By the time I get the amp tuned the DX has left the building. In a pinch that extra few dB can make all the difference, but if I could have just one amp this little AL-80B is a near perfect ham radio amp.

I've been looking at the new Ameritron ALS-1300 to add to the line up. First, IT ISN'T TEN FREAKIN' THOUSAND DOLLARS!! Second it puts out 1200W. Third, it's instant on and it can be made to be band following with the F5K and DDUTIL. This pretty much would make DXing a one click operation. Click on the station and the correct antenna is chosen the correct setting on the amp etc etc. Sounds lazy I know, BUT that would allow me to use my F5K from anywhere in the world, including as a mobile when hooked up to the iPHONE. I'm not quite sure how I would send CW with an iPHONE BUT.... Always something new to fool around with in Ham radio, but I digress.

The 3D2 was working up 2 and working single callers. No pileup at all. In fact hardly any takers. He would sit there and call CQ maybe half a dozen times between contacts. I sent my call sign twice and he came right back. With the spicy after taste of panang still tickling my senses we exchanged our info and I plugged him in the log. This is the kind of stuff you dream about. Flipping on the radio and popping Fiji on 160 first call... No big wall of the west coast to penetrate. No big wall of Japan to penetrate. Just 2 hams mano a mano against propagation and that ever present deep fading of 160. I checked the lightning map and FINALLY there are no red dots out in the Atlantic or in the Gulf. Just a couple dots in Kansas and one lone dot in west Texas. As I sat here typing this, Fiji got bored and shut down about 5:15 Strong signal, low static, excellent operator, and no takers, too bad. The second and third tier could have picked off a juicy one this morning. The east coast slept and my insomnia payed off. Now I'm dreaming of my Latte...


Monday, October 19, 2009


The static at my QTH has been absolutely horrible. Its still bad this morning,but not as bad as the past few days. I got up a little early to spend a few minutes in front of the rig, sip some latte and hear what I could hear. I worked K4M on midway on 30M this morning. He was strong here. It was a strange sensation watching Skimmer as I could see virtually no one calling him, yet his rate indicated he was hearing a nice sized pile up. I could hear K4M on 40 and 80 as well, nice signals. Here is the reason I didn't work them this morning on 160

The 160 Titanex bit the dust.

I worked TX5SPM on the Marquesas Archipelago on 40M. It is such a pleasure to work a DX station that knows how to operate. All of the operators I have worked at K4M and the two TX5 expeditions have been top notch. Unlike those morons out on Glorioso a couple weeks ago. What is the point in spending all that effort in order to do a crappy job?

I did tune down to 160 and heard 3D3KJ on Fiji. He was super weak, barely copyable, and there were a scant few calling him, people with better stations than mine!! With all this Atlantic static I have bee strongly daydreaming about putting up some kind of directional receiving array. Given where my main transmitting vertical is located it would have to be in the front yard and I don't think the wife would be too amenable to that, but we will see.

Here is a shot of the static sources this morning

Nothing across the entire nation except in my skip zone off my east coast!!!

Over the weekend I worked a guy running a home brew 6L6 rig. It brought back memories of 50 years ago when the bands were full of chirpy drifty little rigs running 20 to 50W. Not to complain about my contacts transmitter quality at all, but as I worked these stations this morning I remember my S-40B receiver that covered the entire 40M band in about 5/8 of an inch of dial space, and a Q multiplier for filtering. On that radio I ran the volume wide open and adjusted signal strength using the RF gain, much like I do now with the F5K. To make these contacts this morning I had the filter cranked down to 25hz. At that bandwidth a few hz off freq makes the difference between Q5 and no copy. If you ever wonder at the strides we've made over the past 50 years in ham radio, and you were around in those days its amazing to think of what we are able to do. All of these DX stations are rock solid and do not move. Once I get on freq I can forget about tuning. One of the K4M transmitters is consistently about 10hz low, this is how far we have come. My little 6L6 buddy made me open the filters up to a khz or he would have drifted completely out of my pass band with every transmission. 25hz filters, and being able to track a stations freq down to a hz so using a 25 hz filter is possible that is simply amazing!!

