beehphy

https://www.aliexpress.com/item/1005001438132912.html using https://github.com/mdkendall/NanoVNA-QT

My results:

Update: I made updates and was slow to upload. my new build does not have your cable feed design, but I plan to test differences. I cut a new radiator in brass at the correct size, 68mm. I machined a small piece of copper pipe to reach from SMA to radiator. I recut the original radiator down from 65 to 54mm, to be the first lens disk and closed the original feed spot with aluminum tape. The reception may be slightly better now, maybe, but the screw attachment method seems to have been working okay. I don't have a 3Ghz SWR meter to test this antenna yet (in the mail). In the mean time, I built an antenna testing robot (azimuth 160d, elevation 90d, polarization 360d). It has a GRBL machine controller and nRF24L01 radio system. The 24L01 radio does not have RSSI, but I created a ping reply program to 'Test' the link quality. A python script on a computer aims the robot and stores samples. Later I can analyze in a spread sheet (pic: data of bowtie scan). I plan to test my latest yagi build then convert to coax feed design as you specified above. I hope to quantify the differences. @ Admin, in that white antenna you pictured, what does the electrical connection to the radiator look like? Are the coax feeds connected to different surfaces of the laminated material? (side -> front, top -> back) or are front and back metal sheets connected? (side + top -> front + back)

@Admin: Would single sided copper FR4 function here? or do I need to purchase a ~1mm sheet of brass? How much effect will the fiberglass have (I assume it acts as a lens) and should any dimensions change? Should the FR4 be between the radiator and cup or should it be facing out? @clanon: I may have misspoken on the type I built, maybe double bowtie? After writing, I looked back and noticed that, the maximum for the disk yagi should be 11.5dbi, where the bowtie claimed 13.5dbi. So yes, it should out perform the yagi. I found plans for a single bowtie and just doubled it. I think the original design called for 48.3 x 37.8mm diamonds and a 15.3mm spacing from the reflector. I provided measurements of the thing as built.

My rendition... I accidentally cut disks from the wonky plans on page 1, but later found the improved design and was able to salvage the build in the following manner: Double clad 1.6mm FR4 backplane, with 0.4mm double clad FR4 flange, interior D90 x H15mm Disks cut from 1mm AL sheet. I have a lathe so all diameter are .03mm accurate. Built disk sizes are 65, 52, 39, 37, 37, 37mm. The improved design calls for 68, 54, 38, 37, 37, 37mm. I found 10-32 (sae) rod and nuts in the shop, and that woks out to ~4.7mm rod. Cut down an M3 brass standoff with a small saw, used a screw to hold it at the correct height when soldering to the SMA feed. Used copper washers on the radiator layer to get 5.75 mm spacing. The rest of the disk heights were set by nuts (T3.2mm) on both sides. How bad did I mess up the gain by using the wrong sized disks? I built a double bi-quad that performs significantly better than this yagi. Is it worth re-cutting the radiator and first 2 disks? BTW, Admin, thanks for all you do on this site.