Ducati Multistretta 1098S

[porter_jamie]

old venetian blind on the front of the rad, with the strings on the twist grip - the more throttle you use, the more it opens.

i might patent that idea.

[Mr. Dazzle]

How do you make a Venetian blind?

Poke 'im in the eye!

[millemille]

And, thirdly, after I'd fitted the bearing in the cover, using a socket as a drift on the outer race, it felt a bit notchy when I span it. I had put this down to it being a C3, tight tolerance/high Vref, bearing and it being dry but I'm now wondering if I brinelled it when I fitted it.

The new failed bearing hasn't over heated so lubrication isn't an issue. The crank hasn't got any movement or run out (confirmed by both the DTi and the fact that the alternator cover fits on the alignment dowels with the bearing on the end of the crank, if the the end of the crank was bent/misaligned the cover wouldn't fit on the dowels) so my thinking is either the wrong bearing, a counterfeit bearing or operator induced failure due brinelling the bearing when fitting.

I'll whip the alternator cover off and drop it off with Cornerspeed and have them fit the official Ducati bearing which will rule out all of the above and then put it back together and see what happens.

Oh man, this is turning into more of a trauma than it ought to be !

It is a pain in the fucking hole, that's for sure. It's destroyed another stator. Hopefully it was just my fuck up and having Cornerspeed supply and fit the Ducati bearing* will sort it.



* When I was going through the 1098 service manual last night I noticed that it talks about fitting the bearing with the "shield" inwards. Neither the original failed bearing or the newly failed bearing had a shield.......

[millemille]

Picked up the cover with the genuine, Ducati supplied, bearing fitted.

I think we've got a winner on why the bearing failed. The genuine bearing is completely different, with a far more substantial appearance and with a much more detailed rating, to the one I fitted and which in turn means the original one in there that failed was also wrong. So it looks likely that at some time in the past, prior to my ownership of the engine, some numpty fitted a replacement bearing (or even possibly cover as a result of the crash that wrote the bike off) and then this numpty perpetuated the mistake.

Waiting on the replacement stator, which is winging its way from Lithuania and isn't due for another week.

[Mr. Dazzle]

Or Ducati changed it as part of a CI exercise.

Either way it sounds like you're good!

[millemille]

So, where to begin?

When the multi is working it is bloody awesome, but keeping it working is proving a challenge....

We'll get to the elephant in the room in a moment, but first of all the simpler stuff.

The forks were causing me some handling issues, having to wind in so much preload there hasn't yet been a number invented that is large enough to count the number of turns I had to put in in to get the sag in the ball park. They were also blowing through the travel way too easily, all pointing towards the springs being too soft. Plus I had a nagging feeling that something else was going on in the forks, it might have been my imagination but I thought I could feel an occasional knock through the bars as the forks compressed and rebounded while riding.

The Multi has Ohlins R&T forks, but they are longer and have more travel than standard R&T's. I spoke to Kais Suspension, who are relatively local, about stiffer springs and they said they use standard length R&T springs with an extra long preload spacer. That didn't sound right to me, was there really enough free length in the standard R&T springs that they could allow for the extra travel and not become coil bound?

I, reluctantly, spoke to K-Tech (had several bad experiences with them in the past, but they are local) and they couldn't have been less interested or helpful.

So, I cast my net wider and gave Darren at MCT a ring. Should have done that in the first place as he knew straight away what springs were needed and ordered them for me there and then and we agreed that I'd bring the forks for him to fit the new springs and give them a once over.

Cue a few weeks later and combining a work trip to deepest darkest Essex to visit @porter_jamie's brother machine shop* with dropping in to see Darren on the way home. I can heartily recommend Darren for his knowledge and skill.

Straight away he found that each fork leg had a different amount of oil in them, neither of which was correct, and that the fork caps had been incorrectly assembled by the last person to have them apart (K-Tech) so the damper rods weren't correctly secured - explaining the knocking I thought I could feel - and the damping adjusters would do nothing.

New springs fitted, oil levels evened up and - after Darren understood better what bike they were going in and how it was ridden - some more oil added to decrease the air gap so the effective spring rate increased more as the forks reached the end of the stroke and the caps were assembled correctly.

