Torque arm and Watts Link added...

Discussion in 'Corner Carver Racing Tech Discussion' started by rwleonard, Mar 3, 2018.

  1. SoundGuyDave

    SoundGuyDave This Space For Rent

    1,978
    16
    First off: I do NOT have a Watts on my race car. However, I have driven (and raced, as in competition license raced) a LOT of S197s, with Watts links and Panhard bars on track, so I think I have a decent perspective on this.

    1) I personally don't "feel" enough difference on a full-sized track comparing right and left turns to dismiss the validity of a ROD-END Panhard.
    2) I do not autocross, so have no exposure to slalom-type corners, and will fully admit that a Watts may be superior in quick alternating transitions.
    3) The Watts is, hands-down, heavier and more expensive, and with like bushing construction, doesn't present any drastic improvement in axle location.

    What I *THINK* is happening with a lot of the "Watts Links are awesome, Panhards are devil-spawn" posts and threads is a lot of different factors going into the judgement, which like tolerances, can stack up...
    1) There is usually no comparison made with setups using like-material bushings. IOW, apples-to-oranges. Stock PHB sucks; going to ANYTHING will be an improvement!
    2) There is some confirmation bias involved: It's only natural if you just dropped a grand or more into locating your axle to mentally perceive SOME improvement.
    3) I've never seen any comparison where all other factors were eliminated: Bushing compliance and roll center being the biggies.
    4) With the exception of Terry Fair, I don't recall any "non-learning" driver weighing in on the debate. Not intended as a slam to the less-experienced, but unless you can consistently grind out laps within a couple tenths of a second per mile, the mechanical advantages of any hardware will be lost in the "noise" of lap time variance, which is directly related to driver input errors.
    5) Combining #2 and #4, you easily get more "driver confidence." If you typically use between 80-90% of the "grip budget" of the car, the "OMG I have a Watts!" factor can allow a driver to narrow that variance somewhat which will be perceived as "Watts is better." Never mind that the REAL change was all mental.
    6) Possibly the most telling, the Watts is typically NOT installed in isolation, but is credited with all the change... Watts, shocks, LCA's and springs are a "normal" package to install in the rear, but for some reason the Watts gets all the credit. I would argue that shocks make a LOT more difference than any of the rest of it!

    I can't comment on the street aspects of Watts versus Panhard. To me, though, I have a couple of points to raise as food for thought.

    1) Mathematically, we've established that the Panhard allows around .080" of lateral axle shift over a "normal" range of suspension deflection (with Heim joint ends). We're talking on the order of 1/16" and that's just not enough to even talk about compared to the 1" or more of tire carcass deflection.

    2) There seems to be no chatter whatsoever about the differences in Watts designs. Fays2 mounts the prop bolt on a subframe which effectively doubles the vertical shift in RC location compared to a PHB in pitch. The Cortex/Griggs Watts mounts the prop bolt on the diff, which locks the RCH, but does still allow twice the variance in lever-arm length to the chassis compared to the PHB.

    Please note that I have NEVER said that a PHB is superior to the Watts, only that thought needs to be applied to the decision, not just a knee-jerk "OMGz!!1!!!!1!" reaction. If you're complaining that a stock rear suspension setup kinda sucks, I'll agree with you. If you throw good shocks, springs, LCAs and a Watts at it, you'll have a drastically improved tractive capability, period. If that leads you to push the edge of traction more consistently, then you will also be faster (lower lap times). Don't for a minute, though, think it's all because you pulled one bar off the car and replaced it with two bars, a football and a fixed prop bolt, because it just isn't the case. I will argue that with a rod-ended PHB eliminating the bushing deflection issue, there just isn't enough of a difference in LATERAL AXLE LOCATION to be worth the expense, complexity and extra weight that comes with a Watts link. If you want to talk about altering the roll-center location, then that's a whole different story. I just really hate categorical statements without anything to back them up!

