The relative benefits of a pull rod suspension

Here are our CFD links and discussions about aerodynamics, suspension, driver safety and tyres. Please stick to F1 on this forum.
User avatar
ringo
227
Joined: 29 Mar 2009, 10:57

Re: The relative benefits of a pull rod suspension in 2011

Post

Alienating?
We are not the same, I am a Martian. 8)



We were having a nice discussion until you popped in with some old stuff that is more subjective than anything.
The logic behind the movement of masses and also the narrowness of the top of the gearbox was phasing in, then you come and create a diversion with old arguments about clumsiness and such.

You mentioned talk about long gear box and other pull rod users having clumsly solutions. Follow closely:

Image

here we have a gearbox with a arms attached. We have pushrod in blue and pull in green. All are symetrical to make this a fair comparison.
The dimensions are realistic and estimated from the F150th italia.

The wheel force is being applied through A. The member is AG for pushrod. And member D, oh i left out a letter, call it DC.

theta xz is the angle of the vertical plane of the pushrod with the floor.

both rods titanium. pull rod is solid.

pushrod is tube. inner diameter is 0.9 of the outer. but i'll use solid first just to show equality and prove i'm not rigging the numbers to suit my agenda.

titanium yield strenght = 830 N/mm2. young's modulus E = 1.1x10^5 N/mm2

factor of safety is 1.

Ok,

Assume a force through the tyre equal to 2 times the corner weight = 3200N.

Find the lenght of AG (= DC) AG = 690.66 mm , basic, no eplanation for that.
so AG = 691mm then.

sine theta = 225/691 = 0.3257. the angle is 19 degrees.

pull rod force = push rod force = F/0.3257 = 3200N/0.3257 9825N

Now for the pull rod DC, it's under pure tension, so simple stress theory used.

minimum area: 9825N/830N/mm2 = 11.8mm2,

pull rod diameter = 3.87mm ~ 3.9mm

Buckling calculation. The load cannot exceeded the buckling limit load.

So F = (pi^2)EI/4L^2

solving for I, I = 4FL^2/pi^2xE
= 4 x 9825 x 691 x 691 / 9.8596 x 1.1 x 10^5
= 17308.3

I= pi x d^4/64 ,
d = 23.36 mm

pull rod diameter = 3.87 mm push rod = 23.36mm

mass of titanium :

pull rod mass = 4.5g/cm3 x .118 * 69.1 = 36.69 g

push rod mass = 4.5 x 4.28 x 69.1 = 1332.6 g
TUBE
for the tube d inner 0.9 of d outer.

outer diameter for pushrod = 31.82mm for pull rod tube it's 8.9 mm.



Now these numbers only show that the pull rod is thinner and much lighter. We already know that, and this can only show aero benefit.
At least it put things into perspective. Just in case some one askes why are they so thin, it's because of the safety factor and the load i assumed.

But what can be shown further is that if the distance EG is increased. We will reach a point where the push rod can exceed the maximum allowed width of a suspension element which is 100mm.
Last edited by ringo on 10 Mar 2011, 00:10, edited 1 time in total.
For Sure!!

User avatar
ringo
227
Joined: 29 Mar 2009, 10:57

Re: The relative benefits of a pull rod suspension in 2011

Post

Using the same diagram, lets say mid season ferarri want to push the push rods up further.
If the new rocker arm point is at d2, 550mm away from wheel center line, what happens?
Lets see if a push rod can take advantage of a long gearbox just the same.

New length AG and DC = 776.6 = 777mm

angle reduced to 16.84 degrees.

Force increased to 11,045.9 N.

pull area = 11,045/830 = 13.3 mm2,

New buckling load limit = 4 x 11,045 x 777 x 777/ 9.8596 x 1.1x10^5
I = 2.457x10^5

push rod diameter = 47.3mm , mass = 6.144 kg

pull rod diameter = 4.1 mm, mass = 45.45g

For tube with inner d 0.9 d outer,

Diameter push rod = 61.75mm :o , Diameter pull rod = small. :mrgreen:

So those are real numbers, and it is clearly shown that a push rod is in a constant struggle as the length increases. It doesn't have the freedoms of a pull rod.

It pays for it in it's mass. 6 kg is nothing to smile at for unsprung mass.

And this can get worse if i used a factor of safety greater than 1. No doubt it will exceed the maximum diameter of 100mm if i used this tube.
Though i could use a thicker wall to mitigate that.

Any how you cut it, it's going to get very clumsy place a fat rod about the A arms.
It can't weave trough and out the A arms like redbull did with the pull rod.
It can't weave through, it has to stay within the A arms.
It can even cause mounting problems on the up right and mess around with the brake duct design.

The teams using pull rod now, will most definitely slim them down as the season progresses.
Performance differentiator exposed.
For Sure!!

