The current limiting resistor was chosen to provide a reliable workable output for the FT-857D’s CW keying circuit which is 5V at 1mA. The current on the parallel port pin will be 10-15mA depending on the type of parallel port and voltage.

Please check the specs for your radio, the data sheet for the 4N28 and IEEE 1284 parallel port standards. If your radio requires less current sink than mine, please adjust the resistor accordingly. In most cases you should be able to increase the resistor value up to 2.2K depending on how much voltage and current must pass through the key.

I own a SignaLink USB interface, but that isn’t going to key CW on the FT-857D.

What we need is an interface between either our serial port or parallel port. While searching through various online fora and Google I failed to find many designs that I thought would work without purchasing additional parts I couldn’t find in my own junk box. Also few of the designs I found explained how they worked.

I chose to use my parallel port because I print over my local area network, or via USB and the parallel port is not used.

Ham Radio Deluxe, and other software like CWType (both tested) allow you to assign a pin on the parallel port as the data output pin.

The pin will go to a data high positive voltage as each dot or dash is keyed.

You can test it by putting a simple LED on the chosen pin (the anode or positive lead on the pin, and the cathode/negative shorter lead on ground, which is typically pins 18 though 25. These are the *only* two pins you need. (don’t put more than one LED, and don’t leave it in place like this – the port is only rated for 20ma)

The whole thing can be housed inside a single parallel port male connector – or like me, you will want to use something you already have in your junk box like an old parallel printer cable or DB25 plug.

The interface is made from one switching transistor, a diode, and a resistor. That’s it!

I’m using pin 3 (which is also known as D1 or Data 1 – don’t get confused).

Pin 3 is wired to a 3.3K resistor, which is wired to the base of a 2N3904 NPN transistor. Any NPN switching transistor like a 2N2222 should work fine.

The key input to the rig is across the collector (positive) and emitter (ground). I also used a protection diode, a 1N4001 (any power diode), with the anode wired to negative side and cathode wired to the positive side. This protects the transistor from any back EMF resulting from the relay coil switching in the rig.

The voltage from pin 3 turns the transistor on, which allows the keying to close and pass current between the collector and emitter. The resistor limits base current and keeps the transistor in the correct operating range.

Inside Ham Radio Deluxe (or in similar fashion for your software such as CW Type) open the Digital Master software program, select Tools > Program Options > Modes > CW and on the right hand panel, select Parallel Port Keying, select your parallel port (almost always LPT1 0378) and Pin 3 (or whichever other pin you used). Do not select PTT.

That’s all there is to it. Note some commercial products and other designs use Pin 17 – you may use whichever pin your software allows you to assign.

NOTES: for use with older radios, that use high voltage/current keying circuits – the use of a mechanical relay or some other solid state device capable of higher voltage is indicated. This design will only work with low voltage, low current solid state rigs.

This design also does not isolate the keying circuit from the computer – so the proper use of chokes may be indicated. (see controlling common mode current on a CAT cable)

Test the interface with the BK switch off (so you are just keying side tone and not transmitting).

If you find that the switch interface sticks after switching to transmit (just transmits one long CW), try increasing the value of the resistor.

A more advanced design, suitable for older radios would use something like a 4N28 opto-coupler. I will present a new design using such a device in an upcoming article.

There is an excellent DTR/RTS interface design on VK1OD’s web site, http://vk1od.net/module/DtrRtsInterface/index.htm utilizing an FOD852 which is good up to 300V.

I use my current meter to measure common mode on my transmission lines to help me understand how my antennas are performing.  (I should note that common mode current is not a universal measurement of system health, the current is a standing wave and varies throughout the line, but high measurements are indicative of a problem worth solving)

Recently, I built a top loaded short vertical antenna for 160m. I incorporated a choke into the design which I placed at the feed point to the antenna. The choke consisted of 10 turns of transmission line around a FT-240-K toroid core (available from Amidon Corp among other sources).

When I measured the common mode current at the shack,  it was higher than I would like to have seen – around 125mA when operating with my AL-572 amplifier driven with 85 Watts.

To measure the choking impedance of a choke or current balun, I short the two input leads together, short the two output leads together, and then measure across that whole thing – something like if I were measuring the value of a resistor on a digital multi-meter (DMM). (this is the common mode current path through the choke, in comparison our differential current flows in opposite directions on each individual wire and the sum effect of the choke (ostensibly) is zero (no balun will be perfect, and there are stray effects, particularly capacitance that will have some effect depending on frequency, etc).

