QRP for the FTdx5000

After selling my Elecraft KX3 earlier this year (as planned) I was left with one rig that could be dialled down to 5 watts for entering contests in the QRP category. That rig is a Yaesu FT950, a rig with very dated technology. It used to me my main transceiver for most operating. Although I still enjoyed using the KX3 its use was solely for QRP contest entry after returning to using higher power.

Because its receiver is quite poor I avoid the FT950 other than as a second radio for SO2R contest operation. Until I upgrade shack equipment to be more contest friendly I use it to practice SO2R. Eventually it'll be replaced.

My main rig is an FTdx5000. This 200 watt transceiver cannot be reduced to QRP power level. Its minimum setting is 10 watts. The problem was that I intended to enter the ARRL CW Sweepstakes contest in the QRP category this past weekend. I couldn't bear the thought of using the FT950 all weekend.

I also wanted to improve my modest SO2R skills. This contest is great for that since the pace is more measured and with QRP I don't have high hourly rates beyond the first few hours. I also wanted to improve on last year's score when I might have won had I operated full 24 hours.

My objectives meant borrowing a transceiver or fitting the FTdx5000 into a QRP box. I opted for the latter. Aside from what ought to be the obvious solution that I eventually selected a few other paths were investigated:

• Class A: Running the transmitter in low distortion and low efficiency class A reduces the maximum power to 75 watts. It turns out that adjusting the power does not proportionately decrease the power by ⅝ at all level. If you set the power to 10 watts you still get 10 watts, but a supposedly very clean 10 watts.
• External ALC: There is a negative feedback port on the rear panel to allow an external amplifier to dynamically reduce the power level when it is over-driven. I tried it with a AAA battery in reverse polarity -- the port accepts from 0 to -4 volts. Nothing happened. An internet search didn't solve the puzzle. E-ALC was ruled out since time was running out.
• Lossy coax: Unfortunately I had none suitable that I could connect. In any case the loss would be frequency sensitive.

Not surprisingly I went for the brute force solution: an RF attenuator. It's a simple project that took me only two hours to whip up in the workshop. It has just a few design parameters:

• Knock 10 watts down to 5 watts from 1.8 to 30 MHz.
• Dissipate 5 watts of heat, preferably with 100% duty cycle for ultimate robustness.
• Good impedance match to 50 Ω.

The circuit diagram for 3 db attenuators with 50 Ω ports is easy to find. Calculation is more difficult. From an old ARRL Handbook the series resistance is 18 Ω and the parallel resistance on each port is 300 Ω. These are exact or close to standard resistor values so I went digging in my junk box for suitable candidates. Then it was off to my workshop.

Before visiting the junk box I did some calculation to determine that most of the power dissipation is in the parallel 18 Ω resistance. This isn't surprising since it is by far the lowest resistance in the network, ports included. I didn't bother figuring out the exact dissipation values since close is good enough for this application and solving resistor networks isn't fun.

Construction is incredibly simple. The biggest job was punching the ⅝" holes for the SO239 panel jacks. I used the cheapest jacks I had on hand. The box is new from my stock of plastic and aluminum boxes that I keep on hand for home brew projects.

Although simple there are a few points worth mentioning:

• The box should be conductive to prevent RF leakage in and out and to minimize frequency sensitive impedance variation.
• The wire between the jack shells further improves high frequency performance. It's better than solely relying on the enclosure. Notice the short leads of the parallel resistors the wire makes possible.
• Carbon composition is the best choice with carbon film close behind. For HF use other resistor types can be used but never use wire-wound resistors.
• Resistance values can be achieved with multiple resistors in series or parallel. With the help of an ohmmeter you can get very close to the exact values. Resistors often deviate from the specified value (pay attention the tolerance colour band) so try them all. Multiple small size resistors can have the power dissipation rating of a single high power resistor which is less likely to be found in the junk box.

After soldering it together I plugged my antenna analyzer into one port and a 50 Ω load into the other and swept the HF spectrum. It's very flat with a maximum SWR below 1.1 up to 30 MHz. I didn't test it at 6 meters and above.

The final test was to connect it to the rig and place watt meters at each port. Since I have only one standalone power meter I relied on the rig's PO meter for the transmitter port. Power was reduced to 10 watts and tested at a few frequencies.

The result was very good. Since power meters are often inaccurate I compared the meter reading with what I recall from using my KX3 at 5 watts. Both readings are a whisker above the 5 watt tick. The difference expressed in decibels would be exceptionally small. In any case the true output of any transmitter is never as precise as suggested by the digital display.

As a further test I opened the box and ran the transmitter with a solid 10 watt carrier for a few minutes. There was no smoke coming from the resistors nor was there obvious heat radiating from them. The series resistor pair was a little warm.

There are a few items to keep in mind when operating with an attenuator of this type:

• If your rig has a physical power level control it is easy to forget the attenuator is there after you're done with a QRP event. Transmit at 200 watts and you'll have to replace all the ruined resistors. Look for the software parameter to limit the maximum power to the lowest setting. On the FTdx5000 that is 20 watts. The attenuator will survive dissipating 10 watts.
• Receive strength is reduced by 3 db. On HF you may not notice, more so during the present solar minimum when you are not often on 10 or 15 meters. If it's a problem increase the rig's pre-amp setting. Modern receivers have an abundance of pre-amplification features.
• You don't need a tuner! The increased return loss due to the attenuator greatly lowers the mismatch seen by the rig. For example, my poorly adjusted 80 meter inverted vee has an SWR of close to 3 at 3500 kHz and requires the ATU. With the attenuator the SWR was not much worse than 1.2 so the ATU was switched off throughout the contest. 
• Despite the low SWR seen by the transmitter the SWR at the antenna port is unchanged. Large deviations from 50 + j0 Ω will change the attenuation level to a value lower or higher than 3 db. For high SWR you should use a tuner between the antenna and attenuator to be certain that you are below the QRP power limit.

The attenuator worked perfectly during Sweepstakes. I was going to discard it afterwards (recycle the components) then reconsidered since there are other contests in which I'd like to have the option of operating QRP. One possibility is the Stew Perry TBDC on 160 meters this winter. I'll keep it around.