Over the last few months my 907 has gradully been more and more difficult to start when cold and now it refuses
compleatly to start with the start motor.
Now i have to push start it and then it starts very easily.
As long as the engine is hot there is no problem.
So far i put new sparkplugs, fixet the starter sprag wich have been week for some time and also changed brushes in
the startermotor hoping this would solve the problem.
I am going to look at all the wiring conections to see if there is bad connections anywhere but i would
apprisuate if anybody has any idea of whee to start looking for the problem.
Could any of the sensors gone bad?

Hi Desmo,

As your push-start is ok, I think you don't to worry about sensors.
Maybe the battery it self is the problem.
If you measure the volts of th battery prior and during starting you get a lot of information.
when you measure with the ignition switch "OF" above 12,5 volt it's ok.
Below 12 volt the battery is certainly suspected.
if you turn the ignition switch on a few devices like fuel pump start. Maybe your reading is a little bit lower now.
if you have a significant drop in the reading, you have a short circuit somewhere.
When you try to start and you have a reading below 10 volt, you battery is gone.
if you hook up a car battery, you may try again. if the engine is cranking at normal speed it sure is your battery.
If not,r you have to check the starting wiring closely or the bearings of startermotor itself are defect.
The cables used by Ducati are rather thin (small crossection). a further problem is that the wire thickness is large and therefore not flexible and tend to brake due to vibration. most of the times directly near the connector.
If this is all in good shape, take the startmotor out and look of it spins freely.
If not, you have the culprit in your hand.

good luck

Ducbertus
sratmotor

I'm inclined to agree with Ducbertus; Ducati cabling is crappy but OK if in good condition.

the battery MUST be in good condition for starter motor start - under 11 volts to the ECU and things get pretty sick in the ignition department. the starter will draw a lot of current and may drop system voltage below 11 volts if the battery is suspect.
voltage shoud be 12.5 - 12.7V (as per Ducbertus recommendation)
if voltage drops under starting from 12.5V to (say) 9 volts it could be a crook battery cell.
if the starter motor spins freely for (say) 3 x 10 seconds it should be OK voltage and charge-wise.
don't keep spinning it on the starter motor as they will over-heat [give it a rest between starts]
don't keep cranking the motor on a weak battery - it's bad for the starter motor windings and brushes.

I find that no throttle is best to get it to fire but choke is necessary at almost any ambient temperature - unless the bike has just been running and engine warm.

check the choke lever is engaging the switch; the switch just richens up the ECU fuel supply at idle speeds.
the trim pot on the ECU is a favorite for mechanics to play with that can't tune the bike.
from recollection it re-sets with two turns either direction (there is a faint click that fairys can hear). The trim pot only affects low rev (idle) mix.
if hard starting continues get the exhaust CO analyzed and then stuff chewing gum in the pot so you know if someone's screwed with it later.


unlikely to be sensors - the ECU has "limp home" mode which cuts out any "dead" sensors...
the WTS and ATS are NTC type sensors; yes - they do fail but generally somewhere around operating temp. They are easy to test and expensive to replace with OEM gear. Test them before tossing them.

Mine is a bastard to push-start; I could not do it myself on level ground...
I need all my weight (100kgs ) AND someone pushing (running) if on the flat in 2nd gear to spin it over. [they usually die shortly after this but are expendable]
If you can bump start on the flat solo I'd be inclined to check your compression.
Low compression could cause hard starting...
I'm sure you wanted to hear that...

Thanks for your input all!
I have some good news but mostly bad...
The bike starts and run again so I think i sorted the electric problem, cleaned up all ground connections and all
connections in the starter circuit and found that one of the connections on the starter relay was badly corroded
because in stead of the original brass(?) nut someone had fitted a normal steel nut.

The bad news is I measured the compression and found that the vertical (rear) cylinder had somewhere between
8-9kg/cm2 while in the front cylinder there was almost no pressure at all.
I removed the valve covers but could not see anything obvisually wrong, I know the engine is supposed to be cold
so i cant take any measurements until tomorrow.
I am really surprised the bike runs as well as it is with this kind of compression.

Could only wrong adjusted valves cause the total lack of compression or should I look for more damage like like
piston/rings?


The bike now has about 135.000 km on the meter, heads were compleatly overhauled at about 90.000 km.


thanks again
Håvard

Hi,

what you can do to locate the major leak is to modify a sparkplug, so can connect a air line from a compressor or bicycle pump ( in that case someone els has to operate the pump).
Put the cylinder you want look at, at TDC. in compression stroke (valves shut)
take the lit of the airbox and remove the oil filler plug.
when you apply compressed air in the cylinder, you can hear the leaking into exhaust/airbow (open the trotlle completely, other wise you won't hear a thing) or oil plug (which means leaking pistonring).
This can give a quick scan about the cause of compression loss.

if the valves are set incorrectly also to tight, you probably have burned valves. which cause compression los.

The bike still can run with little compression. in normal driving mode say 4000-5000 rpm
you have abour 40 compressions strokes a second and while you measure compression you run idle say 600 rpm starting rev. this means that time to lose compression during normal driving is much less.

Ducbertus

I'd say that there's no "quick fix" if you're at this stage.

severe compression loss is usually valves...unless the thing is pissing smoke and using oil.

a quick check is to squirt some oil (say 10 cc) into the cylinder and re-test the compression. The oil will seal any piston ring leak-down for a while but not the valves.

generally if the rings are that bad the plug will be fouled and black, often the plug will be so fouled it will short and misfire.

unlikely you have a holed piston - but you can check this by removing the oil filler and determine crank case pressure when running (blow-by for rings is a bit...blow thru for holed piston is a lot...) a holed piston will cause a lot of smoke in the crank case.

if the valves have not been correctly adjusted the seats will be fouled
I'd remove the exhaust and see if there are signs of white (burning / leaking) - do this before any addition of oil testing.

I'd suspect exhaust valve because the intake failing to seat would probably cause backfire thru the manifold...

You might be super-lucky and have an incorrectly adjusted valve or broken spring.

good luck with this...

Before all this, I'd suggest you to measure the valve clearance and repeat the compression test when cold

Hi

Did the pressure test modyfying a sparkplug and could both hear and feel the air coming out the exhaust, the valve had no clearance at all so i decided just to pull the head, this is what i found.
Looks lik i need at least a new valve....

Thank again

Håvard

O.M.G!
Hope you can recover the head!
Never seen such a bad maintenance before!
Maibe new valves (all 4!) and new seats will recover your engine!
Keep us updated,

Wow, I`ve never seen anything like that. I wonder where the missing pieces have gone and how the cylinder walls look like.
If that wasn`t a material defect one could assume a stress problem due to too small valve clearance or skipping valve clearance check intervals.

G.

...with any luck the chips went out the exhaust...

check the cylinder bore; while you're at it check the bore for taper [use gauges or a piston ring].

looks like a new valve and synchro-seat job and you're back in business.

I always push mine through a few strokes before using the electric start when cold.
Here's the procedure:
Select 5th gear - Full choke - ignition on
Bump machine until I hear a POP, POP, POP (about 15ft) then stop the bike and ignition off. Reselect neutral and use the electric starter. Fires up no problem.
Perhaps they breed them tough here in England?