Starting trouble...

Sorry about the long post...
I own a Husqvarna SMS610 and its been giving me trouble starting for
many months now. I'm at the end of my tether with the thing.

It will not start on the electric-start button, no matter how long the
engine spins over, but if I give it a prod on the kicker at the same
time, it sometimes starts.. and when warm, it runs ok.

Thursday night, I stripped the carb and cleaned it, and fitted a new
throttle cable, as strangely, I couldn't get any slack in the old one...
the new one allowed me a little, but not much (These bikes need a closed
throttle for starting)

I also checked all the earths, and connections to the starter-motor and
solenoid - I found 2 earths loose, so cleaned and tightened them up (all
connections are now clean and tight). I cleaned the plug, washed and
re-oiled the air-filter, charged the battery, and Bingo!.. it started
almost immediately, with assisting prod on the kicker of course.
Next morning I started the bike ok, ready for my trip to work, and
warmed it up a little. Then I switched it off while I went inside to get
my gear on, came out, and it would not start!!... Put the thing away
again and tried again that evening after work.. it started.

Last summer I took it into a workshop to have the valve clearances done,
as I they had been neglected and might be the cause of the problem, I
thought. They found a small score in the barrel and had that replated as
compression was low. I also had a new camchain and tensioner, and chain
runners, a new CDi unit (£235!!), and a new battery.

Now i suspect flooding and plug-fouling, as there seems to be nothing
else left to check. The bike does appear to start easier when the plug
is clean and dry, and the battery is fully charged. I've adjusted the
idle speed correctly.. mixture screw is 2 turns out.

What else can I check??

 

I think you *almost* answered your own question. Large single cylinder
motorcycles have a rather long intake tract and it takes several kicks
to draw any fuel/air mixture to the cylinder...

You just don't get much air flow velocity through the venturi if the
slide or throttle butterfly is open the tiniest bit...

Old Brit bike enthusiasts who travailed with big singles would
routinely turn the idle speed knob all the way out before they
attempted to kick start their machines...

The got used to nursing the throttle to keep the engine running until
it was warm enough to re-adjust the idle speed to whatever was needed
to keep the engine idling smoothly...

My electric start Kawasaki KLR behaves the same way. Before I realized
that I would need to back off the idle adjuster all the way, I would
have to crank and crank and crank the engine to get it to fire...

 

Ah, I was just about to ask how to connect my multimeter! Would I
connect at the low-tension connector on the coil, and earth? Why can I
not do this with me having CDi? And how do I test then?

I feel the spark *is* weak... it looks it, and sometimes I get the odd
misfire and pogoing... perhaps I have more than one problem

 

Thanks Krusty. The thing is, something has changed, as the bike used to
be a superb starter whatever the weather.
I was very puzzled that I couldn't get any slack in my throttle cable,
so I assumed the outer sheath had stretched when being pulled through
the tight spots between the hoses and wiring. The new cable is better
but not by much, and I've experimented with different routings.. the
present one is the best I can get, to achieve some slack.

I noticed the throttle-stop screw has slight grooving where the slide
has been resting on it, perhaps a half-milllimetre deep, if that. Could
this have any effect?

 

I not do this with me having CDi? And how do I test then?

The older battery-and-coil ignition systems suffered from voltage
leakage across the spark plug insulator to the spark plug's steel case
because the voltage rise time was too long...

The primary coil voltage was 6 volts or 12 volts in automotive systems,
and it was there all the time the ignition system was turned on...

Capacitor Discharge Ignitions (CDI) charge a large capacitor up to
between 175 and 400 volts, and when the pickup coil or other triggering
device signals it's time to fire the spark plug, the capacitor rapidly
discharges. The needle of analog multimeter set on a high enough DC
scale would just flicker, but wouldn't give an accurate reading of the
capacitor's output to the primary coil...

