Beam axles

[James Hunt]

hahaha, I know..

I read a post written by a automotive engineer I have respect for asking why we don't use beam axles (I think he trolls this forum). His reasoning being;

- reduced development time
- cheap
- four maximum contact patches at all times
- antiroll stiffness wont effect camber
- spring changes won't effect camber
- pitch changes produce minimal effects on camber
- more predictable roll centres

I understand packaging a front beam could be difficult.

Static and unsprung weight would be greater than an independent setup, but I wonder if there is a point of diminishing returns fast approaching in regards to that. Regarding unsprung weight, is contact patch size more important?

How much do FSAE wheels camber in action?

Who has used a rear beam? What design teams have kicked the idea around?

[Kirby]

ACME Racing - Aust. Defence Force Academy.
2009 Car - F&R DeDion

User to contact: oz_olly

Car: yfrog. com/4fe34qj

[Adambomb]

I remember a team a couple years ago with a solid CF rear axle, the car was like 308 lbs and performed well if I remember right. Then there's the Buffalo "Satchel linkage," RJWoods was a big fan of it.

It would simplify kinematics and therefore development time, plus I imagine it also could be done lighter (rear at least).

Although I'm going to have to throw the BS flag on the "maximum contact patch." Actually anything that is described as "maximum" or "maximized" is usually a trigger word for judge-types to ask you some very uncomfortable questions. In other words, to them (as well as many other experienced engineers) "maximized" is another way of saying "it's the best, but I can't quantify why."

And I'll also argue that it is possible to get overall more desirable camber curves with unequal non-parallel arms, if you know what your doing. And I'll also argue that their roll centers are no less "predictable," in that their kinematic locations aren't particularly difficult to calculate, and besides beam axles are just as susceptible to roll center migration due to tire deflection.

Unsprung weight may be an issue. Even with an independent setup and a lot of emphasis on cutting unsprung weight, the unsprung/sprung weight ratio ends up in the ball park of 1/4 in FSAE cars (compared to like 1/10 on a passenger car).

[oz_olly]

Why don't we move the discussion to this thread:

http://fsae.com/eve/forums/a/t...=781103883#781103883 (http://fsae.com/eve/forums/a/tpc/f/125607348/m/781103883?r=781103883#781103883)

I'm always happy to talk about beam axles. We can also talk about what Paul Clausen from Kirby's team did back in 2005 (if my memory serves me correctly).

Cheers

[oz_olly]

I made a post with a link to the 'Beam Axles Front/Rear or Both' thread I started a little while ago. Just waiting for the link to be approved.

[thewoundedsoldier]

I agree with Adambomb on every point.

The inability to gain negative camber during a cornering maneuver is the big downside to a beam axle. However, the reduced development time, cost, and simplicity are big bonuses to an under-staffed team.

The cheapness and simplicity that I love about it is not necessarily in the suspension, but the driveline. You don't need silly cv-joints and splines and yada yada everywhere. This not only provides more power in a straight shot from output shaft to tire patch, but also makes your rotational inertia silly low which helps your angular acceleration. And besides, who needs a diff? The tires slip plenty.

Oh yeah and about unsprung weight, the ultimate issue with it is responsiveness, right?

[exFSAE]

Nothing inherently wrong with one suspension layout versus another. Establish what you want your suspension to do at a "top vision" and requirements level, and then pick what geometry works best. Maybe it's dual a-arm, maybe it's live axle. Regardless of what you pick it has to be justified by some engineering data.

I'd agree with the previous comment of avoid saying any BS like "provides maximum contact patch." If I were a design judge I'd rip that apart.

TBH I'm not sold on lack of camber gain being a real detriment on these cars. On "big boy" racecars that can put some real load into the tires, yes. At FSAE loads.. maybe not so much. But don't take my word for it... look at the tire data.

[James Hunt]

To all those that replied, thank you! I really appreciate the knowledge.

I'm not an engineer, but no one is challenging those who are around my parts.

[Kirby]

Originally posted by oz_olly:
We can also talk about what Paul Clausen from Kirby's team did back in 2005 (if my memory serves me correctly).


