First long-distance HSMM-MESH link

Last week I was able to find some free time to attempt my first "long distance" MESH link.  The link was from my house to the hill next to Nielson Hall at Kirkwood Community College.  I finished modifying two routers to be Part 97 devices by changing out the reference crystal and mounted them in a couple of surplus Transeo TR-CPQ-19 CPE antenna enclosures.

For the link setup, I placed one of the Transeo enclosures on my front porch, bungeed to keep it stationary as I didn't have a second tripod available for this test. For the location at Nielson Hill, as I nicknamed it, I had a tripod with the other Transeo enclosure attached and did my best to aim the patch in the direction of my house:

To get the best chance of a link, I left both the routers set to +19dBm initial output power (which translates to 6.275W EIRP on both ends).  I haven't come up with a good method of aiming the flat patch antennas yet, so I took my best guess.

Lo and behold, it connected right away, average of 12dB SNR margin, not bad for the first attempt:

I then began to decrease the power level on my end of the link to see how far I down I could go and still maintain the link.  Decreasing in 3dB steps (cutting the power in the power each time).  Here is a screenshot at the 13dBm output level.

Then something unexpected happened, the link completely died.  With no-one watching the other end of the link I'm suspecting that the antenna shifted such that the pattern was no longer lined up with the hill.





This test was mainly to test the distance link before heading out to the Pigman Triathlon race, where we're planning on using this equipment to create a link that will be about 0.75 miles in distance.  I am now comfortable saying as long as we have line of sight the link should work just fine.