shim changes and damping curves, based on dyno testing

[kevinstillwell]

There are a lot of discussion on shim stacks.  Everyone is looking for info on how to tweak the shims to get a better ride. 

 

It's helpful to think of how shim changes alter the damping curve.  When we run a dyno test, we test at different velocities to get an overall picture of how the shock is working.  Here is a quick look at a shock compression stack:

- -  The velocities are displayed as inches per second (ips) and the comp force numbers are in lbs.

- -  Looking at the graph of the damping curve, this particular stack looks digressive.

- -  The lower velocities on the test would correspond to lower shaft velocities on the bike, and the same for higher velocities. Dips and g-outs are in the 10-20 ips range, and bump control is in the 50+ ips range.  Most of the 'curve' in the damping curve happens in the first 40 ips.  After that the forces become fairly linear.

 

 

- - - - - - - - - - - - -

 

If we run two tests side by side, we can get a quick picture of the difference in their damping curves.  Here is a graph and the compression numbers from two dyno tests from a YZF 250 shock (KYB 46 piston with the 18 shaft).

 - - These two dyno tests show a considerably different in the compression numbers, and you can see how the damping curve has changed.

 

---> For fun, take a guess at how many shims it would take (and where in the stack) to get this much of a difference in the compression force numbers.  Then tomorrow I will put up the actual difference in the valve stacks.  Don't worry about guessing wrong, it's all for fun.
 

 

Edited April 25, 2014 by kevinstillwell

[4rest]

Here is my wild guess based on the little I know from reading Race Tech's Suspension Bible:  The two curves change at about 5 IPS so I'd say that you'd be adding a shim or two pretty close to the face shims to make the blue stack closer to the red stack.  The fact that the red line is near linear between 10 and 20 IPS means it's probably only one size of shim added.  So my guess is one or two shims behind the face shim of the same size.

Edited April 25, 2014 by 4rest

[ds1434]

I'm guessing clamp change only

[yamaha234]

246 lbs is a lot of force  I guessing it will take at least 8-10 shims to get that much difference

[C-P]

I would guess the change is later in the stack too.  Clamp change sounds good. 

[NastyR1]

For the lower curve remove about 1/2 of the face shims, reduce shim thickness first 2-3 shims under crossover and clamp one size down -

[Clicked]

I think it's easier to interpret dyno data and figure out how a stack is going to “feel” by looking at the damping coefficient curve instead of dyno damping force. So I re-plotted Kevin’s data as a damping coefficient.

 

From 0 to 5 in/sec stack 1 is a little stiffer. Either the clickers were closed a couple of clicks or stack 1 has a stiffer low speed stack.

 

At 5 in/sec stack 2 cracks open and produces a slightly digressive damping force curve through the high speed region. Like maybe a crossover gap blew open. Stack 1 keeps going up and peaks at 22 lbf-s/in and then produces a similar damping coefficient slope through the high speed region.

 

There are a lot of ways to create those two curves. I’d say pull the crossover gap out of stack 2 and you get stack 1? Or, hack on the crossover shim diameter to make the gap stiffer. Or change the clamp, like ds1434 said, to make both the low and high speed damping stiffer. So there are at least three options......

 

Since the increase in low speed damping at 5 in/sec sort-of-kind-of matches the increase at 50 in/sec I think ds1434's idea of a clamp shim change makes a lot of sense.

Edited April 26, 2014 by Clicked

[NastyR1]

Ineresting - comparing the numbers at 10 in/s the stack really starts to open and the reduction in damping is farily consistent at each stroke speed after that at about 71 - 73%.

 

It's always been my belief that regardless of the stack configuration once there's enough oil flow to make it open it acts as one big composite spring with a linear rate - so when a suspension shop tells you they revalved it to be more progressive toward the end of the stroke they're lying. Bottoming resistance is achieved through the combined force of the spring, spring perload, nitrogen pressure,  shock bumper and stack, you cant' rely on the stack alone -

[racerfmx450]

i'm not sure, but for me the stiffer stack looks like a preload shim..

or different crossover settings?

i realy want to know how different the stacks feels on the track..

