Sunday, May 12, 2019

Red: Epic Tank Fail


Today was more about recognizing when you're about to step over that line from bad to worse.

Session started out okay.  The tank had dried nicely but I could still see some black coming through the white coats so I added a few more layers before doing the top coat.

Between coats, I changed Ivy's oil (but not the filters).

Then back to Red... went ahead and added 125 mL of fork oil on each side.  In doing that, I haphazardly went about torquing both the fork bolts and the main steering bolt.  No issues there but realized immediately after that something was wrong and that the forks were not moving at all.  They literally seemed frozen in place.

Tried to reset at that point.  Drained the oil from both sides (the left side drained in a solid stream, but the right only dripped, not sure why yet).  That changed nothing.

Added in the jack stands and started unbolting the forks themselves.  Got the left side undone from the fender and axle and found it moved fine.  Undid the right side and realized that was the one that was causing the problem.  I could get it to move but only with a lot of effort and it would keep getting "stuck" in place.

Undid the lower fork bolt but it was only when I undid the top bolt that everything came loose and started working again.  I must have torqued it all out of order and caused some kind of bind.

Redid everything in the right order and problem solved.  Added back about 100 mL on each side (forgetting that the right side probably still had a bunch stuck in there).

And I thought that was gonna be the hard part of the day...


Then a kind of summit fever hit.  Realizing I was so close to being done, I decided to push for getting the tank installed and petcock hooked up.  There was still more than enough daylight to get everything done and take her out for a maiden run around the block.

Chain of events went something like this...
  • The tank dropped into place nicely.
  • Installed the dual-inlet petcock, no problem.
  • Ran a 3/16" line from the vertical inlet to a gas catch.  Thought maybe I should do the secondary inlet, too (and lucky I did)
  • Poured a cup of gas into the tank (petcock still closed) to test whether I had any leaks out of the base of the petcock itself.  So far so good...
  • Turned the petcock on and here's where all the trouble started.  With the petcock on, I got flow coming from both lines, which I was not expecting.
  • I thought, no problem, I'll just run the secondary inlet to the actual plugged side of the tank and essentially render it closed, but when I did that, I stopped getting flow from the main inlet.  I think it must create a vacuum lock or something.
  • This is exactly the point I should have stopped, packed up, and thought through the problem carefully, slept on it and come back the next day.  Instead, I got stubborn and all kinds of impatient.
  • I tried a couple more combinations hoping to find a way to get flow from just a single inlet, but nothing doing.  Should have been a little more thorough with testing because now I can't even remember exactly what I tried.  Plus, I only tried it on Reserve.  Maybe on plain "On" it would have worked???  I just didn't have enough gas in the tank to try it.
  • So instead of stopping, I thought I'd try to hook up the old single inlet petcock.  But made a gigantic toxic mess as I tried to somehow detach the dual petcock without spilling the remaining gas all over the place.  
  • I succeeded in getting the single inlet one on... only to realize that it was that it was so wide that you couldn't mount the tank with it already installed.
  • That's the point I called it a night.  Made even more of a mess trying to slosh out the remainder of the gas into a bowl.  Screwed the single back in and left the tank open.
  • Ended up smearing some paint on the bottom left of the tank and all the gas dripping out basically melted away everything on the right side near the petcock.

Lesson(s) learned.

First lesson: I should have done ALL of this petcock testing before painting the tank.  That was just plain hubris.

I should have been more patient when first testing the dual-inlet.  I would have realized I could have tried the "On" because obviously that would NOT have required gas flow from secondary inlet!!

I did end up ordering a single inlet as soon as I got back up, which is probably the right way to go anyway.

Do I fix the paint while I wait for it to arrive, or not even bother till I test everything completely?










No comments:

Post a Comment