    In closing, I will NOT say that the Watts doesn't present some advantages over a PHB, even with like-material bushings, because that just wouldn't be true. All I will say is that there just isn't enough of a difference in "feel," or more importantly lap times, to warrant the change in my situation. If you're autocrossing, then the rapid transitions in the slalom sections may put the Watts higher on the list than it is for the big track. When it boils down, there are just too many FAST S197s that run PHB's to categorically state that the Watts is light-years better on the open track. Much better to focus time/money/attention on the lower-hanging fruit: better dampers, and most importantly, the driver-mod!
     
    Sky Render likes this.
  2. stevbd

    stevbd forum member

    100
    1
    Good points as always, Dave. But I'm not talking about smooth race tracks or lap times. I agree with your points in that context.

    I'm talking about the daily driving situation of going over really bumpy and poorly maintained public roads, where the suspension is really struggling to control the 250 or so pounds of unsprung weight that is the rear axle. In the OP's post, bridge expansion joints are an example of this. I have a Whiteline PHB, good Bilstein struts, and P springs and although it is far better than stock there is still room for improvement.
     
  3. kerrynzl

    kerrynzl forum member

    66
    1
    Again if your situation is unstable on bumpy roads the PHB isn't the culprit.
    The main problem is caused by rear roll-steer which feels like "tramlining" in the rear end [or the rear end shifting sideways]

    Now we all understand that the wheels are always perpendicular to the axle centreline, so any change in wheelbase will steer the rear-end.
    I'll use my car as an example. It has 18" trailing links that are 2" lower at the front. If there is 1" of suspension compression the link Arcs forward and shortens the wheelbase 0.140".
    If there is 1" of suspension extension the link Arcs rearward and lengthens the wheelbase 0.080".

    So if there is bodyroll that has 1"compression[outside] and 1"extension[inside] there is 0.220" of wheelbase variation which causes Roll Understeer.
    Roll Understeer is stable on corners but you can certainly feel it on bumpy roads.

    If you fitted one of those stupid Trailing link relocation kits [which lowers the link at the rear end] This would promote Roll oversteer, and also create unstability in a straight line on bumpy roads. It is undesirable to increase the wheelbase with suspension compression.

    Roll understeer needs more steering input to overcome, whereas Roll oversteer needs opposite steering input to correct it [roll oversteer is twitchy]


    As Dave pointed out earlier in his great posting. A watts link isn't the great "fix all" that people think it is.
    The WL is great for adjustable Roll Centres but is limited by the bellcrank radius from going too low.

    On a race car where the designer wants a really low RCH [eg below the diff] they can use a hybrid watts/panhard link known as a "Woblink" or in the case of the Lotus Cortina a simple A frame pivoting off the underside of the diff housing.
     
    Last edited: Mar 12, 2018
  4. cgornowich

    cgornowich forum member

    50
    2
    kerrynzl wrote:

    "I cannot figure out how those torque arms can work properly without binding. There is a long torque arm that pivots at the front, and 2 shorter trailing links at the outer axle tubes.
    They both move on different Arcs so they will bind."

    If the group will indulge me: I installed a Griggs torque arm on my '08 with jack-stands in my garage and learned the hard way that while the torque arm defines the arc that axle housing travels, the lower control arms only dictate how far the driveshaft slip yolk rides on the output shaft of the transmission. The "pivot" end of the torque arm is a slip joint that also allows for sufficient fore/aft movement to avoid binding. This seems to work because the torque arm bolts directly to the axle housing, while the lower control arms are offset from the center of the axles by brackets (roughly 7 inches with the brackets). I saw no "binding" taking place, the axle pivoted freely until I had everything fastened.

    For what it's worth, the standard 3rd link suspension exhibits the same characteristic to a slightly greater extent.

    I'm no expert, just a guy with a toolbox who spent too much on silly upgrades. But some of these have increased my confidence at HPDE events, and that makes it more enjoyable for me. And there are a few that did the opposite... For me, my torque arm falls in the former category.
     
  5. kerrynzl

    kerrynzl forum member

    66
    1
    When I googled "Griggs S197 Torque Arm" I got this image. The front pivot is quite clearly a fixed position.

    If the front pivot could float forward/aft as you mention the TA wouldn't bind

    Regardless of what "Experts" claim , there is no major lift [from pinion torque reaction] at the front pivot.

