Tuesday, December 29, 2020

KK1L Antenna Switch Installed

Antenna selection panel at eye level in station.
Last month, I built the KK1L 2x6 Antenna switch. But to use it in my station, I needed to be able to select the right antennas.

I found a small metal box and a 1 pole-7 way switch in my junk box. A couple of switches, an LED and a 15 volt DC wall-wart, and I was done.

On the right is a power switch with an LED. The center switch selects the antenna relay, or none at all (it has seven selections, and there are only six relays). 

The small switch on the left controls the shunt feed matching network

It works great, and has completely replaced the 6-position B&W switch I've been using for years. In fact, I had to remove the old B&W switch from the operating position because of the habit of looking at it. 

Monday, December 28, 2020

A3S/A743 Refurbished - Rotating Once Again

A3S/A743 QRV @ 15m
Since I took it down back in October, the A3S/A743 is once again proudly back at the top of my tower. It's been something of a long ordeal.

Refurbishing the antenna was delayed when I had to wait eight weeks to get traps out of MFJ -- they, like many companies have been greatly affected by the Coronavirus. Indeed, their biggest problem is getting parts from their suppliers, who are likewise affected.

Even after getting the parts, I had a lot of trouble tuning the A3S somewhere close to the bands. That required a lot of jockeying on the tower. It did give me a lot of time at the top of the tower, something that is always a confidence-builder.

I had a little trouble with the rotator yesterday -- one of the connector lugs came off the pigtail. This meant that while the rotator was in place, it wouldn't turn. I was tired from the long climb yesterday, and fixed it today. 

The good news is the antenna seems to be working correctly. I haven't worked a lot with it, but I'm sure I'll get a chance during the RTTY Roundup this next weekend. 

Hopefully, this antenna should serve me well for many years to come -- the initial installation worked for over fifteen years. Fifteen more would be fine by me.

Next project will be to fix the WARC trap dipole and put it back up.


Sunday, December 27, 2020

Re-writing the K9AY Controller

 I was quite happy earlier this month to get the K9AY Loop Controller working. After using it in the ARRL 160m contest, I decided I needed to make some changes to the firmware.

That presented a problem. You see, while Microchip has made some excellent developer tools available for their PIC series of microcontrollers, like all software, it is updated from time to time, and things change.

When I used MPLAB X three years ago to write the controller firmware, I used the PIC assembly language tool MPASM. That was with version 3 of MPLAB X. I recently updated to version 5 of MPLAB X, and MPASM was no longer supported. Instead the new assembly tool was called pic-as. And, of course, it was syntactically different than the old MPASM.

I made an attempt to edit my existing code into something acceptable to pic-as. While I got it to build, I found it didn't run correctly. Rather than try to debug, I figured it might be easier to start over.

Microchip has a good C compiler. I decided to start there. I'm well experienced with C, but this left open a lot of questions. The PIC has a number of configuration registers that must be set at startup. How does one do that with C? 

This actually turned out to be very easy. Microchip has a tool called the MPLAB Code Configurator (MCC). It provides a number of graphical tools and wizards that specify the chip configuration, down to the pin assignments and interrupts. MCC then generates the appropriate C code. This is a good thing. One of the hard parts of working with the PIC is figuring out how to set all the internal registers. MCC does this part for you.

My second question had to do with interrupts. How do we process interrupts in a C program? This also turned out to be easy, because MCC generates code to handle interrupts, too. When properly configured in MCC, one need only write a simple void function to process the interrupt. My controller uses a timer interrupt every 2 ms to check the state of all the buttons and update the LEDs and relays appropriately.

Re-writing the controller program in C ended up taking less time than figuring out the assembler syntax changes between MPASM and pic-as. 

And the best part is, it worked! 

Saturday, December 19, 2020

Tuning the A3S/A743

Adjusted A3S/A743, ready for rotator.
After adjusting the A3S/A743 on the ground and getting it back on the tower, I was very disappointed to find the SWR dips for each of 10, 15, 20 and 40m were all outside the band. They were all too low. This surprised me. When I measured the antenna with the beam facing upward, the dips were very close to the band. Now that the antenna was on the tower, it was going to be more difficult to adjust.

Rick Dougherty, NQ4I, gave me an important tip. Always measure horizontal antennas horizontally. Don't attempt to make measurements with the antenna vertical. With the antenna 6-8 feet off the ground, you can get a good sense of the measurements, knowing that the antenna will rise in frequency as it moves away from the ground. Generally by about 100-300 kHz at the 30 or 40 foot level. As you move further from the ground, the effect lessons. 

Rick knows his stuff -- he built a superstation in Griffin, GA, and has experienced this phenomena putting up dozens of antennas. 

OK, I didn't follow Rick's belated but sage advice - what should I do now? I didn't want to take the antenna off the tower again, it was a lot of trouble to get it up there in the first place. I decided I could make adjustments similar to how I added the A743 option originally - by tipping the Driven Element (DE) vertically and climbing down the tower.

