Winter Layup - 1995 Ducati 900 Supersport

[OddDuck]

Photo one is the reverse face of the flywheel, showing the one-way bearing. The gear to the starter motor gear train just pulls out.

The large gear at the top front of the motor casing is the transmission to the timing belts - I had some trouble getting this off. A sharp chisel is what's really needed to fold the safety washer back down. I managed to get there with a screwdriver and a drift, though. The impact wrench wasn't needed to get the nut off. Locking the wheel with a 1/2 drive hex bit and extension bar worked, the torque wasn't very high.

The trouble was that the gear had locked itself to its shaft. The motor is full of ash. Really fine black ash has filled up every narrow space going - it acts a bit like threadlock. It's weak, but if you can't get some leverage on something, it's a problem. In the end I found that I could lever it up with two arms off a gear puller, then fit the rest of the puller for the last bit.

The gear shift lever arm comes off with two cap screws and lock washers. No messing around with hairpin springs needed, once the screws are loose, the whole thing comes out as a complete assembly.

[OddDuck]

I had a lot of trouble with the Woodruff key to the timing shaft - it just seemed to be stuck in. Pulling with pliers, levering with a screwdriver, nope. Not happening. Even trying to drive it from the upper end with a pin punch didn't do the trick.

Finally I took a Dremel saw blade and notched the thing, so I could get positive purchase on it. Then I spent half an hour with a screwdriver angled at 45 degrees, hitting the shaft of the screwdriver from the side with a hammer. It really was jammed in. Getting it out (finally, 2 hours down the track) it turns out that the key was deliberately notched before being installed. The notch looks good enough to be official but possibly it's a mechanic's socket used as a hollow punch. I've found similar before with the clutch pushrod, this was deliberately pattern centerpunched so that it jam fit into the pressure plate bearing's cup.

The spacer for the clutch hub also turned out to be difficult. The shop manual just says, pull it by hand, it'll be a bit stiff with the O-ring on the splined shaft. Not a chance. There's nothing for grip and you'd have to have fingers like a gin trap to pull this sucker.

Dremel again, get the beam and pillar bearing puller onto it. No problem. I kept the notches on the flanks minimal, just enough for positive purchase for the puller plates.

The crankcase pressure vent valve came off easy. Too easy. It turns out that it's plastic, someone had already been in before and used silicone to try to seal it up, but the plastic has been shrinking under heat and time. The aluminium band joining pipe feed to valve body is loose. No matter what I do, it'll leak oil slowly but steadily. Time for replacement with something decent from the aftermarket.

The last thing was finding that I needed a specific pin wrench to get the crankshaft's drive cog off. I'd said earlier that you don't need the pricey factory tools - not quite correct. Oops. Might get away with improvising when taking apart, but not a chance when reassembling. Torque figures for anything with a safety washer are all 100+ Nm. A deep 30mm socket is also needed, standard 1/2" drive sockets will foul on the end of the crankshaft.

So - nearly at the case split and being able to see what's happening with the gearbox, need tools to go further though.

[OddDuck]

Had a look online for the wrenches I need last night and it's not good - pricey or stupidly pricey, a couple of weeks away or an unspecified wait before they even send it. If they have either of the two wrenches I need listed at all.

I had a look at my cam pulley tool this morning (bought from the dealer for changing timing belts) and the locking half of it isn't that high-tech. It's a pin wrench, meaning that pins in the wrench face lock into holes in the component. The pins are hardened dowel pins, like these:

http://www.bearingandengineering.co....Dowel+Pin.html

The rest of the tool is black finished mild steel, a handle welded to a hollow diameter. The pins are pressed into drilled holes. The only tricky bit is getting the pin's holes in the right place. So I thought, I know how to drive a milling machine and a lathe, a mate lent me a stick welder... why not?

Pins on order. I took measurements and I'll go looking for steel tomorrow.

In the meantime... pocketing. This is very similar to porting, except that all that's being done is the careful rounding off of any ridges left after manufacture in the valve pockets.

Strips of sandpaper and scrubbing by hand, or more accurately by fingertip. It's tedious, but judging on what was turning up, it was worth doing. The theory is that the pockets (the area of the port immediately behind the valve) are the highest flow part of the engine. The interface between machined surface and original casting is likely to have some shallow ridges on it. Flow will separate at these ridges, reducing effective port area. Rounding the ridges off helps.

The photos were shot with a lamp arranged at the open end of the port, so the light was almost transverse to the surface. Any defect would be highlighted. The ports look pretty rough in the photos, but they've improved in smoothness considerably. Don't have a flow bench for proper measurements, will have to wing it on this one.

