Direct Acting Spring Damper

[ritwikdas18]

1.Z, I checked with the radius of gyration of typical FSAE Cars in pitch and roll (which is 600mm and 300mm respectively) .My values may be wrong(infact they are wrong) but its close to actual values.All these values are anyways ballpark values because I don't even know the exact CG height and weight distribution of my car . Infact I don't even know the exact unsprung or sprung weight of my car.
2.Z & Claude , Could you comment on my damper code selection? Its at the lower right corner of the damping rate spreadsheet.
3.Also one more thing I wanted to say that I plotted a graph of motion ratio variation v/s bump travel. Now its within 1.5% at the front and within 3% at the rear (for 1.6 inch of travel in compression and rebound). Its degressive obviously but as far as I think it wont make much of a difference because the above numbers would lead to a 3% and 6% increase in roll rates at the front and rear(due to (MotionRatio)^2)) respectively and that too when we assume that the travel is 1.6 inch in compression and 1.6 inch in rebound which is too much for my case(<.5 inch dynamic wheel travel). So any comments on the above?

[Tim.Wright]

Non linear motion ratios aren't necessarily bad, but you need to keep an eye on how they change the LLTD balance in function of roll angle. These non linearities will likely change the balance of the car between the linear and limit ranges.

[CWA]

My interpretation is as follows:

Goost's original plots / Claude's descriptions (pretty standard 1DOF oscillation) depict 'wheel travel' change during a bump event; this is unsprung mass motion in relation to the chassis. They are not wrong. But neither are they describing the actual motion of the chassis through space. Or perhaps more accurately, Goost's plots show displacement of the chassis w.r.t to a bump profile, but not the rest of the road.

As Z says, the motion of the chassis through space (hence ride height variation) should correlate with the vertical forces acting on the chassis, which are generated during these 1DOF oscillation events. With a stiffer car (more springs per mass, or more damping per S/M), yes the wheel travel (w.r.t chassis) is less. But the forces acting on the chassis will be much greater during the event, so doesn't it make sense that the chassis moves more in space? That stiffer suspension gives more ride height variation?

If we picture this in terms of net displacements in space, this also makes sense. Greater wheel travel (remember above definition) motion during a single bump event does not correlate with a chassis ride height increase as I believe is being implied. Is the opposite not true; 'bump profile height' minus wheel travel gives overall (upwards) chassis displacement (w.r.t to the rest of the road). So less wheel travel, means more overall ride height variation.

Consider a front axle and the chassis above it, of a car travelling along a flat road which has a sizeable bump in it. If the front suspension is soft, so soft that the wheel travel seen during the bump event is equal to the bump profile in the road, neglecting phase lags, doesn't the chassis remain flat w.r.t the rest of the road? Doesn't the soft suspension facilitate minimal ride height variation?

Now consider a much stiffer variation of the same car conducting the same event. During the event, less suspension travel means the chassis follows the profile of the road more, and is displaced this much in space (w.r.t to the flat road) too (with all associated uncomfortable accelerations). This second, stiffer case of course has very low wheel travel (unsprung mass motion w.r.t chassis), but the instantaneous ride height variation (of the chassis w.r.t the rest of the road) is going to be much higher than that of the first, softer case.

As ever I welcome correction on this, but this is what I believe to be true.

Perhaps what Goost is trying to imply is that on a bumpy surface, the pitch orientation of a race car (which aerodynamics are perhaps more sensitive to?) is less controlled with a softer car. I don't necessarily disagree with this, but I think it should be made clear that calling this a chassis 'ride height change' issue is perhaps not descriptive enough, and the plots posted above to try and make this point are misleading. To make this point, don't you at least have to consider 2DOF side-view coupling of modes (with pitch inertia considered) rather than 1DOF SMD, as Z has suggested?

If I have misunderstood the issue and this is redundant input then apologies.

[Flight909]

Non linear motion ratios aren't necessarily bad, but you need to keep an eye on how they change the LLTD balance in function of roll angle. These non linearities will likely change the balance of the car between the linear and limit ranges.

Also is important the effect when you have downforce because you can change balance with speed. With more compression of suspension the stiffness will change (maybe even balance front to rear), this can also be good if the tires are not well balanced when you load them.

[Claude Rouelle]

For people who do think that there are no need for suspension...... I just want to remind that damping is the #1 parameter that influences tire temperature (and therefore tire grip) is the suspension damping.