Memory is correct. It was certainly one of the more interesting concepts that Adelaide tried over the years.

[PatClarke]

We can also talk about what Paul Clausen from Kirby's team did back in 2005

Built the worlds stiffest anti roll bar?? ;-)

Pat

PS, I have been following the thread on Atlas that James Hunt refers to. What the discussion there misses is that FSAE is a Design Competition, they are talking about building a race car ... Not the same thing at all ;-)

PC

[Test Driver]

Originally posted by thewoundedsoldier:
............
The inability to gain negative camber during a cornering maneuver is the big downside to a beam axle..............

On a solid version maybe but, how about the twist version of a beam axle. It's a tough sell. I know. I tried it. For one of my teams, the elegance of simplicity was lost in the fog of perceived double wishbone sexiness.



And besides, who needs a diff? The tires slip plenty............

Yes! This is one of the better better lines I've read here in over a decade!

[Zac]

Are any teams running VSAL's short enough that they actually allow them to gain negative camber? Or am I just confused?

Personally, I want to see less double wishbones and more sliding pillars and De Dions.

[wagemd]

Originally posted by Test Driver:

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content"> And besides, who needs a diff? The tires slip plenty............

Yes! This is one of the better better lines I've read here in over a decade! </div></BLOCKQUOTE>

+1

[murpia]

Originally posted by Zac:
Are any teams running VSAL's short enough that they actually allow them to gain negative camber? Or am I just confused?

Personally, I want to see less double wishbones and more sliding pillars and De Dions.
I'm with you on the de Dion. I'd love to see a conceptual design study for a 'longitudinal' V-twin, quad-bike diff, pushrod actuated de Dion. I think you could pull the driver right back in the car and maybe even get the steering rack in front of their feet...

Carbon tube for the de Dion? Watt's linkage? Some sort of 'wedge' to adjust rear camber?

Regards, Ian

[MalcolmG]

Originally posted by murpia:
<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by Zac:
Are any teams running VSAL's short enough that they actually allow them to gain negative camber? Or am I just confused?

Personally, I want to see less double wishbones and more sliding pillars and De Dions.
I'm with you on the de Dion. I'd love to see a conceptual design study for a 'longitudinal' V-twin, quad-bike diff, pushrod actuated de Dion. I think you could pull the driver right back in the car and maybe even get the steering rack in front of their feet...

Carbon tube for the de Dion? Watt's linkage? Some sort of 'wedge' to adjust rear camber?

Regards, Ian </div></BLOCKQUOTE>
I like how you think http://fsae.com/groupee_common/emoticons/icon_wink.gif

The biggest issue I found with trying to get the driver far back with an engine behind/under their back is that f/r weight bias gets pretty horrendous. Our car last year was at about 57% rear weight with minimum wheelbase and a longitudinal single quite snug behind the driver. My initial hopes were to mount the steering rack and front ARB inside the front bulkhead, but even with what was a pretty compact car the driver's feet were still well ahead of the front axle (probably 200mm or so), and the angles on the tie rods would've been horrendous.

[murpia]

Originally posted by MalcolmG:
<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by murpia:
<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by Zac:
Are any teams running VSAL's short enough that they actually allow them to gain negative camber? Or am I just confused?

Personally, I want to see less double wishbones and more sliding pillars and De Dions.
I'm with you on the de Dion. I'd love to see a conceptual design study for a 'longitudinal' V-twin, quad-bike diff, pushrod actuated de Dion. I think you could pull the driver right back in the car and maybe even get the steering rack in front of their feet...

Carbon tube for the de Dion? Watt's linkage? Some sort of 'wedge' to adjust rear camber?

