53000 miles: Fuel leak and carb clean

Yellow petrol residue on the alternator

I noticed I was arriving at home smelling of petrol. Although I find a bit of a petrolly whiff has its own appeal, it is not something my wife appreciates. There came a point when it was obvious that something needed to be done. That point was seeing petrol dripping onto my alternator. Not good.

I thought it was due to a faulty float bowl seal on the carburettor for number two cylinder so it was a case of getting the bank of carburettors off the bike and giving them a bit of a birthday. I figured the air filter could do with a clean anyway and, after 53000 miles, I decided it would be wise to change the cam chain (next post). So Ruby was going to get a big birthday.

24 years and 52,995 miles

As it happens, I think it likely that she was made in July or August in 1991. I don't know for sure. She was registered 1st of September 1991 but with chassis number 544, production having begun in March and building up to a rate of about 50 machines per week during the year, my guess is that she was constructed any time from mid June to Mid August. So happy 24th birthday, Ruby. She would be having a reasonable run out afterwards because I had an international trip to do for work, flying from Heathrow, and I had decided to ride there, stowing my bike gear in left-luggage for the duration of my time abroad. FWIW, that trip worked out well - free parking for the bike and £20 for left luggage.

So the full list of jobs were: diagnose and solve fuel leak, clean out carb float bowls, clean and re-oil air filter, change air box for a better second-hand part I'd bought from Trevor, change cam chain and check valve clearances. The first four are here.

I have found that the top of the motor tends to accumulate road grit over time. It surprises me because I have inner panels that sit in front of the exhaust pipes and fill the gap above the oil cooler and below the water radiator. Anyway, it is helpful to brush off the cam cover before removing the plug leads. I remember I'd got the coil low tension connectors muddled up when I first tried to restart the engine back in 2012. So when I removed the coils, was very careful to keep note of which connectors belong to which coil. 

It is always a bit of a fight to get the carbs out. There is not much room between the airbox and the inlet stubs, and the rubber is ageing. I don't rush at the job, straddle the bike so I can get a good grip on both sides, and put a block against the front tyre to guard against the risk that I'll roll the bike off its stand in my pulling and pushing. The air filter was satisfyingly dirty. 

I have had a bottle of RAMAIR foam filter oil for years. I got it for my old GS650 Kat. It has a solvent that evaporates to leave a tacky coating. Good stuff. It just needs engine degreaser to clean it up. I was lucky enough to have a sunny day so the filter dried out quickly and then took the RAMAIR oil very well. 

 The old air box was severely cracked and my attempt to repair it with filler rod and a soldering iron had finally given up. The replacement box was in great shape and, I realised, had a slightly different design from the broken one. It has a plastic frame that retains the filter and this frame has recesses for four long stainless screws that clamp the sides of the box.

Replacement airbox, with filter frame including
recesses for through-bolts on each side. 

 The carb float bowls contained a fair amount of a red dusty residue. The seals showed a bit of alloy fur but nothing that looked like a leak. Perplexing but I cleaned them up and reset the floats 1mm higher than before (15.5 instead of 14.5). The motor had been running a bit rich, from the colour of the exhausts, so I was happy to make this change regardless of the leak.

The carburettor floats being reset to 15.5mm.
Fuel pipes visible beneath 

 To cut a needlessly long story short, I reassembled the carbs, installed and tested them for leaks. The leak was still there - but this time, with the tank out of the way, it was clear that the leak was actually from one of the fuel pipes, just above one of the blue plastic T-pieces I had installed. So I replaced it with some double-walled 5/16 fuel hose. That really did fix the leak.

View of the reoiled filter through the left end of the airbox

Plumbing problems

Plumbing problems beset my ride yesterday. I headed out in the Salisbury direction and found the bike started missing on the Warminster bypass. I slowed down and it cleared. I knew it was the return of the fuelling problem I had before Winter. I was trying to persuade myself it was nothing.

On the way home, she stopped altogether three times. I could see that the fuel filters were dry so there was no doubt it was fuel starvation. My guess is that when the bike heats up, the rubber fuel hoses soften and then close up where they bend. I think it is both the vacuum pipe to the rear of the fuel tap and one or both of the fuel hoses to the carbs. Each time, after the bike had cooled I could see fuel flow into the filters again and so could do some more miles.

So last night I took it all apart again to take out the fuel filters so the hoses would have an easier route to the carbs. And in doing so I managed to snap one of the t-pieces I'd plumbed in to link the two sides of the system!!! So I had to pull the carbs out too so I could remove the broken stub etc.

