Jump to content


Photo

Push rod question


  • Please log in to reply
38 replies to this topic

#1 pbukovca

pbukovca
  • Member

  • 36 posts
  • Joined: October 11

Posted 07 April 2012 - 19:58

Hello Guys.

I'm just curious as to what would happen if the push rod on the front outside wheel fails on a F1 car in the middle of a corner. Will the car probably fly of the track ?

Would the wishbones still keep that tire loaded and the car still stay on the track ? Or would the whole front end of the car bottom out

on the track ? Or would the bump stop prevent that ?

Many thanks.

Advertisement

#2 Greg Locock

Greg Locock
  • Member

  • 6,353 posts
  • Joined: March 03

Posted 10 April 2012 - 23:43

I'm just curious as to what would happen if the push rod on the front outside wheel fails on a F1 car in the middle of a corner. Will the car probably fly of the track ?

Would the wishbones still keep that tire loaded and the car still stay on the track ? Or would the whole front end of the car bottom out

on the track ? Or would the bump stop prevent that ?


My guess - The car will probably spear off the track, because it is already in a ballistic condition (to use Moss's analogy) and there is no spare grip to tolerate sudden unloading and then loading of one tire.

#3 Lee Nicolle

Lee Nicolle
  • Member

  • 11,038 posts
  • Joined: July 08

Posted 11 April 2012 - 07:57

There is a few drivers around who that has happened too. That and broken rockers. And they are still around to tell the tale!

#4 Paolo

Paolo
  • Member

  • 1,677 posts
  • Joined: May 00

Posted 11 April 2012 - 09:35

The car would behave more or less like it had just three wheels. Which, in a corner taken at decent speed, means it would understeer or roll off the track. Bump stops : it depends on where they are placed and how the triangle is hinged. They could prevent rolling, but before that the tyre would be unloaded and the car would understeer off-track.

#5 kikiturbo2

kikiturbo2
  • Member

  • 869 posts
  • Joined: December 04

Posted 11 April 2012 - 09:45

after a pushrod is broken, there are NO bumpstops anymore.. :)

#6 Paolo

Paolo
  • Member

  • 1,677 posts
  • Joined: May 00

Posted 11 April 2012 - 16:09

It is theoretically possible to use bump stops placed near the hinges to limit the triangle movement. If they are on the rod, of course if the rod goes the bumpstop follow.

#7 MatsNorway

MatsNorway
  • Member

  • 2,822 posts
  • Joined: December 09

Posted 11 April 2012 - 18:10

We have bump stops on the arms on RC cars but

that would be ineffcient in F1 during normal condition as you have to design the lower wishbone to take the full load that the damper rod normally takes.

Edited by MatsNorway, 11 April 2012 - 18:10.


#8 bigleagueslider

bigleagueslider
  • Member

  • 1,235 posts
  • Joined: March 11

Posted 15 April 2012 - 00:30

Current F1 cars actually have a bit of bending stiffness in the A-arms since they rely on flexing for wheel travel. But it's not enough to stop roll in the event of a pushrod failure. The likely outcome of a front outside pushrod failure during cornering would be for the skid plank to bottom out. The end result would probably be similar to a front outside tire failure, since the tire sidewall height is much greater than the plank ground clearance.

Frankly, I don't know how you would design an effective bump stop in an F1 suspension for something like a pushrod failure. What would it look like? Maybe a tension strap or cable?

#9 Magoo

Magoo
  • Member

  • 3,703 posts
  • Joined: October 10

Posted 15 April 2012 - 00:57

I think the biggest variables are 1) the speed of the corner and 2) the point in the corner where the failure occurs. At a bus stop chicane, meh, but in a high-speed sweeper it will be nearly impossible for the driver to hang onto it. Whether the car planks out or slams the bump stop probably won't be crucial. Either way the wheel rate on that corner instantly shoots straight up. This used to happen on Indy cars on ovals in a number of ways and the result was typically a wall impact.

