Miata: Stiffness versus Weight Reduction

While stiffness and weight reduction have been discussed at length on their own, I have had trouble finding information on finding a balance between the two.

I have a NB1 Miata that I run in NASA TT5. It has your basic Miata weight reduction mods, a hardtop, and a roll bar. It feels very unsettled in quick transitions despite having the typical Miata formula for good springs/dampers/sway bar. I attribute this to un-damped chassis flex since my roll stiffness is biased heavily to the front of the car. This did reduce my confidence in these turns. 

The car weighs 2360lbs with me in it.

Option 1: At 2400lbs, I can run wider tires (257mm template instead of 226mm), add a little power, and I will have room in the weight to add door bars for chassis stiffness (aftermarket frame rails are not allowed).

Option 2: At 2250lbs, I am at the ragged edge of weight reduction inside the rules and my current power is still legal (could actually add a couple ponies), but I continue to deal with the famous Miata noodle-chassis. 

Option 3: GRM comes up with something I haven't considered.

What would GRM do?

For autocross, you want to be as light as possible even if the doors are popping open mid-run.

Track is a little less clear, though, as you're dealing with higher speeds where it's not all about transitions. It's hard to tell since you're quite vague about what you've actually done, but could it be something as simple as a set of shagged out control arm bushings? Can you expand on "unsettled" and maybe define just what you've done for setup?

Is reaching 2250 lbs plausible given your basic weight reduction mods? That's over 100 lbs of weight loss, which is fairly difficult for a car that's fairly pared down to start. That means you're at the upper end of your weight class already, so trading 40 lbs for an inch of tire width and more power might be worthwhile even without any increase in chassis rigidity.

Are you allowed to do zero weight modifications such as seam welding? That can make quite a difference on the door opening and should be useful on the tunnel-to-bulkhead junctions.

Do you believe the flex you're experiencing is between the bulkheads or could it be ahead of the firewall? You've got a hardtop so you've already closed the box a little on the cockpit and you've got a lot of front roll stiffness. It might be worth borrowing a shock tower brace, doing a 60 second installation and see if it makes a discernible difference.

In reply to Keith Tanner :

Suspension:

• Poly bushings everywhere. 
• Racing Beat 1.125 hollow front bar with the block/bolts brace kit. The rear bar is stock.
• MeisterR ClubRace coilovers 700/400 lb/in (12.5/7 kg/mm) because I am poor and I didn't go noticeably faster in my friend's car on Xidas.
• Alignment is -3.4/-2.7 for camber, 5 degrees of caster (no power steering), and no toe. 
• Konig Dial-in 15x9 with 225/45r15 Maxxis RC-1

Weight reduction:

• interior removed except for the dash cover
• HVAC and p/s deleted
• side windows and moving parts deleted
• plastic corner windows 
• OE hardtop with plastic window, fully bolted to the car
• ultrashield seats
• front bumper beam replaced with bash bar
• hood structure removed and aerocatches installed
• lithium battery
• Wilwood front brake kit

Aero

• 9LR medium downforce kit

On NCM, through turns 21-23 (an aggressive chicane that cannot be straightlined) there was a significant dead spot in the middle of each transition that felt as though the car was no longer connected to the road. I'll be honest, this scared me and I am not one usually prone to fear behind the wheel in a controlled enviroment.

Since my suspension is fairly stiff, as are my tires and bushings, the final area for flex is the chassis tub. Based on the OE stiffness of 6000 Nm/deg, my front sway bar and coilovers have nearly the same roll stiffness as the chassis has torsional rigidity. A good guideline for a normal car is that the chassis should have twice the torsional rigidity as the suspension has roll stiffness, where a racecar should be more like 10 times (source FSAE scholarly papers).

I am allowed to seam weld the car and I intend to do so based on the guide you put out some time ago. Next steps for weight reduction would be for me to lose 20lbs (fewer cheeseburgers), CCP fab hardtop 20lbs, the steering column to lose 16lbs (IDIDIT racing), passenger seat removal for 20lbs, CCP fiberglass trunk 7lbs, alumalite splitter blade 10lbs, and a few other odds and ends (inner door skin removal, cut the lower half off the dash, install headlight blanks, remove unnecessary brackets from the dash beam).

I will definitely try a front strut brace and see what happens since it is easy to do so trackside. 

Watch out when adding up claimed weight savings, you'll find that most are a little optimistic. For example, I'm not convinced you'll have 20 lbs on the hardtop compared to a plastic window OE unit. It would suck to throw a bunch of money at lightweight steering columns and body parts to end up 30 lbs short of your goal.

Given what you describe, I'd be playing with shock settings. That sounds like a damping problem to me. Do you have the coilovers set up for maximum bump travel as well?