    Due to the Outer control arms being lower than the axle centreline [I measured 5"] the accelerating thrust loads from the axle into the control arms tries to point the nose of the diff downward [this is normally controlled by upper link

    A manufacturer can strategically place the control arm [distance] below the axle centreline so that thrust loads can cancel out pinion torque reaction . C&P ratio is a big factor here

    Brake torque loads on a T/A are far greater than pinion torque reaction




    1 griggs_racing_s197m.jpg
     
    Last edited: Mar 29, 2018
  6. Norm Peterson

    Norm Peterson corner barstool sitter

    Age:
    70
    2,669
    28
    I've seen that picture, and it does not eliminate the notion that the chassis-side pickup is rubber-bushed specifically to allow the TA to 'float forward'.

    If it is in fact pinned securely to the chassis and does not allow any 'plunge', then it really isn't a torque arm regardless of what it may look like at first glance. And yes, it would certainly bind. If it matters, a "pinned TA" would really be a torque tube suspension (maybe think half of a NASCAR truckarm suspension) and wouldn't be using any trailing arms/links at all.

    Some approximate TA chassis-side force numbers . . . a 3500 lb RWD car accelerating at 0.5g is generating 1750 lbs of total forward traction. If we assume 27" diameter tires, that's a total axle torque of 1750 x 27/(2 x 12) or a little under 1970 ft*lbs, which is essentially being resisted by the TA alone (not precisely true, but it's close enough for purposes here).

    Now let's assume that the TA is 42" long (3.5 ft). It could be a bit more or a bit less, but that's not going to change the conclusions. The upward reaction at the chassis pickup is going to be 1970 ÷ 3.5, or about 560 lbs (4' long TA → 490 lbs, 3' long TA → 660 lbs). I'd certainly call that significant. These force results are assuming zero bind and are perpendicular to the TA's long axis.


    Norm
     
    Last edited: Mar 29, 2018
  7. kerrynzl

    kerrynzl forum member

    66
    1
    Posting Deleted

    It is pointless trying to actually improve people's knowledge and understanding in this area.

    When they would rather save face instead of bettering themselves.
     
    Last edited: Mar 30, 2018
  8. Norm Peterson

    Norm Peterson corner barstool sitter

    Age:
    70
    2,669
    28
    I think you're overlooking the fact that axle torque must be resisted as well as the tractive forces. Otherwise the axle would rotate in the opposite direction that you're trying to make the tires rotate in (until it either ran the rear U-joint out of angular travel and broke it or pulled the driveshaft completely out of the transmission, anyway).

    Force at the CP is the result of axle torque, not the reaction to it, and the CP force cannot exist without a reaction because axle torque needs a reaction to be in place to work against. Take away the reaction to axle torque and your CP force goes away with it (save for a brief inertia effect).

    There's more than just X-direction forces involved even though that's all you need to consider at the contact patch or use to compute how much X-direction acceleration you should be getting.

    Pinion torque is a different torque entirely. While it is certainly related to axle torque, it's in a different direction and follows a different load path. Pinion torque (or driveshaft torque) in a stick-axle car is what (in most cases) plants the left rear and unloads the right rear (the vertical-ish force at the R&P gear teeth is balanced by the pinion's force reaction against its bearings and these forces are completely resolved internally within the pumpkin).

    The load path for pinion torque goes through the axle to the springs, sta-bar, and (dynamically) the shocks before it gets back to the chassis and from there back through the engine & transmission mounts to the engine. Not through any of the longitudinal axle locating devices - at least not in the S197's case (the Fox/SN95 arrangement with LCA-mounted springs and rear bar doesn't separate suspension functions and reactions nearly as neatly and adds bending and torsion to the LCA design . . . kind of a mess).

    I have no idea why any reference to a 3-link's UCA had to show up in this discussion, although you're still ignoring the axle torque reaction there as well.


    This forces/moments/reactions/etc. stuff was my major in school, and I spent most of the next 42 years getting paid pretty good money for understanding it.