I had a recent correspondence with some one who was carping about DDS. I had written about the recent review of the F3K in QST and about how the people they choose to do the SDR reviews know basically nothing about this kind of radio. The reviewer stated that SR (spur reduction) does nothing. He then proceeded to describe that when he pushed the button that the any spur he had in his pass band moved OUT of his pass band. This is what spur reduction does it moves spurs out of the pass band. So basically the radio was performing perfectly and the reviewer was too naive to even understand what the radio was doing. In my retort I asked rhetorically what other radio allows you to move spurs out of the pass band? I got a response that the K3 does because the K3 also uses a DDS and the comment was this was somehow a backward movement in VFO development. The reason I can have 25 hz filters and talk to stations that have enough stability to use this kind of narrow bandwidth is precisely because of this kind of DDS technology. So the next time you hear some guy quacking drearily about his 75A4 and how it's "as good as" think about what it takes to make a QSO using 25hz filters.


Friday, October 16, 2009

K4M/KH4 on 80

Given the amount of time it took me to work them on 40 yesterday ( a little over an hour), I got up an hour early to see if I could bag them on 80. As you can see the pileup is dramatically smaller, BUT I had storm static that was out of control. My S meeter spent most of its time at -90dBm about the same level as K4M. I live on the central east coast of FL and here is a map of my noise sources:

In a word UGLY!!! I'm virtually surrounded by storms.

It was the first time I needed ALL the noise reduction goodies to make the contact. I turned the AGC-T way down so the static fell into the background and only peaks got through. The Static is limiting so there is no reason to run the static way into AGC. The way you do this is to set AGC-T so the average static level is just audible. Crashes will of course poke through but the static will NOT dominate the AGC action of the radio. When you hear people speak of signals "popping out of the noise" this adjustment of AGC threshold is why. The noise was still too loud however so I kicked on BIN. BIN stands for bin-aural and it splits the audio into a low pass and a high pass so one ear gets the low passed audio and the other gets the high. I was running only 50hz bandwidth on the filters so there isn't a lot of "low" and "'high" to filter, but it did seem to improve copyability so left it active. Next I turned on NR. The NR in this radio actually does something

Here is an MP3 with AGC-T set to 85 (towards the high end)

Here is an MP3 with AGC-T set to 56

An MP3 with BIN added

An MP3 with BIN and NR added

The last 2 are subtle and probably require headphones to really appreciate but each step in the noise reduction scheme of this radio combined to make the signal quite readable through the din. There was considerable QSB to contend with. I made several trail recordings of these settings and kept the ones with peak signals to try and avoid QSB as a factor in the recordings.

During AGC 56 there was a -70dbm signal (aka loud) station calling about 40 hz off freq it can not be heard except for a few of his key clicks.

One thing that was a blessing is that the pile up was VERY well behaved. Almost no tuner uppers and almost no jokers calling on freq, and only one ahole sending strings of dits.

The operator today was running a much more traditional pileup. He was pretty much +2 to +3 and not drifting all over the place, which is why the pileup was nice and compact. He was working a lot of JA's and some Russians in-between US contacts, as the day to night gray line was just washing over Japan and the night to day end of the gray line was still an hour away from the east coast. All in all I had him in the log in under ten minutes TOO MUCH FUN


Thursday, October 15, 2009


Worked K4M on 40 this AM. It took forever. He was moving up and down the pileup obviously tuning with a knob since it was directonal. He would start at the bottom of the pile and rapidly tune toward the top of the pile working what ever he could hear and then start back down to the bottom, so both ends of the pile could be active within a couple of minutes. I tried postitioning my self in quiet spots along he way, I tried tail ending, finally what worked was to plant myself at the top of the pile and just wail away until he got to me.

The pile up was very well mannered today. I was on for a couple minutes yesterday and the noise on the DX's freq was out of control. Yesterday K4M was a true S-9 (-70 dBm) on 40 today he was S-5 (-100dBm), but he was much better copy since the DX freq was clear. I just don't get it. If you are going to tune up, why tune on the DX freq? You are not going to transmit there so why tune there?