#10 gruntguru

gruntguru
  • Member

  • 7,637 posts
  • Joined: January 09

Posted 15 April 2012 - 04:29

Wheel rate shooting up (hitting a bump stop) would be far preferrable to completely unloading the tyre in favour of the plank. Effective rate while sitting on a rigid bump stop would only be approximately double since about half the vertical compliance is attributed to the sidewall anyway.



#11 Tony Matthews

Tony Matthews
  • Member

  • 17,519 posts
  • Joined: September 08

Posted 15 April 2012 - 06:35

Current F1 cars actually have a bit of bending stiffness in the A-arms since they rely on flexing for wheel travel.

Are you sure about that? They certainly have flexure joints at the wishbone/tub interface. (Bit of tech-speak there). I believe the wishbones to be to all intents rigid.

#12 bigleagueslider

bigleagueslider
  • Member

  • 1,235 posts
  • Joined: March 11

Posted 17 April 2012 - 04:34

Are you sure about that? They certainly have flexure joints at the wishbone/tub interface. (Bit of tech-speak there). I believe the wishbones to be to all intents rigid.



TM,

As I'm sure you're aware, there's no such thing as a truly "rigid" structure. Your basic point is only partially correct. The flexure is located at the point where the moment is highest and is made from a material with a much lower MoE than the rest of the A-arm. It is designed to be soft in bending in the V/T plane, but also stiff in bending in the L/T plane, as well as being stiff in tension/compression. The composite portion of the A-arms are only designed to be stiff in tension/compression along the axis of each leg. The reason for designing F1 A-arm structures like this is because the A-arms are only intended to transmit loads in tension/compression. The pushrod is the structure intended to react the bump, rebound and roll forces.

Once again, the only practical place to locate a bump stop on an F1 suspension would be somewhere on the inboard suspension linkage, such as the rocker or dampener. If the pushrod were to buckle, the suspension at that corner would lose almost all of its bump/roll stiffness.

Incidentally, when composite suspension members were first being tried some of them used a metal/composite hybrid construction. The composite took the tension loads and the metal helped with the compression loads. Composite materials are great at taking loads in tension, but are not so good taking loads in compression/buckling. An F1 front suspension pushrod presents additional concerns with regards to buckling failure modes, since it is exposed to potential impacts.

slider

#13 Tony Matthews

Tony Matthews
  • Member

  • 17,519 posts
  • Joined: September 08

Posted 17 April 2012 - 07:10

I believe the wishbones to be to all intents rigid.



TM,

As I'm sure you're aware, there's no such thing as a truly "rigid" structure.

Yes, I am aware.

#14 rachael

rachael
  • Member

  • 118 posts
  • Joined: October 10

Posted 17 April 2012 - 17:49

TM,

As I'm sure you're aware, there's no such thing as a truly "rigid" structure. Your basic point is only partially correct. The flexure is located at the point where the moment is highest and is made from a material with a much lower MoE than the rest of the A-arm. It is designed to be soft in bending in the V/T plane, but also stiff in bending in the L/T plane, as well as being stiff in tension/compression. The composite portion of the A-arms are only designed to be stiff in tension/compression along the axis of each leg. The reason for designing F1 A-arm structures like this is because the A-arms are only intended to transmit loads in tension/compression. The pushrod is the structure intended to react the bump, rebound and roll forces.

Once again, the only practical place to locate a bump stop on an F1 suspension would be somewhere on the inboard suspension linkage, such as the rocker or dampener. If the pushrod were to buckle, the suspension at that corner would lose almost all of its bump/roll stiffness.

Incidentally, when composite suspension members were first being tried some of them used a metal/composite hybrid construction. The composite took the tension loads and the metal helped with the compression loads. Composite materials are great at taking loads in tension, but are not so good taking loads in compression/buckling. An F1 front suspension pushrod presents additional concerns with regards to buckling failure modes, since it is exposed to potential impacts.

slider


The composite suspension arms are designed with a reasonable bending stiffness to prevent buckling under high compressive loading. The flexures are not made from a material with a particularly low modulus - typically carbon or Ti. The flexibility of the flexures comes from their thin X-section.