[kevinstillwell]

 tomorrow I will put up the actual difference in the valve stacks. 

 

OK, here's the answer.  The difference isn't in the shim stacks.  It's from the cadj.  I've mentioned in a couple other posts that no two shocks are alike.  This is an example, and I'll say it's the most extreme example we've ever seen.

 

Here's the scenario as I was testing this shock.

 

1. We got the shock in for a revalve.  I pulled the shock apart, made the shim changes to the main piston and then dyno tested the unit.  I didn't grind the nuts on the cadj to pull it apart.  I knew from previous revalves the stk adjuster was just what I wanted with this shim stack.

 

2. We ran the dyno test and expected the comp forces at 50 ips to be about 650 lbs.  They were 912 lbs.  We about fell over. 

 

3. Ok I tell myself, must have done something incredibly stupid with the shim stack (such as slip in about 5 extra .30 shims). 

 

4. Pulled the shock apart and checked the main stack.  It was ok.

 

5. We only dyno tested this shock, as we didn't have a shock set up with pressure sensors.   If we pressure test a shock, we can get a breakdown of all the forces.

 

6. Pressure testing requires a specially adapted shock (as shown above).  But as mentioned, we didn't have a KYB 46 shock set up for pressure testing, so we simply removed all the reb and comp shims from the main piston. 

 

7. This allows us to test the cadj only.  We expected the comp numbers at 50 ips to be about 150 lbs.  They were 330. 

 

8. Ok, it's the adjuster.  Must be something obviously wrong with it.  We pulled it apart, measured and spec’d it out.  It looked ok.  We reassembled and tested again.  This time the compression force at 50 ips was 391 lbs. 

 

9. The rider was expecting his shock immediately if not sooner so he could race that weekend.  So we just grabbed another adjuster and put into the shock body.  We still had all the compression and rebound shims out, so we tested the cadj only again.  It tested 172 lbs at 50 ips.  That was acceptable in a pinch.  Just like no two shocks are exactly alike, no two cadj are exactly alike.

 

10.  Later, we got the shock back and dyno and pressure tested the cadj.  This time, we had a shock set up for pressure testing so we could run the adjuster with the compression and rebound shims in.  The cadj tested 389 lbs at 50 ips.  Ok we said.  We are going to figure this out.  I spent the better part of a day doing step-by-step tests and couldn't find what was up.  The goal was to test one possible thing at a time until we found specifically what was causing the issues.  Finally ran out of time and put it back in the bag.  Something to save for a rainy day.

 

11.  Now we're ready to get this thing done and out the door.  We set the shock up,  revalved a new (different) cadj and tested it.  The cadj force at 50 ips was 148 lbs. 

 

12.  We have only talked about the numbers at 50ips.  Here are the all compression force numbers showing the differences throughout the entire velocity range.

 - - All the forces are the same except the cadj.  The super stiff cadj is stiffer right from the get go.

 

 

 

 

13. The cadj actually controls the pressure in the shock.  This next table shows that P.c was considerably higher with the stiff cadj.

- - P.c is the 'pressure in the compression chamber', and is directly above the main piston.  This pressure is determined by the shaft diameter and the stiffness of the cadj, hence the super stiff cadj pushed these pressures much higher.

 - - For comparison, pressures in P.c seldom go above 1000 psi.  We actually had to get some 2k sensors so we wouldn't destroy the 1k's.

 

 

 

 

14. You might ask, have you ever figured out where the difference came from? 

 - - Not yet.  Haven't taken the time.  We will basically have to pull two adjusters down to each individual part, and then meticulously measure each and every part, dimensions, angle and whatever else might be the culprit.  Again, whatever the cause, we don’t want to work in haste and test the adjuster, finding out we fixed the problem but don’t know exactly what we did to fix it. 

 

15.  And the funny thing is, I don't recall asking the rider how the shock felt before.  If I remember right, it had already been revalved by someone else.  Of course, the previous tuner had no way of knowing this particular cadj was all funky.  And The rider must have thought it was ok, not knowing what it was supposed to be.  Who knows, maybe he thought it was bitchin.

