Three cylinder XJ550

My 81 XJ 550 SECA only runs on three cylinders. I have checked a number of things and found no obvious causes. Can you help to define next steps?

The long story:
The bike was running a little lumpy.
When I pull the plug cap from plug 1 the engine immediately dies, when I pull of one of the other plug caps, it more or less keeps running. So there is a problem with cylinder 1.

What have I already done:
* Valves were checked and adjusted within the last 100 km, so these are assumed OK.
* Carburators were rebuild last year with new throttle shaft seals. Fuel levels were wet set. Carbs had bench sync and running sync.
* When opening the drain plug, fuel comes out of the float bowl.
* Switched leads from plug 1 and 4, problem stays at cylinder 1.
* Switched plugs of cylinders 1 and 2, problem stays at cylinder 1.
* Measured coil resistances. These are OK, but found that all plug caps have a resistance of 5K in stead of the prescribed 10K. Something to correct later.
* The plug sparks when held against the casing.
* Checked compression. All cylinders measure between 8 and 8,3 bar. New spec is 8,5 bar, minimum is 7, max difference is 1 bar. So all in spec.
* Checked synchronization: Cylinders 1 and 2 are a little lower that 3 and 4, but all measure about the same vacuum.
* When I take the plug out of cylinder 1 it is wet and smells of petrol.

This is where I am now. As far as I can see the cylinder gets air (sync, compression check), fuel (carb rebuild, wet plug) and spark and it should be working.
Only it isn't.

Any suggestions where to look next?
Could I have a blocked exhaust pipe?
Or something with the carbs; but what should I be looking for?

Thanks in advance for your suggestions.
Joris

 

I was going to ask about the color of all of the plugs. Does sound like #1 is funky. Check for air leaks. When it's idling, you introduce some propane from an un-lit torch around the carbs/boots to see if idle is affected.
How are all of the mixture screws? Did you start them all at 2.5 turns out? Did you do much adjustment of the mixtures when doing a running sync? Have you tried resetting #1 to 2.5 turns and seeing if that clears it up?
Also, I suspect if you're pulling #1 plug wire and it dies, that was providing most of the power. I had my bike running on only 2 cylinders when I first got it, then I got the carbs re-built and did the valves, etc.
Sorry, random thoughts.

 

Which of the exhaust headers are getting hot?

 

I had a similar problem on my XS1100 and it turned out to be too much air getting sucked in through the side of the carb to head rubber manifold due to a useless sealing gasket.
Not saying that's your problem but you could start there.
Fire up the bike again and spray some WD40 or similar at the problem manifold and see if it kicks in by listening to see if the engine revs rise.

Ooops, writing as dkavanagh posted, sorry.

 

Are you getting a really good spark on plug number 1? Twisting the cap clockwise onto the lead may help. Mabe check the plug gap. I have dried a wet fouled spark plug with a blowtorch in the past and it worked to remove the fouling.

 

if it dies when you pull plug 1 but not when you pull 2,3,or4 why do you think you have a problem with cylinder 1?
sounds like #1 is carrying the other 3

are the exhaust pipes all getting hot? if you can measure temp are they close?

 

If you have access to a colortune plug then you could see for sure what number one cylinder is doing, if it was me I would put 4 new plugs in . I would remove the tank and swap the coils and see if this follows . Or I would swap caps on the plugs around to verify this is not the problem. Even if you get a spark it could be intermittent .The other item is you could have a plugged jet .

 

Thanks all for the replies.
You're right. I stated this the wrong way: when I pull lead off plug 1 engine keeps running, any other plug and engine dies.
And when starting for tests after about one minute, exhaust 1 is lukewarm and the others are too hot to touch.

The mixture screws were set at 2,5 turns out and not changed since.

I will try some propane to cylinder 1 manifold, but this will have to wait until the weekend.

 

First of all, congrats on doing some very good troubleshooting and a thorough explanation of what your testing routine and results are..........saves a lot of time for others trying to assist.

Although you wet-set the carb fuel levels a year ago, something may have happened to cause the fuel inlet valve on this cylinder to get stuck "open" (or it is leaking thru the o-ring around the float valve seat). A fuel-wet plug typically indicates raw fuel (or somehow overly rich). I would check the fuel levels again, especially in the carb. Also, be aware the spark plugs, once they become fouled like that, may never recover fully.

One other question: since the carb rebuild last year, did the engine ever run properly? Or is this a recent development, and did any other issues or situations develop immediately prior to this "running lumpy" issue?

 

To answer the last question first: In retrospect I think the engine never ran properly, but every time I did something to it, it improved.

I checked fuel level and that was OK.
Then I put 4 new plugs in and now all four exhausts get hot.
So success I guess, but I do not really understand why.

Never mind, now I must just find time to ride and redo the running sync.
And then colortune before I will say that it is really running properly : )

Thanks all for the help.

 

New developments.

I rode the bike to work (20 km) before further syncing the carbs.
First parts in town and on country roads were excellent.
Among others, I overtook a car more easily than ever before: just roll on the throttle and away we went past the car.
Then on the highway (speed 120 kmh / 80 mph) trouble started.
Started to loose speed while the revs stayed the same.
This happened in jerks, that felt like sliding a bit along a rope and than suddenly catching on again.

