Salt Water Dummy Load

Posted on May 8, 2018

After about the fourth time of wishing I had a dummy load, I decided enough was enough and that I needed to build one. After searching for plans for a dummy load and seeing page after page of people putting a bunch of 1k resistors in parallel, I came across an article describing a salt water dummy load. I asked the NFARL TechNet what they thought of it, and I got some mixed responses, so I decided to do the experiment and build one myself.

What is a dummy load?

(For those already familiar with dummy loads, please skip this section). In esscence, all a dummy load is is a 50Ω resistor connected across an rf connector. This is extremely useful when testing radio equipment, as nearly every piece of rf equipment, including transcievers, filters, wattmeters, and coaxial cable, expects to be connected to a 50Ω load. This makes it hard to test these pieces of equipment without a 50Ω load, as some equipment, such as filters and wattmeters will give incorrect measurements, while others, such as transmitters, can even be damaged by an improper load.



For my salt water container, I used the only plastic bottle I could find in my recycling bin, this tall sauce bottle:


To attach my SO-239 connector, I first drilled two holes in the cap of my bottle, one for the center pin of the SO-239 and one for one to connect to the flange on it. I then soldered two wires to the center and flange of the jack, before passing them through the holes in the cap and soldering the wires to two thicker wires. The result is shown below:

Then to prevent the salt water from leaking out or evaporating, I sealed the joints with plumber’s caulk.

The finished bottle before adding the salt water:

Salt Water

Initially my plan for adding the salt water was to use a multimeter to measure the resistance of the water until I reached around 100Ω, then to use an SWR meter to do the final adjustments until the water resistance reached 50Ω. Unfortunately, this didn’t work as the multimeter didn’t seem to do a good job of measuring the resistance of the water, probably due to the formation of a passivation layer on the electrodes via electrolysis. This meant I would overshoot the 50Ω resistance while my multimeter read 2kΩ or more. Eventually I switched to just using the SWR meter, which worked a lot better.

Final Procedure

To adjust the salt solution to be a good 50Ω match, first fill the container with water, leaving a bit of space at the top for expansion, and add a pinch of salt. Then with a transciever on its lowest power level, insert the electrodes into the water, and watch to see where the SWR meter dips. If the SWR reaches a minimum, then rises back up as the electrodes are inserted, then there is too much salt in the water, and some should be poured out and replaced with fresh water. If however, the SWR reaches a minimum when the electrodes are fully inserted into the water, and the SWR is farther from 1 than desired, the water doesn’t contain enough salt and more should be added with stirring.


The bubbles shown are most likely due to dissolved gasses in the water coming out of solution due to temperature increase

The salt water dummy load turned out to be a decent dummy load for the lower frequencies of HF (<14MHz). However, it seems to have a small reactance at higher HF frequencies and above, making the SWR never reach below 1.1 in my setup. I would hook it up to a VNA or antenna analyzer, but I doubt KSU wants me hooking up a bottle of salt water to their $10,000 VNA. On the plus side, it should be able to handle a decent amount of power without overheating. After running 100W into it for a couple of minutes it got a little warm, and by my calculations, it should take 20 minutes for the temperature to increase from 24°C to 90°C at 100W, which is not bad. Additionally, I had no problems with electrolysis or chlorine gas generation, likely because the RF is AC.

SWR vs. Band

Band SWR
160m 1.1
80m 1.1
40m 1.15
30m 1.2
20m 1.2
15m 1.3
12m 1.5
10m 1.8
6m 2.5


If you need a cheap dummy load for 40m or 80m, and want to be able to run a decent amount of power without the dummy load overheating, this seems like a reasonable option.