GSXR 1K fork swap: anyone have any experience with the GSXR valves/shimstacks?

[FrankenDuc]

Thought I'd put this out there and see what others experiences have been and what others have done...

I've just recently thrown a set of 2007/2008 GSX-R1000 forks on my Monster 620 (the Kayaba's, I understand the earlier GSXR1K forks are essentially the same outside of thicker/stiffer outers and adjustable spring preload on the compression shim stack, might be wrong but I think they all apply here...).

The forks are sprung 0.85 (should be about right for my 200lbs), with bone stock valving, sag is 34mm (rear sag is 33mm, feel at speed and neutral throttle is nicely balanced).

All in all, it's way way better than ever before, but if I didn't have any complaints I wouldn't have anything to tinker with, and I'd have to go find another hobby...

My issues are as follows:

Rebound - the highest setting that doesn't completely hammer me around town or on bumpy roads leaves the bike feeling very unstable, eerily unstable, on sweepers and smooth higher speed turns.  About 4-5 clicks back in towards the hard side and the bike begins to stabilize on the high speed stuff, but then the ride is very harsh and I fell like I'm going to hit my head on the headstock after every bump...

Compression - haven't figured this one out completely yet.  I've got LS and HS turned all the way out, and it get's over big bumps OK without getting too upset, small bumps are smooth, and feels like the forks aren't moving at all over rolling lumps. This is the lesser of my concerns since it's so far not bad, but I'd rather be running in a more reasonable working range of adjustment.

Also, on the compression,
- was extremely harsh on larger bumps with the stock (125mm) oil height, I dropped to 175mm as a bare minimum starting point (think 200mm is the rock bottom working height, but I threw in some margin to be safe) and it's helped over the big bumps.  Still have 25mm of unused travel though..
- I have in mind to bore out the base valves' compression paths to see what happens - currently the base compression valve uses three 1.5mm restriction holes, seems a bit small to me so I'm going to try 2.5mm to see if I can't get more tuning range out of the HS (and hopefully the LS as well).  The mid-valves also have what seems to me pretty small holes on the compression side (three 1.5mm holes), but I'm not at the point of pulling out the cartridges yet.

I have new fork seals (all balls), and movement in the compression direction was stiff (even after a couple hundred miles) until I slathered silicon grease in there (did the normal wipe with fork oil on the first assembly).  Otherwise I've gone through great pains to assure fork alignment and minimize stiction (turning forks 45 deg at a time in the triples and checking at top and bottom of stroke with fork caps popped), it's <3mm right now so I'm pretty sure it's not a factor.

I have absolutely zero background in suspension valving, this is my first endeavor into the realm, so I'm learning as I go (and hopefully not ruining too many parts along the way).

I'd appreciate hearing any of your experiences with the GSXR 1K forks, and in general anyone's experiences with valve and shim stack modification, and especially any help I can get in making my forks behave.

Thanks in advance!
~Doug

[FrankenDuc]

OK, no response, so I'm going to just go ahead and post my own findings.

I've spent a bunch of time researching and experimenting, what I found about the stock Kayaba '07-'08 GSXR1K setup is interesting.  

I was focused on rebound, but the rebound valving is really not bad, as I've learned.  The problem is in the compression valving...

What was putting me on the focus of rebound was my efforts to keep the tire on the ground.  What was really happening was moderate to large bumps throwing the chassis in the air, with the forks trying to extend to keep the tire on the ground - the forks in stock configuration do a pretty good, actually pretty amazing, job of this, extending to keep the front tire on the ground, but it's nevertheless up to the compression valving to make sure to not pass too much of the bump to the chassis, to not throw the chassis up in the air too far, and to give the rebound path at least half a chance of keeping the tire on the ground.  And the stock configuration of the KYB fork is very progressive on the compression, turning moderate bumps into mountains, giving the rebound no chance whatsoever to keep the tire on the ground...

I picked up a software tool called "ShimRestackor" that has helped immensely. I can't speak highly enough about it...  It has taken me a little bit to learn how to use, but it's very much worth it.

It turns out, the primary compression valves, the base vales (down on the fork lowers), are set up slightly digressive (with the clicker closed).  But, the mid valves (on the damper rod), though they have a shim stack on the compression side that is held only by a light spring, have only 3x 1.5mm throats, leading to a highly progressive compression damping force.  The mid-valve compression actually overwhelms the base valve at >20in/s suspension velocities (at velocities greater than standard moderate braking chassis upset), and at large pothole velocities forces reach >20,000lb...(as shown by the bends in my stock Brembo aluminum front rim)...  at moderate bump sizes, cavitation occurs (I kept wondering why, despite the fork oil being at a pretty constant temperature ~ 70-100F, the fork's dampening faded during rides)...