You'd probably need an oscilloscope to read the primary voltage *peak*
from the capacitor...

odd misfire and pogoing... perhaps I have more than one problem

Battery and coil motorcycle ignition systems always were rather weak in
terms of output voltage. They would only put out about 10,000 to 12,000
volts, and the voltage buildup time at the plug was critical, you'd
lose much of your voltage if the buildup time was too long...

The addition of CDI to a motorcycle helps to solve that problem...

You really don't *need* 35,000 volts to light the fire in your
cylinder, *if the starting mixture is rich enough*...

Low voltage at the sparkplug makes the engine behave like the mixture
is too lean, and maybe it is too lean...

But you'd probably need to take the bike to a shop that had an
oscilloscope to measure the output voltage...

One thing you could try at home is to widen the spark plug gap a bit,
if you've been running around, say, 0.025 inch gap, widen it out to
about 0.032 inches and see if that helps...

If the starting mixture is too lean, a wider plug gap helps the spark
to encounter more molecules of fuel/air mixture...

 

has been resting on it, perhaps a half-milllimetre deep, if that. Could
this have any effect?

The throttle slide is probably worn more than the stop screw, because
it's aluminum. It would probably be expen$ive to replace the slide, and
I certainly wouldn't replace the slide unless I just had to...

But if you screw the throttle stop screw all the way out and the engine
starts easier, you've eliminated wear on the slide and the screw as a
problem...

 

Not enough voltage to run the ignition when using the starter.
Battery a bit tired ? Corrosion somewhere along the wiring ?

This may be false economy. As it seems you have a weak
spark, put in a new one.

Slightly warm has always been the worst condition you can have
when starting a thumper. Cold is usually OK, hot might require a
slight bit of throttle and can be slightly tricky, in-between is a bitch.

 

I've considered this too...
About a year ago, when I dropped the oil for a change, I noticed it had
turned orange! So I removed all three filters and found lots of bits of
an orange/beige plasticky material in the bag filter right behind the
generator.
I sent photos to the Husky importer who suggested that the material was
excess glue (epoxy perhaps?) that sticks the magnets around the
generator rotor. It had probably just broken off.

Here are two pics.. http://homepage.ntlworld.com/steveo2/Husky/

The shop that stripped my engine agreed with that diagnosis, and told me
all the magnets were intact and everything was as it should be.
Now I'm worrying that something could have been missed, but its another
big job to remove the generator again

I wondered if there's a way to measure its output?

Perhaps, but I've never seen corroded wiring on any bike before.
Connections yes, but not wiring. I've checked them all for tightness and
sprayed with WD40. They all appear to be fine. It's had a new battery.

I've tried several new ones over the months.

Yes, I know this, but this bike was previously a great starter, whatever
the temperature of the engine. In future though, I won't warm it and
switch off - I'll warm it up and go!
Tomorrow I will try Krusty's idea of screwing-out the throttle-stop
screw and see if that helps.

 

the energy available is 1/2*C*V^2, so energy across the gap is
proportional to voltage-squared. Breakdown voltage is roughly
proportional to the gap.

Turbojet engines of the 1950's employed a buzzing spark box similar to
the one that Henry Ford used on his Model T Fords. It worked like an
electric horn or a door buzzer. As the central electromagnet's primary
was energized, it would open a set of contact points, de-energizing
itself and inducing voltage in a secondary winding...

I remember the first CDI system I ever saw, in 1960. It an improved
system for starting turbojet engines. The ignition system doesn't have
to run continuously in a jet, just long enough to light the fire...

The CDI demonstration board had a whole CDI system laid out so youn Air
Force mechanics and electricians could learn how to troubleshoot it...

Whereas a Kettering battery and coil ignition system will make a
"snap-snap-snap" sound as the spark plug fires, the aircraft CDI's
spark plugs would go "whock!-whock! whock!". There was a lot of power
crossing that plug gap, and it would produce big blue-white flashes
inside the glass tube that the spark plugs were in. The glass tube was
there to keep youngsters like me from getting electrocuted...

Kawasaki was the first to employ a CDI ignition system on their
2-strokes to eliminate spark plug oil fouling. Their system did employ
a whining oscillator in an anodized box under the seat about 1969. And
such a high tech system was very expensive...

