Register now to get rid of these ads!

3 link front suspension

Discussion in 'The Hokey Ass Message Board' started by wsdad, Nov 15, 2008.

    Joined: Jun 6, 2005
    Posts: 1,887

    Member Emeritus

    This has gotten somewhat interesting so here is a little more. If you look at some early motorcycles they have a suspension similar to this idea with a leaf spring suspending the forks. Works fine with single tire but as you go to two tire approach as pictured the center roll line becomes important. This flat blade center spring must work in two planes, up and down and torsional. It is the torsional issue that needs to be addressed as the chassis meets a bump or pot hole. The flat center spring will try to avoid a twisting motion, this will cause the chassis to roll to configure this the rise or drop, the sway bars do nothing more than try to balance the wheel load. Adding a stiffer bar will only make the roll resistance stiffer and that looses the ride and a soft bar that is lesser than the torsional twist of the spring just locates the axle. In either case in the driver seat you end up bobbing and weaving from being to soft or to stiff.

    The question as to the effect of friction material for shock dampening is limited in this application. In your picture you are pinching a friction material between the sway bar arms, while might work when the axle operated evenly it doesn't when the axle twists. Friction shocks work on the surface area of the friction material in this case it is so small that it will not do much for motion canceling. The distance that the sway bar actually contracts in wrap would have to be determined so that when the chassis was in roll it did not bind severe enough to lock up. If anything this would just produce a harsher ride. Shock absorbers are a better approach.

    If you want to continue this approach alter your design by substituting twin torsion bars in the position of the sway bar. Both the spring and the torsion arm work on the same plane and the torsion bar rate can act as a roll control.
    Last edited: Nov 16, 2008
  2. Thunderrace:

    I've never seen any of Roth's cars that used that, maybe someone can throw up some pics? It's probably due to the fact (I'm gonna get bashed for this one) that I've never been a big Roth fan so I haven't paid much attention to his cars, and have probably overlooked some good engineering ideas


    I think I may have gotten a bit long winded on my shock dicussion. I work as a product developement guy by day designing dampers and suspension components for an audi aftermarket company, so take that into account, I mean, what's good on a brand new Audi S4 in most ways DOES NOT carry over to a traditional rod. It's good food for thought, and good knowledge to have when you design suspension for anything, but by no means is it gospel.

    I personally think you'd be okay with friction shocks, you will probably get better handling and ride with properly mounted gas shocks but they will not look as cool, will add clutter, and definitely will take away from any FED flare. As far as using a shock that had 100/0 characteristics I would shy away from. Having 0% of control in your rebound movement would just be very bad. Like I said, go with the friction shocks, don't worry about it too much, if down the road you don't like the way the car rides, then you'll have an idea on what to change, but I think the friction shocks will look bitchin the way you're designing them. Sorry if my posts earlier seemed like they were trying to detour you from using them, I just got off on a shock tangent.

    As far as the VW torsion design goes, you wouldn't necessarily need a sway bar, since both sides of the axle would be sprung independently it would offer roll resistance as opposed to the center spring design because the adjuster holds the spring stationary in the tube, and allows them to rotate independently of each other on either side while being fixed in the center (imagine if you took your sway bar and fixed it to the tube right in the center, it would make for two shorter torsion bars giving idependent spring effect).

