Spool + Weight Jacking

[swong46]

Hey everyone, we are currently designing the suspension around a spool setup and want to ask a question on how to effectively use a spool.

So first I said put the rear RC as close to the CG height as possible, this way the weight transfer will purely be geometric WT; thus causing the rear inner tire to lift at corner entry/mid. As soon as the RC migrates, we will generate a small roll moment and the springs will then deflect to prevent erratic wheel lift.
Now with all the weight of the rear on one tire, we use our dynamic toe to maintain peak lat. grip. I am afraid this would have an oversteering effect for the drivers because it should happen rather instantaneously.

Another thing I thought about was diagonal weight jacking, but this deliberately changes loading in the front to effect the rear. This I think is not the way to go as front grip changes (but I guess you can toe in the inner front tire to make up for this load change) but more importantly, I would think this method would be too slow to react.

Whats your opinion guys? Anyone from Monash care to step in? I know their spool setup was done wonderfully. The car was pulling hard throughout the skidpad with just one rear tire through almost the whole thing!

[swong46]

Hey everyone, we are currently designing the suspension around a spool setup and want to ask a question on how to effectively use a spool.

So first I said put the rear RC as close to the CG height as possible, this way the weight transfer will purely be geometric WT; thus causing the rear inner tire to lift at corner entry/mid. As soon as the RC migrates, we will generate a small roll moment and the springs will then deflect to prevent erratic wheel lift.
Now with all the weight of the rear on one tire, we use our dynamic toe to maintain peak lat. grip. I am afraid this would have an oversteering effect for the drivers because it should happen rather instantaneously.

Another thing I thought about was diagonal weight jacking, but this deliberately changes loading in the front to effect the rear. This I think is not the way to go as front grip changes (but I guess you can toe in the inner front tire to make up for this load change) but more importantly, I would think this method would be too slow to react.

Whats your opinion guys? Anyone from Monash care to step in? I know their spool setup was done wonderfully. The car was pulling hard throughout the skidpad with just one rear tire through almost the whole thing!

[Chris B]

diagonal WT is the way to go in my personal opinion. theres alot of reasons for it which i wont go into but to answer your question, everything in my experience with locked rear ends is diagonal WT. and dont worry if the drivers adapt to the locked rear properly the effect will be quick enough and not sluggish.

[Tom Wettenhall]

Here's the other side of it:

Check out go karts for your caster-based DWT. Bear in mind that with a go-kart, you have a short chassis with removeable/replaceable members which can be used to tune torsional rigidity and thus the net amount of DWT. Some SAE cars have adjustable caster, which might be worth a look for a similar effect. All the same, easy cornering, no understeer. Unless you've got the torsional rigidity to work it, don't even bother with DWT.

MUR '10 has little caster and a pretty soft chassis, so its cornering comes from an epic rear roll rate and much right boot. (MUCH right boot)
The advantage is that it's way easier to tune out that weight transfer if you need more grip, and you can make the car a bit more predictable than it otherwise might be 'cos the weight will even up across the rear when you oversteer, whereas with caster you have to centre the steering to plant the rear again.
Also, once you do, it's gonna want to go straight even if you've still got some lateral g, so (Speculation starts here) I suppose you might be able to get cleaner corner exits without DWT. (Speculation ends here) Certainly the promo videos look pretty cool when even your green drivers can do full-bore 10,000rpm four wheel drifts.

My personal opinion would be, for a single with a bit less power you'd probably go for DWT as you don't have to try so hard to minimise the inherent understeer, (It's possible, even with our 2010 car which has been a bit... ah... in the past) whereas with a four like ours, (many herbs) the roll rate method might be the go. (Speculation begins again) From what I understand of the M10's (Monash) dynamics, the rear wing keeps it in line pretty nicely if you overcook it. (End of speculation)

Oh, and DWT is probably the way to go if you're chasing skidpad, because our method sort of plays your four tyres off against each other, rather than the kart method which just takes the inside rear out of the equation and lets the other three do their thing.

General spool advice: Wide front track. Big tyres. Don't even think about running without adjustable sway bars.

[Tom Wettenhall]

Hey, what do you mean when you say 'dynamic toe'?

[wagemd]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by Tom Wettenhall:
Hey, what do you mean when you say 'dynamic toe'? </div></BLOCKQUOTE>

Thinking bump steer... I don't have much experience with spools, but I know they do things like that to make them work a little better.