But Honda, and probably other manufacturers have used a more
cost-effective additional high voltage coil in the alternator to
generate about 175 volts to charge up the capacitor instead of using
electronic circuits to boost the voltage. The high voltage coil's
enamel insulation can be punctured by the voltage and partially short
out from winding to winding, reducing the available current that is
supposed to charge the capacitor...

That red-orange stuff Slope found in his engine may have ominous
portents, it may have something to do with the windings of his
generator...

 

Is there a way to test the output of the generator? Could a proper Husky
dealership have the equipment to perform such a test?

I'm seriously pissed-off now. I started the bike earlier today, but it
cut-out as I was warming it up and it wouldn't start after that.

 

Husky dealership have the equipment to perform such a test?

It's best to consult a shop manual, if you have access to one, and I
would certainly expect a Husky dealer to have such a manual and all
necessary test equipment. But if your closest Husky dealer is far away,
it helps to know how things work and do a little more internet
research...

If you don't hear a high-pitched whine when you first turn your
ignition key on, your CDI is probably powered by a separate coil in the
alternator that produces about 200 volts AC. Then the AC would be
rectified by diodes to make DC to charge the capacitor. If you have an
owner's manual that shows the wiring, you can trace it out according to
color code, looking for loose connection along the way...

Sometimes you can find a seemingly insignificant little connector that
has melted or turned dark colors and that's the source of a high
resistance connection problem...

Loose connections in a circuit like that will burn up a lot of
electricity as heat, you never get the full voltage to the part your're
trying to power...

Typically, checking the output of a generator coil on a bike that has a
battery and coil ignition system is done with the generator output
wires disconnected and the engine is running on the battery...

But if your Husky's CDI is powered by a coil, you wouldn't be able to
check the open circuit voltage while the engine is running, because it
wouldn't run with the CDI coil disconnected...

You could do resistance checks with the engine not running, but you'd
need to know what the specification for the CDI's coil was supposed to
be to tell if the windings were shorted to each other internally. And,
I wouldn't expect the CDI coil to be grounded to the frame at all. If
you get a resistance reading to ground when you check across the
disconnected CDI coil leads, I would believe the CDI coil was shorted
to ground...

And the CDI coil might have its own separate single phase full wave
bridge rectifier to make DC out of AC. If one diode in such a bridge is
blown out, you will see different resistances when you reverse the
ohmmeter leads...

OTOH, the necessary diode bridge might be inside a sealed control unit,
a mysterious "black box" attached to the frame somewhere...

it cut-out as I was warming it up and it wouldn't start after that.

I would definitely look over all the wiring and connectors before
loading the bike up and hauling it to a dealership to have the mechanic
charge me $$$ for troubleshooting. Mechanics like to troubleshoot by
replacing expen$ive electrical parts, and, once installed on your bike,
they are yours forever, he won't return them to stock if they don't fix
your problem...

And, don't forget to check your ignition switch. I have seen a few
ignition switches corrode inside, and get hot enough to melt the solder
attaching the wires to the switch terminals...

 

Tried a replacement ignition switch earlier tonight, and also checked
and cleaned the connections inside the kill-switch. No change
I appreciate your help Krusty.. thanks.

 

Bad language from me. I meant corrosion in the connections.

All that strange oil colour, plastic stuff etc etc does sound scary !
Let's hope it's some sort of adjustment issue. Best of luck !

 

homebrew photo flash type cap, sensitive gate SCR, Yamaha coil. I have
other stuff to figure out, like how to advance/retard with RPM changes
but it works.

If you ever figure that out, put it on your personal website...

A Laverda owner asked me how to add an electronic ignition advance
curve to his Jota, and I was stumped by the question...

The only thing I know about such a circuit is that my Yamaha's digital
spark box counts the number of times one cylinder fires and that
information goes to the electronic tach, controls the ExUP valve and
advances or retards the spark, as required...