He was on 80M yesterday and then 160M today and I could copy him Q5 on 160 but he was weak and the pile up was actually quite large on 160, and growing as east coasters rolled out of bed and flipped on their radios, so I didn't bother.

The new computer is running flawlessly. I increased the frame rate on the panadapter to a ridiculous 30fps. The processing overhead barely budged. It had a subtle change on the way the display plays. It's a bit "smoother" and more persistent. The effect is surely subtle. More fun to play with. This radio never ceases to amaze me. The experience is always evolving, AND MIDWAY ISLAND is n the log thanks to its well mannered brutish horsepower.

Saw my old buddy K3RR at the top of the pile up, and N4PY at the bottom, so they discerned the pattern also and figured out where to place themselves as bait. That's another feature of this system. Over time you get to chronicle DXers and see how effective they are in their strategies. You can see who is hearing the DX and who is not because the people who are not just keep transmitting when the DX is transmitting. You can see propagation by who is sending 599 and tracking those regional data points in your mind For example yesterday on 160M it was all JA and W6 that was dominating the K4M pileup when I listened. Its all VERY interesting.


Monday, October 12, 2009

Noise Blanker

I was playing with my new Computer/Radio. The latency in the new computer is 45 microseconds down 10 times from the core 2 duo machine. As I was measuring latency I was monitoring 160 CW and a terrible noise came on:

You can see the noise as a jagged regular wideband saw tooth of about -90 dBm. I get this noise intermittently. I live out in the country, and we are all on well systems as opposed to county water, and I believe it to be from a well pump motor from one of my neighbors. I flipped on the NB and here is what I saw:

Noise 100% eliminated and you can see stations that otherwise would be obliterated pop out of the noise. Look closely at the noise still visible in Skimmer. It looks like an increase in background snow on the display. The signal that goes into Skimmer is the raw I/Q data not the processed data that is displayed on the panadapter. Hence you can still see the noise in Skimmer even though every trace of it has been eliminated in PSDR. This is the advantage of this kind of radio system. The signals are processed in a kind of virtual time that is "near real" so you can do many things to the signal (like remove all the noise) that enhances the human-system experience. This is one blanker that really works!! Notice that you can not see the signals in Skimmer that you can see in the panadapter with the blanker on. The noise is intermittent enough and the blanker works so well I haven't even tried to spot who's pump is the culprit.

Here is a shot of "normal":

Here you can see Skimmer and the panadapter are seeing the same signal. Very interesting display. With the NB on there is no distortion added so even the weakest MDS signal is unaffected by the NB. Notice the S-Meter reading in each screen shot. These readings tell the story. With the NB engaged there is no difference in my 160M MDS between noise and no noise

I finally have all the basic accouterments loaded on the new computer finishing with Tortoise SVN. I made an image of the partition using Acronis, and now I can return to a pristine condition in about 10 minutes by simply reloading the image of the partition. I like to do that every so often and Acronis works great for that purpose.

Here is a copy of my usual desktop when I am operating:

I have PSDR, DDUIL, Skimmer and 4 of the DX lab suites applications open as well as Firefox, and some tray applets like Google talk. SpotCollector which is the DX lab cluster program is connected by telnet to 3 spotting sources. In the past having this much internet activity (SpotColector, Google Talk, and Firefox as well as email) has caused the % processing utilization to sore. With this rig I have never seen it more than in the low teens.

This computer makes this radi0 and all the rest of the other applications that I like to have open play so smooth, its a delight.

73 W9OY

New Puter

I have been running a Shuttle cor several years n0w. It was a core 2 duo 2.5 ghz with a couple gigs of ram and a SATA 3 hard drive. I Overclocked the box to about 2.8 ghz. It was a nice running machine and had pretty good horse power as far as a ham computer goes. I plugged in a Radeon X1300 512 mb video card so I can run 2 monitors of extended desktop. I chose this particular card because it can run 2 monitors and does not have a fan. I added Windows XP pro, and lived the good life. After several years of clean living about two weeks ago the system started to hiccup and finally died to the extent the machine would not even POST.

