Britbike forum Forums British Motorcycle specific Brands Triumph bulletin board Misfire worsening when warm

Try the simple things first.

Take it for a long enough run so that the problem is evident and release the fuel cap to ensure that the breathing is not a problem.

Likely already seen too but not mentioned.

Yes you are right, as far as i can remember the Hillman Avenger was the first UK production car fitted with EI
in around 78/79. I know Vauxhall cavaliers had EI from early 80's. When i worked for the 'rising sun' company
in the 80's all the company cars i had with them were EI and we produced systems for several car manufacturers.

The old Boyer systems i used were good but then again i've known a few blokes who had problems with them,
same as anything really, some people swear by them, some swear about them.
Later micro based systems have some big advantages but are more susceptible to noise than the more primitive
analog setups, but do not suffer low voltage limitations. Trispark have become a legend in the triple world and have
very positive feedback regarding their timing maps. It's like buying a tv or stereo, take your pick.

Just Google freezer spray, it's an aerosol liquid which evaporates really fast, "freezing" what it contacts. Very common in troubleshooting thermal problems, as you can cool individual components. Not sure about the medical variety"

The ignition switch was on when testing the battery voltage (there is no kill switch on my bike). I've no idea whether the Trispark consumes power when the engine isn't running? If so, maybe it consumes quite a lot? If not, the voltage drop does look large to my completely untrained eye.

To go back to a few other unanswered questions:

I actually didn't try cracking open the petrol cap. I will have a good look at it in a minute. Could it have been that all along?

The alternator is 3-phase. I had a protracted email exchange with Steve Kelley after the first unit packed up, in which he asked me to carry out a lot of tests to his instructions and send the results, along with photos of various of them. Among the photos was one showing the regulator/rectifier, which he accepted was OK with the rest of the electrics.

I actually duplicated the wiring from the coils to the ignition unit (ie: made up two new wires running directly from the coils to the unit, to eliminate any issues within the harness with the original wires). I also duplicated the engine earth, with another wire direct from the head to the spare negative terminal on the Motobatt battery (this bike is negatve earth). The whole exercise eventually proved to Steve that it was his unit, not my bike's electircs at fault, hence the replacement under guarantee. It also gave me a better understanding of things and made me check my wiring very thoroughly!

I'm off out to the shed to see what happens when I check across the battery terminals with the engine running, 3000RPM+, lights on & lights off....

OK, just tested for charging, and petrol filler cap.

I'm getting 14.6V at the battery terminals with the engine at 3500RPM (I couldn't be too fussy about the throttle while also holding two meter probes in place), lights off. Headlight on, voltage dropped to 14.34. I can't understand why the voltage would drop but the drop seems small and certainly nowhere near low enough to allow the battery to discharge unless the battery's buggered. I don't think it is.

The breather hole in the petrol cap was clear.

One development is that it ran terribly just now - sounding retarded and missing more than usual even when cold (I don;t have a choke and it does run slightly roughly for the first minute or so unless the weather's hot, but this was far worse than normal.

I think it's time for a Pazon ignition. I was hoping for something simple, just to save me the money & hassle, but I'm out of ideas and the way it was running just now was worse than ever. It seemed reasonably OK yesterday.

Replace the probes with some alligator clips. They allow you to do other things with your hands while testing.

https://www.4statetrucks.com/air-electrical-safety/semi-truck-10-amp-red-black-alligator-test-clips-w-pvc-handles-2-inch-5-sets_291899.asp?utm_source=bing&utm_medium=cpc&utm_campaign=4ST%20-%20Bing%20Smart%20Shopping%20-%20All%20Products%20-%20360%25%20ROAS&msclkid=e59f67ea58c911ad4a5f04b36d2753e9&utm_term=4576236135768327&utm_content=Ad%20group

"14.6V at the battery terminals with the engine at 3500RPM (I couldn't be too fussy about the throttle while also holding two meter probes in place), lights off. Headlight on, voltage dropped to 14.34"

I would consider that to be satisfactory evidence that the charging system is working as it should, and is not the cause of your problems.

