Chassis Flex: Does it matter?

[Ghopper]

Chassis flex: Does it matter?

I am poking the board. Motivated by some brand promotion for simulation software. (if your company is doing simulation of mechanical systems, consider Siemens Simcenter which has many types of solutions under one brand)

https://youtu.be/Ea-MZfhnUmU

Now Billet and Austin can make comments...

[Lizardracing]

I learned in college engineering class, everything is a spring.
If it's consistent, you can work with it, or around it.
If it's not, things get a lot more complicated with never ending variables.

[billetbirdcage]

Chassis flex: Does it matter?

It should but I will pose this question? Say you take a car at a private test session and start cutting bars out of it (completely cut out at each end) and rerun laps after every cut bar.

I know this may be hard to know exactly what bar I'm describing but I'll give it a shot:
1. RF diagonal bar from mid plate upright to main frame rail
2. LF diagonal bar from mid plate upright to main frame rail
3. Diagonal bar from dash bar to RF upper rail
4. Diagonal bar from midplate upright backwards on right side
5. long diagonal bar from right side upright to frame by lower RR 4 link
6. Bar from RF to LF at very front of car in front of radiator (nothing connecting the 2 front upper rails besides removable engine bar to front shock mounts
7. Diagonal bar over drivers head in halo
8. Diagonal bar from dash bar to RF midplate upright


So how much did each bar change lap times? Take a guess?

Some side notes to help:
1. This was in 2010
2. This was a 100% all moly car, all the car was .065 wall tubing besides the main frame rails
3. This was at magnolia and at the time the track/car was in the low 15 second range
4. Only car there so track was not changing and It took me roughly 1 to 2 minutes to remove hood or whatever and cut each bar with a sawsall then back on track for 3 laps.
5. This was at the end of a 7 hour test so track was black and very consistent.

This obviously isn't very scientific as you really have no idea if the car was flexing a ton to start with or if the car was rigid for the time so you can't really say if I made the car 1% more flexible or 50% by cutting those bars out, shrugs. But I can tell you the car didn't like to take a hit and would bend pretty easy with all those bars out of it.

[Ghopper]

6. This car did not have a sway bar.
7. This car had data? or just subjective driver feedback?

I will guess the car got looser overall, but possibly had better corner entry mid way through the cutting session? It could have even got faster when it got looser.

[billetbirdcage]

6. This car did not have a sway bar. No sway bar
7. This car had data? or just subjective driver feedback? No data, just lap times and driver feedback on what it did to car


I did forget to add: SAS spec package at the time, so steel block 358 with 12" spoiler at 2100#. IIRC we where 2150ish before bars cut and spec motors then where in the 660 to 680HP range

[Austin34471]

I will yield my time with the microphone to Billet to hear the results of this test..

[Austin34471]

My guess is that: the driver said it felt a little tighter on entry, less drive, and *sigh* it went the exact same on the stop watch with every bar cut out because dirt late model

[MasterSbilt_Racer]

While we wait for Billet, I know of a car in the mid 2000s that when it stopped winning, it was time to take the roof off, reweld halo cracks, then it would win again. Scary and true.

[fastford]

the theory use to be , in certain conditions , some chassis flex was a good thing , i personally never bought into this and thought of it as a crutch , I think with todays shocks and springs , the less flex the better ....jmo.....

[billetbirdcage]

Now you need to remember this was a while back and 800# zero point shocks and other really stiff stuff was super common at the time, so obviously that should have an effect on the chassis.


1. RF diagonal bar from mid plate upright to main frame rail No change in times - driver could not tell a difference
2. LF diagonal bar from mid plate upright to main frame rail No change in times - driver could not tell a difference
3. Diagonal bar from dash bar to RF upper rail No change in times - driver could not tell a difference
4. Diagonal bar from midplate upright backwards on right side No change in times - driver could not tell a difference
5. long diagonal bar from right side upright to frame by lower RR 4 link No change in times - driver thought he felt a difference
6. Bar from RF to LF at very front of car in front of radiator No change in times - driver could not tell a difference
7. Diagonal bar over drivers head in halo No change in times - driver thought he felt a difference
8. Diagonal bar from dash bar to RF midplate upright No change in times - driver could not tell a difference


Austin cheated as he has heard this story before. It absolutely made no difference in lap times at all, the times the driver thought he felt a difference in those 2 bars he still wasn't 100% sure what it did and was indecisive about it.