What to do? The Shuttle had given me great service so I decided to try another. The Shutle uses a TI fire wire chip set and has always been perfectly compatible with my Flex gear, so I decided to pop for another Shuttle. I usually don't buy the highest performance stuff, but I do like to upgrade to some extent. I bought the SX48P2 E chassis. Next what processor? I chose this

This is a Intel Core 2 Quad Q9550 Yorkfield 2.83GHz 12MB L2 Cache LGA 775 95W Quad-Core Processor The reason I chose this processor over something a bit cheaper was the size of the cache. Hard to beat 12mb of L2 Next was memory

Mushkin 2GB 240-Pin DDR2 SDRAM DDR2 800 (PC2 6400) Desktop Memory Model 991580

I bought a total of 4gb. The Shuttle can address a total of 8gB of Ram spread across 4 slots, so if I ever upgrade to a 64bit OS I will be able to just pop in another couple of sticks of memory to fill out the bill. As it stands with XP I can access a little over 3 gb. The hard drive is a Seagate from the other computer and the DVD is a read write from the other computer. The computer can be overclocked, but at 95w I decided that was enough, no need to stress the cooling system. It runs at about 40 deg C.

It took about 45 minutes to build the computer, including dissembling the old Shutle and building the new. The Shuttle case is entirely thought out and it is basically put in the component and hook it up. There was no troubleshooting to be done and I fired it up went into the bios screen to set up a few parameters like drive boot order and started loading windows. Windows is another matter. That took the better part of 5 hours to load and bring entirely up to date, including all the Flex stuff and DDUTIL VCOM SKIMMER and DXLAB. I plugged in the F5K and it went through the 3 driver installs and I was on the air!!!! It went smooth as silk.

This computer has a few idiosyncrasies. It does not have PS-2 ports, so you have to use a USB Keyboard and mouse. It also does not have any serial or parallel ports. For that duty I use USB ports as well. I still have the USB to parallel port connector for my SDR-1000. My printer is a network printer so I do not need any ports for that. I use a USB serial for the input from my keyer, and to communicate with my LP-100 watt meter and all of those were installed in mere minutes.

It's performance is significantly better than the old core 2 duo, so all in all the upgrade was entirely uneventful and there were NO incompatibility issues.

I chronicle this build, because people often wonder what computer is compatible with the Flex series of radios. This is not an overly expensive machine to build, so if you are in the market for an upgrade or want some info on a system that works out of the box with the Flex line this sucker is works great. It's also quieter than the previous Shuttle. The system uses heat pipes to pipe the heat to the cooling fans. Very efficient.

73 W9OY

Wednesday, October 7, 2009


Jacek SP5DRH is active as 3D2KJ for the next 4 weeks or so from Fiji. I heard him on 160M this morning, and above is a screen shot of the "pileup". He was mostly working JA's but a few 1 landers were getting through as well. He was Q5 with a noise level in the -114dBm range, except for the static crashes which were running -85dBm.



Monday, October 5, 2009

Things are looking up!!

I had my main Ham computer go down last week. It's a "Shuttle" and I think the main board went south. Over all I have been quite satisfied with the performance of the computer for my Flex radios, so I am in the process of upgrading to a new wiz bang version of the Shuttle with a quad core intel processor and a scad of memory etc etc. In hte mean time I am using an old AMD dual core computer that I have been using as my Linux box. I added Windows to hold me over until the new fire breather is on-line

In the mean time the bands have been really looking up here in Florida. There still is considerable storm static I suspect out in the ocean, but especially on 40M things have been hot especially into the Indian Ocean and Oceania. Last night Madagascar and Saudi and Austral Islands and this morning Mauritius all of them Q5 and second or third call.

The F5K is invaluable in picking the spot to make the call and determining the pattern. This morning the 3B8 was starting at the bottom of the pileup and working his way to the top picking out half a dozen stations along the way and then returning to the bottom to do the pattern all over again. It was simply a matter of putting myself into a "quiet" 300hz window and blasting away. If he passed me I would move up a khz to another clear spot etc etc. With out the F5K and Skimmer figuring out this pattern would have been impossible

Way too much fun!!!!

73 W9OY