User avatar
Ciro Pabón
106
Joined: 11 May 2005, 00:31

Re: The relative benefits of a pull rod suspension in 2011

Post

Excellent, ringo.

Anyway, this has been said, with much less detail: members in tension are lighter than members in compression in the vast majority of cases (well, there are short members that work better in compression, but...).

However, there are other things you already know about the implications of each system that limit the structural advantage.

As usual, if you have a better recipe it doesn't mean you'll make the best dinner, because a dinner involves more things that just the main course.
Ciro

User avatar
747heavy
24
Joined: 06 Jul 2010, 21:45

Re: The relative benefits of a pull rod suspension in 2011

Post

should we include the forces feed through the top and bottom wishbone in your assessment Ringo?
What does it means for the dimensions and weight of these components?
Do we need stiffer/stronger top wishbone pickup points at the gearbox, when we go to pullrod?
"Make the suspension adjustable and they will adjust it wrong ......
look what they can do to a carburetor in just a few moments of stupidity with a screwdriver."
- Colin Chapman

“Simplicity is the ultimate sophistication.” - Leonardo da Vinci

User avatar
ringo
227
Joined: 29 Mar 2009, 10:57

Re: The relative benefits of a pull rod suspension in 2011

Post

Yep, it can be done, but i didn't want the main point to be drowned out.

The wishbone forces will be interesting because of the kind of loading, compression or tension, as well as the joint displacement due to strain.
I guess tomorrow it can be looked at.

a quick look at it;
ED will be compression for the pull rod. GD compression, This will be the heaviest member of the pull rod.

AB tension, AC tension for push rod.

So we have an interesting balance, here. It may turn out that the masses will be close between the total unsprung mass, or it may not.

The only thing that will spoil it for the push rod tension streak with AB and AC is braking forces and lateral tyre forces.

I'll size them tomorrow.
For Sure!!

Jersey Tom
Jersey Tom
166
Joined: 29 May 2006, 20:49
Location: Huntersville, NC

Re: The relative benefits of a pull rod suspension in 2011

Post

IMO it's almost impossible to truly say what unsprung weight is doing for mechanical grip... so I wouldn't play that card too heavily.

And there's a reason why RBR's pullrods aren't 4mm tubes or whatever dimension you have spec'd. Are rods in tension probably going to be built lighter than those in compression? Sure, but not by 10+ lbm in this case.

Amazes me that this thread is still going.
Grip is a four letter word. All opinions are my own and not those of current or previous employers.

User avatar
ringo
227
Joined: 29 Mar 2009, 10:57

Re: The relative benefits of a pull rod suspension in 2011

Post

I am not saying what it is doing for grip, i don't know. I just know less mass is better.

The 4mm diameter is very realistic and so are the mass differences.

It is a simple as the assumptions presume.

I assumed titanium for both rods. Maybe a team uses crfp with titatinum for a pushrod instead to avoid that weight.

I assumed an arbitrary force. I really don't know what the worst case vertical tyre force is in an F1 car, much less lateral or longitudinal.

I said my safety factor is 1. Teams wont use that low of a number.

There is consideration for fatigue. I did not do that.

Then there is impact, I did not do that.

There are many other factors that could been considered.

It makes no sense showcase the details, calculation is hardly done on the forum either, so i'm not sweating the details.
For Sure!!

Sayshina
Sayshina
1
Joined: 04 Mar 2011, 21:58

Re: The relative benefits of a pull rod suspension in 2011

Post

First, Ringo, it is fundamentally unfair for you to scold anyone for including aero and/or transmission issues in a "suspension thread" when you yourself have done exactly that a number of times. They are fundamentally intermarried systems that can only be discussed in isolation on a profoundly limited basis.

ringo wrote:
Sayshina wrote:
n smikle wrote:The first pull rods were nothing compared to the ones today. Just google some photos. Similar to the first EBD's to todays. Very crude.
Are you claiming that the current Pushrods are exactly the same as the first and haven't also benefitted from development?
The truth is that the push rod has hardly changed since its inception.
A push rod 20 years ago is just the same as it is today.
Parts inboard of the car on the gearbox. Maybe a coil spring or torsion bar.

We can't say the same for a pull rod.
Especially in the rear of the car.

I can bet you that you can't find a torsion bar pull rod suspension in the rear of the car, that is mounted inside a gear box.

If you can find that then you have a case.
To both Smirkle and Ringo, wow. Just wow. We've gone from round section steel members (all members) connected by joints to aero section (both positive and neutral) fabricated composites connected by flexors. We've seen 2 dampers with coilovers, monoshock, 3rd spring, 3rd damper, single spring, innerters, mass dampers, do I really need to go on? What, the pushrod itself is still just a rod so there's no change? That's like claiming there's been no changes in brakes since the 1940's because after all, cast iron or carbon they're still just rotors and calipers.

