A Wet Misfire

On its 2005 service, I asked Pidcock's to check it over, they said they could find nothing wrong, it was running as sweet as a nut. I had to agree it was, but then it was dry. (I have no complaint about Pidcock's, they have always been very helpful.)

Symptoms

My normal practice when starting from cold is to let the engine warm sufficiently to push the choke right in before moving off. The symptoms were that, after standing in rain or being washed, although it would start OK from cold using the choke as normal, it would misfire quite badly as it warmed up. Once warm, it tended to hesitate whenever the throttle was opened, including moving from rest. On the open road, if I wanted to overtake, instead of the immediate surge of power when I wound on the throttle, it would hesitate, taking a second or two before accelerating. On the "Hebrides " trip, where we had two solid days of heavy rain, it would sometimes go for long periods on three cylinders.

21th February 2006 Things came to a head when, exiting my place of work, the engine misfired enough to cut. I was already committed to turning, the bike stopped dead. I couldn't prevent it falling over, so smashing the indicator and scraping the fairing and pannier.

The next day, doing exactly the same manoeuvre, it would have cut, but I was ready for it; it went to two cylinders and had virtually no torque. It took a lot of care with throttle and clutch to make the bike move without stalling.

Investigation

24th February 2006 As I am a member of the Yahoo Triumph Trophy group, I put the problem out to them, and got a couple of suggestions, one the more obvious one of damp electrics, bad connections etc, the other that if the air box drain plug was missing, it could suck water into the carburetors.

26th February 2006 I checked the drain plug first, to find that it was in place (albeit invisibly pushed up the drain pipe).

One Problem

One check that the Haynes manual suggests is measuring the resistance of the coils. This I did, to find that both primaries measured about 0.5 Ω (the book said 0.6 to 0.8); I would accept this as OK. However, the secondaries measured very high. The left when I first measured it showed over 800 kΩ, I measured the right as 365 kΩ. Re–measuring the left gave a reading of 465 kΩ. I have no explanation for the difference. The book suggests 10.5 kΩ. (Curiously, the two ends of the secondary are connected one to each of two plugs, there is no connection to earth, so the plugs are effectively connected in series, joined by the coil secondary. This is the 4 cylinder model where there is a "wasted" spark in the plug at the end of the exhaust stroke.)

Replacement Coils

2nd March 2006 Coils came into Pidcock's within two working days. When I collected them, I measured their secondary reistance. Each was about 2.3 kΩ, much less than the 10.5 kΩ Haynes says, but that figure was probably for the triple's coils. Or, hopefully, they've increased the wire diameter of the secondary. It certainly looks healthier than the several hundred kΩ of my originals. The primaries measured about 0.5 Ω which seems OK.
I fitted them the same evening (and the replacement indicator). Ran the engine to make sure everything worked.

3rd March 2006 Used the bike to commute the next day. No problems, but the weather was dry (and cold, –7C, 19F; I wouldn't have used it at that temperature if there was wet about; my local Council doesn't salt my road). The bike ran OK, but after it was warm, tickover was no smoother than before.

Coil Test

5th March 2006 Washed the bike on Sunday. The weather was still very cold (there was snow on the ground where the sun hadn't reached) but it was very dry; the snow was evaporating before melting.

Took the bike out for a spin to dry it off (as I usually do after washing). Disappointingly, the problem was still there, it missed at tickover after starting and when opening the throttle. So, back to the investigation.

Further Investigation

So, if it's not the coils, perhaps it's water getting where it shouldn't. I took the tank off again to see if there was any sign in the HT area (plug leads etc). Could see nothing untoward. The options (still assuming it's an ignition problem) are the pickup coil and the igniter. The Trophy's igniter is behind the front faring. I took this off to reveal the igniter. It was absolutly dry and clean.

Igniter

Water on igniter connector

There was a little water on its connector. This can just be seen on the photo where the white wire goes into the plug. The amount, and its position, I don't believe are a problem, particularly as the white lead is actually the earth wire.

I then decided to look for any signs of bad wiring between the connector and any of the other parts of the ignition system. I disconnected this connector, then measured various resistances from the exposed loom socket (using the Haynes manual for reference), including:

Left coil primary
Right coil primary
Pickup coil

All the above were as expected.

I tried to measure the safety circuit (neutral, cutoff, side stand), but I got a reading of about 7.5 Ω. I would expect near zero or open circuit. Switching off the cutout switch made it open circuit. Further investigation revealed the ground connection of the connector was itself about 7.5 Ω from the frame.