On the tower adjusting
the now vertical DE.
One the first climb to make adjustments, I could already see part of the problem - some of the DE joints had slipped outward while the antenna was waiting to be installed pointed vertically. The extra weight of the A743 parts probably had something to do with this. 

Loosening the U-bolts for the DE and the support mast, the DE tipped over easily and I was able to climb down the tower and adjust. I had to make some of the adjustments somewhat blind, since I could not easily put a measuring tape against the elements. 

I ended up with the 10m traps out about 83.5" (A), the 10 and 15m traps spaced as 6" (B), and the 15 and 20m traps at 18.75" (C) -- this was the lower limit, since the tubing ends were contacting each other inside the support tube. the 40m portion I extended to about 66". These measurements are for each side of the DE.

This took two climbs, and ended up with the 10, 15 and 40m SWR curves well centered in the bands, and 20m right at the bottom of the band. This isn't perfect, but it is usable. Note I did not make any adjustments to the Reflector or Director, as they are well out of reach.

Next time I set up an A3S/A743, I think I will adjust the 10 and 15m portions to the MID-band settings, and split the MID and CW settings for 20 and 40m. That's roughly where I ended up anyway. 

I'll also consider replacing a lot of those stainless steel worm clamps. After being out in the weather for 30 years, a few of them would not tighten the way I would like. 

The A3S/A743 should be ready to go after one more climb to raise the antenna and install the rotator. I also plan to tweak the support mast a bit higher to keep water from accumulating in the 20m (TK) traps, since that's likely how they failed in the first place. 



Wednesday, December 16, 2020

4U1UN (and 4U75UN)

A little more than a year ago, I was looking at the Most Wanted list in ClubLog. I noticed that the United Nations HQ entity was number 30 on the list. In all my years of ham radio, I had never heard them on the air, much less worked them. I wondered if I'd ever have the opportunity to work them at all.

Well, the 4U1UN club has been busy, and late fall 2019, they finished their installation which allows them to remotely operate a transceiver near the top of the UN towers from a room on the ground floor of the building. Since that time, they have been sporadically active. 

I don't know what the rules are for operating 4U1UN, but the station has been active at least some time every month during the past year. They have gone from an entity you'd wonder if you'd ever work, to one that can easily be worked by anyone with a bit of determination.

Over the past year, I've kept a running DX Watch query going to see when they were on the air, and worked them any time I could hear them. They've been easy to work even with compromise antennas.

They confused me in September, as I thought they had gone silent. No, they just switched callsigns. They are celebrating 75 years of the UN with the call 4U75UN. There's a special certificate for working them. They should continue with that call until December 31st.

They'd be pretty easy from the USA on 160m, but I don't believe anyone has ever operated on that band. I doubt they have an antenna, and I head the noise level in NYC is extraordinarily high.

They QSL promptly through LoTW (be patient, it may take a week or two to be uploaded). I've gone from wondering if I'd ever work them to having them confirmed on nine bands (80m-6m) and three modes: CW, Phone and Digital. 

Keep an eye peeled on the spotting networks, and perhaps you can work a new one.

Wednesday, December 2, 2020

K9AY Controller and Antenna

K9AY Controller at operating position.
In 2007, I put up some K9AY loops. They were a little less than half-sized, about 33 feet in each loop. They worked pretty well on 80 and 40m, although they were a little weak on 160m. After about three years, they stopped working. At the time, I was using a rotary switch to change directions, but I really wanted a push-button controller.

In 2015, I designed a controlled based on a 74LS175 Quad D-type flip-flip. It was pretty simple in concept, but, it didn't work. The bounces from the switches changed directions rather randomly.

Which lead me to write this article, about a solution - using a PIC micro-controller to solve the bouncing switch problem. 

I built that prototype, and wrote about it three years ago. One problem was that the wiring appeared flaky, it wasn't always dependably switching. I figured I needed to design a PC board. 

While I've been working on that, I had a thought a couple of weeks ago -- maybe I could piggyback a second piece of perfboard to my prototype and keep the board from flexing, and that would avoid the flaky wiring. Easy enough to try.

K9AY Antenna box
at base of tree used as
support
Sure enough,  doubling up the perfboard helped. I determined that the wiring wasn't that flaky. Part of the problem was that the NW and SW switches and LEDs were reversed. Pressing the SW button sent the NW voltage, and vice versa. 

With that fixed, I couldn't find any reason not to put the K9AY back up. I used 42 feet of wire in each loop. This is a good compromise length for 160, 80 and 40m. 

Some on the air testing shows that it works pretty much as expected, and the push-buttons make it really easy to change directions. 


Monday, November 30, 2020

Schematic for the Little 40m Receiver

I'll admit that it has taken far too long for me to publish this schematic. I wrote the original article about a small solid state receiver in January of 2009. I promised a schematic, but I never published it. 