Two years ago I simply jumped on one FSAE car at Formula North. Feeling and looking at the decrease of the amplitude for a given input, stepping on or out the car, I did not need any sensor, just a bit of experience, to know that the dampers were far away from critical damping. I went with one of the team members through some basic calculations and found that their damping ratio was about 0.2 or 0.3 (I do not remember exactly). I suggested that they tune the damper to go to 0.7. They had the damper curves and they were able to do that. The lap time improvement was really significant: 2.5 seconds (again if I remember well) on a circuit of 45 seconds (again, if I remember well) but the tire temperature increased something like 20 or 25 degrees C.

Later on, that winter, that team tried different front and rear roll centers and the difference in the lap time and the car feedback was HUGE. According to the team comments just raising one roll center 10 mm and the other 30 mm made the car 1.5 seconds quicker

I am sure that those team members will recognize themselves in this thread: up to them to comment and confirm with exact numbers.

As a consultant I once had a high end single seater car customer who told me that damper did not count. OK I said and I move the bump and rebound knobs and significantly changed the damper curves. The driver went out on new tires and completely destroyed the tires in 4 laps while he normally was able to run 50 to 60 laps with the same sett of tires. You draw the conclusions.

0.7 of damping ratio in heave is not a bad idea to start with on a FSAE car. But often that will give you way less that 0.7 in roll and pitch. Up to you to decide what damping is the most important.

[ritwikdas18]

Claude Sir, I would like to say one thing ,
For equal ride rate and damper coefficient front and rear and cg at (1/2) of wheelbase, the ratio of (damping ratio in heave / damping ratio in pitch)=Pitch dynamic Index=(Radius of Gyration in Pitch)/(Wheelbase/2) For FSAE cars or race cars for that matter the pitch dynamic index is much lesser than 1 . In my case it is ~0.78 . So Damping Ratio in Pitch =(1/0.78)*Damping ratio in Heave.=> Damping Ratio in Pitch for Race cars/ FSAE is usually greater than damping ratio in Heave. This is however based on the assumption that I have but it stands that damping ratio in Pitch is greater than that in Heave, my excel figures also tell me the same.

https://docs.google.com/spreadsheets...it?usp=sharing

Also Claude ,
I don't want you to give me the answer but could you comment on my Damping Rates (Damper Code Selection) & Ride and Roll Rates.

[Z]

Ritwik,

Per your numbered points top p4.

1. Your estimated Pitch and Roll Radii-of-Gyration are close enough. The Yaw-RG will be a bit bigger than P-RG. It is the one that all Teams should work hardest to reduce (ie. for "agile" car), but it is also NOT relevant here...

2. Damper code selection? This depends on money available, but I would be inclined to buy 4 x sets of damper-shims in each of the #3, #4, and #5 ranges (Edit: Oops, see correction at bottom of post...), for low/high-speed and bump/rebound. If you can afford this, then there is no problem having many shims sitting in a cupboard in the workshop. If this too expensive, then buy 2 x sets of (Editted->) stiffest and second-stiffest, and mix-and-match between F&R. But see note* below.

3. The best FS/FSAE driver in the world will not be able to feel any change from your 3% MR change. Especially not when this 3% is over your quoted +/- 4 cm range. As you note, the practical movements during cornering will be more like +/- 1 cm, giving ~0.007 x MR change. And since these changes are likely same F&R, then NO change to "SS-balance".

* Your most important goal now is to get a well-detailed and ROBUST car built ASAP. Then straight into MUCH TESTING and DRIVER TRAINING. A moderately well-trained driver will make much bigger gains (ie. be much faster!) than any amount of tweaking that you now do to the "theoretical" suspension design.

During said testing you will also have a great deal to learn about the best (???) tyre-pressures, toe-angles, and camber-angles. Only AFTER you have these three working sort-of-OK should you try a different set of carefully chosen spring-rates and damper-shims. Then new tyres (you should have already worn out one set) ... and more adjustments to tyre-pressures, toe, camber...

Once again, most important to fast competition times is much driver training time.