My red transmission line is short enough, and close enough to my coax’s characteristic impedance (Zo) of 50 ohms. One could also use 50 ohm coax – but the teflon coated silver wire here is very resistant to voltage breakdown and the effects of heat. I’ve measured the differential mode impedance change through this choke, and it is minimal. (SWR is unchanged as measured with a 50 0hm resistor for a load). Keep the leads as short as possible. You can see even here I am introducing capacitance into the system – although at 1.8MHz it is less critical than it would be at a higher frequency. Do an impedance sweep from somewhere below to somewhere above the frequency of interest – and find the point where Xs is near Zero and Rs is very, very high. (the high impedances will result in very unstable measurements – however you can average several sweeps in Excel. You can still see the trend and interpolate the point of self-resonance, and where common mode impedance is greatest, despite the somewhat erratic measurement. We need a relative indication of high resistance to common mode current more than a perfectly accurate one in the this case – if you have a laboratory class analyzer, ever better). Excel is a tremendous tool, and also capable of very sophisticated charting, interpolation, trending, etc…

Here is the response of my original 10 turn choke, which I averaged impedance readings from 4 different sweeps in Excel. You can see peak response occurs too high for it to be optimal as a choke on 160m. (as a side note  air wound coaxial baluns, or ugly baluns are even less broad banded than this – randomly coiling coax is a hit or miss situation for good chokes, and at 160m even 25 turns of coax on an air core wouldn’t even come close – G3TXQ’s measurements make for interesting comparison http://www.karinya.net/g3txq/chokes/)

I also began to notice that the menu on my FT-857D kept changing into fast-tune mode, and also noticed eccentricities in the keying performance of the radio while sending CW.

I measured my choke, and realized 10 turns wasn’t giving me as much common mode rejection down on 160m as I thought – and through measurement with a RigExpert AA-54 analyzer I discovered that 16 turns gave me the wanted response on a FT-240-43 core (because that was all I had available).

This reduced the common mode current measurement on the transmission line to barely measurable levels on my 100mA scale. However, the issues with the FT-857D persisted.

IT WASN’T A COMMON MODE ISSUE ON THE TRANSMISSION LINE! (but how would I have known this without having the ability to *MEASURE* it!!!)

I targeted the computer – control line (CAT cable), and discovered that at only 100 Watts output, my CAT line had 400mA of common mode current flowing on it!

I unplugged the USB cable from the computer, and the current went away. I plugged it back in and the problem returned. Clearly, there was a common mode path between the computer and the radio.

I made sure all my equipment in the shack was grounded to the same buss.

Eventually, I took my remaining unused FT-240-43 core and wound 16 turns of the CAT cable before plugging it back into the computer.

Problem solved… Common mode current on my CAT line is now under 10mA at full power.

I know the internal jumpers are set correctly. I’ve made many successful PSK and JT65 QSOs with the interface and the FT-857D.

What I discovered tonight, is that the jumper behind the Rx pot was loose. Reseating that jumper made a huge 10db improvement in the audio levels at the computer.

I know there are a number of threads in various fora online about this low output problem.

In my case, a very light crimp with a pair of needle nose pliers to tighten that jumper did the trick.

The tuner is rated for “1.5KW SSB Power” – but these ratings are misleading for two reasons: Power given is amplifier INPUT power, not RF OUTPUT power – and the power rating assumes some range of impedance mismatch under which the device won’t fail. An extreme mismatch, even below the rated power specification, can still bring this tuner to it’s knees.

The actual power handling capability is more like 800 watts PEP output on SSB and even less for CW/RTTY/Digimodes – but even then an extreme mismatch at a high voltage antinode will result in damage. There is no way to predict any individual situation without measuring – so the best advice is to at least follow the manufacturer’s ratings.

I didn’t follow MFJ’s specification and repeatedly used my tuner at power levels up to a KW. I paid the price by destroying my tuner.

This is what happens when you break the rules – particularly with a device made by a company known for manufacturing items at attractive price points (meaning there ain’t no wiggle room for user error – it breaks easily).

What to do? Well – one can purchase replacement parts from MFJ or send the item in for repair. I made a project of it and replaced the damaged part.

My capacitors were in good shape. My switches were all in good shape. No other signs of flash over or heat damage inside the case. So I set out to find a good roller inductor – something better than anything provided by MFJ.

The MFJ uses a long, 140mm coil, which is 50mm in diameter and a 3mm pitch. Using the HamWaves inductance calculator I calculated an inductance of approximately 26uH.