So, basically, the stock compression is both digressive (harsh on small bumps at low speed), and progressive (extremely harsh on any bumps at higher speeds requiring higher suspension velocities), and to top it off, prone to taking on air and fading...

knowing my forks only have ~34mm to extend (sag with my lard on it), and some 110+mm to compress, I decided I should probably soften the compression, particularly at higher velocities.

What I decided to do was to leave the base (compression) valve alone, and to take the mid-valve and bore out it's compression path.  The ports on the compression side of the mid-valves are actually quite large despite the restrictor paths being very small, so I bored with a 1/8" ball-end bore through the three ports - below is one of the three ports bored out:
https://docs.google.com/open?id=0B1DE0ZygiWDycFk5eTZseXZXVFE

(don't mind the burrs, good thing I'm not a machinist for a living )

In conjunction with increasing the flow on the mid-valve compression, I decided to keep a bit of the compression response in the mid-valve, and to build a very light shim stack in place of the spring loaded blow-off valve.  I'm using a 0.15mm thick 17mm, 14mm, and 11mm diameter on the shim stack (don't know if they'll hold up, might need to go to 0.1mm and 17/16/15/14/13/12/11mm on the diameter to get a reliable stack).

The benefit here is to have a valve that responds more quickly than the base compression valve, since the base valve is responding to oil displaced by the damper rod, where the midvalve is responding directly to suspension travel... To give at least a little compression damping response at high frequencies of rod movement, regardless of velocity.  Also, the vastly increased flow in the midvalve (reduced pressure differential) reduces cavitation.  This configuration actually shouldn't take on any air, shouldn't fade at all..

While I had the mid-valves out, I lightened up the rebound shim stack just a bit (~30%) - with the focus being to allow running the clicker near it's closed position, giving a more digressive curve, with less high speed damping - to achieve higher rebound velocity under fully unloaded conditions (ex. recovering from a buckle in the road), with higher damping at low velocities and/or at low forces, so after hitting a buckle in the road the tire quickly reaches for road, and when the tire hits the ground after bouncing over the buckle in the road, and the forces equalize, damping force increases to hold the tire on the road...

results?  Well, as far as I'm concerned, completely amazing!!!
bumps, juts, ruts, buckles, etc., don't upset the chassis nearly as much, and the rebound - I was very concerned increasing fully unloaded rebound velocity by ~30% - is actually much kinder and gentler... the bike feels overall still very firm. Dive under extreme braking is just as large, but much more controlled, it happens slower.  Rebound is exactly as I wanted, the tire shoots out as if to hammer the ground, but seems to stop short of actually hammering, and consequently stops short of hammering my wrists...

Most importantly, it's easier, at any speed, for me to control the beast!

Will scour through my notes once more an post anything I've forgotten.

Cheers  
~Doug

[MonsterHPD]

Great work, it seems you have learned a lot in a very short time  

If you feel you have excessive brake dive, maybe some more oil or some more low-speed compression would help, but I suppose you have ideas about what to do if you feel it is necessary, right?

I´ve never heard about the ShimRestackor SW before, and on first look it seems great, thank you very much for the tip, I´ll sure look into it thoroughly. Which version do you use, the free version or the pro?

[FrankenDuc]

It's gotten my head spinning, and yet I feel I'm still only scraping the surface of the subject!!! 

I went ahead and sprung for the pro version - it's much less expensive than a typical engineering textbook, and, though it's taken me maybe a few [...uh...hundred...] runs to start getting the hang of how the shimstacks and valves work, it's more direct to the point and visual than a typical engineering textbook - much friendlier to my patience (or lack thereof )

For what it's worth, if you know what you're looking for, how the shims work (and you're apt at scaling), I think the free version would do the job. But for me, I think knowing what I'm looking for has been actually my biggest struggle.  Having the a tool to simulate at will (along with a lot of test riding over the same bumps, sweeps, etc., at various speeds and so forth) has helped me a lot in building a picture of what [I at least think] I want.

Funny you should mention oil level, I'm very glad for the reminder - That's where I started with all of this (I'd since forgotten, but at first I was hoping I could just suck a bit of oil out and solve everything for free). Dive is not at all terrible, I can now almost bottom it out by braking hard enough to hold the rear off the ground for a little stretch, but that's not far south of 600lb's force on the front (not including the braking force vector itself), which is a bit more than the springs' force at somewhere near full compression.  Stock I'd pulled a massive amount of oil out and dared not to move preload in just to get a little bit of extra desperately needed softness, but I do feel I'd like a few mm of spare travel (roads are kinda rough out here), and given the current setup I think higher oil level will probably work best for tuning that in (could use preload instead, but then I'd be giving up extension on the forks, and since I like riding on the rather irregular surfaces that pass as roads out here, that's probably not the best choice...).

time!
~Doug

[Slide Panda]

Keep going - I know you're not getting many response. But I'm following along. I'd respond if I had anything more useful than encouragement - but for now I'm just reading along.