The original sketches of the schematic are too embarrassing to publish directly, so I had to re-draw the schematic neatly.  I started on it last summer, but I finally finished it very recently. 

Just click on the image to the right. 

A couple of notes on the schematic.

The IF Out label on the right side of the page connects to the IF Out label in the lower middle of the drawing. 

I didn't have a 7808, so I used a 7805 regulator with a stack of four 1N4001 diodes that gives about a 7.5 volt supply. You could use a 7808. In fact, I purchased one a while ago, but I've never modified the rig to use it.

You might have to adjust the values in the S-meter circuit if your meter movement isn't a 600-ohm 250 uA device like I used.

I don't have a block diagram, but the sections are straightforward:

  • SA612 Mixer / Oscillator
  • Crystal ladder filter
  • MC1350 IF Amplifier
  • SA612 Product Detector / Oscillator
  • LM 386 AF Amplifier
  • LM 358 AGC Amplifier / S-meter driver
  • 2N3906 Mute driver

I built the rig in stages, starting with the AF Amplifier, then the Product Detector, IF Amplifier, Crystal Ladder filter, Mixer. I then added the AGC Amplifier and the S-meter driver, and finally the mute driver. 

The Crystal Ladder filter is based on 4.915 MHz crystals. The Oscillator is 7 MHz above that, nominally tuning 11.915-12.065 MHz. I've been thinking about adding a switch with a different oscillator and bandpass coil to cover 30m. 

Friday, November 27, 2020

Improving the Shack Desk

Finished, standing against wall.
Thirty-five years ago, I asked my father-in-law to be for advice on lumber for an operating desk. He suggested I find a solid core door. I bought an 80x30" door and have used as my operating desk since. It has held up very well, surviving two moves, and three re-finishings. 

Originally, I bought the door and two small two-drawer filing cabinets. With the cabinets on either side of the door, the result was a very stable platform, strong enough for me to stand on. 

However, it wasn't the most perfect arrangement, as I wrote four years ago. Adding a 1/4" radius edge to the front, was a  dramatic improvement, mainly because the operating surface was just too tall. 

This year during Contest University,  Doug Grant K1DG gave a talk on Optimizing Your Station For Contest Operations. One of the salient points in his talk was that the operating desk needed to be at the right height for the operator. 

The door across the filing cabinets was 30 7/8" off the floor. Doug's talk indicates that the optimum for my height should be closer to 28 1/2". The micro-shack desk was about 30", and the luncheon table that serves as my workbench on the other side of the shack is 29" off the floor. What I needed were shorter supports. 

Box frame.
It is not possible to make the filing cabinets shorter, so I needed something else. It had to be strong -- the door itself is heavy, and the linear amplifier and other bits of equipment add up. I also needed to be able to get it through the door to my 6x15 foot room that is the shack. 

What I came up with is a set of box frames aligned by some cross members. The box frame is a "U" shape made out of 2x4s, capped with a 30" 1x4. A couple of 1x4s act as a shear web to keep the box frame square. It's held together with glue and screws. Three 58" 1x4s serve as the cross members. The cross members are each in a different plane, and give plenty of room for a chair to roll underneath.

Finished stand.
It's sufficiently strong that I can easily sit or stand on it.

And, here's the best part. It's on six castors, so I can move it away from the wall to get behind the desk to wire things. No more having to do my rig wiring blind!

I was worried that the tile floor might make it hard to roll the desk, but it's actually pretty easy. 

This design is pretty simple, and it came together quickly. The 30" 1x4s on top have a single wood screw on each end into the door, so the surface can't slide around. And the cross members are held on only with screws, so they can be removed when in order to get the whole assembly out of the shack.

Voila! Away from wall.
The door already has a couple of power strips mounted on the underside, which makes plugging in equipment that much easier. 

Based on what I was thinking four years ago, the next improvement will be the equipment hutch that goes on the desk. That's also a relic from 30+ years ago. I've already purchased the lumber for that project....



Monday, November 23, 2020

KK1L 2x6 Antenna Switch

Completed relay unit.
I've been meaning to incorporate an automatic antenna switch in my shack for years. Indeed, I bought 20 RTB14012F relays 11 years ago just for such a project (and they were $2 each back then!). But, I'd never gotten around to designing and building one.

Somehow, I ran across a link to KK1L's 2x6 Antenna Switch board, and looked at the Relay Unit. I loved the elegance of the design, and the board was a beauty. I wrote to Ron asking him about the pricing for the board, and if my RTB14012F relays would fit.

Ron answered me promptly. He quoted me a price of $33 for each board, and he could mail up to five in a priority envelope for $7.15. He also informed me that my relays wouldn't fit -- that the pin spacing was incorrect. I was disappointed and set aside the idea for a couple of months. 

Board top.