#15 saudoso

saudoso
  • Member

  • 6,776 posts
  • Joined: March 04

Posted 17 April 2012 - 20:43

I also have the impression that multipart articulated joints have been obsolete for some years now. At least at the front.

Edited by saudoso, 17 April 2012 - 20:45.


#16 bigleagueslider

bigleagueslider
  • Member

  • 1,235 posts
  • Joined: March 11

Posted 19 April 2012 - 03:28

The composite suspension arms are designed with a reasonable bending stiffness to prevent buckling under high compressive loading. The flexures are not made from a material with a particularly low modulus - typically carbon or Ti. The flexibility of the flexures comes from their thin X-section.


rachael,

While I understand the intent of your post, I might quibble over the specifics of it.

There is a complex relationship between bending and buckling. The nature of composite materials make this relationship even more complex. You can have two composite A-arm legs of equal length, section profile, wall thickness, resin and fibers, the only difference being that one leg is hollow and the other leg has a core of foam or honeycomb. Both legs will have a similar bending stiffness, but the leg with the core will have a much higher compressive buckling margin since the core acts as a shear web that stabilizes the laminate.

When discussing the relative material properties of carbon-epoxy composites and titanium we both need to be careful. Metals like titanium have isotropic properties, while carbon fiber-epoxy laminates are mostly orthotropic. The MoE of titanium is less than half that of a typical UD carbon fiber laminate in tension. So describing titanium as having a "low modulus" in comparison to a UD carbon fiber laminate material is ultimately subjective.

Best regards,
slider

#17 rachael

rachael
  • Member

  • 118 posts
  • Joined: October 10

Posted 19 April 2012 - 20:12

rachael,

While I understand the intent of your post, I might quibble over the specifics of it.

There is a complex relationship between bending and buckling. The nature of composite materials make this relationship even more complex. You can have two composite A-arm legs of equal length, section profile, wall thickness, resin and fibers, the only difference being that one leg is hollow and the other leg has a core of foam or honeycomb. Both legs will have a similar bending stiffness, but the leg with the core will have a much higher compressive buckling margin since the core acts as a shear web that stabilizes the laminate.

When discussing the relative material properties of carbon-epoxy composites and titanium we both need to be careful. Metals like titanium have isotropic properties, while carbon fiber-epoxy laminates are mostly orthotropic. The MoE of titanium is less than half that of a typical UD carbon fiber laminate in tension. So describing titanium as having a "low modulus" in comparison to a UD carbon fiber laminate material is ultimately subjective.

Best regards,
slider


Slider,

You said in post 12 “The flexure is located at the point where the moment is highest and is made from a material with a much lower MoE than the rest of the A-arm”

The suspension legs away from the flexure are made from a mix of cloth and high stiffness ud such that the bulk axial modulus is ~160-180GPa. For a composite flexure there is a similar mix of cloth and ud but both are of a lower stiffness such that the bulk axial modulus is ~120GPa in the flexure region. A Ti flexure will have a modulus of 110GPa so I think it is misleading to state that the flexure modulus is “much lower” than the rest of the leg.

You are correct that cores filled with honeycomb will have a higher buckling stiffness than unfilled but this is not due to shear connection but rather the core maintaining the laminate section like a plumbers spring inside a copper pipe. Aero requirements dictate that F1 suspension members are placed unfavourably for geometric stiffness so most legs tend towards solid construction with no core which means the flexure section often limits the buckling load.

Rachael.

#18 gruntguru

gruntguru
  • Member

  • 7,637 posts
  • Joined: January 09

Posted 19 April 2012 - 23:30

Nice collection of specifics there rachael. I hope slider doesn't find too much to quibble over!