And as far as googling early cars, I was around then, I'm perfectly capable of remembering them. What you've really seen is a gradual development of pushrod, granting the illusion of little change, whereas pullrod has gone directly from then to now with nothing in between. It only looks like pullrod has changed more because it's happened overnight.

Both pushrod and pullrod F1 suspensions have only ever existed in an aero world, and were developed for aerodynamic gain. In recent years we have seen a number of cars that performed quite well (relative to their position in the pecking order) on high speed stuff, only to display horrible mechanical grip whenever they dropped out of their aero envelope. Those teams have often chosen to ignore their lack of mechanical grip (where they were bad) in favor of chasing aero gains (where they were already good). This is a clear indication that in current F1 thinking, mechanical grip and suspension dynamics are a very low priority in comparison to aerodynamics.

In recent years we have seen VERY different transmission layouts, going to and then away from transverse units as but 1 example. In all cases, up to and including todays RBR and Williams units, it is clear that aerodynamics are not only pushing but defining transmission shape and size. It then follows that current F1 thinking considers the transmission to be an aero part which must also perform some secondary functions.
I can bet you that you can't find a torsion bar pull rod suspension in the rear of the car, that is mounted inside a gear box.
Explain to me please what the mechanical benefit is in having any of your suspension components inside the gearbox. For that matter, explain please the mechanical benefit you think derives from using a torsion bar or coil spring, pick whichever you think is superior and I will be happy to argue the other side. Keep in mind that a coil spring provides most of its resistance in torsion anyway.

These things are not suspension advances. Going to torsion bars did not make for a mechanical grip improvement, it was done for "packaging purposes", which is to say aero.

During last season and the one before, the pullrod layout was decidely inferior to the pushrod layout, perhaps through no fault of its own, but still a fact admitted to by your lord god allmighty, Newey himself. The Red Bull was clearly quick despite this handicap.

Transmissions are thin walled box structures with internal baffling. Increasing 1 dimension so that you can reduce the other 2 has a rather large and negative impact on the rigidity of your structure. Punching holes in the walls of a structure also nagatively impacts rigidity. Punching extra holes in a structure just so you can put say suspension components in there, and by the way feed suspension loads in from shall we say odd angles, will have a decidedly negative impact on rigidity.

From both a reliability and a vehicle dynamics perspective, there will be a minimum acceptable rigidity. In both cases one would normally assume more is generally better. What we remove from structure we need to either accept as a loss, which seems unlikely in this context, or replace the only way we have left, by adding matereal. It normally takes quite a bit of matereal to make up for a compromised structure.

Ringo, in your attempt to demonstrate a clear weight advantate for pullrod vs. pushrod, you've failed to include one tiny detail. Any pushrod capable of handling its compression loads will be perfectly at home when it finds itself in tension, while a pullrod designed only to handle its tension loads will rapidly aproach failure when it finds itself in compression.

Both rods see compression loads. The actual wheel movement on track tends to be significantly more violent in bump than rebound, and because of this there are theoretical weight gains to be realized from a pullrod setup. However, as I mentioned a few pages ago, these theoretical gains tend not to actually happen in the real world.

By the way, of all the load paths from suspension to transmission, you focus on the 1 path that sees damping and think the location of that is going to be the defining factor in where matereal is going to be placed in the transmission housing? One would expect there to be more matereal down low in a pullrod setup, in comparrison to pushrod. But the difference won't be nearly as much as you seem to expect.

You do realize that if gound effect does return that will most likely be the end of your holy pullrod, don't you? But then once Newey stops using it will it even be holy anymore?

User avatar
Ciro Pabón
106
Joined: 11 May 2005, 00:31

Re: The relative benefits of a pull rod suspension in 2011

Post

Holy cow, Sayshina. Is there nothing sacred anymore for you, kids?

Image
Ciro

User avatar
Ciro Pabón
106
Joined: 11 May 2005, 00:31

Re: The relative benefits of a pull rod suspension in 2011

Post

Jersey Tom wrote:...

Amazes me that this thread is still going.
I did this specially for you, JTom...

Image
Ciro

User avatar
Ciro Pabón
106
Joined: 11 May 2005, 00:31

Re: The relative benefits of a pull rod suspension in 2011

Post

ringo wrote:...
There are many other factors that could been considered.

It makes no sense showcase the details, calculation is hardly done on the forum either, so i'm not sweating the details.
Well, ringo, I rechecked your calculations.

The formula you used is valid only for axially loaded columns. You did not calculate for bending moment at the center of the beam. Why? Which member is taking the load perpendicular to the pushrod? The spring? I want to say thanks in advance for any explanation anyone can provide.
Ciro

User avatar
ringo
227
Joined: 29 Mar 2009, 10:57

Re: The relative benefits of a pull rod suspension in 2011

Post

Pivoted joints cannot transmit moments; assuming the pivots are frictionless.
So ideally there is no bending in the rods.
I don't have much knowledge about f1 joints, so i don't really know the details, but i am assuming the rods are pivoted and frictionless.