    Needing a sway bar or not will really come down to how hard you drive it in the corners if there is enough roll resistance or not, and it doesn't sound like you're building a canyon carver by any means. Look at any straight axle hot rod out there,, I'm gonna be you're gonna see very, very few sway bars on them.
  3. Dolmetsch wrote, "I agree though it is not a street set up, just an interesting experiment."
    Now that you've had some seat-of-pants-feel experience, what specifically is it about this set up that makes you leery of using it on the street? I loved the videos, by the way.
    Thanks, They were as fun to make as to watch.
    Nothing from driving it would discourage me. It is just that in a street car I and most are looking for something different than a drag race car. In the dragster because it carrys the load in the centre front of the frame it does not in any way fight the rear tires but allows them to go where ever they need to following the profile of the track. My only concerns on cornering were that if I got into trouble it would stay together while i fought it. I am satifisfied it will however I dont think it would do well on a road course and I never intended it to. For regular driving and cornering stabilty, the springs or at least pivot or load points should be as wide apart as possible. The three link with a centre 1/4 eliptic is the exact opposite of this. BTW the shocks are snowmobile shocks and were short enough to do the job nicley. I laid them on an angle to get some control in both planes . I will use them in the next car too.
    In reality the chassis probably thinks it is a tricycle . Maybe that explains it better. It suits my application or a drag solid rear application well but that is probably the best place to use it.
  4. I have seen several autocross cars built using some very trick mountain bike shocks. No they don't look traditional, but should work well on short vertical legs off your sway bar arms. If the sway bar is stiff enough, the center spring shouldn't be a problem, as Dolmetsch has verified. I'm using a similar setup on the rear of the rod project I'm building, only the rear end is located by ladder bars, which will take all the roll out of the equation. You might think about using hairpins instead of the sway bar and arms. Put some good hardware and heims on and let the axle tube handle the roll resistance. Just build it!
  5. 90 /10 shocks
    Before I played withdragsters I was a door slammer guy. We played a lot with shocks over the years. The 90 /10 deal to me was not exactly as it seemed. the 100 /0 idea is closer than you might think. The real need is to allow the front end to lift quickly (instantly if possible) unhindered in anyway. The settling 90 is more because we need some kind of shock for control and also the rules say we have to have them. In fact when we made our own shocks from new ones by drilling through the outer tube and inner tube at the top then plugging the outer hole with a carefully tapped 4/40 screw all without losing any fluid we got a very good working shock for drag racing. Did this for many many years. Also did it for other racers . The ninety tens didnt quite get the job done for us but of course were much better than 50/50s . Automatic (torque converter) cars react quite differently than manual trans cars. Ideally we want a quick front end rise , short travel and a small wheelie all the way through first gear. We were able to do that. Suspension is as important as the engine. In fact as long as the engine runs and makes decent power the rest of the game is in the suspension. MMMMMMMMMM how did we get here?
    oh yeah 90 /10s
    sorry but I find this type of stuff interesting. Discussion often brings up a point i may not have considered before.
    Last edited: Nov 17, 2008
  6. fiftyv8
    Joined: Mar 11, 2007
    Posts: 5,157

    from CO & WA

    Model AA trucks tried it but they still used a full length spring to achieve the 1/4 eliptic principle. The only other vehicle I know of that also used the spring as kinder a suspension arm was a German pickup called a DKW that was front wheel drive and used the transverse front spring as the upper or lower suspension arm.
    We are talking maybe 1940's here.
    But still however you look at it it takes one spring per wheel and 1/4 of a spring alone is just not quite enough to do the job unless it is part of a full semi eliptic spring.
  7. How about this?

    Or this. Sort of an ill-handling little piece of shit...:D
  8. DKW Deutsche Kinder Wagon .
    I take exception to your analysis of 1/4 eliptics. Like torsion bars they are just another way of doing the same thing and can do it well. We in North America are more familiar with the 1/2 eliptical . One thing I have learned in my career is that as long as the car thinks it is right, ie it has the right deflection and load carrying abilty then it is fine. Now if you want to tell me your prefer a normal as you call it full leaf spring that is fine. I think the 1/4 eliptic is an interesting variation that offers possiblilties I intend to explore. And I have only one 1/4 eliptic for two wheels BTW on the senior dragster and it is nicely sprung even on my bumpy field. I sorted through the spring pack combo till I thought the deflection was about rght and then finalized it. There is a tendency to dismisss what we dont like or are not used to as no good. As i grow older i try not to think that way. After 42 years swinging wrenches and racing I have seen some very interesting deals that worked a lot better than one would think. Sometimes the crowd is mistaken. Especially if the Germans thought it wasa good idea I wouldnt be in a hurry to poo poo that. course that could be because of my square head but I like to think not. Technically they are pretty darn good in most cases.
  9. wsdad
    Joined: Dec 31, 2005
    Posts: 1,258


    That's brilliant! It would eliminate the weight and complexity of the sway bar. Plus, no binding issues where it attaches to the frame. Plus, practically no change in caster (as pointed out by Dick Spadaro).