#19 rachael

rachael
  • Member

  • 118 posts
  • Joined: October 10

Posted 20 April 2012 - 06:19

Nice collection of specifics there rachael. I hope slider doesn't find too much to quibble over!

Let's hope he doesn't want fibre and resin type, stacking order, fibre orientation and cure temperature! ;-)

Advertisement

#20 gruntguru

gruntguru
  • Member

  • 7,637 posts
  • Joined: January 09

Posted 20 April 2012 - 07:10

Are you holding out on us? :)

#21 bigleagueslider

bigleagueslider
  • Member

  • 1,235 posts
  • Joined: March 11

Posted 21 April 2012 - 01:48

Slider,

You said in post 12 “The flexure is located at the point where the moment is highest and is made from a material with a much lower MoE than the rest of the A-arm”

The suspension legs away from the flexure are made from a mix of cloth and high stiffness ud such that the bulk axial modulus is ~160-180GPa. For a composite flexure there is a similar mix of cloth and ud but both are of a lower stiffness such that the bulk axial modulus is ~120GPa in the flexure region. A Ti flexure will have a modulus of 110GPa so I think it is misleading to state that the flexure modulus is “much lower” than the rest of the leg.

You are correct that cores filled with honeycomb will have a higher buckling stiffness than unfilled but this is not due to shear connection but rather the core maintaining the laminate section like a plumbers spring inside a copper pipe. Aero requirements dictate that F1 suspension members are placed unfavourably for geometric stiffness so most legs tend towards solid construction with no core which means the flexure section often limits the buckling load.

Rachael.


Rachael,

Sounds like you have a better knowledge of current composite A-arm construction than I do. I was only considering the case of a metal flexure bonded to a composite A-arm. While I have worked in racing as a design engineer in the past, it's been quite a while, and back then A-arms were all TIG welded chrome-moly steel.

I currently work as an aircraft mechanical systems engineer, and my technical knowledge of composite structures is only that which I have absorbed by "osmosis". I work with metals, not composites. So in short, I will respectfully defer to you on this particular topic of discussion.

Best regards,
slider


#22 rachael

rachael
  • Member

  • 118 posts
  • Joined: October 10

Posted 21 April 2012 - 08:07

Rachael,

Sounds like you have a better knowledge of current composite A-arm construction than I do. I was only considering the case of a metal flexure bonded to a composite A-arm. While I have worked in racing as a design engineer in the past, it's been quite a while, and back then A-arms were all TIG welded chrome-moly steel.

I currently work as an aircraft mechanical systems engineer, and my technical knowledge of composite structures is only that which I have absorbed by "osmosis". I work with metals, not composites. So in short, I will respectfully defer to you on this particular topic of discussion.

Best regards,
slider


Slider,
No need for such a deferential sounding post. This forum should be about friendly discussion and knowledge sharing, I'm not interested in crowing or belittling others.

Regards,

Rachael.

#23 NeilR

NeilR
  • Member

  • 623 posts
  • Joined: October 09

Posted 21 April 2012 - 09:40

The composite suspension arms are designed with a reasonable bending stiffness to prevent buckling under high compressive loading. The flexures are not made from a material with a particularly low modulus - typically carbon or Ti. The flexibility of the flexures comes from their thin X-section.



If you were using a steel arm could you use a thin section of spring steel?

#24 Kelpiecross

Kelpiecross
  • Member

  • 1,730 posts
  • Joined: October 10

Posted 22 April 2012 - 03:08

If you were using a steel arm could you use a thin section of spring steel?


Again without researching - didn't some late 50's F1 Coopers and maybe earlier Cooper-Bristols use a leaf spring as the top suspension arm and spring combined?

#25 John Brundage

John Brundage
  • Member

  • 309 posts
  • Joined: July 08

Posted 22 April 2012 - 03:30

Again without researching - didn't some late 50's F1 Coopers and maybe earlier Cooper-Bristols use a leaf spring as the top suspension arm and spring combined?