[MonsterHPD]

Well, your experiences mirror mine while working out the "Showa Blues" issue some years ago .... complete dis/re-assy cycle, test ride, repeat .... I also did not know more than bare basics when I started out. After a while it gets pretty annoying that the midvalve is absolutely as far into the fork as you can get.... no shortcuts

Having finally worked out what was wrong with the stock Showas, I´ve just used the shimming I got with the Öhlins or K-Tech shimstacks as that works great compared to stock.
However, beeing pretty light and not liking the rattle you get from sharp-ish road irregularities (sort of same roads here, apparently ...) I´ve slightly modified the compression shimming on my Öhlins forks to soften the high-speed compression damping, enabling me to dial in some more low-speed damping with the adjuster screw to counter brake dive.

I run about 120 mm air cushion in the Öhlins (and my Showas, for that matter); I´ve never been able to use the last 10 to 15 mm of travel no matter how I´ve tried even though I can not run lighter springs as that would ruin sag settings. However, more oil will add pregressivity; if you log in to Ohlins and search out a manual for their RT forks, there are some curves there that basically applies to all forks of roughly similiar dimensions.

The shimming changes I´ve made have been pure, slightly educated guesswork, but the Shimrestackor would obviously take much of the guesswork out.

Friday night, time for 

    
    

[Speeddog]

Looked over the ReStackor website, and read just about all of the info.

Got such a huge brain boner I bought the ReStackor Pro.

[MonsterHPD]

Well,
I´ve tried to get the demo vesion to work on my Win7 computer, no luck so far 
Anyone else got it to work on a Win7 machine?

[Speeddog]

Well,
I´ve tried to get the demo vesion to work on my Win7 computer, no luck so far 
Anyone else got it to work on a Win7 machine?

Yes, works fine on my Win7 laptop.

What is it doing?

Are you running with Excel or OpenOffice?

[MonsterHPD]

Yes, works fine on my Win7 laptop.

What is it doing?

Are you running with Excel or OpenOffice?

I run Excel, and now it works. I thought I had the decimal sign configured to "dot" since that is what my excels show, but apparently the configuration was "comma" anyway; when I changed that, it ran OK.

Will take a while to understand what is going on, though ... and it seems most results are shown in "new world units", like lbf, F, in .... or is it just nother setting to discover?

Anyway, seems to be pretty useful, so I´ll probably give in and order the pro version.

[Speeddog]

All of the measurements of physical parts are Metric, Output is Imperial.

So a mixed bag, just like real life. 

[FrankenDuc]

Yeah, I've tripped myself up a few times putting in a "comma" instead of a "dot", I can only imagine how hard I'd scratch my head over a dot/comma config swap!

I do love the metric/imperial mix the ReStackor-metric.xls has outputs in metric (and inputs too!), but the metric/imperial mix works since all of the valve pieces are simple metric values, and the scale I'm using to weigh the bike and everything shows pounds

[MonsterHPD]

I´ve heard the US is going metric, inch by inch .... 

[FrankenDuc]

MonsterHPD, ... I've gone to great measures to find shortcuts in getting to the midvalves, now I just yank the cartridges out with the forks still on the bike
https://docs.google.com/open?id=0B1DE0ZygiWDyaGllcGlVemNtVWM

I'd read up on your Monster "Showa Blues", really amazing work!!!  (was going to point you to it until I realized it was your work in to begin with )

For anyone else with Monster/ST Showas:
http://www.ducatimonsterforum.org/index.php?topic=55303.0
If I had these forks I'd immediately chuck this in the "mandatory mod" bucket!

[FrankenDuc]

Slide Panda, Thanks! Suspension is a tough subject, a very personal one too.

Through this, I even worry a little that the settings/changes I share might be taken by someone as the right way to go for them (all while I'm still figuring things out myself!!!).

On that note, and going back over my notes, I thinking it might be helpful for anyone struggling with the same problems if I share how I got to making the decisions I made.

I started off thinking maybe I could do some quick static force analysis and figure out the damping from there...  Then I started thinking about what type of bump to optimize for, where in the bump, what speed, chassis upset, tire deformation, dv/dt, da/dt, steam started coming out of my ears, and... Then I must have hit my head or something and realized if I just could find a point where the suspension was working well as is and I could shape the damping curve around that point (KISS theory).

Compression:
For me this was the difficult one.
I'm blessed to live on a street made of concrete slabs that seem to keep shifting relative to one another (thanks I think to the Loma Prieta fault line that passes nearby).  There are many square steps of various sizes on this street, some pretty exact at 1 inch, that allow me to test my bike's suspension and compare it directly to stock data available online.  Worst case, there are curbs of pretty precisely 4 inches in some spots if I need them.