Also, that front bumper beam is a significant source of rigidity in the nose as it's the only thing connecting the two front arms, and that's where all your front end loads go. A lot of people ignore this. Unless you've chopped something up, I'd bolt the stock piece back on and see if you feel a difference.

In reply to Keith Tanner :

I agree that the weight from the hardtop is optimistic. The other numbers have a bit more science behind them. 

The dampers are run full stiff per the manufacturer's recommendations for track use. I have a healthy amount of bump travel set up and I have found the limit of travel a couple times on the bigger curbs at Barber where I get more aggressive due to experience there.

The front beam was cut out to facilitate engine removal in my low garage, but has been replaced by a removable piece of steel box section in the form of a bash bar. I did not notice the car handle differently after this change was made, but it would be fairly easy for me to add a vertical aluminum plate here to tie the front chassis legs together even better without adding much weight. 

For reference, this is the bar I am using: https://www.miataturbo.net/miata-parts-sale-trade-5/radiator-support-bars-more-airflow-cooling-99508/

The designer claims it is stronger than stock and several people are using it the same way I am without any negative comments on chassis stiffness.

Cage it.

From your mods, it looks like it's a pretty dedicated track car at this point, so you might as well cage it and solve all your chassis stiffness problems and improve safety on the track while you're at it.  I was amazed at how much better our NA drove after building the rollcage... you can feel the suspension working as designed, without input from the chassis flex.

Sell the OE hardtop and buy a race top and you won't even add all that much weight overall.

EDIT:  FWIW there are lots of folks running similar 'crash bar' replacements (including me), and I've never heard anyone complain about them, but then again most cars are caged by the time they reach that point.  I didn't experience any issues with it prior to caging our car.

That's what I was thinking, cage plus whatever weight reduction is necessary to get back to 2400. Seems like it should be doable.

In reply to Rodan :

I will call my cage builder and see what he says.

In reply to dps214 :

How much different would this be than adding good door bars in terms of performance?

Back to the original question which was brought on by advice from a guy who holds track records in a similar car, which is better light and floppy or heavy and stiff?

I think a little weight is okay as long as it adds performance, but my colleague asserts that I should let it flop, make it as light as possible, and learn to drive through it. I don't care if I am right, I just need to pick a direction and I am beyond my experience at this point.

Do you need a hard top if you have a cage? Could you just run arm restraints and get by without a top at all? If you race in rain that wouldn't be comfy, obviously. I have raced with arm restraints before and once you get the hang of putting them on it's no big deal and they disappear once you're on track.

In reply to YoursTruly :

Floppy and light might be the correct answer for AutoX, but a stiffer car is going to be more predictable on track at higher speeds, and that should contribute to consistency.  Fighting the car will be unlikely to reduce lap times.

FWIW, our caged NA with similar prep to yours (it's a track car, so no street stuff) is 2050lbs without driver with a race hardtop and SPM S1 aero.  15x9s/225s, big Wilwoods.  An NB will be a little heavier in similar prep, but I'm honestly unsure of the exact delta.

In reply to dculberson :

The weight penalty of a race hardtop (~18lbs) will easily be made up by the aero improvement on track.  AutoX might be more of a wash.

In reply to Rodan :

My car is 2100lbs without me in it. I am one of the heaviest single components in the car.... 

In reply to YoursTruly :

All else equal, light and floppy might be better. But in this case heavy and stiff also comes with more power and wider tires. And some added safety that might allow you to feel comfortable pushing a bit harder.

I don't have any numbers, but a cage adds a lot more structure than just adding single door bars to a roll bar.

That bash bar is not going to be as stiff as stock. I would address it, even if you just bolt the stock one back on. The legs of the chassis are trying to go in different vertical directions and a piece of bent 1" bar stock is not going to resist that the way that the factory piece does. You've got that weight on the car, make sure it's working for you. If you're not sure, ask Dave Coleman about it. 
 

Also, try running those shocks softer. It costs nothing and I suspect the car will feel less nervous and more planted. Go way soft and increase the damping until you feel the springs come under control. Make sure your available bump travel is as much as possible, not just "healthy". With those spring rates and a reasonable ride height, you should be able to get very little bump stop interaction. 

In reply to dps214 :

More power and tire are definitely something that would be nice. I agree that the cage adds more structure, but I just got off the phone with my cage builder. Door bars are $300 and a cage is $3500.

For $3500, I can part this car and buy a more modern platform that requires less reinforcement to begin with.

In reply to Keith Tanner :

The stock beam in an NB cannot be bolted back on once it is cut out from between the frame rails. The only bolted piece is the plastic bar that supports the bumper skin in 5mph scuffs. I will have to fab a proper brace that can tie the two chassis legs together in the vertical direction, but this will get in the way of my radiator ducting and badly cramp my current engine air intake. It is what it is. 

