shock friction 7

[bhart]

Another recent thread mentioned the problem of friction in shocks due to the side loading imparted on the shock when a coil over spring buckles under compression. I have been curious about how big of an effect this is (I would like to see some numbers). The main problem that I see w/ quantifying this is that I don't know what would happen to a damper dyno if you tried to run it w/ a spring mounted w/ the shock -- maybe people do it all the time. I know a lot of racecar teams are spending a lot of R&D $$$ to try and reduce all forms of suspension friction (shocks included) and I assume this is even why some F1 teams have gone to a torsion bar setup.
I believe that this friction can and does significantly affect the performance of the suspension, but I would like to see some numbers.
Thanks in advance,
brandon

 

[mloew]

Brandon,

I can't quote numbers directly, but in the early stages of a production vehicle module design, the measured side loads were in the 1000 - 2000 N range.

A product of interest to the race car side of the industry might be the Hyperco/ICP Hydraulic Coupling Device:

http://www.propartsusa.net/products/hypercoil/hcd.htm

http://icpcitation.com/shockperches.htm

This devise allows the spring seat to pivot to reduce the side-load of the spring.

There is an article on this system in the April/May 2003 issue of Racetech Magazine (

http://www.racetechmag.com).

Additionally, The following SAE International (

http://www.sae.org)

technical papers all include information on the phenomenon of side-load in springs:
2002-01-0317: The Effect of Rubber Seats on Coil Spring Force Line
2001-01-0496: Modeling of Coil Springs Using Parallel Mechanisms
2001-01-0497: Development of L-Shape Coil Spring to Reduce a Friction on the McPherson Strut Suspension System
2000-01-0101: Optimization of Force Action Line With New Spring Design on the McPherson Strut Suspension for Riding Comfort
2000-01-0727: System Modeling of a Damper Module
970100: Optimization on Mac Pherson Suspension With a Spring
960730: Approaches to Minimizing Side Force of Helical Coil Springs for Riding Comfort
940862: Side Load Springs as a Solution to Minimize Adverse Side Loads Acting on the Mcpherson Strut

For automotive damper struts, the springs can be intentionally designed to produce side-loads to minimize system friction. For damper modules, the side-load needs to be minimized.

Hope this helps.

Best regards,

Matthew Ian Loew

Please see FAQ731-376 for tips on how to make the best use of Eng-Tips Fora.

 

[GregLocock]

Matthew's post is correct, around 1000 N is a typical figure.

Thanks for the papers, I'll have a read, since I am stuck in McP strut hell at the moment. Grumble grumble.

When we get into trouble on this we strain gauge the tube so that we can measure the forces. Given the lower shock velocities and high spring loads a race car sees I imagine this is rather important. Another option, if you have the room in an SLA or double wishbone, is to terminate the shock at a ball joint, integral with the lower spring tower. This would stop the spring loading the shock in bending. I've only just thought of that, does it make sense?




Cheers

Greg Locock

 

[mloew]

Greg,

I thought so. Have a look at the formulation that relates the spring seat position on the damper body to the side-load resolved at the rod-guide and piston.
____F_____
0[ [==]=========0
A B

Dimensions:
L = length between bearings
DL = length of damper body (between lower bearing and rod-guide)
X = position of F from lower bearing
BS = bearing span (between piston and rod-guide)

Forces:
A = force acting at lower bearing (assume + up)
B = force acting at upper bearing (assume + up)
F = side-load force acting at X (assume + down)
F[sub]p[/sub] = Force at piston
F[sub]rg[/sub] = Force at rod-guide

Eq 1) A + B - F = 0
Eq 2) F(X) - B(L) = 0
Eq 3) F - A + F[sub]p[/sub] - F[sub]rg[/sub] = 0
Eq 4) F[sub]p[/sub] + B - F[sub]rg[/sub] = 0


I am too lazy to solve this right now.




Best regards,

Matthew Ian Loew

Please see FAQ731-376 for tips on how to make the best use of Eng-Tips Fora.

 

[GregLocock]

I was just going to model it and see if it broke!

I am 99% confident it is a good idea, it just may be a bit too complex for a production car, and too long for a racer.

Actually there are endless variations possible on this, there is definitely a strong case for replacing the bottom bush on a shock with a ball joint. Might even improve micro shake.









Cheers

Greg Locock

 

[pmwltd]

Most race shocks come with spherical bearings mounted in upper and lower eyes. Although the claim is that it is to reduce compliance, which is present in rubber or urethane bushings.

Even with strut suspension we convert the upper pickup point to a sperical bearing.

Dave

 

[patprimmer]

Greg

With some modern "bearing grade" engineering plastics, it might be possible to design a high volume spherical joint at a price that might be viable.

I have recently seen some very high performance from a Bronze Filled Nylon, with mechanical and bearing properties rivaling sintered metal.

i don't want to get further into it here, as I do have a commercial interest, and this is not the place to pursue it



Regards
pat

 

[mloew]

There are several automotive bushing manufacturers that offer a "slippery" bushing. It is a rubber mount that has conventional radial and conical rates in the compliant material and a bearing type arrangement to allow for "free" rotation (once the friction is overcome). Pretty slick for overcoming some of the moment generated at the lower mount.