Regarding the lower voltage when the lights are on, a simple way to understand it is as follows (numbers are for illustrative purposes only and not intended to be accurate).

At any given rpm (say 3500) the alternator produces a certain number of amps (say 10).
Say the ignition draws 4A, leaving a surplus of 6A. The regulated voltage varies according to the surplus. The bigger the surplus the higher the regulated voltage up to a plateau of about 15 or so volts, so in your case 14.6V at 3500rpm lights off.

With lights on,the surplus is reduced by whatever amps they draw (say 3A) leaving a surplus of 3A, resulting in a lower system voltage (14.34V in your case).
If you were to add more consumers that draw another 3A, there would be no surplus, the generating would just balance the demand, and the system voltage would be the normal battery voltage (~12.6V).

Hope that makes sense!

Yes, that does make sense Koan, thanks! As you can see, electrickery isn't a strong point in my knowledge, though if I have good components I can make a good wiring harness up.

Croc clips is a good idea for the meter probes too. Surprising they don't come with them as an alternative really.

If only we still had local Maplins for simple things like this! I've bought several meters over the years from them in the £10-15 price range, all fine with no emi issues.

Croc clips and leads are readily available on ebay.

The only meter I have now, bought 4 years ago for ~£20 is this one:

https://www.ebay.co.uk/itm/403812463214?hash=item5e0519326e:g:ZzYAAOSwFZpi6eR8&amdata=enc%3AAQAHAAAA4MrQysmK93ZHu3BYCNMBSYgTun1%2FACgU7bZkQFwW9fX5fyzJC8nRlJqleU2Kl883hNqGxYGpezCnDUnBb0rbpSrSJIPXnSCJZWSJP9kz8zZiqXH1MYcvmFl8o9c97cKmAUCWCH7sp7%2BdAkNKKJCR8IbQ7LGG5KVez00vcfLrHV5aLi1Eg6Oe3rOEa2VddevS8UwIwKWfC9MMcG5fu3P0%2FQD1KG0nb4Q45vUcxJMMiKbfOcHJeYgudu1wNijKh4YfkhDuPAxzhFyO9xjLtonlOxyMJ6UYtG9Lo0xvDYm1Rs0Y%7Ctkp%3ABk9SR8rg9NiCYQ

It comes with normal and croc leads, and a temperature probe. Still pretty good value at £25

The test for ignition interference is simply to run the Tri-Spark
off of a good charged battery ( with the voltage regulator disconnected )
... and watching that the voltage while running stays above 12 volts

If the engine problem disappears ... the Tri-Spark is good
But lacks sufficient buffering for this particular voltage regulator .

The simple solution is tri-sparks own overpriced aftermarket
Ferrite-inductor-thingie
( they should maybe send this part out for free to any customer who asks for one )

I don't think it is anything to do with the regulator/rectifier unit. Steve Kelley ruled this out during our discussions when the first one failed. If it was a matter of interference from the reg/rect unit, wouldn't it show up long before several thousand trouble-free miles?

I do see what you're saying about running total loss from a good battery as a test though - useful for someone whose ignition plays up from the start.

just wondering ,
when did you add the LEds ?
This may have significantly changed how the
Regulator operates .
The three phase stators are "higher capacity"
but the Leds do not need it ... so the same regulator will
will clamp down more of each ac phase angle ... to meet an easier to achieve dc voltage point .
...

I had an LED headlight in for about two years but it wasn't much good (cheapo from Ebay). I changed it for a better one this summer. I guess I covered 400-ish miles with it fitted before the misfiring issues began (the engine ran normally throughout this mileage). I had wondered about there being a connection (simply because the LED headlight was the only thing to have changed) but pretty well dismissed it because turning the lights off didn't make any difference. Do you think there could be a connection?