[Tom Wettenhall]

Ah...

Do things like that: Yeah, sure. Two-position toe link pickups. ~0 bump steer, or moderate roll oversteer. Simple.

Work a little better: It's a nice idea, but it severely limits the amount of power you can put down if your rear wheels are facing outside the corner. It's essentially wet weather in a can in my (limited) experience.
Seems like spool setups nearly always have to be driven lairy and hairy to really work, so your driver needs to have a great deal of confidence in the rear end because they're always going to be pushing it, unlike say an open diff where your best speed comes from clean driving with no wheelspin.

On the other hand, maybe the Aussie teams who run locked diffs (Or damn close, like Swinny) have just seen too much of the V8 Supercars...

[wagemd]

Oh, not advocating, just defining what I think Swong was talking about.

I'm somewhat partial to my diff... http://fsae.com/groupee_common/emoticons/icon_smile.gif

[swong46]

Hey guys, thanks for the inputs. Wagemd is correct, bump steer is what I meant.
We will be running a single cylinder so perhaps DWT is the way to go. But I like the idea of tuning with roll stiffness rather then caster/scrub and all that.

Tom, I totally agree with the driver style; a go-kart wont "work" unless pushed hard and it definitely takes a lot of confidence.

But as for my ideas on design, am I on the right track? Keep the feedback coming hah

[Tom Wettenhall]

Wagemd, I wasn't intending to have a go at you mate, I was just trying to elaborate for Swong's benefit. Sorry about the confusion.

Swong: I can tell you what I know, heaven knows it's damned hard to find information on designing for a locked diff, but you need to evaluate how you want to build your car based on your own understanding and the priorities of your team.
For instance, you may want to use caster to create DWG, but can the team afford the extra weight of tubing that has to be put in to your chassis to create the necessary rigidity?

A couple of things from your OP which may be errors: First, unless you have physical movement, such as excessive roll, which makes your suspension run out of droop travel, your inside rear wheel will not leave the ground. Weight transfer alone will completely unweight it but will not remove it from contact with the ground. This is a good thing.
Second, apart from the tiny movements caused by body roll and other stuff like fuel burn, your CG is basically static. Did you mean RC?
Finally, remember that changing load conditions at the front as a function of steer angle can be corrected dynamically and in kind with Ackermann geometry. That's mostly what it's for.

[Chris B]

its not too hard to design the car to have adjustable if you do it to begin with.

but totally agree with the driving technique, in order to get a spool to handle well you need to drive it hard. so if you go with a spool and your drivers complain, tell them to sack up and drive it harder :P

[StevenWebb]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by swong46:
Hey everyone, we are currently designing the suspension around a spool setup and want to ask a question on how to effectively use a spool.

So first I said put the rear RC as close to the CG height as possible, this way the weight transfer will purely be geometric WT; thus causing the rear inner tire to lift at corner entry/mid. As soon as the CG migrates, we will generate a small roll moment and the springs will then deflect to prevent erratic wheel lift.
Now with all the weight of the rear on one tire, we use our dynamic toe to maintain peak lat. grip. I am afraid this would have an oversteering effect for the drivers because it should happen rather instantaneously.

Another thing I thought about was diagonal weight jacking, but this deliberately changes loading in the front to effect the rear. This I think is not the way to go as front grip changes (but I guess you can toe in the inner front tire to make up for this load change) but more importantly, I would think this method would be too slow to react.

Whats your opinion guys? Anyone from Monash care to step in? I know their spool setup was done wonderfully. The car was pulling hard throughout the skidpad with just one rear tire through almost the whole thing! </div></BLOCKQUOTE>

We design our car around tight australian tracks, where turn in is more important than getting the power down (slaloms... so many slaloms). this means we have a lot of weight on the front and a heavily rear biased weight tranfer distribution, sometimes a bit of rear toe out as well(gasp!).An underpowered car on an open circuit might not need as much corner exit grip aswell so this may still apply. Your tyres will also play a large role as less load sensitive rear tyres mean you won't loose to much overall grip when throwing all that weight on the outside rear tyre, but your rear tyres might not like doing it for a whole endurance

To get the weight transfer distribution right we do a bit of everything, higher rear roll center (still much lower than the cg!) for initial turn in, a STIFF rear bar for mid corner and alot of caster which is a corner radius sensitive effect (the understeer moment caused by the spool is also corner radius sensitive..).