When I arrived at work, the bike sounded rough again.
I did not do any investigation then, but decided to ride back home in the evening while avoiding the highway.
That went OK but it was clear that the bike was much less smooth than it had been in the morning.

So far I have had little time for further diagnostics.
I did remove the plugs (new before this ride).
They looked like this (with and without flash):




After I put the plugs back and started the bike, no 1 exhaust stayed cold once again.

My conclusions so far:
* I probably need new clutch plates; haven't replaced them since I bought the bike. That will be a job for this winter.
* Plugs of cylinders 2 and 4 look OK / lean. This makes 3 rich, I think.
* I do not know what to think of the plug of no 1 cylinder. It looks and feels like it is covered in some greasy layer.
* The plugs would probably be more equal after a running sync, but I do not know how to sync when one cylinder isn't cooperating.
* Just wondering about one other thing: some time ago I replaced the fuel line with a new line with a fuel filter, both from XJ4ever. I used the complete 30" of fuel line so that I can keep the fuel connection between the tank and carburetors. I just lift the back of the tank and that gives me enough room for syncing the carbs. Would this extra length of fuel line in some way deteriorate the fuel flow?

I would really appreciate your opinion, specially on the looks of plug 1.
Thanks.

 

Number 2,3,4 are Very lean number 1 is close to the light brown you want . Are your mixture screws backed out 2.5 turns , a color tune plug you would be able to see better where to adjust mixture screws, ( please note the screws do NOT have to be set the same ) if it were me ( of course I have the color tune plug) I would back out the mixture screw a half turn ,ride the bike 15 miles at a steady speed pull over and pull plugs to check color of the electrode. If bike is running real lean , it is running HOT and this can also can make it feel like it is losing power . I suggest trying what I said above , one other question have you checked valve clearance , this can be another source of power loss .

 

+1
The number 1 plug is the only one that's in the ballpark.

 

Im with the crew here... Other than the "greasy" comment, 2-4 look real lean, 1 is the only one looking even close.

Good plugs should run tan brown to slightly black depending.
Id check that that long fuel line isnt kinked somewhere. Also, if you got a quality filter and your fuel is dirty, it may be plugged already. Also, have you checked your petcock screens.
It sounds like your running out of fuel, and the bikes overheating.

 

Thanks all for the replies.

Good to hear that I have at least one good plug
I think I want to check if I have some air leaks.
I will let you know how things progress when I have found time to work on the bike.

 

I worked on the bike, solved one problem (I think) and found a new one.


I have completely torn down the bike in the past and at that time replaced int intake connections between the carburetors and the head with better (less cracked) ones of a parts bike.
I suspected that I had not put a gasket between the manifold and the head.
On inspection It showed that the old gasket had stuck to the manifolds and were in place.
To be sure, I replaced these gaskets with new ones.
This was not enough to get the bike to run on all four cylinders, so it probably was not an simple air leak.

Next steps were:
* Re-torq head bolts. These seemed not to move up to the specified torq.
* Check valve clearance. All clearances were in spec.
* Inspect the carburetor for the number 1 cylinder, where the exhaust stayed cold before.
I noticed two things:
First the passages for the pilot circuit in the carb body seemed blocked (not the little passages in the fuel bowl or in the long suction tube).
After removal of the pilot air jet and the pilot screw I was able to flush the passages out.
Second the connection between the float and the fuel needle seemed like it had moved.

I repositioned the fuel needle to the normal orientation and bent the wire a bit so that it could not move back to this position.
After putting back the carburetor, the fuel level was set.

With this work done, I reassembled the bike.
It started right away and ran on all four cylinders.

However, after a few seconds I heard a sound like gears that were forced to run at very high speed.
It lasted a few seconds and then disappeared.

When I started the bike up later to warm it up for syncing, the noise came back and did not go away in seconds, so I stopped the motor.
I am afraid that for some reason the motor engages the starter clutch in reverse and drags the starter.
Has anyone else ever seen something like that? And what would be the way to solve it?
I don't want to run the motor like this because I am afraid to damage the starter.

 

I have had starter spin due to a bad diode block and low oil. I was changing the oil, when I was filling it with oil I turned on the key and starter started cranking.
when full of oil it did not crank by itself.
The oil bulb test is hooked to the blue wire that runs to the starter button

the motor can not make the the starter clutch spin in reverse when motor is running.

 

The starter clutch can fail in such a way that it will continuously turn the starter motor - here is one example where the cause was the starter clutch bolts had loosened causing the failure.

https://xjbikes.com/forums/threads/starter-motor-issues.67016/page-2

Electrical issues can also occur as XJ550H suggested - add to that a sticking solenoid or starter switch can keep the starter spinning. I am thinking you can tell the difference between a mechanical and electrical failure by checking the voltage at the starter or at the solenoid output side. Normally, the voltage at these two points should be zero with the bike running.

If the voltage is at or near 12V then the solenoid is engaged because of some electrical or component failure - could be a stuck solenoid, stuck starter switch,or other electrical failures.

If the voltage is higher and changes with engine speed then the starter clutch is mechanically bound and spinning the starter so it is acting as an unregulated generator.

 

Thanks for the pointers.
I think it was indeed something electrical.

But of course when testing, the error did not come back.
I will go through the starter circuit when the problem returns or else next winter.

Now it is time for a short drive to get the engine warm so I can get a proper synch and color tune.