 

 

- - - -

Edited April 27, 2014 by kevinstillwell

[Vietze]

Very interessting, thanks for sharing!

 

Just for clarification:
The two cadjs you used were both of the same design, right?

The YZs had a few different designs with different portsizes too.

But I assume you work with huge scrutiny and you are aware of that.

What type of cadj. have you used on that one?

The newer one with 2.7mm ports and "longer" spring (the latest design so to speak...)

[demzsolt]

Thank you Kevin !!!!!!!

[mog]

This just worries me more and more that shocks are not the same lol , thanks for sharing Kevin great info

[James I]

Thanks Kevin, that was eye opening. I've found chunks of aluminum casting large enough to cause big problems at high shaft speeds stuck between the res and main body a few times; but, never imagined that there could be so much difference between compression adjusters without having something visually apparent.

If you discover the problem please let us know. Thanks again.

[kevinstillwell]

Vietze, The adjusters were the same design. 

 - they were both the KYB long adjuster like those found in KYB shocks from 2003 - 2014

 - both had the same cadj spring and flsprs (floating spring seat)

 - both had the 6 - #44 (2.15 mm) holes in the piston

 

 

Mog, sometimes it's better not to know.  Of course, I'm showing extremes, but the differences are there.

 

 

James I, that's actually what I first expected.  Something to be stuck somewhere in the adjuster.  That is why we worked so carefully when analyzing the adjuster.  If something fell out of a nook or cranny and we didn't notice, then we'd never know what the issue was.  This is one of those things where you just want to know.

 

 

- - -

[GHILL28]

Did you measure the free length and the rate of the spring?

 

Shim thicknesses and total stack thickness?

[kevinstillwell]

I think it's easier to interpret dyno data and figure out how a stack is going to “feel” by looking at the damping coefficient curve instead of dyno damping force.

 

Clicked,  I like your idea of the damping coefficient.  I haven't spent much time looking at the numbers in this format, so they don't have as much meaning as the format we are used to using. 

 

We have devised a very intricate testing system, and have thought long and hard about how to record and analyze the data.  We've run thousands of tests, and this has become engrained in our brains.  It's almost like we learned a new language.  Now when we look at the data, it almost speaks to us.  We don't just see numbers, we see what they represent. 

 

But the damping coefficient numbers look like a very good way to get a different perspective.  We'll have to add that to our language banks.  And to me, that's what this testing is all about.  Gaining knowledge and getting a new take on what's available.

Edited April 27, 2014 by kevinstillwell

[kevinstillwell]

Did you measure the free length and the rate of the spring?

 

Shim thicknesses and total stack thickness?

 

Yes, we measured all that. 

 

Because the forces were so high on the funky adjuster, we figured it might be a combination of factors that contributed to the high force.  But all the obvious things checked out ok. 

 

Currently the adjuster is sitting in a box, completely torn down as far as it will go.  This is where I left it from the last series of tests.  The next step is to reassemble it and see if it's still super stiff.  This adjuster was so far out of whack, I keep thinking it's something stupid that I did while installing it.   Something I missed when putting it together.  I almost expect it to just start working normally one of these times, and I will never know what the problem was.

 

 

- - - -

Edited April 27, 2014 by kevinstillwell

[cj_wai]

I had read once about a small and hard to notice difference in these adjusters.On the piston there is the small shoulder where the shim stack and then the shaft clamps them down.On one piston that shoulder was flush to the outer surface and on the other it was slightly recessed into the piston.

 

The difference was hard to see until you lay a straight edge across it.That change would have an effect on how far the shims check and also effects the bleed they generate when closed.

Edited April 27, 2014 by cj_wai

[MRW]

Very cool info Kevin!

 

  Gatta be the spring.

 

 

 

 

I had read once about a small and hard to notice difference in these adjusters.On the piston there is the small shoulder where the shim stack and then the shaft clamps them down.On one piston that shoulder was flush to the outer surface and on the other it was slightly recessed into the piston.

 

The difference was hard to see until you lay a straight edge across it.That change would have an effect on how far the shims check and also effects the bleed they generate when closed.