Jump to content

Dyno data –And what it tells us about how to tune a shim stack and control the shape of the damping force curve

Recommended Posts

High Speed Stack

Enough theory, lets look at some dyno data for practical suspension tuning problems

 

Loosen up high speed damping by pulling the even number shims from the high speed stack, unless you be fast – then pull the odd.”

 

That statement is spread all over suspension forums. Others say shim stacks don't work that way, the high speed stack effects damping force everywhere, not just at high speed.

So whats the deal? Do high speed stack changes make a difference or not?

 

MXScandinavia posted some dyno data on TT showing what modification to the high speed stack do in terms of changing the shape of the damping force curve. This first example keeps the same low speed stack and replaces five 0.25 mm shims in the high speed stack with softer 0.20 mm thick shims. That mod keeps the same number of shims in the stack so the shim surface area and friction should be the same for both stacks.

1-mx3c.png

By the thickness cubed rule 0.20 mm shims have 51% of the stiffness of a 0.25 mm shim. The stack has ten shims in the stack taper and this mod replaces five of them with 51% softer shims. So half of the shims in the stack taper were replaced with shims that were half as stiff. So maybe that loosens up the high speed stack by 25% or so?

 

Does softening the high speed stack loosen up high speed damping or not?

Share this post


Link to post
Share on other sites

Yes but it affects the low speed by a smaller margin

Share this post


Link to post
Share on other sites

Tapered Shim Factors

 

To get a guess on how the dyno test might turn out here is some tapered shim factor info. Both stacks have the same number of face shims so the low speed stiffness of the two stacks should be the same, at least some think that way. The high speed tapered section of the mx4c stack uses thicker 0.25 mm shims. Sum up the shim factors for the high speed stack and the mx4c configuration has a tapered shim factor of 126.

The modified configuration, mx3c, replaces the 38 through 30 mm shims in the stack taper with thinner 0.20 mm shims. That drops the shim factor for the high speed tapered section to 88. Compare that to 126 for the baseline stack and the modified high speed stack should be 30% softer.

2-mx3c.png

Sum up the entire stack and the tapered shim factor for the baseline stack is 219. That gets reduced to 180 with the replacement shims in the mx3c stack making the overall stack 18% softer.

Tapered Stack Shim Factors

  • Same low speed stack

  • High speed stack 30% softer

  • Overall stack 18% softer

Nothing subtle there, those are some big changes to the stack. Does loosening up the high speed stack soften high speed damping and change the shape of the damping force curve or not?

Share this post


Link to post
Share on other sites

Dyno Test Results:

The goal here was loosen up high speed damping by replacing five 0.25 mm shims in the stack taper with softer 0.20 mm shims. By shim factors that should soften the high speed stack by 30%. The dyno shows damping force was only reduced by 9%. Shim factors missed the change in high speed damping by a factor of 3.

 

Sum up shim factors for the entire stack and the overall stack stiffness should have been reduced by 18%. The dyno shows 9% softer compression damping. Shim factors missed the change in overall damping force by a factor of 2.

 

What about the shape of the damping force curve? The dashed blue line on the plot below takes the mx3c dyno data points and multiplies by 1.088. That dashed line runs right through the mx4c data points and the FEA curve for that stack. The data shows no difference in the shape of the damping force curve between those two stacks. Modifying the high speed stack just made the damping force softer and it was softer everywhere, not just at high speed.

3-mx3c-dyno.png

The general webology thinking is replacing those five 0.25 mm shims in the stack taper with softer 0.20 mm shims should have really loosened up the high speed stack and high speed damping for better wheel compliance. The dyno shows the 30% shim factor change in the high speed stack only produced a 9% difference in damping force and the damping force got softer everywhere, not just at high speed. The dyno data shows the high speed stack works a whole lot different than the general webology thinking of high speed shim stack tuning.

  • Like 2

Share this post


Link to post
Share on other sites

High Speed Dyno Data

The dyno data from that test goes out to shaft speeds of 120 in/sec. For a 3:1 shock linkage that shaft speed works out to a suspension wheel velocity of 360 in/sec (9.1 m/sec). From the bump velocity plot we looked at earlier (linky) a dyno shaft speed of 120 in/sec is equivalent to hitting an 8” bump at 30 mph, or a 4” bump at about 50 mph. That's definitely in the high speed range where modifications to the high speed stack should have shown up in the dyno data. The dyno did not show that. Instead, the dyno shows modifying the high speed stack just makes the damping force softer and it was softer everywhere over the entire speed range, not just at high speed.