2nd Problem

Showing serrated washers
on earth bolt

I traced the ground wire from the connector into the loom, it went right back to a ground point under the saddle. Measurement showed one of the earth tags under the bolt was connected to the igniter earth wire, but didn't go to the bike's frame, even though it looked good.

I tried re–tightening the bolt, this brought the resistance down to about 3.5 Ω, still not good. So I put some serrated washers between the bolt head and the top tag, and between the tags (see picture). I put some petroleum jelly all around and between. Now I've got a good earth. Note I've not put a serrated washer between the lower tag and the frame. All this would do is encourage rust. I rely on the bolt and its thread for the electrical connection.

Went out again for a five minute run. It all behaved impeccably - except when I exited a roundabout, rolled the throttle open, then had to rapidly shut it again as the back started going sideways. There had been a little sleet falling, and with some nearby construction works, there was a layer of slimey mud on the road. This feels much more like the torque there should be. However, it's still essentially dry (and it's too cold and late to wash it again), so I will await the wet weather expected later in the week. But, fingers crossed for now.

Still Wrong

6th March 2006 Commuted on Monday. It sleeted at lunchtime, the bike stood out in it. Starting up to go home, I got a bit of misfiring, even a backfire. It settled out reasonably, and after a bit of a warm–up, I rode home without incident.

7th March 2006 Commuted again, this time there was rain all morning. I had no trouble going to work, the bike was dry from overnight in my garage. It actually felt very good, with all the throttle reaction I would hope for. I am writing this at lunch–time, it is still raining. I am looking forward with some trepidation to this evening's home run ...

... came to start it. It was still raining. The bike started somewhat reluctantly, then missed a bit. Had to hold it on the choke for longer than I would have expected. I waited until it would run on all four, seemed to take ages. It seemed to run faster on the choke than normal as it warmed, but was reluctant to let me push it off. It stalled once. I restarted it, and let it warm some more. Eventually it settled to a fairly even tick–over. I set off home. It once hesitated when I opened the throttle, but apart from that seemed OK. Its characteristic has changed, but it is still not right. It's as if it has a better spark that is still masking some bad behaviour.

8th March 2006 Morning start (after overnight in garage) is nearly normal, I could gradually push in the choke with no excessive revs or feeling of imminent stalling. In the evening, after considerable rain, it was very similar to yesterday. I didn't let it stop during the warm up, but had to be very careful with the choke, holding high revs (without any throttle). Attempts to push in the choke made the revs drop, with misfiring, threatening to stall.

More Electrical Investigations

Sidestand switch connector

In front of the battery
(battery box removed)

Alarm connector with link plug

Alarm connector with link
plug disconnected

Light switch contacts,
starter switch

Cut-out switch exposed
(and out of focus)

11th March 2006 Located the sidestand switch, disconnected it, inspected it and put it back. Removed the tank, battery and battery box. Checked the wiring in front of the battery. The alarm connector was dirty, on opening it the connections looked a little corroded. I put some petroleum jelly on them, and worked them a bit to clean them. I then wrapped the connector in a piece of polythene, tie–wrapped it in place, so that water could not run into it.

I removed the light switch with the cut–out switch, disassembled it all. All contacts looked perfect.

I checked the HT leads to the plugs. They all looked clean. There was some water–marks on the left hand cover (presumeably that side because of the bike's lean on its side–stand). There was no sign of water in the vicinity of the spark plug area of the cylinder head.

I inspected the air intakes, no signs of leaks. Looking into the back of the airbox showed no signs of water (some bits of grit, but that's what the filter is for). All carburettor connections looked in place. The only thing I could find amiss was the throttle cable guide wasn't quite angled as it should be, so there was a slight kink in the end of the cable. Nothing to do with the problem at hand.

Maybe I will have to look at the pick–up coil, though a on brief inspection it looked OK. I didn't check its connector. After the service ...

It's going for its service on Monday. I shall tell them all about it.

A leak

Mended?