Chassis stiffness vs weight transfer distribution does some interesting things, you would think a less stiff chassis would lessen all the diagonal weight transfer (certainly castor, but the castor effect will still be significant). But when we tested a super floppy chassis, our mid corner rear weight transfer went up! (most other teams will get the opposite result, but we have something they don't....)

as far as getting the wheel off the ground anyone who saw our acceleration setup would have seen how much droop travel our shocks have and we definately don't roll that much. Did I mention our STIFF rear bar?

Skidpad, that's a completely different story for us. our skidpad setup is about getting as much downforce as possible (more rear wing), using the car setup to keep the car balanced at skidpad speed and strategically making the change from skidpad mode to acceleration mode as quickly as possible

[Tom Wettenhall]

I hadn't thought about the prospect of making your rear bar that stiff that it'd hold the whole bloody car up!

[Adambomb]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by Tom Wettenhall:
I hadn't thought about the prospect of making your rear bar that stiff that it'd hold the whole bloody car up! </div></BLOCKQUOTE>

Our baja team did something similar like 4 years ago, only a step further. They designed an incredibly intricate system of linkages and bellcranks that did nothing more than tie both rear wheels together with a dual a-arm suspension. They wanted to get the roll stiffness of a swing arm while also searching for those mysterious "innovation points" baja always talks about. As you can imagine it was a complete disaster in every possible aspect...it consumed 250% of their design time, wrecked them on cost, was heavy, expensive, took a lot of time to build, and broke during competition. I told them if they really wanted quasi-infinite rear roll stiffness they should have just used a "really big" rear bar...or just run a swing arm like they did before.

Anyway, with that strategy, to go with an arbitrarily low front roll stiffness and very high rear roll stiffness does seem to be the best for baja with a spool, 500 lb car with arbitrary weight distribution, and (less than) 10 hp Briggs with 25-30% driveline loss (our baja team loves piggybacking as many different final drive units inline as they can, I think it gets them like 2 more "innovation points," and because it's what rock crawlers do).

So basically, it sounds like a good way to start with that, although because you do care about both ultimate cornering and also less than ultimate cornering response, I would say you would absolutely want to have a wide range of bar adjustment (like +/- 100%), and then train your drivers to deal with it, as I'm pretty sure the biggest challenge will be that, as has been mentioned, if the car is balanced for max cornering at the limit it will have pig understeer when driven at moderate speeds (think about all those newbie drivers you don't really want to have attempting to take the car to the limit right off the bat...).

Unfortunately I don't have a lot of first-hand experience with getting a locked diff car to corner well, but based on what I do know it seems like the best strategy is to get it set up so that it unloads the rear inside tire at lower lateral loads so it is still driveable at lower speeds (I'd say this is very important, as NONE of our drivers are consistent enough to tolerate a lot of variation), then just deal with lifting the rear more and more as lateral load and roll increases, and get it to work with three wheels on the ground. Just my $0.02.

[swong46]

Hey guys, im bringing this thread back up. I thought I had a good idea of how to piece this all together but I guess not.

After talking to our technical director, he said to have the RC at the ground or below the ground even. We are now stuck on what is the ball park for roll center heights?

My OP is incorrect as I said purely geometric WT would throw the car over lifting the inner tire, but it should say to use purely elastic WT (RC at the ground) so the body can roll and droop limiters + big rear ARB to hold the inner tire up. Is this thinking correct?

And last note, I was told that the rear RC should never be below the front RC. Not sure why this is so bad though, except for more roll and higher dependency on springs and ARB to prevent excessive roll.

[swong46]

Anybody?

[Adambomb]

If you want more rear jacking, you would want the rear RC closer to the CG, not farther away. Also, different things happen when the RC crosses from above to below ground while driving, and different things happen when it is below ground to begin with. Not the direction I would want to go.

Looking at some of the earlier posts regarding adjustable roll steer, that is something our baja team did last year, and after talking with them they were quite happy with it. It was as simple as having a long inboard tie rod tab with 4 sets of holes that you could adjust. I've never driven it, so all I can comment on is that while following it it appears to have a comical amount of bump steer; you can watch the dynamic toe. Also worth noting that they have built cars with negative caster, and thought they drove "just fine" as well.

[swong46]

Sorry, I meant the front RC. I understand that you want a higher rear RC but not sure where to start at the front.