The early Cooper Monaco rear transverse leaf spring was also the upper arm.


#26 NeilR

NeilR
  • Member

  • 623 posts
  • Joined: October 09

Posted 22 April 2012 - 05:39

I was thinking more a 30mm long section of steel.

#27 Canuck

Canuck
  • Member

  • 2,384 posts
  • Joined: March 05

Posted 22 April 2012 - 06:48

Slider,
No need for such a deferential sounding post. This forum should be about friendly discussion and knowledge sharing

Are you sure you're in the right forum? :D
I believe it is mandatory to rub it in in the technical forum. Here, let me show you. :rolleyes: Whatever - aircraft are not F1 cars, they're just composite tubes with protrusions.
Also mandatory - when you're wrong (which I'm not implying you are), you must defend your erroneous position vehemently, ultimately abandoning the technical argument and moving into personal attacks.

On a more serious note, I love this forum for the wealth of knowledge that flows so freely. (but it's not truly seamless!)



#28 Greg Locock

Greg Locock
  • Member

  • 6,353 posts
  • Joined: March 03

Posted 23 April 2012 - 00:41

Here's Christine, our 1987 solar car that uses a transverse leaf spring as the control arm.

http://new.aurorasolarcar.com/Cars

It is bloody heavy, I think in retrospect coilovers on the shocks (which are inboard and run off the lower arms) would have been better - not that I was involved back then.











#29 gruntguru

gruntguru
  • Member

  • 7,637 posts
  • Joined: January 09

Posted 23 April 2012 - 02:18

Here's Christine, our 1987 solar car that uses a transverse leaf spring as the control arm.
It is bloody heavy, I think in retrospect coilovers on the shocks (which are inboard and run off the lower arms) would have been better - not that I was involved back then.

A "steel" leaf spring? Perhaps a composite leaf spring would offer the best of both worlds.

#30 Magoo

Magoo
  • Member

  • 3,703 posts
  • Joined: October 10

Posted 23 April 2012 - 02:31

Again without researching - didn't some late 50's F1 Coopers and maybe earlier Cooper-Bristols use a leaf spring as the top suspension arm and spring combined?


Yes indeed. Cooper, Tatra, countless others down through the years. The transverse spring can also serve as the lower control arm, as in the Planar IFS system developed by Delmar G. "Barney" Roos for Studebaker and then Willys-Overland. Or the transverse leaves may serve as both the lower and upper control elements, with no separate control arms. (With the added advantage that multiple leaf elements damp each other.) The Blood Bros. Cornelian driven by Louis Chevrolet in the 1915 Indianapolis 500 used such a setup both front and rear. Has often been billed as the first car with four-wheel independent suspension in the 500.




#31 bigleagueslider

bigleagueslider
  • Member

  • 1,235 posts
  • Joined: March 11

Posted 23 April 2012 - 02:46

Slider,
No need for such a deferential sounding post. This forum should be about friendly discussion and knowledge sharing, I'm not interested in crowing or belittling others.

Regards,

Rachael.


Rachael,

Thanks. I agree that these forums can be a great place to gather some tidbits of technical knowledge that might be hard to find elsewhere (such as details of composite A-arm construction). I enjoy sharing whatever knowledge I might have on a particular topic, for what it may be worth. I also try to go out of my way to be friendly and deferential when someone else corrects or disagrees with something I may have posted, rather than being argumentative or confrontational.

Unfortunately, it just seems to be the nature of many intellectuals like engineers to enjoy playing the game of "gotcha" by simply criticizing posts rather than engaging in thoughtful, polite discussion. I don't want my posts to give that impression.

The abbreviated format of these forums also makes discussion of complex technical issues difficult. The structural mechanics of a composite A-arm could fill a book. Doing the topic justice with a few sentences in a forum post is an impossible task!

Best regards,
slider


#32 Greg Locock

Greg Locock
  • Member

  • 6,353 posts
  • Joined: March 03

Posted 23 April 2012 - 04:42

A "steel" leaf spring? Perhaps a composite leaf spring would offer the best of both worlds.