Took them home and fitted them and WOW! A revelation, better in every single way. One more thing on the snagging list crossed off....









*https://www.iporterltd.co.uk/ heartily recommended by me for both personal and work related design and machining

[millemille]

Coolant temperature and apparent lack thereof...

Having taped up the radiator and got the coolant temperature up to what I thought was in the ideal range I was giving thought to how to produce something more permanent to restrict air flow (not involving louvres/venetian blinds/curtains etc.) but there was a nagging thread of doubt running through my deliberations.

Is the Aprilia radiator I've used REALLY that much more efficient at cooling?

Does the engine not being behind fairings REALLY make that much difference to running temperature?

Now, it just so happens I've got a friend out in Ireland who has totally, bog stock, standard 1098 of the same year as my engine so I asked him if he could do me a favour and over a couple of weeks note the coolant temperature and ambient air temperature in a variety of riding conditions.

Guess what?

My multi is actually running at the same temperature, or even slightly warmer, than a stock 1098. That's without any tape on the radiator.

Problem solved then.

[millemille]

All big V-twins take a lot of work to start when cold, big v-twins with high compression pistons even more so. The multi engine is, effectively, now a 1098R engine and they are well known for being c***s to start when cold.

Fuel injection systems have cold start/enrichment trim tables; these modify the injector duration and advance and ignition advance from the main throttle position/rpm tables, referencing coolant temperature, because low cranking speeds and cold engine when starting doesn't encourage efficient atomisation of the fuel in the cylinder.

The screen shot below shows the cold start trim tables for a 1098R, top, and standard 1098, bottom. Coolant temp is "Y" axis and cranking rpm is "X" axis and the values in the cells are the changes to injector opening duration, in microseconds, the trim table makes to the values in the main rpm/throttle opening table.



When I took the bike to CJS for dyno set up I asked Chris to look at the cold starting and he took a load of ignition advance out of the cold start map to give the starter motor an easier time so the engine would spin over faster and draw less current while cranking. This appeared to improve starting BUT the problem is unless you wait hours for the engine to cool down between each attempt you're not truly doing cold starting.

And it hadn't actually made it better, in fact it had made it worse. And the problem with trying to start big capacity, high compression, v-twin engine over and over again is the sprag clutch takes an absolute hammering.

If you look at the two trim tables above you can see there is a massive difference in the amount of fuel that is being added while cranking on the 1098R compared to the standard 1098. The 1098R is absolutely hosing fuel in at the kind of coolant temperatures we see in the UK when cold starting.

So I bought one of these....



...and downloaded the Guzzidiag reader and writer software from here...

https://www.von-der-salierburg.de/download/GuzziDiag/

..which allows me to connect my laptop to the ECU and copy the hex files on the ECU and upload new hex files back into it. So the plan is I copy my ECU's hex data, email it to CJS, he'll modify the starting table to match the 1098R (because it needs CJS's own software to modify the hex data) and email it back to me and I'll upload it into the ECU and we'll see what difference it makes.

This may go backward and forward a few times before we get it right....

[weeksy]

My 1098 wouldn't start either. I kept it 8 days.

[millemille]

My 1098 wouldn't start either. I kept it 8 days.

Bog standard 1098?

[weeksy]

My 1098 wouldn't start either. I kept it 8 days.

Bog standard 1098?

Yeah. Bought from a dealer, when it ran it was incredible, but I only got it to run twice in a week. I wasn't prepared to put money, time and effort so sent it back to him

[millemille]

My 1098 wouldn't start either. I kept it 8 days.

Bog standard 1098?

Yeah. Bought from a dealer, when it ran it was incredible, but I only got it to run twice in a week. I wasn't prepared to put money, time and effort so sent it back to him

100% down to poor previous maintenance and set up.

There are many self styled Ducati Specialist's out there but very few genuine Ducati experts.

My multi came with a full service history from a Ducati specialist and it was a cunt to start. Took it straight to Nelly/Cornerspeed and told him to go through it with a fine tooth comb.