Ironically, I could see that the carb throats and inlet rubbers were wet with fuel. It is ironic bearing in mind the problem I was trying to solve was not having enough fuel. That suggests I set the float heights too generously. So I'll be adjusting them by 1 - 2 mm whilst I'm at it.

On the plus side, with the carbs out I could do something else. And so I have a nice new heavy duty lead from the starter solenoid to the starter motor ;-)

On the road again

I'm rolling again. The experience is startling after pootling about on my 900. It feels like instant urge. It feels just great.

I used two T pieces to plumb in a link pipe between the petrol feeds for the left- and right-hand pair of carbs. I decided to leave the fuel filters in place as well so the result is on the crowded side. I might ditch the filters if this becomes a problem but after all the trouble I've had with dirty fuel, I'll leave them there for the time being. 

Cap masked off, mounting plate repainted

Tank fittings ready for remounting.

The mounting plate for the fuel cap was very tatty so I took the opportunity to paint it. I used alloy wheel paint. It is pretty robust and can be used on aluminium without a primer. 

 21st of April, late at night, on the road again.

Failed tank liner

I have been having intermittent problems with fuel starvation. The effect of this has been running on two cylinders, cutting out, then restarting after a few minutes at the side of the road. I know it is a fuel starvation problem because the bike would run on either cylinders 1 and 2 or 3 and 4: both combinations are served by both coils so coil failure could be responsible. I have had trouble with one or other of the petrol pipes doubling over and cutting off the supply to the left or right pair of carbs. Also, I have fitted two clear fuel filters into the petrol pipes and so I could see them empty of fuel when the bike has come to a stop, and then gradually refill. I figured that flakes of the liner were probably wrapping themselves around the filter that is integral with the petrol tap, thereby restricting the flow of petrol through to the carburettors. I had previously had a go at removing all the loose liner from my petrol tank, including fishing in it with a spring-loaded grabber tool and then draining it entirely. I managed to get a lot out and thought the remainder was sound. However, the return of the starvation problem suggested otherwise. I was pretty sure that once again the failed liner was at the root of the problem so I decided to try getting some more of it out.

I used my grabber tool again and soon found it wasn't hard to fish out large flakes of the liner that had come loose inside the tank and was floating about in the petrol supply.

In the photo above, you can see the size of these flakes in relation to the petrol filler.

So after about 30 minutes of work, I managed to get a considerable amount more of the failed liner out of the tank. I'm pretty sure there aren't any other large flakes there right now but I'm equally sure more will come looses and foul the fuel outlet once again.

Fuel cap

I wanted to make sure my fuel cap was functioning properly. I have had some problems with the breather on the identical item on my 900. It has occasionally failed to admit sufficient air into the tank to allow fuel to exit, especially after a period of high speed running.

The cap is an aircraft-style flush filler cap, operated with a key in spring-loaded mechanism. The key hole is itself hidden under a sprung cover. This cover was loose, not snapping shut, and the key was not centering after the cap was pushed home. Not a big deal but I thought I'd see if I could clean out the components and restore some snap to the action.
The main components of the cap are visible above, including a lid with rotating centre that is hinged to an alloy ring; a retaining tang which is drawn back by the two pegs on the rotating centre; a small spring which forces the tang out to retain the lid by clipping under the forward edge of the tank's filler hole; the main fuel seal which is held against the tank's fuel filler hole by four small springs; a tang retaining cover, and steel retaining disc and two screws which clamp the tang holder and seal in place.

The fuel tank breather is a small air passage that runs from a little hole in the tang retaining cover, through a casting that contains the lock barrel, and through a pair of diaphragms to another little hole next to the hinge. The tank emits a rather attractive humming, whistling noise when in the sun or cooling down after a run. The noise is made by fuel vapor escaping, or air being drawn through these diaphragms.

The little holes are indicated by the tip of a screwdriver in the pictures above. The left-hand picture above shows the cap without the main fuel seal in place. The four small springs visible around the rotating body apply pressure to the main fuel seal to ensure it works effectively. The right-hand picture shows the cap reassembled.



The cap assembly is retained by four screws into the top of the tank. One is under the lid, preventing the cap from being removed without unlocking it first, the other three are allen-headed screws that pass through the periphery of the hinge ring. Two of the allen-headed screws are just decorative. A rubber dust and dirt seal fits into the top of the tank first, followed by the cap hinge ring.



I cleaned off all the corrosion from the components and lubricated the rubber seal with rubber grease, and the spring-loaded tang with copper grease. I replaced the inner screws with stainless items. Job done.