Which reminds me, what competition are you going to?
~~~o0o~~~

CWA,

Your understanding is the same as mine.

Z

(Edit: Just checked Kaz-damper-graph... It seems they have shim-kits #s 1-2-3-4 getting progressively STIFFER, but then #5 is the SOFTEST!? So, in my above comments read it as "buy the three STIFFEST ranges of shim-kits"...)

[ritwikdas18]

We are an Indian team and we get the dampers from US.US dollars when converted to INR are really too much to afford, Maybe, maybe we can afford different spring rates if more money comes in at the end but not now. Also about the event:
We are required to complete the CAR and submit its running video with pictures showing that we abided by the rules by 5th October .Without that deadline completed we will not be allowed to participate/register at the event.We are participating in Formula Student India 2016.(FSI). So obviously everyone at the event will have tested their car a lot simply because the rule requires them to do so. About our past result ,In FSI 2015 we stood 8th place among ~40 teams .

[Goost]

No, "mass", in itself, CANNOT "transmit the sudden impact".

Again, I said arguing over this is not fruitful. If you apply a practical force to anything in real life, the only way it can truly instantaneously react is by its mass. A spring requires displacement, a damper requires velocity.

The important part of this is that this input IF A Force only adds a certain amount of energy to your system.
But ah now you agree we mean displacement...

where a known input Displacement occurs at the wheel, and it's fast enough you might even say it was 'instantaneous'.
In this case, yes stiffer spring with same displacement means More Force (or more energy).
If the car had no dampers - would the spring rate change the amplitude of oscillation at all? or would the stiff one 'launch'?
If it 'launches', you have got to say what part of the model was ignored before anyone can move forward.

~~~

By contrast, "SD" is so-stupid-simple that I calculated suitable numbers without even using pen or paper (see earlier post).

Yea I saw that. unfortunately, thanks to these calculations Ritwikdas18 has springs (as of the time of this post) that are about twice as stiff as a 'good soft' FSAE car.
Oops.

~~~

Goost's plots show displacement of the chassis w.r.t to a bump profile, but not the rest of the road.

This is what I mean if you like. Really it's just an under-damped, second order spring-mass-damper.
I think ~85% of automotive springs and dampers can be explained with that model.
The things happening in the plot are useful when discussing these concepts, but useless presented to someone wanting to argue.

It also shows Claude and Z were both right about the response with varying damping (the initial question), only Claude's description was more practical at that point in the discussion.

The rest of this is right, which we all agree on and I probably didn't need to repeat in section at top.

~~~

Trying to get to the point...

Claims that the chassis 'launches' anywhere with stiff springs make crummy points for argument as it's a hyperbolic example.
Choosing the softest spring that doesn't bottom out in your suspension to soften ride - depending on the travel you have that's either hyperbolic or simply wrong.

Neither of these examples gives a criteria for choosing anything about your springs and dampers!

it doesn't bother me a bit to ignore the chassis at first and just look at the tire contact with the ground.
There you will find an actual spring rate and damping rate that are ideal for the things you cannot change about your suspension - the Masses and the Tire Spring Rate.
But what defines 'ideal'? I know how I do it, how do you? whether it 'launches' over a bump is not my criteria.

Some things I have learned from this reasoning;

1) why (non-aero) racecars have stiffer springs than road cars, but often not by much (it isn't suspension travel)
2) that choosing a damper somewhere between 'ideal for ride' and 'ideal for grip' is a very good compromise
3) The actual effects of unsprung mass
4) why automotive damper 'curves' are not symmetric in compression/rebound
5) why low speed damping ratios > 1 are completely reasonable

~~~

ritwikdas18,

it would be much easier for anyone to give you advice about your dampers if you reduce it into damping ratios.
You will want the ratios later anyway.

My advice without more numbers would be to let your supplier help you choose them, they are always helpful and usually know what they are doing.

Damper code selection? This depends on money available, but I would be inclined to buy 4 x sets of damper-shims in each of the #3, #4, and #5 ranges

How can you say something like this after discussing for weeks about how to properly pick springs??

[ritwikdas18]

@ Goost
1.The damping ratios are included in the lower right corner of the damping rates sheet. You can also check the damping ratios in heave,pitch and roll similarly and leave a comment
2. Also if you go on the ride and roll rates sheet , and decrease frequency to make it a good soft car then my roll gradient increases a lot to around 3-4 deg/g .The only way to decrease that would be to use ARBs. A lot of time will be used up in designing and manufacturing one because we have never used it before.So I prefer not to use them given that 5th October is the deadline of car completion. If there is mistake in my calculation or anything that you feel is wrong / too stiff then I would request you to suggest me what to do with them. If you feel ARB's are absolutely absolutely necessary given my vehicle information , then please tell me the same. One more reason the spring rates are high are due to low Motion ratios of 0.76(F) and 0.58(R) (I changed the rear one).Since ride rates/wheel rates are important , could you comment on them?
Thanks a lot.