Old gasket

Looking into a bad (left)
and a good plug connector

25th March 2006 Picked up the bike from Pidcock's. Had a long talk with Mick, the service manager. The conclusions he had come to can be summarised as follows:

The head gasket leak was nothing to do with the misfiring. Triumph have changed the gasket type because there is an occasional problem with these engines, though not usually with the Trophy variant. Apparently the differential expansion between the steel liner and the alluminium block can cause the gasket to fail. I have been using my bike after night-time temperatures well below freezing, so I have given it many relatively extreme temperature cycles.
While the bike was parked at Pidcock's the leak got significantly worse; it's nothing to do with the "cold storage" suggestion, though it is to do with the differential expansion of the steel liners and the aluminium block leading to a gasket failure with engine heat cycling.
The wet misfire was almost certainly due to water running down into the spark plug wells. There were obvious signs of corrosion on the plug cap connections (I hadn't look at these, only into the wells). He changed the leads.
After completing the work (head gasket and a service), they'd left the bike in standing rain for two days, at the end of each day the bike started and ran OK. They also (as part of a very good valet) aimed water up under the fuel tank with no signs of a problem.
The carburettors needed synchronising.

27th March 2006 Went to work on the bike today. It's lunch-time, it's raining steadily. This-evening's run home is crunch time.

Except...

The fuel consumption has increased dramatically. While it was misfiring and I was commuting, I could understand this. But it appears to be very poor on a couple of runs, including a visit to the Triumph factory in Hinckley, and one of my "standard" runs into Derbyshire. I phoned Pidcock's, and after a chat, we decided I would take a look at the plugs.

My 4 plugs and
an old for comparison

My no. 1 plug and
an old for comparison

This I did, to find all four were well sooty. I sent pictures via e-mail to Pidcock's, and phoned them, arranging they would have the bike in to look at it. He hopes it's only the idle settings on the carburettors. I have my doubts because it's obviously high consumption when travelling at speeds of 50 to 70, but, as the man said, the throttle is barely opened at these speeds.

22nd April 2006 So I've taken the bike back. I've had to leave it with them because they couldn't book it in for about 3 weeks, they are so busy. They'll give me a ring when they've fixed it

24th April 2006 Got a ring to say they'd done it. Can't get it until later in the week.

28th April 2006 Picked it up. They said they'd cleaned out the carbs and adjusted the mixtures. Fuelled it, awaiting a trip to judge the consumption.

30th April 2006 Went out for a "fuel consumption" run, then took the tank off and looked at the plugs. They were, if anything, showing too weak. I went to a filling station, and with the bike having done 69 miles, I put in 8.45 litres. This equates to 37.12mpg, still not good, but half a litre difference could make quite a difference. I shall reserve judgment

Comments

Maintenance Essentials

Various things about this come to mind, in random order:

Last year, the bike started misfiring and generally running rough under any conditions, it sounded as if it was running rich. I took it to Pidcock's , got them to change the air box. This didn't cure it, they then stripped the carburettors, and told me they cleaned a lot of debris from the float chambers. This cured that problem, but left my "wet" problem. I don't know if there is a connection. (Water getting into the float chambers/airways?)
Ignition coils: The original ignition coils must have had their secondaries burnt out. The resistance I measure is probably carbon tracking in the "extra" spark gap where the winding is open circuit. These were probably not the real problem; they may have made a difference, but replacing them was not a cure.
Bad earth: I would have expected an open circuit in this earth to have stopped all ignition. It seems the residual 7 Ω or so was sufficient to allow a nearly good enough spark.
Somewhere there must be another device using this earth that provides the 7 Ω to ground. I have no idea what it is; most loads when switched on would put 12V onto this line and cut the ignition totally. This never happened. I might have investigated this further, but I didn't have time. (The only thing I can think of is my heated grips , but I've had these switched on without losing engine power.) On second thoughts it can't be these; when they are off there is no path from ignition to earth through them.
This has been an ongoing problem for well over a year, but it is not easy to diagnose. It is probably not helped by there being more than one fault.
It looks as though the "final straw" problem was the plug leads. This had been suggested to me, but I failed to look into the plug connectors. I did measure their resistance. I inspected the leads themselves, and concluded (probably correctly) that their insulation was good. In fact, the prime problem was water running into the cylinder head plug well.
Assuming the bike is now cured I will have to keep an occasional eye on the plugs to ensure water doesn't run into their wells (wells? there's an appropriate word!).
Head Gasket:I hadn't noticed the head gasket leak prior to taking the bike to Pidcocks. I'd been using it for a few weeks for a short, trafficky commute; this was the first time for quite a while that it had been on a 20 mile run, most at 70mph. Perhaps it was an insipient problem ... I now think it was a known weakness caused by differential expansion between the steel cylinder liner and the aluminium block that causes the gasket to fail. Recent commutes have been when the night-time temperature has gone well below freezing, so giving many extreme temperature cycles. I'm told the new gasket is constructed differently to alleviate the problem.

A Misfiring Trophy

History