Yup steel. Bigger organisations than Aurora have run into severe problems with composite leaf springs. Incidentally just because it is black and on an F1 car don't assume it isn't steel. Particularly at the start of the season.

#33 Kelpiecross

Kelpiecross
  • Member

  • 1,730 posts
  • Joined: October 10

Posted 23 April 2012 - 07:10

Yup steel. Bigger organisations than Aurora have run into severe problems with composite leaf springs. Incidentally just because it is black and on an F1 car don't assume it isn't steel. Particularly at the start of the season.


Chevrolet claim to have never had a fatigue failure in the fibreglass leaf springs on the Corvette - which some models have on both front and rear suspensions.

#34 Kelpiecross

Kelpiecross
  • Member

  • 1,730 posts
  • Joined: October 10

Posted 23 April 2012 - 11:50

Yes indeed. Cooper, Tatra, countless others down through the years. The transverse spring can also serve as the lower control arm, as in the Planar IFS system developed by Delmar G. "Barney" Roos for Studebaker and then Willys-Overland. Or the transverse leaves may serve as both the lower and upper control elements, with no separate control arms. (With the added advantage that multiple leaf elements damp each other.) The Blood Bros. Cornelian driven by Louis Chevrolet in the 1915 Indianapolis 500 used such a setup both front and rear. Has often been billed as the first car with four-wheel independent suspension in the 500.


"Countless others" includes the Fiat Topolino which had this type of front suspension. This was copied by Cooper for the 500cc cars and then used on (I think) all the early racing Coopers for quite a few years.

#35 Greg Locock

Greg Locock
  • Member

  • 6,353 posts
  • Joined: March 03

Posted 23 April 2012 - 23:07

Chevrolet claim to have never had a fatigue failure in the fibreglass leaf springs on the Corvette


I saw what they did there, did you? I've never had a /fatigue/ failure in the carbon fibre mast on my sailboard, but for some reason I keep having to buy new ones.

Sure, you can develop robust fibreglass springs, if you have the time and resources. Aurora has very limited time, and very limited resources. Compared with GM so does a F1 racing team.

#36 pugfan

pugfan
  • Member

  • 177 posts
  • Joined: August 09

Posted 24 April 2012 - 02:18

I saw what they did there, did you? I've never had a /fatigue/ failure in the carbon fibre mast on my sailboard, but for some reason I keep having to buy new ones.


Bloody hell, what kinds of winds are you going out in?

I'm 6'4", 96kg and I've never snapped a sailboard mast yet and I'm not one to shy away from catapults.

Booms however are another story.

Edited by pugfan, 24 April 2012 - 02:19.


#37 Kelpiecross

Kelpiecross
  • Member

  • 1,730 posts
  • Joined: October 10

Posted 24 April 2012 - 03:44

I saw what they did there, did you? I've never had a /fatigue/ failure in the carbon fibre mast on my sailboard, but for some reason I keep having to buy new ones.

Sure, you can develop robust fibreglass springs, if you have the time and resources. Aurora has very limited time, and very limited resources. Compared with GM so does a F1 racing team.


What did GM do there?
I always thought that the fibreglass springs were just a sales gimmick to go along with the Corvette fibreglass body image.

#38 Greg Locock

Greg Locock
  • Member

  • 6,353 posts
  • Joined: March 03

Posted 24 April 2012 - 04:47

What did GM do there?


They specified /fatigue/ failure. Sounds techie, but it doesn't include any other failures.

Yeah, catapults. Oh and a very ugly one at Sandy Point, the mast speared into the beach and decelerated the board and me rather rapidly. Luckily i rolled into the water and missed the carnage.

#39 NeilR

NeilR
  • Member

  • 623 posts
  • Joined: October 09

Posted 25 April 2012 - 01:34

I thought the mast was supposed to point up...or is this the full contact sailboarding combined with lacross?