Saturday, December 23, 2017

Ducati 43mm non adjustable forks 1999 - 2005 ish - The ones with an awful lot of low speed damping.


There's some 43mm non adjustable forks fitted to the smaller engined Monster and SSie from 1999 to 2005 or so, made by Showa and Marzocchi and someone else who possibly makes them under license from Marzocchi (maybe?) that have quite amazing amounts of low speed damping.

The easiest way to recognise them is to strip them and try to get the oil out of the cartridge.  It would appear that the only way for the oil to get out is between the rod and corresponding opening in the top of the cartridge.  A few drops at a time.  I really don't understand that.

As an aside here, after this style fork they went to an externally visually similar (I'd say probably identical) fork which has rebound damping only on one side and compression (allegedly) on the other.  These are the forks in the S2R800, 695 and late 620 and 400 models.

I don't see many of these bikes for fork oil change services, but I did an oil change on a 1999 750SSie a couple of years ago and tried going down to 5 weight oil to reduce the damping.  It didn't really seem to make much difference, and that's as far as I got with that bike.

The oil spec for those forks is Showa SS-8, which I found a spec for on the transmoto.com.au/comparative-oil-weights-table of 36.8 cST @ 40 degrees C.  That's a tiny bit lighter than Maxima 10 weight (32 cST on the linked table, but 37.4 cST in the Maxima blurb), but generally typical for the available range of 10 weight oils.  The Maxima 5 weight is 16.2 cST, which is a big difference that gave little change.

The later non Showa ones are listed as using Shell Advance Fork 7.5 in the 2001 M400/600/750 manual, whose viscosity is 22 cST in the Shell specs.

Many years ago there was a posting on the Ducati Monster Forum from a fellow from NSW who had Shaun at D Moto have a play with his M800ie forks, which would have been of this style.  I rang Shaun to ask him what he had done and he said he'd drilled some holes in the cartridge above and below the piston travel to reduce the low speed damping.

So, recently, when I had a 2003 M400ie come in for some fork seals, I took the oppurtunity to have another go at making them work.

The photo below shows the marking on the inside of the lower leg.  Not sure which company this is.  The cap has a 19mm hex.




removed the cartridges (undo bottom screw) and tried to get the oil out of them, the total unsuccessfullness of which confirmed I had the forks I thought I would.  This is what the cartridge looks like (RH end is the bottom)



The hole you can see at the RH end is where the oil enters the cartridge to fill it on the unlikely occasion it is empty.  This is a non serviceable cartridge, unless you cut it open and then weld it up again.

I made a guess on where I thought the compression valve at the bottom would end and where the piston would be moving, and drilled a couple of holes to allow the oil to bypass the shim stacks.  This is exactly what most adjusters do on adjustable forks - open a hole and allow oil to bypass the technologically superior shim stack via a very technically inferior hole.

Oil flowing through holes is how damping rod forks (old style) work, and that's why they have no low speed damping and often lots of high speed (depending on the diameter of the hole and what weight oil you run).  At some point the hole, which provides no resistance to the oil as it flows through it at low speed, effectively becomes solid as oil is forced through it at increasingly higher speed.  The bigger the hole, the higher the speed at which it becomes a restriction.

The shim stack is very variable and adjustable (if you can get to it) and a much better idea to modify.  If it is any good anyway - if not, you often end up bypassing it with an old school hole.

The holes I drilled were 55mm and 230mm up from the very bottom of the cartridge, as below.



The bottom hole could have been 50mm up I'd think - with the drill bit through the hole you could just feel the end of the piston hitting the bit with the rod fully compressed.  It wasn't blocking the hole though.

The holes shown are the initial 1mm holes.  I used some ti-nitride drill bits and my air drill, which was the only one I had with a chuck small enough to hold the little drills. They drilled through nicely without pressure on the bit pretty much.  The last time I tried drilling holes in some cartridges I broke a couple of drills and it all got a bit messy.  This time no worries at all.