The only reasonably priced roller inductor I could find was a Johnson ceramic, used, with little data advertised on EBay. I was able to deduce an approximate value for the inductor from rough measurements provided by the seller.  I realized my replacement would only be around 10uH – but at that time I was only using my tuner on higher frequencies, 20 meters and up – and the additional inductance was unnecessary.  I installed it and the new ceramic roller inductor performed well.

Here is the replacement Johnson ceramic roller inductor.

I have found I can get close to a perfect tuning on the band edges of 160m with my short vertical even without the missing inductance. However, I can not run more than 100W output at the band edges using this tuner (the input switch and the caps will flash over due to the high voltages involved matching this load on 160m). I can run the tuner’s rated power near the frequency where I tuned the antenna, which is on the CW portion of the band.

The tuner works well for my other needs on all other bands through 10m.

The repair also required a 1/4″ rod for the shaft, and a 1/4″ coupler. All available on EBay or any good hardware store.

I had enjoyed using computer control of the FT-857D, but sacrificed that option in order to plug in the relay control cable for my PA.

I looked at some options that used the microphone jack for CAT control, but these looked cumbersome and unattractive to me.

What I finally realized is that the two pins required for the external amplifier relay, TX and TX GND, are both available when the jack is switched to *EITHER* CAT or LINEAR in the menu options.

Here is a diagram of the jack with both menu settings:

TX GND is *NOT* used in CAT control of the radio – at least it is not used in the USB version of the CAT cable I am using to control my radio from my PC.

So my solution was:

1. Purchase one 8pin mini DIN male to male cable on EBay.
2. Purchase three 8pin mini DIN female panel jacks (I ended up using the version that mount on cable ends and adapting them to a bit of perf board housed in a shielded box).

The mini 8 pin DIN jacks are labeled “Radio” “Amp” and “PC”. Use short, straight through wiring from the “Radio” jack to the PC jack, but do *NOT* wire the “TX GND” pin. Next, from the “Radio” jack to the “Amp” jack wire the “TX GND” pin, and a common ground wire between all three jack’s “GND” pins.

House the above in a shielded box, plug the “Radio” jack into the rig, and the PC and AMP jacks into their respective cables.

It works a treat and I now have both CAT control, and switching of my amplifier – with no danger to the computer from switching voltages.

It is worth noting that if the TX GND were required by the computer, one could use simple power diodes (the rectifier diode assortments from Radio Shack are fine) between the amp and computer pins to protect the PC from the switching voltage coming from the amplifier. One anode goes to the PC and the other anode goes to the amplifier, and the two cathodes connect to the TX GND pin. I use this method to switch an MFJ-1025 or MFJ-1026 noise canceler and my amp at the same time.

Soldering these things can be quite challenging. The pins are extremely small and will not accept even a 24 gauge wire which is too large. I used a small soldering vise with a magnifier (these can be found for low cost at some “dollar stores”, Job Lot, Odd Job, etc…). I used stranded cable, tinned, and then “touch” soldered the ends. Since the jacks are securely fastened, the wiring bears no stress – if these were cable ends a more secure soldering method would be required. Check for continuity and shorts with your multi-meter. You can also check the data signals on your scope if so inclined.

Also note that you may have some 8 pin connectors with cables attached in your junk box. The problem is that many do not have all 8 pins wired. You may also be able to carefully remove the jacket and get to the internal wires of your existing CAT cable, but here you also risk not having all the pins available – as well as damaging the expensive CAT cable. This is why I purchased an 8-pin Mini DIN male to male cable separately.

Here’s a screen shot of Ham Radio Deluxe and the FT-857D courtesy of KC9HZC who has also constructed his own CAT cable – one could construct his cable and build a second mini DIN jack for amp control easily. http://www.kc9hzc.com/index.php?entry=entry071022-222608

Centerbrook Technology LLC will provide expert IT consulting and technical support to business and residential customers in the Centerbrook, CT Middlesex County area.

Visit our website for more information: http://www.centerbrooktechnology.com

Here’s the new Guanella type balun I wound:

After experimenting with an 80m half wave dipole, I found a shorter 31m length was a better performer on most bands for me with 13 meters of 600 ohm open wire feeders. The dipole is about 13m high. It is similar in length to a G5RV, however the open wire feed has a characteristic impedance of about 600 ohms. The balun sits outside and feeds a tuner in the shack via approximately 8 feet of coax.

I also built a quarter wave vertical for 40m and will be making a relay controlled matching network for 30m soon. There are about 40 radials on the ground presently with more planned.