While I was working on another order on Mouser.com, I couldn't get the KK1L relay unit board out of my mind, and checked out adding eighteen RTD14012 relays to my order. I decided I couldn't possibly do any better than Ron's design, so I wrote him back and ordered a board. I also used Ron's Mouser.com project, which included parts like the screw terminals, chassis, diodes, capacitors and resistors.

Electrically, this is a very simple design. There are six switching cells each with three relays, two diodes, two capacitors and one resistor. The "A" ports activate two relays, the "B" ports activate only one. 

Board bottom.

I had a lot of parts already. Besides the board, I ordered the relays, chassis, and screw terminals. I already had the SO-239 connectors, diodes, capacitors and some 1/2 watt 50-ohm resistors. 

[ UPDATE: I recommend using the 50 watt resistors specified by KK1L instead ]

Building the board was easy. The diodes and capacitors go in first, because they are under the relays. I used some 0.1 uF multi-layer ceramic caps I had on hand. These were smaller than the diodes and laid flat against the board. Relays went in next, and the screw terminals. I put the resistors in last. Then I had to remove them and put them in again, because I ran them through the wrong side of the board.

Now came the hard step - drilling the chassis. I'll be the first to admit I'm not a master machinist. I have a drill press, which helps. Precisely aligning all the holes necessary for the eight SO-239 jacks to pass perfectly through that beautiful circuit board was intimidating. I decided I only needed two screws on each jack to pass through the board. That limited the project to 24 holes, plus a couple of mounting holes, and a cable hole - 27 in all. 

I started with the printed template, but found that it didn't align perfectly with my board. I ended up using the board itself as a template and marked each hole on the chassis with a sharpie. 

The instructions say to use #6 hardware on the SO-239 jacks, but I have never seen a SO-239 that would pass #6 hardware. I used #4 hardware instead. I drilled all the holes with a #27 drill in any case, which gave me some leeway. The center holes were a little harder, I widened the pilot holes with two different drill sizes, then used a step drill to get to the final size. 

Chassis ready to install  board.
The instructions call for mounting the SO-239 jacks inside the chassis. I knew I had some hole alignment problems, so I opted to mount them outside the chassis. The disadvantage in doing this is that you can't remove the board from the chassis without unsoldering the antenna connections. But, it would hide all my ugly machining.

I used two short and two long screws for the SO-239s. For the longer screws, I used a jam nut and a couple of washers below the board, to give enough clearance from the chassis. I drilled the holes for the short screws after mounting the jacks and aligning to the board. I decided I only need two of the four mounting holes -- the two that were opposite the A and B connectors. The other end of the board is supported well by the three SO-239 jacks.

After everything was drilled and deburred, I took everything off the chassis and gave it a quick buffing with a wire wheel. This gives the chassis a little bit of sheen and hides any large scratches. The until will eventually be mounted on the wall as part of the single-point ground, so it didn't have to look too pretty.

Fully installed.
Once I had the board mounted, I realized I made a mistake. I mounted the screw terminals so the holes are facing the chassis (outside), which makes them difficult to access. Unsoldering would be a pain, so I decided I could live with the error. 

I did a little bit of testing. First test is to see that all antenna points read 50 ohms. Then I checked to see that all the relays activated correctly. Everything worked great. 

Next step will be to incorporate it into my station.



Sunday, November 8, 2020

K60XV - The Hard Way.

K60XV Built and Installed
Last September, I was somewhat distressed to learn that Elecraft was no longer offering the K60XV -- the 60m and transverter interface for the K2 transceiver. I had always planned to purchase this kit, because I intended to build a 6m transverter. 

According to Elecraft, the kit was discontinued because they no longer had a source for the bottom adjusting variable capacitors at C1 and C2. While I understood, I was still disappointed. 

While reading the Elecraft mailing list, I stumbled on a message from a fellow named Chuck. He said the managed to obtain the parts for a K60XV and put it together. I followed up with an e-mail, and Chuck indicated that he obtained the PC board, processor and a few other parts.

K60XV "Kit" - a collection of parts
I wrote to the Elecraft parts department, to check if certain parts were still available. I was delighted to find the K60XV PC Board and processed were still available in limited quantities. So, I put together an order:

  • E100193 - K60XV, PCB $24.93
  • E610018 - MCU, K60XV $6.60
  • E640001 - Relay DPDT LATCHING $2.60
  • E620005 - 8 pin FEMALE, 0.1LS $3.65
  • E620009 - 3 pin FEMALE, 0.1LS $2.08
  • E620076 - 8 pin MALE, 0.1LS, Elevated $2.49
  • E620077 - 3 pin MALE, 0.1LS, Elevated $0.92
  • E700090 - Nylon Stndff,3/16D,9/16L, Hex $1.18
These were all parts that were either exclusively from Elecraft, or those that I had trouble finding elsewhere.