Running on two cylinders

When I started my Trophy 12 for its MOT, I was disappointed to find that she dropped onto two cylinders after a couple of minutes. She was running beautifully the night before. Lovely even beat. What could have happened? I had my appointment booked though and thought it might be some muck in the carbs or maybe something that would clear as the engine warmed up. 

Not so. Warming up didn't make any difference. Maybe one of my coils had failed. Each coil controls a pair of cylinders so it was possible. However, the two cylinders that were not running were numbers 3 and 4. The pipes on these cylinders were cold. One coil runs cylinders 1 and 4, the other runs cylinders 2 and 3: it could not be a coil/ignition problem. So attention turned to the fuelling side of the equation.
 
I had thought that the two fuel lines were to ensure an adequate feed of fuel at high speed, and that they fed a common gallery for all four carburettors. This was an assumption. Looking at the carbs, I now see that one fuel pipe feeds cylinders 1 and 2, the other feeds cylinders 3 and 4.

To cut a long story short, I found that I had pinched one of the two fuel lines that runs from the fuel tap to the carbs. The previous night, I had not bolted the fuel tank in place. Of course, I wouldn't take to the streets with the fuel tank insecure so I had fastened it down the following morning.

To remedy the problem, I took a close look at the routing of the fuel lines. One of them pushes on to the rear of the tap, the other pushes on from the right-hand side. I had used much thicker fuel lines to replace the hardened originals. This meant that they need more room to physically fit in tight spaces and require bends with a wider radius. I had inadvertently routed the bike's wiring loom along the main frame tube exactly where the rear fuel line would need to turn when the tank is fastened. So I solved the problem by pulling the wiring loom further around the frame tube and securing it with an additional cable tie. That gave the fuel line sufficient clearance to bend around the frame tube without too much difficulty.

The result: all working beautifully. The engine's creamy smooth thrust was restored, allowing me to start enjoying the bike in the next few days. It was still very much in shake-down mode for fault finding and rectification. A great feeling.

Carburettors and fuelling

I was confident that my Trophy's misfire was due to an ignition fault. However, there was a chance it could be related to fuelling. My general thought was to do the best I could to get all the engine systems in tip-top condition, from the starter motor through airbox and carbs to spark plugs. The age of the machine (approaching 21 years) probably meant there'd be plenty of muck in the carbs and that the floats would benefit from attention.

The 125BHP 1991 Trophys are fitted with a bank of four Mikuni BST36 flat slide constant velocity carburettors with the serial number 1240010T0301.

• Cylinders one and four both have SE60-3 jet needles and 112.5 mains.
• Cylinders two and three both have SE77-3 jet needles and 117.5 mains.
• Needle jets (0.8) and pilot jets (40.0) are common to all four carbs.

There was evidence of fuel seepage, a golden-brown glaze, on the edges of the float bowls and staining underneath, and of course plenty of grime everywhere in the induction area.

The carbs strip down without much drama, as long as the bank is firmly supported so you can get a good grip on a Philips #2 screwdriver to release the float bowls. I found that corrosion had crept between the float bowl seals and the carb bodies throughout. I cleaned it off with a blunt screwdriver and WD40. The float levels needed minor adjustment only (a millimeter or two). There was a fine rust-brown deposit in all the bowls.

There are two plastic mesh filters fitted into the fuel line spigots as standard. These showed no evidence of contamination. I think the stuff that had collected in the float bowls was too fine to be stopped by the plastic filters. In fact, the petrol tap also incorporates a fine plastic mesh filter - still too coarse to have kept this silt from the carb bodies. An old toothbrush and carb cleaner sorted this out in a thoroughly satisfactory manner.

The carbs are fed by two fuel lines, each being 7mm bore preshaped plastic pipes. Both had hardened over time - quite understandably - making me think it would be unsafe to reuse them. I bought a meter of 7.6mm bore fuel injection hose and some new hose clips to replace them, along with two inline fuel filters with paper elements.

The fuel injection hose is double walled which means it is robust and will resist kinking. The downside is that they won't tolerate tight bends. I allowed about 5cm of extra length on each pipe, plus the fuel filters each being about 5cm long, to accommodate the wider turns the injection hose would need. There is plenty of room around the front and rear sides of the petrol tap. Unfortunately, the same is not true to the rear of the tap where one of the hoses and a vacuum line attach. Judicious juggling and careful routing of the main wiring harness are needed to make it work. More on that later. The additional length of the hoses means that I can uncouple the low fuel sensor wire, release the front of the petrol tank and rotate it 180 degrees to rest on the rear subframe for engine testing and tuning.

That's right: the startup moment is upon us.