Tuesday, March 31, 2009

Schematic for the Step-start

Dennis N2RIT wrote to ask me about the schematic for the Step-start circuit. Well, Dennis, it's pretty simple. I used the design posted by Rich Measures AG6K here. For the AL-80A, I used the +12 volt supply, which is about 15 volts unregulated. The coil of the relay I used is about 360 ohms, so to get the voltage about right, I used a 100 ohm dropping resistor.

I also added a 1n4148 diode across the relay coil to absorb the back EMF when the relay opens. Probably not needed, but it seemed like cheap protection. 

Friday, March 27, 2009

Step-start for the Ameritron AL-80A

For many years, I did not own an amplifier. I ran barefoot at 100 watts. You can work a lot of stations with just 100 watts. When I contest, I generally enjoy competing in the 100 watt category. But, a four years ago, I bought a used amplifier.

It had been owned by one of my fellow contest club members, K4GA. I didn't know Archie that well, aside from the e-mails we exchanged on the club reflector. After he passed away, his widow wanted his radio equipment to find a good home. I made an offer on the amplifier, and it was mine.

K4BAI later told me that this Amplifier had a bad trip to Barbados once -- it had been damaged in shipping, but had been repaired by insurance. When I got it, it needed a little TLC. The cover was on backwards, and several of the cover screws were missing. The meter switch had been replaced with one that had a shaft about an inch too long. 

I fixed the meter switch with a few minutes work with a hacksaw. I bought all new screws and put the cover on the right way. The meter reading for the high-voltage was a bit low -- this turned out to be a problem with one of the divider resistors. I had much higher-quality replacements in my junk box. That fixed, the meter for high voltage read exactly as it should. The open-frame antenna relay would sometimes leave the receiver antenna disconnected. This was easily taken care of with a little contact cleaner.

This amplifier has given good service in the last four years. I've used it pretty heavily in RTTY contests, running about 400-500 watts out. I've had only a couple of complaints.

One was the jarring THUMP that would sound when I switched the power on. It didn't happen every time, but often. All that inrush current couldn't be good for the power supply components. It didn't take long to figure out I needed a step-start circuit.

Step-start is simple. Low-value resistors are placed on the main power leads. They limit the inrush current when the unit is switched on. Once the capacitors charge to a certain point, the resistors are shorted out by a relay. For the AL-80A, this is easily accomplished with two 10-ohm 10-watt resistors. Selecting the relay was a little trickier. I found a nice 12 volt relay, an Omron G2RL-24. This is a DPDT relay with contacts that can carry 8A at 250 volts. This sealed relay was only about $3.00 from Mouser Electronics. 

I designed a simple circuit board for this project. It was also my first experience with TEC-200 film. The first
 version of the board didn't come out too well, as I tried to flood-fill to leave as much copper in place as possible. The result was pretty ugly, because all that toner didn't stick well to the board. I probably didn't have the heat setting right. I redesigned the board without the flood-fill, and I also beefed up the size of all the traces. Getting the TEC-200 film to transfer the toner with just an iron is going to take more practice, but I'm pleased with the results.

Note from the design -- I layed out this board in 2007, but I didn't get a chance to build it until recently. I will admit that I drilled some of the component holes a little large. I'll have to remember to use the smallest of my numbered drill bits next time I make a circuit board.

The one nice thing about designing your own board -- it's guaranteed to fit your parts. There's only five components on this board, and it goes together with a few minutes of soldering.

Getting the board in the AL-80A took more doing. For one thing, it is heavy. Moving it around is not easy, and must be done with great care to avoid damage to the amplifier and also to myself.

The next problem was getting in to wire the board up. The power connections are on a barrier strip that's close to the power transformer. Fortunately, one can remove the screws and unsolder a few connections and the back panel lays down flat. 

Figuring out how to connect the board was tricky. Like I said, this AL-80A had a bad trip to Barbados. The original transformer had been replaced with an AL-80BX transformer -- which also has buck/boost windings. 

I ended up hooking the board from the connections from the fuses to the barrier strip. 12 volt power from the auxiliary jack drives the relay. 

While I had the amplifier open on the workbench, I also added a glitch resistor to the B- lead from the rectifier stack. This is a 10-ohm 20-watt resistor. If the tube were to become gassy and short out, the glitch resistor will help dissipate the energy stored in the capacitor bank. I borrowed a couple of unused lands on the rectifier board in order to mount the glitch resistor.

Buttoning it all back up, then came the smoke test. Fortunately, I kept the smoke in. Step-start works great. No more loud thump.

Tuesday, March 17, 2009

The Tribander Experience

Most hams start off on HF with modest antennas. Perhaps a simple wire dipole, or maybe a trap vertical. These simple antennas can work, and often work well -- when installed at the proper height and with the appropriate number of radials, respectively.

But those starting off often don't have the experience to do things quite right. My first ham antenna was a simple 40m inverted V -- it followed the roofline of the house on 6" standoffs. The apex was all of 25 feet up -- and the ends were only a couple of dozen inches from the ground. It worked, but not well. With the 50 or so watts I coaxed out of my novice rig, it did OK. I also tried various dipoles strung between trees and buildings, random wires, even a vertical made out of a slinky.