4-hs-mods.png

Both of those stacks had a crossover, so you'd think changes to the high speed stack would have been separate from changes to the low speed stack. They are not. Changing the high speed stack changes the damping force everywhere, not just at high speed.

It is possible to setup a crossover to separate low speed and high speed damping. To separate the speed ranges you have to use a special crossover configuration Jeremy Wilkey (linky) on dirt rider defines as a non-interactive crossover. TT has a bunch of dyno data comparing different crossover configurations and the effect different crossover styles have on the shape of the damping force curve. We will look at that dyno data later and try to figure out what crossover configuration modifications you have to make to separate the speed ranges.

Share this post


Link to post
Share on other sites

It makes you wonder why we bother with cross overs that don't in any way separate high and low speed ,and for the most part are acting like a single stage ?

Share this post


Link to post
Share on other sites

It makes you wonder why we bother with cross overs that don't in any way separate high and low speed ,and for the most part are acting like a single stage ?

 

 

I do believe the ultra low speed is affected - and it may provide an "out" for friction...separating the upper and lower enough to remove the friction component some.

If the "low" speed cross over is smaller than clamp shim I also feel it's more pronounced of a cross over.

 

Lastly, We run preloaded stacks in some cases - however to precisely get a small amount of preload that does what we want is hard the way i do it.  Instead I feel it's more consistent to run a little more preload/taper for machining purposes but then use a cross over to prevent the entire stack from being preloaded.  Of course this requires the cross over thickness be sufficient to leave room for the upper stack to flex the preloaded amount before hitting the lower stack - and in many cases the cross over be smaller than the clamp shim.

 

 

I've actually found a very good level of success with "soft" stacks with decent preload for faster riders in outdoors...getting the balance of high speed bump absorption but enough area under the damping curve to handle these guys speed is tough without boosting the early parts of the curve - but not the bleed circuits which will often translate to harshness.  I think forks for fast guys are different - Decent low speed bleed (soft) but limited peak mid valve deflection adn relatively stiff midvalve seems to work well.

 

Share this post


Link to post
Share on other sites

So you are creating a digressive curve? Are you achieving this with mod to stock piston? Or, in the stack itself? Preload rings are something ive used in shock stax. In forks, controlling initial port growth, while respecting the limited amount of fluid to be "metered" in the base is always a challenge with softer multi staged stacks.imo. speaking to the modern kyb/showa pistons.

obviously the mv is key. But an over dampened mv can create harshness even with a light or open bleed. From my testing.

  • Like 1

Share this post


Link to post
Share on other sites

Good discussion. From the dyno data it is not clear if the crossover gap in those two stacks does anything. Maybe, on an ultra high speed impulse blow, the gap allows the outside edge of the stack to blow off. But the dyno curve shows no evidence of that and no evidence of the stiffer stiffer high speed shims making the damping force stiffer at high speed.

 

Tall Stacks Are Stiffer (Sometimes)

The “tall stacks are stiffer” theory is worth thinking about here. The first MXScandinavia dyno test replaced a stack of 14x40.2 face shims with a shim factor equivalent stack of 4x40.3 face shims. That reduced the overall shim stack height by 29%. The shorter stack was supposed to have the same stiffness but tested to be 20% softer at high lift on the finger press. That gave some credibility to the fulcrum theory, stack bowl theory and the idea that “tall stacks are stiffer”.

 

The second test replaced eleven 0.20 mm shims in the stack taper with softer pairs of 0.15 mm shims. Based on shim factors that stack was theoretically 5% softer but dyno tested to be 3% stiffer. The change in stack height for that test was 22% and gave a second example verifying the “tall stacks are stiffer” theory.

 

The dyno test above replaced five 0.25 mm shims in the high speed stack tapered section with softer 0.20 mm shims. By shim factors that should have made the stack 18% softer. The thinner 0.20 mm shims also made the stack 5% shorter. The shorter stack should have been even more than 18% softer based on the “tall stacks are stiffer” theory.