If they were fixed at the ends then bending could be considered.
In the case of the flextures on the wishbones, anti dive, etc. bending could be considered too.

In the civil engineering world you guys deal with a lot of fixed ended columns.
I see you have the urge to bolt the push rod flange to the A arm. :mrgreen:
For Sure!!

User avatar
ringo
227
Joined: 29 Mar 2009, 10:57

Re: The relative benefits of a pull rod suspension in 2011

Post

Sayshina wrote:
Ringo, in your attempt to demonstrate a clear weight advantate for pullrod vs. pushrod, you've failed to include one tiny detail. Any pushrod capable of handling its compression loads will be perfectly at home when it finds itself in tension, while a pullrod designed only to handle its tension loads will rapidly aproach failure when it finds itself in compression.

A drooping wheel wont cause much damage. That is a dynamics problem and i am not sure of the force a drooping wheel will hit the ground with; that has to be found if you know the damping, spring rate, weight of the car, height of wheel off the ground and the down force.
Both the rods have a free lunch in rebound i guess. So no i don't think one will rapidly approach failure.
Dave W could probably add to the rebound thing. I have no clue at the moment.
Both rods see compression loads. The actual wheel movement on track tends to be significantly more violent in bump than rebound, and because of this there are theoretical weight gains to be realized from a pullrod setup. However, as I mentioned a few pages ago, these theoretical gains tend not to actually happen in the real world.
Why should I believe you? Becuase you said real world? What theoretical gains, Is more violent 100 times more force, a million ?
Where is you supporting evidence.
By the way, of all the load paths from suspension to transmission, you focus on the 1 path that sees damping and think the location of that is going to be the defining factor in where matereal is going to be placed in the transmission housing? One would expect there to be more matereal down low in a pullrod setup, in comparrison to pushrod. But the difference won't be nearly as much as you seem to expect.

You do realize that if gound effect does return that will most likely be the end of your holy pullrod, don't you? But then once Newey stops using it will it even be holy anymore?
Maybe. I don't care for the pull rod. I just like to see teams make the best use of the existing regulations. It's not much a big difference between the 2 choices, but there is a finite difference.
If ground effects come back i would like to see rocker arms with torsion bars. Rocker arm suspension uses no rods at all. And torsion bars would eliminate the need to have big coil springs.
Full exploitation is the Tao.
For Sure!!

User avatar
747heavy
24
Joined: 06 Jul 2010, 21:45

Re: The relative benefits of a pull rod suspension in 2011

Post

ringo wrote: A drooping wheel wont cause much damage. That is a dynamics problem and i am not sure of the force a drooping wheel will hit the ground with.........
This may or may not be the case.
I hope you don´t consider the drooping wheel, the only instant when the pull rod is subject to compession.
BTW how is your force assessment for the rest of the suspension members going?
"Make the suspension adjustable and they will adjust it wrong ......
look what they can do to a carburetor in just a few moments of stupidity with a screwdriver."
- Colin Chapman

“Simplicity is the ultimate sophistication.” - Leonardo da Vinci

DaveW
DaveW
239
Joined: 14 Apr 2009, 12:27

Re: The relative benefits of a pull rod suspension in 2011

Post

ringo wrote:...If ground effects come back i would like to see rocker arms with torsion bars. Rocker arm suspension uses no rods at all. And torsion bars would eliminate the need to have big coil springs....
I really don't think you would, ringo. Some rough & ready estimates of installation stiffness for the front suspension of open wheeled race vehicles:

The last Dallara push rod IRL: 3 KN/mm.
The current Dallara pull rod IRL: 1.5 KN/mm.
A fairly late F1 vehicle with a beam rocker suspension: 0.9 KN/mm.

Why was the pull rod lower than the push rod? I don't know, but could guess.

Why was the beam rocker so poor? Because suspension loads are carried by the rocker in bending. Think about it. A locked damper will lock the wheel relative to the chassis only if the rocker has infinite bending stiffness. The example quoted had a hub mode with only 15% of critical damping, largely because the hub was able to thrash around without troubling the damper.

A beam rocker layout is likely to be heavy, have a high c.g., & it will lack hub control.

p.s. Just to complete the story of a beam rocker suspension, It is almost inevitable that it will have a high motion ratio (it was 1.46 for my example). Now, if a typical damper has an installation stiffness of 3.5 KN/mm, this will translate to a value of 3.5/(1.46)^2 = 1.64 KN/mm at the wheels. Thus the overall installation stiffness for the suspension will be around 0.58 KN/mm, or less than 1.5*tyre stiffness (for my example).