Regards, Ian </div></BLOCKQUOTE>
I like how you think http://fsae.com/groupee_common/emoticons/icon_wink.gif

The biggest issue I found with trying to get the driver far back with an engine behind/under their back is that f/r weight bias gets pretty horrendous. Our car last year was at about 57% rear weight with minimum wheelbase and a longitudinal single quite snug behind the driver. My initial hopes were to mount the steering rack and front ARB inside the front bulkhead, but even with what was a pretty compact car the driver's feet were still well ahead of the front axle (probably 200mm or so), and the angles on the tie rods would've been horrendous. </div></BLOCKQUOTE>

57% is not so bad, I think you could work with this. Locating pretty much everything 'moveable' (ECU, battery, fusebox, catchtanks, electric pumps) very forward in the sidepods would be the way to go. I would look into running the exhausts forward too. Yes, there will be a weight penalty as all the wires and hoses get longer but you could trade off total weight against forward bias.

Would need a very up-front design process too, something I'm in favour of!

As a related point, how 'kart like' can you get a 95th percentile driving position? With the high knees and straight back typical of a kart I think you could shorten the cockpit by a few hundred mm? Anyone tried it?

Regards, Ian

[exFSAE]

As an aside... discussion of kinematics is to some degree a moot point with many teams that have the misconception that as you throw more spring at the car it will become super responsive.

When I hear teams say that they're running wheel rates in excess of 200 or more lbf/in... what's the point of having dampers or kinematics? Your suspension isn't gonna f'in move!

Not that I knew any better in undergrad. Had no concept of rigid body dynamics and just dove headlong into springs, bars, and kinematic curves thinking it was the end-all-be-all of handling.

[Mike Cook]

Originally posted by James Hunt:
hahaha, I know..

- reduced development time
- cheap
- four maximum contact patches at all times
- antiroll stiffness wont effect camber
- spring changes won't effect camber
- pitch changes produce minimal effects on camber
- more predictable roll centres



The cambers would pretty consistently be 0 wrt to the ground. Is this the best camber? For most tires, no. Actually the tires will probably have positive camber since most fsae cars have a lot of compliance.

Unsprung gets heavier (prob not the end of the world).

Assuming we're not talking about a kart style solid axle, the axle needs to move which requires some kinda linkage (4 link, etc.). This isn't that much simpler then a double wishbone setup.

I think the main reason not to do this is because one wheel disturbance affects the other wheel.

I'm not going to even open the can of worms of no diff vs diff. That has been discussed a lot, but in my opinion, it is really difficult to get a car with out a diff to work very well without a lot of tuning.

Mike

[MalcolmG]

Originally posted by murpia:
57% is not so bad, I think you could work with this. Locating pretty much everything 'moveable' (ECU, battery, fusebox, catchtanks, electric pumps) very forward in the sidepods would be the way to go. I would look into running the exhausts forward too. Yes, there will be a weight penalty as all the wires and hoses get longer but you could trade off total weight against forward bias.

Yep, we did a lot of that, exhaust running forward up the side of the car, battery, fusebox, ECU, CDI and datalogger under the driver's legs/ass. When the CoG of your two heaviest components are closer to the rear axle than the front, it gets very tricky to get weight forward - particularly if you're concerned enough about yaw inertia that you don't want to go putting everything in the nose. I decided the ideal would be a longitudinal flat twin laid down under the driver's legs...

Back on topic, Mike there's nothing stopping you from running a small amount of camber, I'd think around 1-1.5 degrees of static camber on the rear would probably give you better camber under cornering and drive than a short FVSAL double wishbone arrangement.

Unfortunately I've never had the chance to see any other FSAE tracks other than the Aus one, but that probably has few enough bumps that I don't think you'd be hugely disadvantaged by the lack of independence in the rear suspension (although I could easily be wrong on that).

[oz_olly]

I am hoping that our beam axle car will be well enough sorted by the competition to demonstrate some of the potential of the concept rather than our operation of the car holding it back.

A way I would like to try and quantify the impact not having independent suspension would be to have two tracks, one bumpy and one smooth and test the beam axle car versus an independent suspension car.

With the rear suspension being a de Dion twist axle we get some amount of camber gain in roll which is good. The design also allows for 0 to -2 degrees static camber through shimming the uprights.

I think a wishbone suspension favours stiffer vertical, camber and toe compliance but there are plenty of teams out there having trouble designing a stiff system.