    Norm
     
  9. Senna1

    Senna1 Junior Member

    29
    0
    I'm just a beginner when it comes to this stuff, so I'm happy to be corrected, but something below just doesn't seem to jive with anything I've read about dealing with the 3-link (unless I'm misunderstanding):
    I don't take issue with your technical description of roll steer, but I don't think I've seen a single person recommend running these cars with the trailing arms/LCAs angled up towards the axle end (lower in front). The "stupid" relocation kits are dogmatically recommended as pretty much required for ANY lowering of an S197, as I understand it, precisely to prevent that condition.

    What am I missing here?
     
  10. kerrynzl

    kerrynzl forum member

    66
    1
    Roll understeer is more desirable [it is more controllable to driver input]

    If you go back and crawl under any leaf spring regar-end car [Chryslers are a good example] you will notice the rear spring eye is a lot higher than the front spring eye.

    My car is built from the factory [as a racing car] with the LCA lower at the front.

    If the LCA is level when static it would arc forward in both compression and extension of the suspension [this is desirable]
    What is undesirable is having the LCA's higher at the front.[except in dirt/drag cars where they dial in "anti-squat" to increase forward bite and tolerate the twitchiness]

    A good example of Roll-steer is following an overloaded single axle trailer. They sway back and forth as the weight goes on one side and makes it steer opposite [causing oscillations]

    Years ago I had a trailer manufacturing business [before I started designing and building Racing Cars] and I experimented with reversing the shackles to the front and having the fixed hanger at the rear on single axle travel trailers.[basically we turned the suspension around]
    This cured all the swaying problems instantly............but the problem was $$$$$$ [no profit in this conversion] It was easier to sell the customer a torsion axle.

    The same applies here.....People would rather buy some bolt on [unnecessary] rubbish instead of thinking about what they're doing.
    In all my years of involvement in racing I found there are better gains by unbolting shit off a car than bolting shit on.
     
    Last edited: Mar 30, 2018
  11. Norm Peterson

    Norm Peterson corner barstool sitter

    Age:
    70
    2,669
    28
    The US domestic car scene is probably 90% about drag racing and straight line stuff - for most people it's the first thing they become aware of and the first thing they gain much understanding of. The only thing, in many cases, so it's an easy parts-upgrade sell. "Dogmatically" is a good way to put it, just like exhausts have to be loud to be any good.

    Even in corner-carving, it's easy to look at maximizing acceleration off the corner without understanding what you might have given up turning into and driving through it.


    I'm with kerry on this one.

    Roll oversteer is inherently an unstable condition. That does not mean "uncontrollable", but it does mean that accurate control inputs - and frequently, corrections to them - are demanded, which can make the car tiring to drive (do you really need that during your 5th session on a hot day? Or in the wet at any temperature?). It's not that all roll oversteer geometries threaten to instantly spit your car tail-first into the nearest ditch either, nor does it suggest that all roll understeer geometries are guaranteed to keep you happy.

    Roll steer can be quantified, and a small amount of axle roll understeer (perhaps 2% - 3%) does make the car easier to drive. Over the years, I have made that suggestion on various forums, but without tying it to any particular LCA inclination. And FWIW, you can feel the difference between 1% and 4% on corner entry.


    Norm
     
    Last edited: Mar 30, 2018
  12. Pentalab

    Pentalab forum member

    2,370
    70
    How do you prevent wheel hop, with the Lca's 2" lower at the front ?
    How do you get any traction?
     
  13. Norm Peterson

    Norm Peterson corner barstool sitter

    Age:
    70
    2,669
    28
    You wouldn't want them 2" lower than the axle side for cornering to begin with (gives something like 10% roll understeer, which is way too much).

    But 2" lower wouldn't have cost you enough AS% from stock (around 30%) to matter as long as you're not needing to make hard launches (the SVIC moves rearward as well as downward, so the slope from the contact patch through the SVIC - this slope being a visual representation for AS% - doesn't change much). Within the realm of 'corner carving', hard launches is really limited to Pro-Solo and some regular autocross course layouts, and only at the start there.

    Hop is basically a matter of bushing compliance (and probably the back side of the tires' mu-slip curve as well). LCA and UCA bushings, and the engine and transmission mounts all store and release energy once you blow clear through the available tire grip.