    I think a better idea than a tube axle would be to use an OEM Ford axle.
    It's shaped like an "I" and is made to twist. As one wheel hits a bump, the other doesn't necessarily have to raise, although it's encouraged to - due to the "twistability" of the axle. Wallah! The front axle becomes a sway bar! It's torsional strength can be adjusted by drilling lightening holes in it. :cool:

    Exwestracer, your idea is simpler, sturdier and lighter. I like it!!! I'll try to redraw it later tonight.
  10. Warpspeed
    Joined: Nov 4, 2008
    Posts: 532


    Great idea, but will a skinny beam axle with lightening holes handle the bending loads with just one central quarter elliptic spring ? It obviously works with the original transverse half elliptic arrangement, but then the bending loads on the beam would be quite low.
  11. Yeah I have to admit that I am slightly less believing in the hairpins coupled to this single spring, because it now means the axle has to twist (or bow slightly) as it moves vertically, because it's now constrained at the ends and trying to follow the arc of the spring's deflection under load.
    Ok we could put a link between the end of the spring and the axle which would then stop us trying to change the shape of the axle as it goes up and down, but then that loses the whole neat sideways location which the one central spring was giving you,
    which kind of ruins the thing.
    I still think the basic game plan was ok as you were going, the link ether side rather than hairpins, that way you are not trying to deform the axle, if you are having brakes on those front wheels you might have to add a rose jointed upper link out near the ends of the axle coming back and pivoting on the chassis as close to the effective centre of flex of the spring as possible, just to act as a 'torque link' to resist braking forces, but other than that I still see this idea as working real well.
  12. wsdad
    Joined: Dec 31, 2005
    Posts: 1,258


    Here's another crude drawing so just ignore the poorly thought out frame. The main focus is the suspension.

    I show the leaf spring lying on top of the axle with a cover bolted on. A better idea would probably have been to slit the axle and run it through the middle. The spring is not actually attached to the axle. It is allowed to slide in and out like a tongue through a donut (Let's keep it clean).

    Henry made the axle to twist, according to some information I've read. Its torsional stiffness can be adjusted by drilling holes or slits in the axle or by the length of the hairpins.

    The hairpins could be replaced by much stronger arms with friction shocks sandwiched between them and the frame. The arms would have a lot of leverage on the shocks so the frictional surface would have to be somewhat larger than typical. Brake pads might work or even a small clutch cut in half for each side. This would give it a very clean, simple, uncluttered look - especially if you put some sort of cover over them.

    Any thoughts, improvements, criticisms?

    Attached Files:

    Last edited: Nov 23, 2008
  13. 29nash
    Joined: Nov 6, 2008
    Posts: 4,544

    from colorado

    All of the above reasons that I don't like it.

    Not even okay for a go-cart.

    Unless you are a mechanical engineer, there are too many flaws/misconceptions on your part to press on.

    FUNCTION comes before simple/clean/uncluttered for 'look' concepts.

    No shocks? Why? It'll ride like a go-cart too.
  14. wsdad
    Joined: Dec 31, 2005
    Posts: 1,258


    I agree with you about function before looks. I don't want this to end up trailered to shows or worse, hurting someone on the road or strip.

    First impressions are sometimes very accurate, however, you are too vague to do me any good. Can you be more specific about the flaws and misconceptions that you see or are you pressed for time?
  15. The sliding contact at the end of the spring,
    I suspect like the tongue and donut it's likely to end up messy,
    full front weight of car on a thing that's got to slide,
    it's ether not going to slide or it will and it will wear bits of metal off fairly quick.
    you would need a bearing of some kind, with rollers so it does not bind up,
    then you would have to try and keep the dirt out....