[OddDuck]

Steel stock metal is going to be a problem. Couldn't find anywhere with a cutting service for the hollow bar I need for the wrench. Might have to go with a sketch on a piece of A4 and a trip to the local engineering outfit.

Alternator wrench: has to be non-magnetic. Aluminium or non-mag stainless only. Similar story here getting it made, draft something up and send a DXF to a water jet cutting shop, then put a collar on the wrench to keep it on the rotor. It's got to handle 200 Nm torque so I've had to do some stress calculations for my design.

Not having much luck with aftermarket valve guides. AV&V aren't being very good about returning communications to the local agent, and he's gone quiet on me too. I want guides and seats and seals, that's it... simple enough request. AV&V say they make them, so what's with this sort of distribution? So I spent way too much time digging and came up with this:

http://www.valvoleivam.info/en/

There's a downloadable PDF catalogue (massive), it's all for cars, tractors etc... I took detailed measurements of the valves used in the bike and I'm crossing my fingers that Ducati simply used a catalogue item, or a design adapted from one. The engine timing belts apparently came off a car, perhaps the valves and guides did too. If I'm really lucky there'll be aftermarket goodies on Rock Auto.

(Update: nope, no luck. No match. It turns out that Ducati valves are unusually short, particularly for the head size. The desmo closing arm doesn't need as much stem length as a spring would. So it's likely that the valves are done on special order and there is no equivalent off-shelf generic part.)

And back into the garage, briefly... Checked the chrome plating thickness on the valve stems. They didn't plate the collet area. Measure one, measure the other, take the difference and halve it and that's the chrome thickness. I've yet to go through them all properly but it looks like they're about halfway through their service life.

The impact wrench took the crankshaft's drive gear nut off without any issues. Trying to pull the gear off didn't go so well, though. I'll have to try heating the gear or using a bigger puller.

[pete376403]

If the valves are shorter than usual then you should be able to find something longer and get them shortened. You can make most things fit one way or another. (I put Chev 350 valves into a jawa speedway engine. Reamed out the guide to suit, reduced the diameter at the top for the jawa collets, worked well)

[OddDuck]

Found out a couple of things about the gear on the crankshaft, via the forums.

First: it's on a taper, not a parallel shaft like everything else. That does make sense, it's transmitting power continually and you really want it locked to the crankshaft. No fretting or catastrophic loosening, please. The flip side is that after twenty years, it isn't coming off without a fight.

Second: it has a tendency to come off with a bang. Anyone who's changed a ball joint on a car will know what I mean. Wear eye and ear protection and don't be in line with it when it goes. A basic precaution is to put the nut back on, turned back by a couple of threads.

I had another try with the bearing puller anyway, tightening up and then tapping upward on the puller legs and down through the shaft to use shock to loosen it, but no joy. It's as tight as the puller can take and still nothing. A three jaw puller was tried, but the crankcase wall is just too close for it to go on. Heat and PB Blaster didn't work either. Like it or not, it's serious hydraulic puller time, maybe with a custom fitting for the end of the crankshaft and the nose of the puller.

For the homemade wrench: dowel pins received, wrench body sketched up, first engineering shop tried (no luck, but I know another which can do it).

The M10 x 1.5 tap was used to clean old loctite / gunk / whatever out of the cylinder stud threads, preparation before reassembly. Initial threading was bare tap and fingertips only, making sure the threads were picked up properly. I didn't have the proper handle but the ring on a 1/4" spanner worked.

[OddDuck]

If the valves are shorter than usual then you should be able to find something longer and get them shortened. You can make most things fit one way or another. (I put Chev 350 valves into a jawa speedway engine. Reamed out the guide to suit, reduced the diameter at the top for the jawa collets, worked well)

Now that's an interesting idea. Thanks for that, sorry if I've given the wrong message - the valves themselves are actually still workable. I was hoping to find a match for the valves so that I'd find a match for the guides, and possibly for the seats as well. I'd have gone looking for guides and seats directly but can't get a decent measurement of diameter, length etc while they're still in the head.

Oversizing valve stem diameter and reaming guides would work, but at that point (for me) it's easier to just cough up and get new guides.