 

But the dyno did not show that. The dyno showed the difference in damping force was only 9%. Not the 18% estimated by shim factors, or an even greater difference based on the “tall stacks are stiffer” theory. Apparently, a 5% change in stack height is not enough to bring the “tall stacks are stiffer” theory into play compared to the 29% and 22% stack height change of the first two tests. There is something else going on in the stack structure of those two shim stacks that makes the stiffness difference less than expected by shim factors and in the opposite direction expected by the “tall stacks are stiffer” theory.

Share this post


Link to post
Share on other sites

Basically mostlywhat we think about stacks is not based in reality ,softer tall stacks are not softer ,stiffer short stacks are not stiffer ,softer hs stacks only give us softer overall stacks ,kevin stillwell tried to show this many times ,but I think most tuners are so stuck in the idea ,a softer hs stack gives a softer hs ride , that's why we can struggle to get really good stacks ,we need to establish firstly what part of the damping is incorrect (very tricky )then we have to change the stack correctly to fix it ,even harder as the ideas we have are fundamentally flawed

Share this post


Link to post
Share on other sites

Basically mostlywhat we think about stacks is not based in reality ,softer tall stacks are not softer ,stiffer short stacks are not stiffer ,softer hs stacks only give us softer overall stacks ,kevin stillwell tried to show this many times ,but I think most tuners are so stuck in the idea ,a softer hs stack gives a softer hs ride , that's why we can struggle to get really good stacks ,we need to establish firstly what part of the damping is incorrect (very tricky )then we have to change the stack correctly to fix it ,even harder as the ideas we have are fundamentally flawed

 

That pretty much sums it up. The only way to figure that out is:

  •  Dig out the ideas of shim stack tuning webology

  •  Compare them to dyno data

  •  Try and figure out what it means

 In this case changes to the high speed stack looks to effect both high and low speed damping. That is different then the high speed stack tuning expectations of webology. The differences looks to be caused by interactions of the crossover with the high speed stack. Getting a grip on that gives a little insight into shim stack tuning and harrperf's comments on stack preload above gives a little more insight into how to actual go about changing the shape of the damping force curve.

TT has some data demonstrating the effect of stack preload. We will get into that dyno data later.....

 

High Speed Stack Stiffness

Here is a second dyno test (linky) trying to demonstrate the same thing. This second test goes the other way making the high speed stack stiffer by replacing four 0.20 mm shims in the high speed stack with stiffer 0.25mm shims. That change should stiffen the high speed stack and by webology thinking stiffen high speed damping.

5-mx9c.png

Sum up the shims in the high speed stack and the stiffer 0.25 mm shims have a tapered shim factor of 116. That is a 40% stiffness increase over the baseline stack that sums up to 83.

Sum up the entire stack, including the face shims, and the modified stack sums up to a tapered stack shim factor of 246 compared to 212 for the baseline stack. That is a 16% increase in overall stiffness.

 

Tapered Stack Shim Factor high speed stack tuning:

  • 40% stiffer at high speed

  • 16% stiffer overall

Share this post


Link to post
Share on other sites

I wrote a long post only to have a TT issue and it get lost.
 

 

But I heard in an interview the JGR shock  guy mention they run a specific shaft diameter, piston diameter, and of course CADJ set up to get the pressure balance they desire.

I assume this interaction and target pressures requires a "known" or desired damping curve they settled on via testing in rider preference.

But it must play a crucial role in getting the shock to achieve what is desired - more than the stack itself - as shown by clicked.

 

Share this post


Link to post
Share on other sites
Clicker bleed, the shaft diameter, compression adjuster and mid-valve have to all work together to pressure balance a shock. I don't understand why their flow circuits could not be retuned to get the shock to work with different shaft diameters.

 

For the shocks here all of the configurations are pretty close to pressure balanced. I'm sure we will find a few that aren’t as we look at more dyno data.

Share this post


Link to post
Share on other sites

 

Clicker bleed, the shaft diameter, compression adjuster and mid-valve have to all work together to pressure balance a shock. I don't understand why their flow circuits could not be retuned to get the shock to work with different shaft diameters.
 
For the shocks here all of the configurations are pretty close to pressure balanced. I'm sure we will find a few that aren’t as we look at more dyno data.

 

 

 

He quoted a specific combination of all three - meaning they run a specific shaft diameter and piston diameter.

My guess they want a very specific bleed through the shaft and the compression adjuster - and a very specific transition into the stack that requires a quantity of fluid flow and velocity to achieve that.

