A surprising gas leak on the PC800

Tonight when I came home from seeing The Hateful Eight at our local Alamo Drafthouse, I smelled some gas. I went down and checked on my fuel injected PC800 to find a small puddle of a mix of gasoline and oil under the bike. I immediately started draining the tank while I removed some plastics to try and determine what was going on.
 
No leaks around the fuel sensor. No leaks around the fuel pump/filter/gas tank outlet. No leaks around the piping going up to the throttle bodies. No leaks on the injectors. But wait a minute…
 
When I took the airbox off, some gas sloshed out of the airbox! It turns out there was quite a bit of gas in the airbox. Some of you might remember that I connected the vent line from the high pressure fuel pump to the airbox. It appears that the airbox and air intake were sucking gas up from the pump the last time I ran the bike (a couple weekends ago). This gas stewed in the airbox eating away at some RTV sealant until it leaked. Or maybe the gas got into the engine oil via the crankcase breather hose and leaked out somewhere lower on the engine (I’ve got a slow oil leak that’s been with the bike for the last 30,000 miles).
 
I’ll do a little more sleuthing tomorrow once the residual fuel evaporates around the bike to find the place where the leak came out to make sure it was some RTV that I had used to seal a hole on the bottom of the airbox and not something else.
 
Luckily I have a California-style gas cap (the one with the hose barb) laying around that I can plumb the vent line from the fuel pump to. That should solve that problem, assuming that was the problem.
 
Always something exciting in my garage!

The complete fuel injected PC800

I made a 20 minute long video walking through all of the many modifications that I have made to my PC800 over the years including the custom fuel injection system running on a MicroSquirt ECU. I’m contemplating selling my bike or the fuel injection system to get ready for the next big project that I have in mind.  If you’re interested, please get in touch with me.

A partial list of the modifications includes:

Fuel Return or Vent Line on the PC800 Fuel Injection Conversion Project

Back in the spring, I had my PC800 up and running on fuel injection with a few minor problems.  Then a few things intervened to keep me from making further progress.  I bought a new-to-me Toyota 4runner that kept me enthralled all summer with 4x4ing in the mountains and the hose barb that I had JB Welded onto a modified fuel gauge sensor snapped off, dumping gas all over my garage.

Fast forward to the last month or so… With winter approaching, I’ve decided to do some major work to my 4runner (solid axle swap, etc etc etc).  To be able to do that, I need garage space.  Thus I’ve been cleaning and reorganizing the garage.  I took about four 4runner loads of stuff to Goodwill and a few more loads worth went to the garbage man and recycling guy.  Not too bad.  Today I finally reorganized my garage enough to have both motorcycles (my other is a 1979 Yamaha XS1100 Special) parked where I can access both, get both in and out of the garage, and work on both of them (the XS1100 needs a little work on the rear brake pedal).  Thus finally I was able to fix the fuel return line on the PC800.

Looking into the VTX1300 Honda Fury’s manual, I stumbled upon some information on page 6-36.  The rest of the manual lists a “fuel return line” which up until this point I had plumbed into the fuel tank in a variety of ways.  On automotive fuel pumps, there is generally a return line to keep the fuel pump cool (for external pumps).  Based on what I knew about this VTX1300 pump and what I saw in the manual, I thought it was the same thing.  But lo and behold!  Page 6-36 listed it as a fuel return vent line.  A little more sleuthing and I had fairly convinced myself that this is actually a vent line for the fuel pump.

To test this hypothesis, I needed to get the bike up and running.  Since the spring it hadn’t run because of the previously mentioned reasons.  Thus it was finally time to get the fuel gauge sensor remounted and leak-free.  That ended up being a bit of a process.  I had marred the gasket mating surface on the tank and the tabs on the fuel gauge were bent a bit.  I used a pair of pliers on the tabs on the gauge to bend them back and I used a ball peen hammer to slowly smooth the metal back out and push the metal out a bit to get a good seal.  It took several tries and adjustments before the seal seated properly and kept fuel from leaking out.  Until I fill the tank completely and let it sit for a while (including some thermal cycles), I won’t know for sure everything is good but at least with my checks of leaning the bike over on its side, it looks good.

I left the supposed vent line from the fuel pump open to atmosphere and cranked the bike over.  After sitting since May, the bike fired right up on the first try.  Not too bad!  I let it idle for a couple minutes in the garage and then revved the engine a few times.  No fuel came out of the vent line.  I’m pretty well convinced now that it is actually a vent line and not a fuel return line.

As a result, I decided to connect the vent line to an intake port that I had previously installed in the air box and left dead-ended.  Now the vent line goes into the air box (like the crankcase breather hose) where it can safely vent into the air being sucked in by the engine where it will be burned in the combustion process.  If I find out that the suction isn’t advantageous for the pump (like if I start sucking up large amounts of gasoline), I will instead dead-end the vent line into a little air filter that would sit up in the gas cap filler area (for monitoring purposes).

Unfortunately it doesn’t look like I will be able to test ride the PC800 before winter really sets in.  With work this week, I won’t get a chance to get out before we get a big dump of snow.  Unless we get lucky in December with some warm days to melt out the driveway (it’s perpetually in shade and gets very icy when it snows), I probably won’t be able to test ride until April.

The test ride will include two main things: 1) make sure the fuel system is all setup and okay; 2) verify that the noise I was experiencing on the ignition sense line has been fixed.  I wrapped the entire line in copper foil that is grounded to the main ground point for the ECU at one end to make a poor man’s coaxial cable.  If I still have noise, I’ll change that wire out for a real coaxial cable and ground one end of the outer braid.  The noise was causing me to get erroneous RPM spikes that were flooding the engine.

I also will do a little fuel injection wiring harness rerouting to make it easier to put the plastics back on the bike.  The main harness goes along the right side of the bike and where it goes out and around the subframe that the faux gas tank plastics attach to, it makes it difficult to align the side plastics.  I can put that bundle of wires inside that subframe and thus should have a much easier time putting the plastics back together.  Before all the plastics go back on though, I will do a test ride to make sure all is right in the world.

Someday it would be nice to get a functional idle air control on the bike (I have air taps installed for that purpose that are currently dead-ended) but it seems that I can get the bike to start pretty reliably without it.  Considering all the headache I went through before trying to get an IAC motor to work (spoiler alert: stepper motor control isn’t available on the MicroSquirt V3 board) using PWM control, it’s probably not worth the trouble at this point unless I expect to be starting the engine in below freezing weather often.

I’ve already started thinking about the next big Honda Pacific Coast project on the horizon: fully electric PC800!