Weather conditions were perfect, it was a beautiful summer evening with no wind. When arrived at launch site everything was setup within 10 minutes, the rocket was filled with 600 ml water and placed on the launcher. I pressurized the rocket with the bicycle pump, at 40 psi I listened an watched for leaks, there were none so I got on with pressurizing. Just before reaching 80 psi I heard a sudden "BANG!", I looked up and watched the rocket tumble through the air, it tumbled because it hadn't any fins on it, they were still on the launcher along with the lower bottle, the join between the bottles didn't hold the pressure.
When I got home I watched the video and tried to imagine what exactly happened. At first I thought the cap wasn't tight enough but then I found a formula to calculate the forces that act to the join (yes, I know I'd better done that before building the rocket) and I was very surprised by the results, the bottles tried to separate from each other at a force of 780 N or almost 80 kgf which, I think, was way too much for the cap to hold. I still thought it would be possible to make the rocket hold higher pressures, it just needed a way to reduce the force on the caps.
A couple of weeks later I got an idea that could work. The idea was to tie 2 cords at the nozzle, lead the cords over the top of the rocket back to the nozzle again creating 4 lines along the side of the rocket, then when the rocket stretches by pressurizing, the cords would tension and so take over some force from the caps. I had some old lines in the shed, left over from my "power kiting" hobby, rated at 125 kg they were a little overdone but for testing they would do the job. While adding the lines I realized that the rocket had to stretch somewhat to make a good seal between the cap and the washer so I added rubber rings, cut from a bicycle inner tube, in the lines.
In the pictures below you can see the lines along the side of the rocket. At the top of the pressure vessel is a piece of another bottle that keeps the lines in place.
It all seemed to work well when testing on the launcher, at 50 psi there already was a lot of tension on the lines so it was time to test it in the field. I decided to start at 90 psi for the first launch and, if there were no issues, increase pressure to 100 psi for the second. Will it work?
Launch 1
Launch 2
It worked!...