A point to make at this time - after drilling the holes, it's a very wise idea to not move the rod at all if possible.  That oil inside the cartridge, previously impossible to remove, is now very keen to spray out all over you or whatever else the hole is pointed at.  Old fork oil is not a nice liquid to coat one's self with.  You'd almost go as far to say that wearing a hundred ml or so of fork oil is justification enough to throw clothing in the bin.


Once the 1mm holes were drilled I refitted the cartridge, filled the leg with Maxima 7 weight oil (26.7 cST), bled it up and had a assessment of the result.  The low speed rebound was much reduced, and felt somewhat normal-ish. The reduction in low speed compression wasn't anywhere near what I was hoping for, so I went to 1.5mm and then 2mm to reduce it further.  With the 2mm hole the compression felt on what I would call the high-ish side of normal, but much, much less than originally.  Normal is also a relative term - often the Ducati forks seem to have almost no compression damping at all.  The later comp only fork leg has a hole larger than 2mm about 100mm up the cartridge from memory, and no compression damping until the piston has moved down past the hole (the last third or so of travel).  

A couple of points:

1/ The manual calls for 7.5 weight oil, which will have been chosen for some reason.  I used Maxima 7 weight because I wasn't sure what the impact would otherwise be on the high speed damping.  All the hole drilling is concerned with is the low speed.  The high speed will be set by the oil weight.

2/ I'm no suspension expert, nor do I have the skill required to make an assessment of high speed damping and requirements thereof.  So the decision regarding oil weight was more to reduce the number of changes being made, and assuming there was validity in the 7.5 weight spec.  It may turn out that there is too much or too little high speed damping and that a different oil weight would be more desirable.  Or it may have a mismatch and ideally require different weights for each direction, which is impossible to achieve.

3/ This bike is an M400ie commuter, so it's a pretty soft target in terms of making a big improvement without causing problems.  The crappier things are to start with, the harder it is to make them worse.  Well, usually.

In terms of the hole size chosen, it was a guess (I actually think it was the conversation with Shaun, and someone else from somewhere) and trial and error.  You could also vary the number of holes.  A 2mm hole is 4 times the area of a 1mm hole, so you could also try 2 of 1.5mm or 4 of 1mm holes which would effect the low speed similarly, but reduce the impact on the mid and high speed damping.  The 1mm hole will effectively go solid much earlier than a 2mm hole.  As well as being 1/4 the size, the circumference, and hence boundary layer effect on flow, is proportionally higher.  Four of the 1mm holes should work much the same at very low speed, but will give more mid speed damping than 2 of 1.5mm holes, which in turn would give more mid speed damping than a single 2mm hole.

Someone who specialises in suspension mods would have a much better idea of hole sizing.

At this point I had a 1mm bleed hole for the rebound damping and a 2mm bleed hole for the compression damping with 7 weight Maxima Racing Fork Fluid set to 135mm.  The oil height specified in the manual is 80mm, which I know from my previous oil height experiments gives a very aggressive air spring effect.  As I was also going to make a spring rate change, I likewise made an oil level change.

The owner of the bike weighed around 75kg, so I didn't want to go too hard with the springs.  If it was mine, I'd be going at least 0.90kg/mm.  I tend to cut down the original springs if I can, and use a spring rate formula to work out how much I need to cut.

The spring rate calc is (G x d x d x d x d) / (8 x N x D x D x D)

where G is Young's Modulus (material property, 76.9GPa)
          d is the wire diameter
          N is the number of working coils
          D is the spring mean diameter

The original spring is 292mm long, 4.8mm wire diameter, 38.6mm outside diameter and has 22 working coils.  It is a typical Ducati dual rate spring, where it has a tighter wound section with constant coil spacing over each section.  See the top spring in the photo below.  The spring rate given by the formula is 0.606kg/mm for the total spring in its initial travel.  Too soft.



