Completing the right fuel tank
TLDR
- Conducted fuel leak test multiple times (Right tank)
- Fixed leaks (Right tank)
Details
This is tracking a lot of small actions across a long month.
I sealed the right fuel tank at the beginning of August, and waited a week and a half to do the first round of leak test. Got a leak, tried different ways to find the leak, fixed, waited, then tested again.
Test setup
Many people build their own manometer to test air preassure change in the tank. I hear it’s more accurate but I never quite figure out how to build the jig. So I decided to go with pre-made PSI gauge.
I bought all equipments from Amazon:
- Clock with capabiilty to show temperature
- AN4 end cap
- An6 to 1/4 NPT fitting (need 2)
- Low reading pressure gauge, measing up to 3 PSI
- 1/4 NPT bike valve (need 1)
- Bike tyre pump
Set up:
- Use AN4 end cap to seal the fule overflow port
- AN6 to 1/4 NPT connect to both fuel return fitting and fuel inlet fitting
- Bike vavle to 1/4 NPT
- Pressure gauge to 1/4 NPT
Process:
- Pump air into the tank until the gauge reads 1.0 PSI
- Disconnect the pump and wait a few hours
- Pressure remain the same after it stops fluctuating in the first few minutes
Incident before first test
When I tried to connect the gauge to fuel return fitting, it does not screw into the fitting nicely. The threads felt like it’s jammed.
I asked Sling technical, and they suggested to get a new return fitting or rethread them. I don’t know what rethread means so I ordered 2 new fuel return fittings (and of course they still haven’t arrived when I write this blog).
I then started to research rethreading, and bought (mistakenly) an AN6 rethread die. It looks like a regular bolt, and when you thread it to the fitting it cuts away a bunch of metals from the fittings old thread and forces a clear channel for the new thread.
I tried to use it on the fuel return fitting, the die immediately jammed and required a huge force to turn. I understand how it worked but really was not comfortable turning that hard.
I returned the rethreading die resumed my research. And apparently there is something called thread chaser. It works in similar ways as a rethreading die, but is much less intrusive. I bought a set of thread chaser from Amazon. When it arrived, I first tried on a good AN6 fitting. The chaser die turned onto the good fitting smoothly, and did not cut anything. I was happy with the result, then I started to use it on the return fitting. It shared away so metal, and then I was finally able to install the preassure guage.
If anyone run into similar issue, DO NOT buy rethread die. Try a thread chaser first. If it doesn’t work, then consider rethreading. But don’t do it form the get go!
The test resumes.
First result: leak
When I first set up the process, I only pumped to 0.5 PSI as a trail run. It’s immediately clear that I have a leak. The pressure drops to 0 within 20 minutes or so.
Following what most other people do, I mixed a bottle of soapy water and sprayed to every fitting and every rivet on the tank. I could not find where the leak is.
I tried a few different ratio/mix of soap and water over 2 days period. None of them worked.
Frustrated, I bought a few new tools to help: fluresent dye, UV light and a product called snoop.
Snoop turns out to be answer. It’s some kind of premade soapy water and can detect tiny holes.
I sprayed the snoop juice on the tank, and within 3 minutes I found a huge of bubble coming out of th fuel fitting.
Looking at the fitting, I don’t know what I was thinking. I did not put proseal around the washers!
Fixing the fitting is standard. I mixed a bunch of proseal and smeared it to both inside and outside of the tank, covering the washer and amost the bolt entirely.
Gasket upgrade
The fixed required removing the sender cover and the gasket, so I had a chance to upgrade the gasket.
A fellow builder Greg is nice enough to manufacture 2 new gaskets using Garlock 3000 material for me. The stuff is much more durable than the cork material that came with Sling factory kit.
When I removed the old gasket, it’s clearly not resuable. Many edges had already cracked, and the holes were enlarged.
Second test
After having done all these work, I started the second round of test 2 days ago.
Friday afternoon 6:30, I pumped the air to 0.95 PSI. The gauge dropped to 0.90 within 15 minutes.
- Theory: air coming from the pump is heated so it takes a while to stablize the pressuring.
Over the next fiew hours, the pressure kept dropping. By 11:30pm, it dropped to 0.8 PSI.
- Theory: Night temperature drop significantly affected the pressure in the tank.
Next morning 7:30am, the pressure was 0.6 PSI.
- Theory: Same as above, overnight temperature and pressure fluctuation
Afternoon 6:30, the preassure came back to 0.95 PSI.
The third day, the pattern repeated and I ended the test at 7:40pm, with PSI AT 0.93 which is the same value as the day before at the same time.
Pressure change over time
Installing on the wing
With the results above, I concluded the tank is good.
I clecoed the tank to the wing with some help with my wife, and we riveted it on the next day.
I chose to use S.S. rivets near the wing root 50cm length out of precaution. This is sugggested on S.B 14 for Sling 2 but not for Tsi. I don’t think it hurts though, so I just did it. I don’t have duralac or tel-gel as anti-corrosion agent on the steel rivet, so I just dabbed them in proseal. This is acceptable per the KAI.
With everything done, I moved the wing onto the rack. Time for next wing :)