    Norm
     
    Last edited: Mar 31, 2018
  14. Norm Peterson

    Norm Peterson corner barstool sitter

    Age:
    70
    2,669
    28
    I suspect that kerry and I would not be in disagreement if the S197 LCAs and UCA were installed in a birdcage arrangement. kerry - there's a disconnect somewhere that's not simply a difference in north american vs down under terminology (f'r'instance, crown wheel makes perfect sense to me).

    Anyway, with the usual birdcages you'd need a second upper (obviously), and a separate link or arm of some sort to resist the axle torque (to keep the pinion pointed straight ahead instead of somewhere up in the sky), but the LCAs and the UCAs would then not go into tension during forward acceleration and the forward force accelerating the car would divide up as kerry has described. Whether the LCAs and UCAs would share any of the axle torque reaction would depend on the details of that additional link or arm.

    A trailing link added to the top of the pumpkin would add compressive force to birdcage links as their part in resisting axle torque, but a trailing arm in bending (kind of a TA) would not add either tension or compression (not as long as the direction of the force at the chassis-side pickup was perpendicular to the birdcage links, anyway).


    Norm
     
    Last edited: Mar 31, 2018
  15. kerrynzl

    kerrynzl forum member

    66
    1
    Firstly, Wheel hop is not caused by LCA angle but kinetic energy being released from bushings or springs.

    lowering the front LCA and front upper link lowers the instant centre [a side affect of lowering the car]
    This instant centre is the theoretical point of acceleration or thrust

    When you lower the instant centre below the CGH [centre of gravity height] you end up with more weight transfer via overturning moment [called squat]


    Note: the same applies to lowering the roll centre ,you get more body roll [when cornering]

    The squat is controlled by stiffer rear suspension but the side effect of this is a stiffer rear roll couple.
    Ford removed the rear anti roll bar on the FR500C because of the stiffer springs.

    On my FR500C if I use "steer it and stab it" driving techniques , it will understeer on corner exit due to the weight transfer onto the rear [and the stiff rear springs planting the tyres hard]

    If drag racing is your poison, and you race in a traction limited class then having a low instant centre and high CGH helps [hence the high stance of 60's gassers]
    Drag cars use softer rear springs to promote more weight transfer [the load and springs always have a point of equilibrium]
    .
     
  16. Norm Peterson

    Norm Peterson corner barstool sitter

    Age:
    70
    2,669
    28
    I think you're the only other person who has made the connection between lateral and longitudinal effects. At least publicly. But I think you're at least as interested in the CP through SVIC construction line vs sprung mass CG height, both taken at the front axle line, as the SVIC height by itself.


    Norm
     
    Last edited: Apr 1, 2018
  17. 2013DIBGT

    2013DIBGT I Hate Wheelhop

    333
    1
    Lol...gone since last summer and I am happy to see that the TA/WL battle rages on :rock:
     
  18. rwleonard

    rwleonard Junior Member

    10
    2
    Not for me! The TA/WL have achieved decisive victory, internet arguments about how many angles can dance on the head of a pinion not withstanding.

    (I think that was pretty damn good, especially considering I am only half way through my first cup of coffee!)
     
    frank s likes this.
  19. kerrynzl

    kerrynzl forum member

    66
    1
    Yep! This battle will go on forever as long as internet "experts" confuse Axle torque reaction [thrust] with axle housing torque reaction [pinion torque reaction]
    And they don't factor in the axle bearings that separate the 2 reactions.

    It is easy to convince people by telling them half the truth [salesmen, politicians etc have been doing this forever]
    So some novice buys some piece of crap and is totally convinced by the "snake oil effect" based on the facts as presented to him [by the salesman]

    This whole "corner carver" section should be about improving knowledge , but the section has stalled.
    You can only debate so much about the merits of a "dead pedal" or "which tyre I bought"

    The truth is, you can get better gains on the track by selling shit off your car than buying crap and bolting it on.
     
  20. Joe88xj

    Joe88xj forum member

    74
    0
    this has been interesting reading!