    I doubt very much that Ford designed the axle to flex, I would guess more like it flexed so so ford claimed they made it that way !
    Seriously though yeah sure they do flex, but it's not actually a good thing,
    I liked your original sketches for this three link because you seemed to have a neat way of using a single 1/4 spring to locate and spring the axle with out any horrible twisting of the axle which hairpins lead one to.
    The stock ford arrangement of radius rods was fairly horrible, but I don't think it reqiured the axle to twist.
    And while I can see that using the axle as effectively a sway bar is very tempting, why add a seperate torsional element when the axle is as good a torsional spring as it is an axle ? I guess yes I can see that.
    Ok it's just the sliding thing at the end of spring then, the tongue/donut interface.
  16. 2manybillz
    Joined: May 30, 2005
    Posts: 827


    This worked on FEDs where the weight was biased way toward the rear. My impression is that a street car with engine weight much closer to the front wheels a set up like this would corner just like a tricycle.

  17. wsdad
    Joined: Dec 31, 2005
    Posts: 1,258


    Idea 1. Insert the spring through the middle of the axle as planned. Insert a "U" shaped brass or nylon bushing below it to contain the leaf spring and give it a replaceable surface to wear.

    Idea 2. Mount the leaf spring solidly to the axle. Let it slide in and out of the frame instead. There'd be more room to get creative with rollers and bearings. You could also hide any ugly nylon slides to keep the Tradition Police at bay.

    Idea 3. Mount a locating mechanism to the axle (see picture). I drew it with two bearings on each side of the spring and a nylon slide underneath it. Maybe it would be better with one or two rollers under it instead of the nylon slide. (The bolt heads are drawn round. You'll have to use your imagination and pretend they're hexagonal.)

    Attached Files:

  18. wsdad
    Joined: Dec 31, 2005
    Posts: 1,258


    ...Unless you have a really stiff sway bar. I'm starting to repeat myself.
    Last edited: Nov 23, 2008
  19. All it needs is a well cross-braced shackle, just like any leaf spring uses. I've got one sitting out here in the shop right now, only it uses 2 springs side by side and a double shackle. NO side to side motion whatsoever.

    My concern with drilling the axle would be it bending under load. Keep in mind all the weight is supported at the center rather than close to the ends. On a light car with a lot of setback this shouldn't be an issue... but if it wheelstands...:eek:!
  20. Yikes! No need to reinvent the wheel... Keep in mind if you're using hairpins, the only thing the spring is doing is taking weight and side load. A sturdy boxed shackle between the spring and axle should do fine.
  21. Couple of pics of what I'm talking about on mine. The front has boxed shackles down behind the axle. The rear pic was taken before the shackle was mounted to the center section. The rear shackle is shaped funny because the boxing plate doubles as a license plate holder. Yes, the front shock mounts are ugly, but they won't be seen when the front bodywork goes on. I still think you should check into small telescoping shocks, but they won't look right...

    Don't ask about the car, I'm not trying to hijack your thread, just give you some food for thought.

    Attached Files:

  22. I kind of like it.

    1) But you need the one with the integral sway bar otherwise it won't work. I'm not sure the I-beam axle by itslef would have enough anti-twist to give you the anti-sway you need.

    2) Not sure you'd have enough damping with your friction shocks at the pivot. To make it stronger with that long arm arrangement you could do the alternating internal / external spline thing like the clutch packs in a TH350.

    I still like the general concept. Please keep us posted on what you do.
  23. wsdad
    Joined: Dec 31, 2005
    Posts: 1,258


    Sometimes I over missincomplicationalizate things. A boxed shackle is just what it needs. Simple, light, strong, effective and brilliant! Thanks for the suggestion.