 

I think tuning at early stack opening and bleed circuits pays the most dividends for rider feedback - and the high speed areas tuned correctly offer better control than if not.

But the percentage of time the shaft spends at low to moderate speeds is far greater than the percentage of time at the peak velocities it encounters - so I feel this area of tuning obviously contributes to the majority of rider feedback.

Share this post


Link to post
Share on other sites

Hopefully the image loaded up. Cool hsc over stock one.

Ill upload spec later this weekend. At bar now lol.

Happy America's birthday!

20150527_150752.jpg

Share this post


Link to post
Share on other sites

One week and no word from GD. He must have gotten lucky, or really unlucky, at the bar.

Here is where we are at: Replaced five shims in the stack taper with thicker 0.25mm shims. Based on shim factors the replacement shims should increase the stiffness of the high speed stack by 40%.

5-mx9c.png

FEA Analysis

So what about FEA analysis? If you ran FEA on the two stacks what does that say about the difference in stiffness between the two stacks?

FEA computes the shim stack deflection as a function of force applied to the stack. At a given force you can integrate around the perimeter of the valve port seat and get a measure of the fluid flow area between the shim stack and valve seat at the computed stack deflection. Do that over a range of stack forces and you get a curve showing how the fluid flow area changes with stack force.

For the stacks here FEA shows the shim stack deflection profile is pretty linear for both stacks. The difference in stiffness is also nearly constant. That means there is not going to be any difference in the shape of the damping force curve between the two stacks. The mx10c stack is just stiffer and produces about 10% less flow area over the deflection range. That's a whole lot different than shim factors that expected a 40% increase in stack stiffness at high speed.

6-mx9c-fea.png

The FEA calculations show the mx9c stack is about 10% softer. That means when the shock is producing the same damping force, with the same pressure drop across the valve, the softer mx9c stack is going to deflect to a 10% larger flow area. With 10% more flow area the shock is going to have to operate at a 10% higher shaft speed to produce 10% more fluid flow to fill that increased area. With that thinking the mx9c stack should match the damping force of the mx10c stack when the mx9c stack is operated at a 10% higher shaft velocity.

FEA Stack Analysis

  • 10% change in shaft velocity for same damping force

Tapered Stack Shim Factors

  • 40% stiffer at high speed

  • 16% stiffer overall

That’s the theory anyway, what does the dyno say?

Share this post


Link to post
Share on other sites

Dyno Test Results:

This second dyno test shows the same thing. Increasing the stiffness of the high speed stack increases the damping force everywhere, not just at high speed. Shim factors expected the high speed stack stiffness to increase by 40%. The dyno shows no increase in high speed damping. Instead the damping force just got stiffer and it was stiffer everywhere across the entire speed range. FEA calculations show the same thing, stiffer everywhere.  

7-mx9c-dyno.png

  • FEA analysis expected the shaft speed would have to be 10% higher for the mx9c stack to match the damping force of the mx10c stack. The dyno data shows that result.
  • Shim factors expected the mx10c high speed stack to be 40% stiffer and by webology thinking that was supposed to really stiffen up high speed damping. Instead, modifying the high speed stack just made the damping force stiffer everywhere, not just at high speed.

Dyno results of this test, and the previous test, both show the same thing. Increasing the stiffness of the high speed stack increases damping force everywhere, not just at high speed.

Share this post


Link to post
Share on other sites

Something of note - while it is 10 percent everywhere - let's make a point here - 

 

1000000 * 0 = 0

 

10 percent of nothing is nothing.

So at lower shaft speeds - the actual FORCE difference felt to the riders is minimal.  Yes it's 10 percent - but visually until 40inch/sec the reality is a rider could make clickers adjustments and get a very similar feeling.

Whats odd here - I have run FEA on some simple stacks using ansys - and if I recall correctly the thickness cubed rule held very true.  What have I missed?

Share this post


Link to post
Share on other sites

sorry, i'll get that spec out. I have rediscovered the weber, and food comas are commonplace.

Share this post


Link to post
Share on other sites

18.2,17.2,16.2,15.2,14.2,13.2,12x1mm plate,(4)12.25,12.3,9.3,10.1,11.1,12.1,13.1,(2)14.1,8.25

the shock had a piggy-back nitro canister as well.

shock was a pro circuit piston with the larger mx comp ports. I replaced the piston band and oring, as they fade faster than the stockers.

Edited by GDI70

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Reply with:


×