I also ordered:
  • FWKAT100 - KAT100 F/W Upgrade to 1.05 $49.95
  • E700002 - TO220 Thermal Insulator, Adh $1.96
My KAT100 was older, and I needed the newer firmware for it to work on 60m. Installing the K60XV requires you to remove the heatsink panel, and I wanted to make sure I didn't destroy the thermal insulating pads.

For the rest of my parts, I combed through my junk box and also placed an order with Mouser. I didn't have any difficulty finding the rest of these parts. 

Board with parts in place.
The K60XV is not a complex option. If you've built the KSB2, this is piece of cake. The board goes together easily. I skipped installing C1 and C2. I had a couple of top adjusting 7-40 pF caps I planned to use. I waited to see what kind of clearance was available.

The main RF board requires a couple of modifications in order for the K60XV to function. This is probably the hardest part of installing this option.

If you were lucky enough to have a serial number of 3000 or higher, this job is much easier. For those of us with Rev A RF Boards, it's a little more difficult. 

D19 and D20 install.
The Rev A boards don't have a place for D19 or D20, so you have to mount them beneath the RF board. 

Hooking up the RF connection is a little picky as well. Take time when prepping the coaxial cable. Some folks have used heat shrink on both ends of the coax, but I didn't find that necessary.

I ended up being pretty happy with the RF board mods. 

RF line install finished.

These RF Board modifications require the heat sink panel to be removed. My early K2 (from 2002) didn't have the transverter holes on the heat sink panel, so I replaced it with the newer one. This requires the thermal insulating pads and the serial number sticker. I was actually a little nervous about this. I purchased spare pads, but I only had one shot at the sticker.

I shouldn't have been nervous. With a sharp knife, the sticker came off easily and transferred to the new panel with ease. The thermal insulating pads came off easily as well, and I had no trouble moving them.

I fitted the K60XV board and considered my options for installing C1 and C2. I considered installing my caps upside-down using short lengths of wire. However, there appeared to be plenty of clearance between the top of the K60XV board and the KPA100 shield. As I measured it, there was about 2mm of space between the top-mounted capacitors and the shield. So, I opted to mount them on top of the board. My caps were a perfect fit.

Rear view of the finished installation. Looks great.
Adjustment was easy. Set the D19 menu to Y, then adjust the C1 and C2 caps to peak 60m. 

Buttoning it all up again, I'm very pleased with the result. I don't know how long Elecraft's supply of parts will last, but if you really want a K60XV, you should inquire with Elecraft's parts department and see if they still have these parts available.

Sunday, November 1, 2020

Rebuilding the A3S/A743

I've had the A3S/A743 down for several weeks now. I've gotten it mostly rebuilt, but I'm waiting on parts. More about that in a moment. 

I wrote previously about my inspection of the traps. The 20m TK traps are completely toast. But the TA (10m), TB and TC (15m) traps appeared to be fine. All the plastic end caps, however, were toast.

This particular A3S/A743 has been up for a while. I originally purchases the A3S in 1991. It spent three years in the air at the old Stone Mountain QTH, then several years in my basement in Gwinnett. Then it spend a couple of years on W1YM's tower while he was waiting for his Skyhawk to be delivered. And nineteen years at the Gwinnett QTH. All told, that's twenty-four years exposed to the elements. The A743 40m option has only been up fifteen years. 

A few years ago, I bought another A3S at a hamfest for $100. Such a deal! This unit was manufactured in 2007 and does not appear to have been outside for any length of time. The plastic end caps on the traps look practically new.

During the rebuild, I decided to use the TA, TB, and TC traps from Cushcraft #2 on the Cushcraft #1 tubing and vice versa. Cushcraft #2 isn't going up any time soon (perhaps when I move to another QTH and can build a 70 foot / 35 foot stack). I figured I would use the least weathered traps on the antenna that I'll be using. 

I used No-Al-Ox liberally when I assembled the A3S nearly twenty years ago. After nineteen years in the weather, much of the original application had evaporated, but the sections came apart easily after unclamping. The traps from Cushcraft #2 had never seen a drop of No-Al-Ox, unfortunately. So, as I intermixed the parts, I made sure I applied some to every joint. 

I ordered all new trap end caps, boom and element end caps from MFJ, along with new TK traps. Laird communications sold off Cushcraft several years ago to MFJ. I was happy that parts were still available. However, the ordering system left much to be desired. MFJ is in the middle of a web site re-design, and there's no way at present to order individual parts. Their older web site would list parts on the same page as the product, and you could click through to the parts link to order each one. Why they would switch to a new web site that doesn't have this capability is beyond me. 

A call MFJ with all the proper part numbers, I ordered everything pretty easily. Then it got weird. The gentleman on the phone didn't tell me the total of my order, he just guessed it would be $130-140, which is about what I expected. He estimated my order would ship in two weeks, and that I should see an order confirmation email and another email when the order shipped. All good, I thought.

Then I noticed my credit card was charged $169.90 almost immediately after I got off the phone. I waited a few days, but I received neither a order confirmation or shipping confirmation email.