Best regards,

Matthew Ian Loew

Please see FAQ731-376 for tips on how to make the best use of Eng-Tips Fora.

 

[patprimmer]

I'm talking about an injection mouldable very hard thermoplastic that rivals sintered metal as a bearing, except at high temeratures of course, but shock mounts can't get that hot, unless they are real close to exhaust or brakes

Regards
pat

 

[GregLocock]

I may need to talk to you Pat, in about six months time. Biggest problem I can foresee is topping loads, and brake heat. But if a rubber bush can handle the heat I'd a thunk your material would be OK.





Cheers

Greg Locock

 

[GregLocock]

OK, we've just got serious about this. Pat can you email me at greglocock at optusnet dot com dot au? It is a long shot, but if you can show cost or performance advantages over a spherical bearing we'll listen.



Cheers

Greg Locock

 

[patprimmer]

It's a new product and we don't have a lot of data organised yet, and we only provide raw materials, not bearings as such, but being injection mouldable, the price should be much lower than spherical bearings, provided you have the numbers to amotise the mould. As I know you work at Ford, I am sure you have the numbers, unless you are only planning a low volume special for this.
I will do some homework and send what I have to your e-mail

Regards
pat

 

[PTwizz]

Gentlemen, in answer to some of the queries above: Certainly side loading induced by assymetrical spring behaviour increases friction in dampers. We have found this to be a significant issue where very stiff springs are used, such as those on IRL and CART oval setups. Our answer to this has been to use linear recirculating ball bearings in our dampers. The reasons for using torsion springs in F1 are more driven by packaging and load distribution factors, it is possible to make very small dampers when there are no spring loads to be considered.
Almost all shocks supplied by us have spherical bearings at both ends, although it should be noted that these are outside the spring/damper sideload system and have no impact on that issue.
There is no problem with running dampers with springs on the dyno (provided the spring load plus max damping force does not overload the dyno). The spring force is a straight line (disregarding hysterisis) which can be "torn off" from the damper curves.
We have a dummy damper (linear bearings and no internals) specifically for calibrating dyno load cells against a known spring.
Pat. I would also be very interested in your plastic bearing material. What do you call "high temperatures"? some of our dampers run up to 150°C (heat soak in the pit lane being the highest temperature event). Does the material work as a liner, or is it strong enough to make entire spherical bearings? Contact PTwissell at Multimatic dot com if you wish.

PT

 

[Uber]

Dear bhart

Actually I am a shocks engineer for 9 years, so I think I can add some comment here.

Shocks friction is an issue of vehicle ride performance especially for passenger vehicles. Currently the competitive shocks for passenger vehicles, which is normally twin-tube shocks, the friction of the shoks does not go over 50N without any side load. The test speed is important factor for measurement of the shocks, it is from 0.0003 to 0.0052 m/s which is somewhat variant between auto makers.

This case is for smooth ride oriented shocks, there I have no idea how race teams treat the friction of the shocks.

 

[exige]

If you want to reduce friction inside a shock absorber, you have to make a compromise. To lower the friction you can shorten the guidings, and use softer material for the dust/oilseals. This will ware out the parts much quicker, so the shockabsorbers have to be serviced more frequent.

And for the springs, the idea of letting the spring twist on bearings is very good. Because when the spring is compressed, it wants to twist. Using bearings, the spring can twist very easily. By doing this, the traction of the car will improve, because the shocks can now follow the little bumps in the road, while shocks with more friction aren't able to move. and this will make the car very bumpy, especially at low speeds.

 

[Tmoose]

&quot;Another option, >snip< is to terminate the shock at a ball joint, integral with the lower spring tower. This would stop the spring loading the shock in bending.&quot;

It might remove the moment created at that seat, but If we are still talking coil-overs or macPherson struts I think if one end of the spring is still on a seat affixed to the the shock body it could exert a moment (if it felt like it), unless the ball jointed spring seat could move laterally.

I picture the playground ride consisting of coil spring welded to a plate on the ground, with a kid sitting in a saddle on the free end. When the kid leans forward the seat may pivot significantly, but eventually the spring resists more translation.

 

[GregLocock]

You can buy spring platforms like that, basically they are large thrust plates that can articulate. Not much use in production, but good for defining your problem.

You are right, the ball at the bottom does not eliminate all of the moments in the shock, if it has an integral spring seat. In practice though it does seem to improve microshake possibly because it provides very high axial stiffness and low coning and torsional rates - possible in a rubber bush, but easier in metal.

I notice Mercedes have this on their double ball joint (virtual king pin) suspension.

It would be nice to move away from coil over shock to free up the shock, but I can see no nice way of doing it with a coil spring, especially given that we normally have to protect for, but never design, a 4 wheel drive. No, we can't go to torsion springs, or lever type shock absorbers!



Cheers

Greg Locock