[billetbirdcage]

Now, let me go back and say: Do I feel chassis stiffness matters? Yes, I do but I haven't had the equipment like Ghopper to get simulation numbers or any time of real number to see how much change is needed to see a difference and when that difference is noticeable and where.

I actually just revisited this 2 months ago, on a chassis that is much stiffer then the one I did back in 2010. While, I'm gonna keep those results to myself this time it did show on the stop watch with a single bar removal and a real change was felt by driver and we could see it also.

[Ghopper]

I am impressed that Billet cut that whole car up! Must have been one destine for the shelf.

We are setting up for some tests this season. We have measured on-track deflection and are quantifying the stiffness changes for several braces. Next do blind studies with a driver of using those braces in different pre-measured configurations. Data system with load cells on some key locations.

It may not yield any noticeable changes. In that case, additional stiffness is better for repeatability. In the opposite case, complementary suspension changes could be necessary and will need understanding.

Why is this an issue at all? Because some drivers say car to car variation matters. That points to a manufacturing tolerance that needs specification and validation test for each new car. Such that new cars are within a window that is smaller than a driver can discern.

What is a dirt car vehicle stiffness window? Around 2000 lb*ft/deg to 4000 lb*ft/deg. Now you can populate your models or do some napkin calculations. Maybe this can be your teams specification to your chassis builder when you order your next car. "I would like 2500 lb*ft/deg with +/- 100 lb*ft/deg over 40 midwest 1/4mile nights and max 0.125' plastic deformation". Or chassis builder could specify for you "this design is one of our favorite in the 2500 lb*ft/deg variety, with notes of strong restarts and smooth finishes". (corny)

[billetbirdcage]

I am impressed that Billet cut that whole car up! Must have been one destine for the shelf.

Why, no bar that was cut out couldn't be replaced with some new tubing, grinding and welding? How bad do you want to win and how far you willing to go?

No, this car wasn't getting scrapped. It was about 10 races old at the time and after that test I put certain bars back in and that car raced for another 5 years. Won, tons of races.

I will be curious how much multiple cars of the same make and year vary. I have tested tubing crudely for how much it flexes and it varies a ton, especially between brands of tubing. I know 15 years ago I rockwell tested several scraps of tubing (dom and moly, from my drop pile and several friends that build cars too) that where different brands and same brands but years apart on manufacturer dates, needless to say they varied a bunch. Dom was the worst, even the same brand but different lot had a lot of variance. While not very scientific, it did show me some of the problem with a good car and not so good car could be coming from the tube itself. Hell back then you could go to one of the big LM manufacturer place and look at the tubing rack and you would see 5 different brands of tubing. Getting the same brand of tubing every time was and still is sort of impossible unless your ordering 6 months to a year in advance to if you have to wait for a certain brand of tubing it doesn't hold you up. I found one brand of moly that was the most consistent on Rockwell numbers so I built all of the cars out of that, right or wrong but the cars seemed really consistent for the small sample size of cars I built.

[billetbirdcage]

Also how big the gaps and fit of the notches makes a big difference on how much the joint/weld shrinks when it cools and thus pulls on that bar and may preload the chassis one way or the other. So how consistent the builder does things, imo is gonna have an effect. Now is that enough to make a difference, shrugs

Tons of variable here, lol.

Disclaimer, I'm just a dumbass and nothing I say is worth 2 cents

[fastford]

with todays load that is placed on the right front , do you guys think anything could be gained by using the down tubes from the halo to the upper front shock mounts ? the old TNT chassis were this way and worked pretty well in there day , but they stopped doing it after a while , of course then there was not as much dynamic load on the RF like today , just wander if there could be something there that could help with todays set ups , one thing i have noticed is the cars today seem to hop or bounce more through the turns , just thinking undampened flex could cause this ....