I used to run the shocks at 50% and they were badly underdamped. I can try backing off to 75%, but MeisterR specifically said that the correct damping for track use is achieved when turned all the way up.  I intend to replace these at some point, but good dampers cost more than twice my mortgage. 

"Healthy" up-travel in my case is 80% of shaft length exposed at static ride height. I have separate adjustable spring preload and damper length to achieve this. On other cars, I typically go for 60% as long as the springs don't become loose at full droop, but Miatas don't seem to use as much droop as other cars. The only time I found the bump stops was on Barber before I had aero and I was smashing the curbs in the kink after the Museum with wild abandon. With aero, I have the grip to turn a little bit more and keep even more speed.

Adjustable preload is complete crap used to market suspensions built out of generic shock cartridges. It's simply not a thing in cars. You want as much bump travel as possible, the correct shock length is "fully engage the bump stops just before the control arms make contact with the subframe" unless you happen to have a very high ride height or extremely constrained shaft travel. That's how Miata-specific suspensions like Xidas and Fox are designed, and it's where the two-piece setups will work best. MeisterR doesn't understand suspension geometry and cannot get their heads around how preload works, which is why I don't automatically follow their shock damping suggestions either. There's a big difference between 50% and 100%, play around with that. Fully cranked feels super-sporty but can mess you up on fast transitions. Even the expensive dampers aren't any good if you don't tune them properly, and turning those knobs doesn't cost a cent and can be very educational.

You have a better idea of what's going on in the front of your car than I do, but if I was chasing rigidity I'd be concentrating on tying those front arms together well. On my V8 track Miata I was able to make it work with some welded-in pipe even with the relocated radiator and the Corvette intake.

YoursTruly said:

In reply to dps214 :

More power and tire are definitely something that would be nice. I agree that the cage adds more structure, but I just got off the phone with my cage builder. Door bars are $300 and a cage is $3500.

For $3500, I can part this car and buy a more modern platform that requires less reinforcement to begin with.

Can you though? In this market...

The cage is really $3k once you sell the roll bar. And the chassis stiffness difference is probably at least in the ballpark of the price difference. But I don't know how much of that extra stiffness you really need. To keith's point, that money might be better spent on door bars and a set of dampers that were designed by adults.

I will reiterate what Keith said:

Soften up the shocks, that doesn't me they are soft just softer. That will help with the transitions

As for the front end.....................you need to rework whatever you need to rework to tie the front arms together. The flex won't be consistent and your already struggling with it as a driver. As you make the car faster the problem will only get worse. Sheet steel with flared holes could be added between the arms and that might not mess with your intake or ducting in the manner that a beam does.

If I have the choice between a rigid car and a 100lb lighter car I'll take the rigid one every time. Personal opinion here but I find flexi-fliers (of any degree) to be utterly useless when it comes to chassis tuning.

In reply to dps214 :

I have a friend who supplements his income by selling used Miata parts and our total is about $11k to fully part out the car. That makes the budget $14k to start over, which is a decent used NC Mx5 with a roll bar, a seat, and very nice coilovers. I often have to let this setup pass me when I am at the track. The biggest problems with this plan are the uncertainty, the work involved, and the year or so of missed events it would take to get every single item sold to maximize budget and buy the next car.

Better dampers are in my future, but I need to save up for a while.

In reply to Tom1200 :

I definitely plan to try them softer, because why not? It is free to try. 

I am trying to figure out the best brace design for minimum weight, non-interference with current parts, and my capabilities with the equipment I have. I think I want to keep the main bash bar, but build off its existing backing plates. 

YoursTruly said:

In reply to Tom1200 :

I definitely plan to try them softer, because why not? It is free to try. 

I am trying to figure out the best brace design for minimum weight, non-interference with current parts, and my capabilities with the equipment I have. I think I want to keep the main bash bar, but build off its existing backing plates. 

Sounds like an excellent plan.

Tom1200 said:

I will reiterate what Keith said:

Soften up the shocks, that doesn't me they are soft just softer. That will help with the transitions

As for the front end.....................you need to rework whatever you need to rework to tie the front arms together. The flex won't be consistent and your already struggling with it as a driver. As you make the car faster the problem will only get worse. Sheet steel with flared holes could be added between the arms and that might not mess with your intake or ducting in the manner that a beam does.

If I have the choice between a rigid car and a 100lb lighter car I'll take the rigid one every time. Personal opinion here but I find flexi-fliers (of any degree) to be utterly useless when it comes to chassis tuning.

Tom.  The Morgan had a wooden chassis.  And was able to be extremely quick for the modest power it had.  
The "trick" with any flexible chassis is a stiff suspension to make it react predictably.