Often, with the Ducati springs, the secondary rate is about what you want, so you cut off the tight wound coils, square and grind the end and away you go.  But, in this case, there are 22 total working coils with 14 open and 8 tight.  The open coils are 10.3mm apart, the tight coils 4.4mm apart.  This means that when the spring has been compressed 22 x 4.4 = 96.8mm, the tight section will be fully compressed and the open section will become the working section.

Unfortunately, by the above formula, the open section being 14 working coils gives 0.953kg/mm.  For a rider weight of 75kg on a Monster that's too much.  Appropriate for an ST, but not an M.  Also, if we have 14 working coils with 10.3mm between each coil the available spring compression is around 144mm.  These forks have around 120mm of travel and you usually have 15mm or so of preload, requiring at least 135mm of compression.  I'd not like to run a spring to within 10mm of coil bind.  So, for this spring, cutting is not an option.  You could cut off less coils, but it would still be a dual rate spring that became a single rate spring of 0.95kg/mm much sooner.

The spring rate I wanted was in the 0.80 - 0.85kg/mm range.  As it happens, the oem rate of an ST series spring is 0.83kg/mm.  I tend to have a few old ST series springs kicking around as I replace quite a few of them with 0.95kg/mm springs, so I grabbed a pair from the old spring stock and checked them against the originals.  As you can see in the above photo, the linear rate ST series spring is 2mm longer than the original Monster spring and preload spacer.

There were a couple of issues I had to attend to to make them fit.  I had to machine about 0.4mm off the spring inner guide (a ribbed plastic sleeve that goes over the cartridge piston rod and inside the spring) to allow the ST spring to slide over it.  Had I used some of the aftermarket springs available I may not have had to do this, but using s/h springs from the pile I have to hand not only recycles a processed piece of natural resource and shares the love, it also knocks about $200 off the job so we can afford a little lathe work.

The ST springs were about 2mm longer than the original springs and preload spacers. The original set up had given 16mm of spring preload as assembled, which I wanted to replicate.  But one issue was that the stamped C plate (see photo below) that holds the spring in didn't sit evenly on top of the ST series spring due to the spring inside diameter.  Well, it didn't sit nicely on top of the original spring either, and the forks had been incorrectly assembled previously 
with the preload tube under the spring as it came apart.  Ideally I needed a piece of the original preload tube to sit on top of the spring under the C plate, so cut a couple of 6mm slices off one of the original preload tubes.  To allow for the increased length (now 8mm over original), I shortened the square aluminium nut that locked the top cap onto the rod by 8mm so that the stamped C plate would sit 8mm higher.  Simple.  Again, see the photo below for the 3 mentioned pieces: nut, spacer, stamped C clip.  Should have put the spacer below the C clip, that'd make more sense.



Back together again, with the bottom on the floor I could almost fully compress the legs with a big bounce of my weight, which seems to be a pretty good test of oil height impact, etc on any fork I find.  It's amazing how much you can feel the impact of 30mm of oil height in a simple test like this.

The top of the cartridge has a recess that a bush on the rod goes into, which when the recess is full of oil becomes a hydraulic bump stop of sorts at full compression.  I though that was kind of cute, although it did spray me with oil a couple of times.




So that's where I ended up.  On the road on my usual road test loop, which is fairly undemanding in a dynamic sense, it felt quite nice.  Before I'd started playing with it (leaking fork seals were the stimulus), the front end didn't really move much. Not too stiff, just reluctant.  Now feels like a normal front end. The owner is very happy with it.  Even though the spring rate is 33% higher, the fact the forks previously didn't move very much due to the excessive damping has lead to them feeling softer and simply more compliant with the road.

It'd be nice to have someone who knows suspension give feedback on the result, as it could undoubtedly be improved upon further.  But, as a simple starting point, and one that's easily repeatable in the back yard, it's pretty good.