My cursory mathematical studies showed me that I would have to live with some chassis movement - if I damped low enough to prevent the chassis movement to any sized square step at any reasonable speed, the ~40lbs static unsprung mass on the front end of my Monster would not decelerate in time to keep the tire roughly in contact with the pavement.

Bearing this in mind, I took a 1inch step at various speed, trying to find the speed where chassis movement was noticeable but tolerable.  This would be the best case scenario - the highest damping force to keep the tire on the ground along with a tolerable amount of chassis movement. The speed I found was ~25mph, which I believe equates to somewhere close to 100in/s suspension movement.

Then, I had to characterize the suspension outside of that point:
1) below this speed, the suspension was softer, and put me at risk of losing contact with the ground
2) above this speed, the suspension was too stiff, and bucked the chassis

Neat trick, don't know this is really valid, but I set up the rebound to overdamp to the point of "packing", then ran the bumps.
- If the front felt like it falling immediately after the bump, I figured compression was not enough and allowed the wheel to go airborn and lose contact with the ground - the excessive rebound then let the chassis fall to force the wheel on the ground.
- If the front felt steady at the bump, the front compressed just enough...
- If the front kicked up then fell at the bump, too much compression - the compression damping forced the chassis up.  Think the whole assembly goes airborn, then the whole assembly falls to the ground.

At any rate, I had my goals - soften it above 100in/s, and stiffen it below 100in/s. Next, how much?  

Thinking through this, I think linear damping above 100in/s would be just right - suspension velocity increases roughly linearly with bump size, and (I think) roughly linearly with forward velocity, and my thought is therefore the deceleration required to keep the tire on the ground, and thus the force required, also increases linearly...  Don't really know, but just my guess.

Below 100in/s - well, think I got a case of "more is better".  At slow suspension velocities, my though is that gravity will take care of things, but higher damping will help with brake diving and mid-corner brake/acceleration adjustments (I never make mistakes on corner entry speed... ).  Really, I'd have pushed the low speed damping curve higher if I though I could do so without impacting the rest of the curve.  The tradeoff here is that city streets get more harsh at low sped and some road features (i.e. slow lumps) become more noticeable (I think that's all, but...).

Here's how I shaped the compression damping curve to fit my needs (yellow is OEM, thick light blue is the new curve):
https://docs.google.com/open?id=0B1DE0ZygiWDyYkw5RDZyYVdXWmc

(note, flat coefficient curve = linear force-velocity relationship, so my hope is the damping force-velocity relationship that works well at 100in/s also works well at 1000in/s)

Rebound:
For me, this was much easier - rebound force is bounded by the sum of the sprung force plus weight (force) of the unsprung mass, regardless the size of the dropoff, and the resulting damped velocity corresponds to this force.  

So, what I wanted to do was find a rebound setting on the original setup where a total unloading of the suspension (square dropoff in the road) didn't result in the tire hammering the road (hammering the handlebars), and set this as my new full stiff position.  As an additional step, I did want to make sure the rebound was not underdamped at this setting (since I was planning to make it my heaviest damped setting), so I ran some sweepers to make sure the front was not unstable/fidgity, and if anything pushed/tucked a little.

With me on the bike the front weighs in at just about 250lbs, so 250lbf became my fulcrum for the original valving and setting vs new valving at full closed clicker, and my goals became:
- above 250lbf, I want the wheel to extend faster, this will help recover from upsets more quickly (I think...)
- below 250lbf, I want to dampen more

Original valving and clicker setting in solid blue, new valving at full closed in yellow:
https://docs.google.com/open?id=0B1DE0ZygiWDyRnAzN0g2Q0dpRGs


Beyond this, my thought here is that as the tire hits the ground and forces from the road begins to counter forces from the spring, chassis, unsprung mass, having higher damping (lower resulting velocity) below 250lbf will slow the unsprung mass more quickly and result in less bounce, fidgitiness, etc., overall allowing me to run the clickers out a little to get both higher unloaded rebound velocity and more stability.  For this, my though was that targeting a ~210lbf crossover would work (250lb full - 40lb unsprung). 

Basically, my thought is I can run less damping even at the static 250lbf level, and get more of both grip and stability, if I can run more low speed damping.  I don't know if this is true or not, or if it even makes sense, but that was my guess. So, at 4 clicks out:
https://docs.google.com/open?id=0B1DE0ZygiWDyXzY3ZHlhZWgweWs


Compression isn't killing me anymore, grip is great at 4clicks on rebound, which also feels very comfortable and stable to me, so I'm very happy.  But, I've got to put some good miles on this setup - I might be entirely wrong on my assumptions and/or goals...

!!!