He found recalls and updates not done, sensors not set up correctly and general maintenance - that any Ducati expert worth their salt knows to do - not done.

For example; Ducati stopped the fuel filter in the tank being a mandatory change item on servicing, irrespective of condition, on all models around 2008. They went to change on condition. How do you know when they need changing you ask? Well when the bike struggles to start....

...because as the fuel filter starts to get clogged/break down the fuel pump draws more current. Drawing more current drags the battery voltage down and once the ECU supply voltage drops below 9V the coils won't fire and the bike won't start.

[millemille]

And now the large grey pachyderm lurking in the corner....

The bearing in the alternator cover. At the moment I cannot stop it failing every 250 miles, bang on, every time I replace it.

3 times now, 3 times.

I think I have found the cause but I'm not sure if I've fixed it and the only way, at the moment, I can see to be sure is to ride it and wait for it to fail. Which is a proper pain in the hole.

To recap, on the V-twin Ducati's - since time immemorial - there is a bearing that is an interference fit in the alternator cover and which slides over the end of the crankshaft and supports the end of the crankshaft to prevent any whip as a result of the flywheel not having absolutely perfect balance.



I've had genuine Ducati OEM and non Ducati aftermarket bearings fail, both at the same mileage. I've fitted the bearing in the cover and Nelly has fitted the bearing in the cover, both failed at the same mileage. After the 3rd failure, which wipes out the stator every time, I had to stop fucking around and get to the bottom of this.

Now, back in a previous career as an engineering consultant tribology used to be a specialism of mine. So I dug out an old FMEA I created for investigating bearing failures when the remains of the bearing don't allow detailed examination to determine cause of failure, like this one....



I went through everything it could possibly be and ruled most of them out - for example, going to the lengths of checking the balance of the flywheel, as it had been lightened, to make sure that wasn't out of balance and hammering the bearing - and one of the possible remaining causes could be excessive parasitic axial load.

The bearing used is a regular cartridge ball bearing, and bearings like this cannot cope with any significant difference in sideways force being exerted on the inner race compared to the outer race, and vice versa.

So, the bearing is fixed in place in the alternator cover and the alternator cover is, in turn, fixed in place to the crank cases. And the crank shaft is fixed in place in the crank cases as well. But there are tolerances in all of these parts, so to deal with these tolerances - and them not putting excessive parasitic axial load into the bearing - the inner race of the bearing is supposed to float from side to side on the crank so that there is no side loading at all.

This is the crank sticking out of the crank cases (the flywheel is removed, it is normally on the splines on the crankshaft. Ignore the mangled, off centre, thrust washer sitting on the face of the timing shaft drive gear, it's a common sight on Ducati's. Even out of the factory). The bearing sits on the smallest diameter part of the crankshaft, closest to the camera, and you can just see the chamfer on the end of the crankshaft that leads the bearing in.



I had the inner race of the last failed bearing and was idly looking at it and the crank and thought I'd slide it on the crank (something that you can't do normally because the bearing is in one piece and fitted in the alternator cover). I expected the race to slide along the crank until it hit the increase in diameter where the crankshaft stepped up to the flywheel nut threads.

Nope.

It stopped with about 5mm to spare and the lead in chamfer only just sticks out of the inner race.

Hmmmm, thinks I. That can't be right can it?

Remove the inner race and run my fingertip over the crank and there's a burr on one section of the crankshaft that is preventing the bearing from sliding on any more. Could that be it?

It makes sense. Because the flywheel is magnetic as soon as you start sliding the alternator cover on the stator - bolted to the inside of the cover - is grabbed by the flywheel and the cover is pulled on, normally taking it out of your hands. You then snug down the cover using the M6 bolts all around the edge of the cover. How would you be able to detect if the bearing inner race is snagged on the crankshaft and can't slide on as far as it needs to?

You can't. So the bearing is sitting there with the cover trying to press it into the engine, but the burr on the crankshaft is stopping it sliding on and hey presto, there's your excessive parasitic axial load.

I don't know for definite how the crank got the burr. If I had to guess I'd say that whoever fitted the non genuine Ducati bearing in the cover when I got the bike didn't have the correct tool for removing the cover and keeping everything nice and square/parallel..