This is not right. Indeed, this may be illegal. You're not supposed to charge someone for an order unless their goods have shipped. That's how it is supposed to work. I called and asked about it, and they said they corrected my email address and I should get a confirmation. My order hadn't shipped yet, because the TK traps needed to be manufactured, and that would take at least two weeks. Fine. But, then I didn't receive any order confirmation emails. A week after my original order, I called again, had my email address corrected again, and asked that a confirmation email be sent. This time, I received it. At that time, I was told my order would ship in ten days. 

And, sure enough, ten days later, they shipped something. I was surprised when I received a very small box -- clearly not large enough to contain the two TK traps. Opening it, I found my entire order of plastic end caps. But no traps. 

It's now been four weeks since I placed my order, and no indication they are shipping the traps that take them two weeks to manufacture. There have been at least two good weekends for putting the tribander back on the tower, but without traps, I can't finish the assembly.

Keep your fingers crossed I actually see these traps sometime soon.




Sunday, October 4, 2020

A3S/A743 On The Ground

A3S/A743 on the ground

I've known for some time now that my A3S/A743 has problems. Back in March 2019, I noticed that one of the 20m traps was arcing. Then a couple of months ago, I could see that several of the trap end caps were either missing or damaged. 

It's taken me several tower climbs to bring this antenna to the ground, but this weekend, we were finally able to successfully lower it down without incident.

My daughter Lauren acted as my ground crew and did an excellent job. I couldn't have done it without her.


Antenna mounted on tripod

Once the had the A3S/A743 on the ground, the next process was to carefully mark, measure and inspect the antenna. I didn't want the antenna lying on the ground. I had a roof tripod I'd picked up somewhere along the way, and with a short piece of mast, it allowed me to work on the antenna horizontal at ground level.

As soon as we had the antenna on the tripod, I marked all the traps. The original Cushcraft trap markings weather off rather easily. I used a sharpie to indicate the element, trap band, and an arrow pointing toward the boom. It will be important to reassemble the antenna with the traps in the correct direction.

DE traps showing disintegrating
end caps.

Next step was to mark all the element distances. This just required a quick sharpie mark at each tubing junction. 

After this, I got out a clipboard and a measuring tape, and took measurements of all element segment lengths for all three elements. I wanted to see how the antenna measurements matched up with my records. As it turns out, even after 19 years in the weather, the director and reflector elements were spot on. The driven element was a little asymmetric -- that likely happened when I added the A743 kit in 2005. 

DE support rope covered 
in lichen.

With the marking and measuring done, it was time to inspect the antenna. Overall, all of the aluminum tubing appears to be in fine shape -- nothing bent, warped or cracked. The traps were a concern. All of the trap end caps were either missing, damaged, cracked or weathered. Three of the element tips were missing, even the boom end caps bad broken. 

My daughter Lauren noticed the lichen growing on the antenna. Lichen covered a few trap end caps as well as the entire DE support rope.

I was surprised to find the bead balun in such good shape after 19 years. The feed line it surrounded seemed rather weathered, and it is my intention to replace it.

Next, I disassembled the antenna. Because I used No-Al-Ox 19 years ago, separating the sections was easy. Once a clamp was loose, they slid right off. I removed the traps and outer sections, leaving the center part of the elements intact. The outer sections I stored in my basement, and the traps went to the workbench. The boom and center part of the elements, I lashed to my back porch for now. 

With the outside work done, it was time to inspect the traps. I pulled all twenty-eight end caps off. Maybe four or five were intact, but even those are heavily weathered. I blew all traps out with compressed air, finding surprisingly little material come out, which I took as a good sign.

The two 20m (TK) traps were a concern. These traps I saw glowing back in March 2019. One of the end caps had melted. With the end caps off, I could see the outer end insulator on both traps had melted.

Only a little char on this one.
To know how bad it was, I had to disassemble the 20m traps. Disassembling these traps is a difficult proposition. The outer end of the trap cover is dented inward in four places to hold on to the end insulator. Because of this, it can only slide outward. The inner end of the trap cover has a double-z bend to make connection to the element tubing. If you undo the bend, the cover will slide off outward. But undoing the bend will likely break the aluminum. 

A lot of char on this one...

For the 20m traps, the outer insulators were toast anyway, so I didn't have any trouble sliding the trap cover inward. I just had to remove the screw at the double-z bend first. 

What I saw inside was not pretty. Both traps were damaged due to arcing and will need to be replaced. 

I also disassembled one of the 15m DE traps. Of course I broke the double-z bend in doing so. However, the inside of the trap looked clean, and, unlike the 20m trap, the coil windings were protected with resin. I didn't think it was worth disassembling any more traps and breaking more double-z bends.

I believe the 20m traps may have suffered from too much droop on the elements and collected a lot of moisture inside the trap. Once fired up with RF, it arced and toasted the traps. 

