I’ve been asked by a number of people about my MLA design that I’ve built for QRP operations. Therefore, I’ve decided to do a quick write up on the concepts behind my design.
First, a word of sensible warning. Magnetic loop antennas create very high voltages across the capacitor, and large magnetic fields. Enough of a magnetic field to light up a fluorescent light bulb when placed close to a transmitting magloop at qrp levels. Like all RF exposure, you (and your electronics) should have considerable distance between you and a magloop when transmitting at more than QRP levels.
It all starts with a plan. I used this excellent small magnetic loop calculator to test a number of ideas. Basically, I went as long in circumferences as I thought would be tolerable for portable operations while keeping capacitor’s limits in mind (see below for capacitor information) . The final design for me settled in at 8.5′ in total length. I decided on LMR400 for the loop element for three reasons: first, ease of coiling up for portable operations (not easy with copper pipe). Second was the fact that I gained electrical length (the velocity factor of the LMR400 is around .86). Third is that the breakdown voltage between the shield and element is high for QRP operation, north of 4,000 volts.
I didn’t want to sink money into a high voltage air variable or vacuum variable capacitor because this was initially an experiment. We’ll see how it works with the final outcome, but I went with a 1kV 22-360pF air variable capacitor on eBay. Based on my calculations using the link above, the capacitor should be good to use on 40-15 meters and up to about 20 watts SSB. Because the capacitor exhibits it’s full range with only 180 degrees of rotation, I figured a reduction gear would be necessary in ensuring accurate tuning for such a high Q antenna. Boy was I happy I bought this. This excellent option only cost me around $25 shipped. I’ll be sure to go back to them when it comes time to play with high voltage capacitors. Of course, I stopped off at Gene’s (KJI Electronics) to grab a few SO239 chassis mounts.
I tend to wander the great halls of Home Depot, looking and thinking about how to accomplish my objectives, rather than precise pre-planning. I ended up walking out with an electrical conduit box, cover, and 10 feet of 1 inch PVC pipe (including the elbow fitting, cross fitting and three “t” fittings) for about $25 or so.
The conduit box needed some of the plastic removed to allow the capacitor to fit inside of it. I used a drill bit to drill out a lot of the screw mounts, and used a Dremel to remove what little was left. A quick PVC cement job on the base elbow joint to ensure the loop support remains vertical, and we were hauling the mail. I wanted to mount this to my Manfrotto camera tripod, so I ordered this. I drilled a hole through the electrical box, and tightened down the fitting with a wrench.
I used a drill press and a hole saw to create the mounting holes for the PL259 chassis mounts, and screwed into place. Calculating the diameter of the loop from the circumferences of 8.5′ gave me a 2.8′ diameter. I cut PVC to length using a hacksaw, and hand tightened the PVC fittings into place.
Mounting the capacitor was easy, considering that the reduction gear provided mounting holes on it. You can see two small machine bolts used to hold the reduction gear in place. Coupled with the fact that the capacitor fit snugly between the top and bottom of the box, it needed no further support to stay in place.
The LMR400 coax was cut to length, and I used my handy DXengineering stripper to get to work. After folding the braided shield back, the dielectric foam underneath was cut away with a box cutter, and the braid was twisted onto the inner copper clad steel conductor and soldered. PL259’s attached and soldered (tip only) into place.
To form the coupling loop, I used some 1/4″ soft copper plumbing pipe. The coupling loop should be about 1/5 the size of the circumference of the main loop, which worked out to be around 1.7′. I bent this into shape, leaving the pipe inside the PVC T connector. I took some RG58 coax, and soldered the center conductor to one end of the loop, and the shield to the other. I used some heat shrink to ensure that the shield and center conductor wouldn’t short out.
Now that all the pieces were in place, I set everything up and tried to tune on various bands. This configuration allowed me to tune 40,30, 20, and 17 meters. Still no 15 or 10 meters, but beggers can’t be choosers. I’ll build another shorter length in the future for the higher bands.
How does it work? Well, it works extremely well, considering that this thing is less than 3 feet in diameter, and doesn’t need a mast to prop it up high. I had a chance to use it on JT65 over the past weekend, and worked all over the US and Europe on 30 and 20 meters. With the antenna inside the house. Mainly stateside contacts on 40 meters, but considering the efficiency on 40 is north of -10db down(about 1/10th of the power is radiated), I’m happy with it. All contacts were made with anywhere from 5-15 watts. Another advantage of the antenna is that it is very “quiet” on receive, and that the antenna displays deep nulls (upwards of 30 dB), allowing you to null out local RFI. As far as its receive/transmit pattern goes for skywave RF, it appears to be pretty much omnidirectional.
Extremely small (considering the size of normal HF antennas), lightweight, and really packs a “wow” factor – the magnetic loop is a fun experiment and an excellent choice for portable operations, apartment dwellers, and HOA restricted hams.
-Rob Fissel, K2RWF
I have just read your write-up. In fact I read it twice. I really liked the content and your writing style. Thank you for sharing your walk through Home Depot.
Any update as to when the next batch of Mag Loops will be available? Your web page says Fall 2021 so just wondering. 73 de WA1LBG
Available now. See updated magloop page
Great write up! I love these types of websites. I would just query the idea that the secondary loop gains in any way from the velocity factor of the coaxial cable. Others may differ, but as I see it, the coaxial cable screen is being used as a linear conductor arranged in a circle and not as a transmission line. I don’t believe that the velocity factor is relevant in this application.