    That's a very interesting car you're building there. Wish I could ask you about it. Especially the rear suspension. I was also thinking about using a 3 point on the rear of my build, but I didn't want to complicate the discussion any more than it already is.
  24. The torsional stiffness of an I-beam is almost nothing compared to a tube. Thats why we use tubes for driveshafts and not I-beams.

    I'm not sure what modeling program you are using, but in might have the capacity to calculate the "moment of inertia" of a cross section of you axle. Compare that value with a similar cross section for a 1-5/8"OD 1/8"wall tube. I think you'll find the moment is about 1000 times greater for the tube.

    This is why the single center spring works for tube axle dragsters like Don's car, but would function poorly for a I-beam axled car.
  25. wsdad
    Joined: Dec 31, 2005
    Posts: 1,258


    The modeling program I'm using is Microsoft's Paint. It's similar to Arby's Napkin, but easier to email. Paint is woefully lacking in absorbency though, so you can't blow your nose in it as with Napkin. It's all a trade-off.

    Seriously though, I can probably find the published torsional stiffness of various tubes. And I have an I-beam in the garage that I can measure. I don't know how long it would take or how much effort is involved, but if you have access to such a program, I would be eternally grateful if you could figure out how much torsional stiffness I'll need (ballpark). I know there are a lot of variables so ask me what ever you need. I've never used one, so I don't know if what I'm asking is unreasonable.

    I was going to build it with the I-beam, drive it, and adjust from there. I was guessing, based on the weight of the car it came from, that the I-beam may be too stiff for my FED T-Bucket and I'd have to drill it. You're saying it won't be stiff enough and I should use a tube. I hope you're right because I've heard that when brakes are applied on some T-Buckets during an emergency stop, you can see the axle flex. It makes the front wheels hop. This is more of a problem on dropped axles because they have more leverage.

    Thanks for your input!
  26. What are the dimensions of an I beam? In particular the top and bottom plate thickness and height, overall height and center web thickness.
  27. Mart
    Joined: Mar 3, 2001
    Posts: 4,402


    I'm sorry but I think it's not a good idea. You need two well spaced springs, (Ok, henry got away with one, transversely mounted) to provide some resistance against roll. The dampers in the ends of the radius rods would not generate enough angular rotation to provide any sort of shock absorber action - they would have to be virtually solid and that would create a lot of stress.
    Imagine if you built the car with that setup and jacked the rear of the car up by the diff housing. It would just want to fall over because the front suspension would not stop it.
    On the street you will have to go around corners.
    Imagine building a 3 wheeler. With one wheel in the middle at the front. That would provide the same resistance to roll as your proposed setup.
    It's good to have a new idea, but what you might just about get away with on a dragster isn't necessarily great for street use.
  28. Warpspeed
    Joined: Nov 4, 2008
    Posts: 532


    It may be easier to measure the actual torsional stiffness, than calculate it, because if the unusual, and changing cross section. Just a couple of fairly long arms, one reaching forward, the other to the rear, and some barbell weights to put some twist into the beam. I would be far more worried about the bending stiffness (from the central spring). The real question is, how long can a seventy year old piece of steel, continue to flex, before it begins to fatigue.
  29. wsdad
    Joined: Dec 31, 2005
    Posts: 1,258


    Mart wrote, "The dampers in the ends of the radius rods would not generate enough angular rotation to provide any sort of shock absorber action - they would have to be virtually solid and that would create a lot of stress."
    I think that you are probably right. You could reduce the length of the arms and take away some of the leverage a bump would have on the shock, but then you'd get into problems with the front end jacking up during hard braking. A little trial and error may be needed to see if they can both be accommodated.
    Any tips on hiding the shocks? What have you tried or seen that's worked?
    Last edited: Dec 7, 2008

Share This Page

Register now to get rid of these ads!


Copyright © 1995-2021 The Jalopy Journal: Steal our stuff, we'll kick your teeth in. Terms of Service. Privacy Policy.

Atomic Industry
Forum software by XenForo™ ©2010-2014 XenForo Ltd.