Some of the traps had corrosion evident around the hold-down screw on the trap covers. I removed the screw and put a thin layer of No-Al-Ox between the z-bend and the trap element tubing. This should limit further corrosion in this joint.





Monday, September 21, 2020

80/40/20m Trap Dipole

80/40/20m dipole strung from tower.
I've had good experience with trap antennas. I built an 80/40m trap dipole a decade ago, and it served me well at four different locations. That is, until it met an untimely demise when a rope was cut and the mowers encountered the antenna on the ground.

Wanting to build a three-band dipole, I did some experiments with computer modeling. I found that there was a lot of interaction between the values used for the traps and the wire lengths used. I fussed around with a model for a while. Eventually, I decide it would be simpler to just build one.

I started with the traps. They are made from short sections of 3" schedule 20 PVC. (Having an outer diameter of about 3.25") The exactly frequency of resonance doesn't matter terribly much, just so the traps are identical.

For the 40m traps, I used 9.75 turns of #12 wire with a 56 pF 6 kV disc ceramic cap. This resonates at about 6.88 MHz, just below the band. I started out with 10 turns, but they resonated around 6.5 MHz, which was too low.

On 20m, 6 turns of #12 wire with a 27 pF 6 kV disc ceramic cap. Resonating at about 13.84 MHz, right below the band.

Early version of traps, showing construction.

I then put it together to form a dipole with the following segment lengths on each side:

20m: 17 feet
40m: 17 feet
80m: 20 feet

This made the whole antenna nearly108 feet long. The ends of the 80m segment are looped back on themselves by about a foot and twisted to anchor to the end insulator. I strung it between the tower and some trees in the back yard. It's roughly 10m high. 

The dipole is fed with good quality RG-8X using 16 Fair-Rite #43 1/2" ferrite beads as a current balun. 

Then began the tedious process of trimming to resonance. I started with the 20m segments, then worked out to 40m and finally to 80m. Patience is warranted here, because it is too easy to cut too much off.

I ended up with these segment dimensions, resonance points and SWR:

20m: 16' 4" - 14.125 - 1.45:1
40m: 10' 4 - 7.165 - 1.2: 1
80m: 21' 8" - 3.76 - 1.1:1

And, yes, I added 20" to the end of each 80m segment. The resulting antenna is about 96 feet long. It covers all of 20m and much of 40m with a 2:1 or better SWR. 80m, by comparison is very narrow. I didn't have much of a chance to use the antenna, but it appears to hear very well. 

I've got a second one I've cobbled together for the Fulton County QTH. I'll need to wait until the tree lose their leaves so I can get it up to operating height for final trim. 


Sunday, August 16, 2020

Bringing Comfort to the Heil Proset



Heil Proset with Brainwavz earpads

I've owned a Heil Proset for several years. While I love the headset -- it sounds great, and has the HC-4 mic element, which, inexplicably, Heil no longer sells -- it has one shortcoming. After about an hour, it gets very uncomfortable -- less like a headset and more like a head vise.

The stock ear pads are the problem. They are not even 1/2" thick. Plus, the cloth covering that goes over the ear pad tends to press the outer ear (the pinna) against the head. For 30 minutes or so, this is tolerable. After an hour, your ears start to hurt. After wearing them for 30-plus hours during a long contest weekend, your ears won't feel right for days.

I'd largely given up on the Proset for long contest use, opting to use the more comfortable Koss SB-45 or Yamaha CB500 headsets (or even a pair of old Sony headphones, if I don't need the mic). But the Koss or Yamaha headsets are designed for good fidelity, not for communications. The Proset is more listenable.

I read some old threads on the CQ-Contest mailing list about a company called Brainwavz that makes replacement ear pads for the video gaming industry. Many gamers wear a headset for hours while playing multi-player games. Perfect. Bob WA1FCN recommended them to me in a private email. 

Brainwavz, unfortunately, doesn't make a product specifically to fit the Heil Proset. I measured the earpieces of the Proset and found them to be 4" (101mm) tall and 3.5" (88.5mm) wide in a rounded triangle shape.

Comparison of thickness with
the stock earpads

Brainwavz replacements are designed for oval earpieces. I worried a little about the fit, since the Proset earpieces are triangular. It seemed worth a try. WA1FCN recommended ear pads designed for certain models of ATH headsets, their dimensions being 110mm x 90mm. I ordered them off Amazon for less than $25.

Removing the old ear pads was easy - they pull off with a little tug. A plastic retainer ring fits under the ear pad edge and snaps into a plastic ring on the inside of the ear pads. Then the whole assembly snaps on to the headset. Given the size of the Brainwavz ear pad edges, the plastic retainer wouldn't be able to hold them. And I couldn't remove the plastic ring from the original ear pads without destroying them. I opted to remove the plastic retainer from the ear pad and place it on the headset, then fit the Brainwavz ear pads over the entire earpiece body.