We have nothing overhanging the drive shafts and a very small distance between the front and rear sprockets (using an 11 tooth front). During the design process I tried very hard to get the weight distribution further back and ended up with almost 50:50 with an 80kg driver. The trailing arms for the rear suspension made it difficult to mount the radiator and exhaust as far back as I would have liked. The driver can't really go any further back as the seat back is almost on the cam cover of the engine. So from that constraint any attempt to recline the driver will only push the cg forwards.

Next year we will follow a similar concept using but taking the track width down from 1250mm to 1050mm whilst maintaining the same cg height to track width ratio to ensure the weight transfer is the same. By reducing the track width in the rear the engine ancillaries can be mounted further to the back. The smaller WR450 engine (over GSX-R 600) will allow the driver to be reclined further and seated further back.

[MalcolmG]

Originally posted by oz_olly:

Next year we will follow a similar concept using but taking the track width down from 1250mm to 1050mm whilst maintaining the same cg height to track width ratio to ensure the weight transfer is the same. By reducing the track width in the rear the engine ancillaries can be mounted further to the back. The smaller WR450 engine (over GSX-R 600) will allow the driver to be reclined further and seated further back.
I like your thinking http://fsae.com/groupee_common/emoticons/icon_cool.gif

[thewoundedsoldier]

Originally posted by Mike Cook:
I think the main reason not to do this is because one wheel disturbance affects the other wheel.


If you count one wheel sweeping a larger radius than the other wheel as a disturbance, then I think you're right!

Judging from what I've seen on other cars and how our car is shaping up, I think that system compliance allows just as much roll relative to the chassis as an independent setup would. I've gotten to the point with the design where I don't even care what the numbers look like anymore because there is so much damn compliance in every system of the car. Tested numbers are gonna be the only way to prove anything. This seems especially true of the rear axle debate.

[thewoundedsoldier]

Oz, your setup for next year sounds very interesting. We're a formula-hybrid car, so we only have to package a 250cc engine (nee. Ninja or CRF). This allowed us to tighten the back track down to 40". The reduced track really helped with torsional strength of the axle, so we are no-joke trying to run our system with a piece of 1.25" OD x 0.095" wall thickness 4130 tubing.

I believe we spoke a while back about a failure you experienced with a solid axle due to torsion? I cant remember if you sheared drive pins or what, but I'd be very interested to know what happened and what your axle looks like this year. Every time I walk through our shop and see that piece of chromoly I giggle inside. Hopefully it works =/

[oz_olly]

We've never had any drive axle failures. Our biggest failure was a front hub that was grossly understrength for the cornering load case. The hub flanges each failed and the wheel departed company with the car. The wheel almost jumped a fence onto a busy road but luckily it ran out of steam.

At the moment we use 4340 quench and tempered drive shafts with the TRE tripods as it is a de Dion style beam axle. The beam has a slot down it's length to make it torsionally flexible so it's not kinematically over constrained.

Malcolm, it's easy to see why you like my thinking as you were the inspiration behind the idea. I remember talking to you about weight transfer amongst other things at the Optimum G seminar. Then I read your team's article in the Race magazine which locked down my plan. The concept with the current car was large track width and beams seeking high mechanical grip on smooth surfaces aka Hoppers Crossing. The best part of the new concept is the geometrically faster route through slaloms. Just means we have to be as anal about cg height as weight (but we should be anyway). The CAD model has a CG plane in it and any component placement above it requires the approval of the Technical Director.

Have a look at this link:

http://www.fsae.unsw.adfa.edu.au/car.htm

There are some reasonable pics of our car but nothing square on from the front or back. The page is an exact copy of our design report. We figure we're not competitive enough to really worry about other teams seeing what we wrote. I think it's in the spirit of the competition to throw it out there, who knows we might even receive some feedback either way.

Cheers

[oz_olly]

Bump... Now that my post has finally been approved.

[thewoundedsoldier]

Hmmmm for some reason I thought you were someone else, using a SBA rear design. My bad.