...and in levering the cover off on the piss they damaged the crank shaft.

I've deburred the crank shaft and put it all together. But at the moment I'm still not certain the bearing has gone on far enough, the difference between correct and failing in 250 miles is going to be less than a mm. I may take it all apart again and give it some more thought....

...and fixed the cold starting issue.

[Skub]

Man,this is some non smooth road for sure.

This is where I realise I don't have the heart for building a real special,because it takes special people.

[MingtheMerciless]

You’re obviously not using a LARGE enough hammer!

#railwayfail

[Bigyin]

I am amazed at your

A) Knowledge of all things listed above that would have had 95% of us or more saying "Fuck it"
B) Patience
C) Perseverance

Your Ducati Testastretta engine experience makes mine look like the blandest Jap reliabilty

[Mr. Dazzle]

I love shit like that Figuring out what is actually going on right down in the weeds.

Silly Q maybe....but can you fit an alternative bearing type which can take the axial load? I suppose you really don't want to be driving end load into the crank via that bearing, that'd be even worse! Which is why it is just a ball bearing to start with....

[millemille]

Done a hundred miles or so on the latest bearing with no problems so far, but it's too early to tell whether it's no longer an issue. I'm going to strip it all down again and check out the bearing to see if there's any signs of abnormal wear and also fit a refurbished stator*.

However, other events have overtaken this....

Been working on the cold starting problem, Chris at CJS has been incredibly helpful in knocking up a number of different sets of maps with different cold starting fuel and/or ignition advance trim tables in them. Trying them out is a long, drawn out, process. The engine has to be stone cold, the coolant at the same temperature as the ambient air temperature and once the engine has been running it takes a good 3 or 4 hours - or longer - for the coolant to cool back down.

So it's try starting it on the original map, for a base line to compare to, and then let the engine cool down for half a day. Upload the new map to try starting it, see if it's any better or worse, and then let the engine cool down and revert to the original map and compare the starting again to make sure it's consistent in the change in performance. So you can try one new map in a day. And it's hammering the f**k out of the starter and sprag clutch and battery.

I'd tried three modified maps and the last one I tried caused the engine to backfire through the inlet so badly it blew the throttle bodies out of the inlet clamps....

So today was stripping the bodywork and tank and air box down to refit the throttle bodies and I took the opportunity to check the underside of the tank for any cracks or wear/rubbing...

And it's a good job I did...

Rubbing on the corner of the rear cylinder head.



Almost through the weld...



Rubbing on the bolt head securing the steering damper/front fairing stay...



Very nearly through the metal...



And rubbing on the rear cylinder head above the belt cover...



Only minor damage here, but if left will go through...



So everything else is on hold until I've worked with Martin the fabricator to modify/repair the tank underside to get it clearing all the rubbing points. It's going to fuck the paint on the underside, but none of it's visible and I'm hoping we can keep the heat damage down to a minimum. The bigger issue is making sure we get rid off all the petrol fumes before we even begin to consider the possibility of welding it....

I think this will probably take the bike off the road for this year, so I may as well sort out the bodywork at the same time as well as some other niggly bits I was going to leave until next winter.



*refurbished by Westcountry Windings. You know? Those good old boys sitting around drinking cider, when they're not winding wire, in Cornwall...or Devon....or maybe Somerset? Who are actually in Essex.....confused? You will be.

[Mr. Dazzle]

I used to work for an engjne design firm (Integral Powertain in MK), one of the things we did a lot of was cold start calibration for OEMs. We did one a day with overnight cold soaks in between. So one every 3 hours is rapid

[millemille]

I used to work for an engjne design firm (Integral Powertain in MK), one of the things we did a lot of was cold start calibration for OEMs. We did one a day with overnight cold soaks in between. So one every 3 hours is rapid

I'm sure if it hadn't been so hot over the last couple of weeks it would have taken a lot, lot longer for the coolant temp to drop to ambient air temperature.

I did wonder for quite a while whether doing in the summer was the right thing to do and whether I should wait until winter.....