That took a bit of struggle. I started at the wide end (the top) and worked the edges of the earpads around with my fingers. The material does not stretch much, but it was enough to fit around the earpiece body. Once in place, I did not worry about it coming off. Just work slowly and use minimal tension to get the job done.

Brainwavz pads offer over 3/4" of room

These ear pads give the headset a different look -- much more cushiony.  But do they work?

The headset rests comfortably around the ear. It's slightly nicer than either the Koss or Yamaha headsets.  I gave them a trial run during the NAQP CW. I operated about six hours, and felt no discomfort at any time using the larger ear pads. 

Definitely a worthy modification. 







Sunday, August 2, 2020

OK, That's Not Good

Noticed today while I was mowing grass that two of the traps on the A3S/A743 -- the 15m trap on the director and the 20m trap on the driven element -- both have end caps that have come off the traps. 

This means the traps are exposed to the weather, insects and what not. 

That's not good. 

Indeed, from the ground, this antenna isn't operating correctly, either. The SWR is over 2 to 1 on 15 and 10m, and the antenna doesn't seem to be that effective on those bands, either. Only a little bit better on 20 and 40m.

The A3S has been up since 2001. I added the A743 in 2005. That's been 15 to 19 years exposed to the elements. It's not surprising this antenna needs maintenance.

I've got the gin pole on the tower, just need a couple of days at the Gwinnett QTH with a little help to get that antenna on the ground.

Sunday, February 16, 2020

SoundBlaster X G1

I'm always on the lookout for good USB sound cards. I've found a few. Some were expensive, like the M-Audio Transit. Others were cheaper, like the Startech ICUSBAUDIOMH. Both of these are 96 kHz, 24-bit, stereo input and output sound cards.

And neither of them work any more. You see, these devices, although the hardware is perfectly capable, are no longer supported with current drivers. As such, they no longer function with the latest operating systems. It's a very disappointing situation.

So, I continue my search to find good sound cards. I look for 48 or 96 kHz, 24-bit devices, preferably with stereo input. Sadly, this last function is hard to find.

The SoundBlaster X G1 seemed promising. It was advertised as 96 kHz, 24-bit, stereo device. It was only about $30, so it fit my criteria as inexpensive. I received one at Christmas time and checked it out.

The SoundBlaster X G1 comes with a TRRS mini phone jack, which should have been a clue that it didn't have stereo input. It comes with an adapter that exposes the connections as a separate headphone and microphone jack. The microphone input is connected to the ring terminal, the tip having no connection. This is apparently pretty standard for PC microphones.

When I first tested the device, I was disappointed that it only had 44.1 kHz, 16-bit input and output. At least, that's the only function it would perform out of the box. While the input appeared on the computer as stereo, both channels are hooked to the single input channel -- giving you two copies of the same input signal.

Turns out, the SoundBlaster people have created a device that has multiple "profiles" and can appear as a device with different capabilities. The default profile is intended to be compatible with the Sony Playstation, and has minimal capabilities. Using a special Windows program, I reconfigured the profile to the Generic profile.

On the Generic profile, the device conforms more closely to it's specifications. 96 kHz operation is restricted to output-only. Input maxes out at 48 kHz. But both input and output are 24-bit.

Once configured, the SoundBlaster X G1 makes a decent sound card, although it is restricted to a single input signal.

SoundBlaster also makes the SoundBlaster Play!, which appears to have similar capabilities 96/48 kHz, 24-bit, stereo out, mono in. The Play! is about 20% cheaper, but both are inexpensive.

Friday, January 10, 2020

Ingenious Fan Holder for K2/100

Binder clip. I should have thought of this earlier!
The K2/100 is a nice little radio, but when running RTTY or other full-duty cycle modes, it lacks sufficient cooling. I couldn't run more than about 30-35 watts without the KPA100 amplifier heat sink becoming alarmingly hot.

I solved this problem a years ago with a small muffin fan on top of the heat sink. The fan is mounted on tiny rubber feet. It blows air upward, resulting in airflow across the fins. Placed just above the finals, it does a good job of keeping everything cool.

This fan is noisy at full speed, so I used a small resistor in series. The resulting half speed airflow is more than sufficient.

This worked great at my Gwinnett QTH for years, where the radio was raised above the desk a couple of inches by a shelf. The K2/100 sat almost flat, and the fan stayed in place. However, when I moved the K2/100 to the parsonage QTH, the radio rested on the desk, and I used the KAT100 tilt bail. Because of the tilt, the fan had a tendency to drift to the back. Every once in a while, it would fall off.

While I was readying the K2/100 for the 2020 ARRL RTTY Roundup, I wanted to make sure the fan stayed put. I came up with a very inexpensive solution. A small binder clip attaches easily to the heat sink fins. The ears of the clip rest against the fan body and prevent it from drifting.

Worked about eight hours over two days in the RTTY Roundup, and the fan never budged. I really wish I had thought of it sooner!