Anyone who uses these simple antennas often dreams of something better. I thought that the guys with the tribander at 50 feet had the high-end installations.

When I bought my own house, I wanted to put up some good antennas. First was a 300 foot longwire at about 15 feet high. Fed with an L-network, it could load up on all bands -- even 160m. It did not work well. For a while, I used a "Loop Skywire" -- a 80m full wavelength loop positioned horizontally. This was about 15 feet up -- just barely higher than the longwire. It worked OK, certainly better than the longwire. But really, none of these were any better than my novice antennas.

Somewhere along the line -- I had an epiphany: for horizontal antennas, the most important single dimension was the height above ground in wavelengths. I built an 80m dipole and got it up in the trees about 45 feet high. This antenna worked great -- much better than the Loop Skywire -- and it only required two supports instead of four.

I eventually put up a beam. First was a Butternut HF4B. It was mounted on a roof tower at a height of about 35 feet (10m). Certainly not optimal for a tribander. I eventually replaced the HF4B with a Cushcraft A3S. I've written about this antenna before -- it is probably one of the best of the small trapped tribanders.

When I moved to my current QTH, I decided not to repeat the roof tower experience. It took nearly seven years before I could save up enough to put up the tower. In the meantime, I used a number of dipoles at successively higher heights, a trapped vertical (the Cushcraft R7000), and even a couple of two-element delta-loop wire beams in the attic for 15 and 10m.

During this time, I had the fortune of being able to guest op at W4AN's superstation near Dahlonegah, GA. NQ4I also invited me to come and operate at his Multi-Multi station. These stations have multiple mono-banders for each band, often at heights much greater than your typical tribander-at-50-feet, and many times stacked mono-banders to certain areas. Operating at a super-station is pretty amazing.

Moving from a simple dipole or vertical to a tribander is an eye-opening experience. With a tribander, the band opens earlier, stays open longer, you have directivity that can bring stations out of the noise or null out unwanted signals. Bands you thought were dead come alive with signals. Pileups that were too big and crowded with a dipole are easily busted with the tribander.

It's curious that going from the tribander to monobanders or even stacked monobanders isn't as dramatic as the shift from a dipole to a tribander. Using a monobander or stacks is much like the tribander, only better. It isn't a sea-change.

Not every ham can afford towers and stacks, but every ham interested in HF ought to consider putting up a modest tower with a tribander. It will make a huge difference.

Tuesday, March 3, 2009

Amateur Receivers circa 1950s

I recently read with some interest K2TQN's column about the HBR Receiver in the February issue of QST. I had never heard about this receiver until this column -- this design likely pre-dates my involvement in amateur radio by 10-15 years. 

When I got into ham radio in the early 1970s, the transceiver was beginning to come of age. In the 1960s, SSB had replaced AM as the primary mode for voice operation. Transmitters and receivers for SSB have many of the same circuits -- so it made sense to share these in a single box. Before then, most hams had separate transmitters and receivers.

Looking at the HBR receiver stage descriptions, the basic design seems foreign to our modern designs. Selectivity is provided by high-Q LC circuits in the low-frequency IF near 100 kHz. But such a low IF frequency wouldn't result in very good image rejection, so a first IF near 1700 kHz comes first. 

Now, in the 1950s and 1960s, the amateur bands were simpler. 160m was covered up with strong LORAN A signals and was virtually unusable. The 30, 17 and 12m bands hadn't been invented yet. So, most radios only needed to cover five bands. The HBR accomplishes this through plug-in coils. 

From a modern view, the lack of bandswitching and sharp IF filtering stand out. It's hard to imagine changing bands by opening up the top cover and swapping out coils -- particularly when a couple of them have potentially dangerous plate voltages on them. Perhaps a band switch was something of a luxury.

The IF filtering confuses me. My little 40m receiver has excellent IF selectivity from just four crystals. Hams in the 1950s could have built ladder filters in their receivers using crystals from 2-10 MHz. Although, I'm not sure the ladder filter design was invented until the 1970s. Perhaps the cost of the crystals was prohibitive -- crystals are often expensive, even today.

In the same issue of QST, there was a reference to an all-transistor receiver design by W2TGP. I looked up the article in the ARRL archive. It was very odd to see a rig using all PNP transistors, especially with a +12 volt supply. (The power goes to the emitters, and the collectors connect to ground -- somewhat opposite of today's convention)

Other than the use of transistors, the design is a curious mix of older and modern elements. The rig is small, so there's plenty of room for a bandswitch for five bands. Selectivity comes from a 455 kHz mechanical filter --  455 kHz is a bit low for good image rejection, so the first IF is around 2000 kHz. 

This transistor receiver likely outperformed the HBR in terms of selectivity, but the HBR likely had better dynamic range. The single-ended mixers of the 1950s can't hold a candle to even simple Gilbert-cell mixers today.