[OddDuck]

Oops. Ducati make a specific tool for the primary gear pull and there are aftermarket copies too:

http://www.stein-dinse.biz/Complete-...ml#prettyPhoto

http://www.hdesausa.com/PRIMER-GEAR-PULLER_p_30.html

http://www.moto1.nz/dragon-stone-pri...-puller-ducati

http://www.gbmotorcycleproducts.com/...2092-170-p.asp

A 5 or 10 ton hydraulic gear puller is the alternative but the difficulty is keeping it straight, as I was finding with the normal puller.

[BMWST?]

i think melbar engineering in wingate has a flash machine or two

[pete376403]

Now that's an interesting idea. Thanks for that, sorry if I've given the wrong message - the valves themselves are actually still workable. I was hoping to find a match for the valves so that I'd find a match for the guides, and possibly for the seats as well. I'd have gone looking for guides and seats directly but can't get a decent measurement of diameter, length etc while they're still in the head.

Oversizing valve stem diameter and reaming guides would work, but at that point (for me) it's easier to just cough up and get new guides.

Worn guides can be internally knurled and then reamed back to size so you stay with the standard valve stem diameter. It COULD increase oil flow down the guide as there is now a spiral track (knurling is a bit like a screw thread but it doesn't cut metal out) but if the guide seals are good it should be right.

[OddDuck]

Got it.

Hydraulic puller borrowed from work, quite a bit of re-jigging the legs and then winding it up, got the sucker off. The legs flexed a lot more than I expected, no wonder these things are built so massively. It wouldn't fit perfectly on-center since the puller feet were wide enough to foul on the crankcase wall, but in the end it didn't matter.

I used a bungee cord and string to catch the puller as it came off - I'd read a few stories about the puller plus transmission gear coming off at high speed. It did pop off with a smell of burning oil, but it wasn't as dramatic as I'd expected. Possibly the PB Blaster soaking a few nights earlier helped. Towel's on the concrete floor as a cushion for if or when things fall, luckily the gear stayed with the puller this time.

I got the screws holding the crankcases together out, then replaced two of them (lightly) for the transfer of the engine to the bench. Then it was split the cases, which wasn't hard. Some tapping with a rubber mallet, alternate crankshaft and then driven shaft, RH side of engine, opened it up.

[OddDuck]

A quick comment about the mechanical two and three leg pullers around... make sure the side plates are riding on bolt shoulder, not bolt threads. I'd tried winding my two leg puller up and watched it start to banana a couple of times, then I realised that the original bolts are a bit too short.

Changing bolts for longer ones, with spacers, kept it straight. Not enough force though.

[Ocean1]

A quick comment about the mechanical two and three leg pullers around... make sure the side plates are riding on bolt shoulder, not bolt threads. I'd tried winding my two leg puller up and watched it start to banana a couple of times, then I realised that the original bolts are a bit too short.

Changing bolts for longer ones, with spacers, kept it straight. Not enough force though.

Good work, nothing like hydraulics for the ultimate "do as you're fucking told".

Pays to keep a very close eye on proceedings though, it's hard to have a good feel for what you're doing with that much horsepower.

[OddDuck]

A while ago I'd posted about excess heat transfer from the cylinder heads to the inlet stub pipes. Recap: Heating the fuel and air charge before it's even drawn into the engine is generally bad. Inefficient, noisy, robs power, leads to super hot combustion and the engine running stinking hot. You need enough heat to keep the petrol vaporised, but no more.

Anyway, I'd run some numbers, worked out (in theory) that about 70% of the heat was conducted through the mounting studs. The paper washer was only about 30% of the heat conduction, despite being only 0.4mm thick. So it's more important to insulate the studs than it is to install some kind of phenolic spacer plate.

Had some time today to get onto a lathe and make the insulating washers. These were turned down from a piece of Tufnol-copy fabric and phenolic resin bar, sourced from LEP Plastics (they'll do online stock material purchases).

The inlet stub pipes were a little awkward to drill out from 8.5 mm to 10 mm, since the pipe shadows the flange. In the end I rotated the table for a drill press upside down, checked levelling, and then held the pipe flange up against that surface and drilled that way. A bit rough but it worked. I haven't checked fit to cylinder head yet, probably I'll have to do some re-drilling or filing to make it fit.

[OddDuck]

And it was worth pulling the engine to bits after all. Aside from anything else, the crankshaft galley plug was floating. It turned out to be completely loose, not even fingertight. It's shiny because it's been skimming against the stationary face of the crankshaft bearing.

A closer look at the tapers for the transmission gear showed why it was so damn tight. The interface between the two surfaces of the taper is packed with ash. Quite a bit wiped off with a rag, but there's still a bit bonded on. Anyone know if 2000-grit paper isn't a good idea?