Yeah, I miss having Radio Shack available for electronic items. These days even something as simple as proper crimp-on connectors must be ordered from somewhere. The local stores stock nothing but bodger connectors, and semiconductors must be ordered from China.

Is the bike just running rich? It will go well when cold but bad when hot.

Dave

I don't know how you can test coils for failure because they get warm. Replace the coil with a known good one?

Can you load test the battery. Poor man's is turn lights on and then off and test recovery of voltage.

I have had a dodgy key switch do the same or a kill switch because of vibration.

Maybe start with an online spark tester which can be seen while riding. Rule in electrical or out e problem.

Good luck

I just saw flickering light comment..check for ac across the battery when charging. Lights don't care if ac or DC. If ac then rectifier and charging problem suspected.

Do you mean, switch the meter to AC and test across battery terminals while running? I can do this if I know what I'm looking for. Presumably, there should be no reading, as there shouldn't be any AC current to/from the battery?

I've already tested with the meter on DC and it seemed good - would the readings not fluctuate if the battery was getting unrectified AC ?

Sounds like a dead red herring to me, you’ve already demonstrated that the charging system is working normally.
There will always be ripple in the charging voltage due to the generating peaks as the rotor poles pass the stator coils.
Ripple (always in the same voltage polarity but varying a little) is not the same as AC (which varies between voltage pos and neg).

Single phase systems have inherently much more ripple than 3-phase, in amplitude (voltage variation) and also 3 times the space between pulses.

In the few years I was running batteryless I found that installing the 3-phase allowed me to run the Rita without a capacitor, whereas the RM21 did need a capacitor to run the EI satisfactorily.
I think this speaks volumes of the low level of ripple in the 3-phase system.

I’m doubtful that a multimeter set to AC could measure anything useful on the DC side of the reg/rec (unless the rec was shorting, which we know its not). I think you’d need an oscilloscope.

I’m pretty sure that you’ve eliminated the charging system, and so the EI is the issue.

From the world of modern vehicles comes stories strange electronic problems cause by a tiny amount of AC getting into system....I have no idea about the problems, just a mention...
My friend's T140 had a misfire on one cylinder at higher rpm.Other than that the engine ran well,proper jetting and sparkplugs..... we swapped a known good ignition system and it still misfired...Proper valve adjustment, spring pressure, lash and cylinder compression..
So while changing plugs he put in a used NGK #9G from my race bike... mis fire cleared up........

Yup and yup. Do you blow headlight bulbs? It is an easy matter to switch the meter to ac and see what you get. My thinking on this is if ac then battery is not charging. The ac will cause the lights to flicker too.

Look up the case of mysterious blowing headlights, my name.

I haven't seen any reason to think this is TN's problem, and there's little reason to think its a reality anyway.
Can you describe how a rectifier or a reg/rec would fail in a way that would lead to such a situation? And yet the charging system still works fine?

TN's charging system is fine, lights are not flickering, blowing of headlights is irrelevant as far as I know.

I also don't think that the lower draw of LEDs has anything to do with it, the system must be designed to work without any lights on at all.

In my humble opinion, forget the charging system (it is fine) and look elswhere (EI being the obvious next call).

Have you checked your valve clearances ?

There hasn't been any problem with blowing bulbs, in fact only one bulb has failed in 10 years and nearly 16000 miles (a dip filament blew about 5 years ago).

The flickering of the LED headlight only happens on main beam - on dip it is perfectly steady. I suspect this is a fault with the LED.

I'll do the AC test and see what the meter tells me (if I can get it to start - last time I ran it, it was quite rough - I think whatever the cause of the problem is, it is getting worse quickly). I've bought a Pazon Sure-Fire kit which I'll fit soon, but it would be good to figure out exactly what has gone wrong first.

KevRasen - thanks for the suggestion but this isn't a valve clearance issue.