Chassis Flex: Does it matter?

[Lizardracing]

Based on internet findings on a metal sales website

Chrome-Moly Tensile Stength = 90,000 psi
Diameter (in) Thickness (in) ID (in) Cross Sect. Area (in^2) Tensile Strength (lbs) Weight (lb/ft)
1.75 0.120 1.51 0.61 55,303 2.089 lb
1.75 0.095 1.56 0.49 44,453 1.679 lb
1.75 0.083 1.58 0.43 39,119 1.478 lb
1.50 0.120 1.26 0.52 46,821 1.769 lb
1.50 0.095 1.31 0.42 37,738 1.426 lb
1.50 0.083 1.33 0.37 33,253 1.256 lb
1.25 0.120 1.01 0.43 38,339 1.448 lb
1.25 0.095 1.06 0.34 31,023 1.172 lb
1.25 0.083 1.08 0.30 27,386 1.034 lb


DOM Tensile Stength = 70,000 psi
Diameter (in) Thickness (in) ID (in) Cross Sect. Area (in^2) Tensile Strength (lbs) Weight (lb/ft)
1.75 0.125 1.50 0.64 44,668 2.169 lb
1.75 0.095 1.56 0.49 34,575 1.596 lb
1.75 0.083 1.58 0.43 30,426 1.478 lb
1.50 0.120 1.26 0.52 36,416 1.769 lb
1.50 0.095 1.31 0.42 29,352 1.426 lb
1.50 0.083 1.33 0.37 25,863 1.256 lb
1.25 0.120 1.01 0.43 29,819 1.448 lb
1.25 0.095 1.06 0.34 24,129 1.172 lb
1.25 0.083 1.08 0.30 21,300 1.034 lb

HREW Tensile Stength = 40,000 psi
Diameter (in) Thickness (in) ID (in) Cross Sect. Area (in^2) Tensile Strength (lbs) Weight (lb/ft)
1.75 0.120 1.51 0.61 24,579 2.089 lb
1.75 0.095 1.56 0.49 19,757 1.596 lb
1.75 0.083 1.58 0.43 17,386 1.478 lb
1.50 0.120 1.26 0.52 20,809 1.769 lb
1.50 0.095 1.31 0.42 16,772 1.426 lb
1.50 0.083 1.33 0.37 14,779 1.256 lb
1.25 0.120 1.01 0.43 17,039 1.448 lb
1.25 0.095 1.06 0.34 13,788 1.172 lb
1.25 0.083 1.08 0.30 12,172 1.034 lb


If 2 chassis were identical, the Moly car would weigh less and be stronger over all but flex about the same.
If you go to mix matching tube sizes and types, there's a lot more math involved.

[MasterSbilt_Racer]

Based on internet findings on a metal sales website

Chrome-Moly Tensile Stength = 90,000 psi
Diameter (in) Thickness (in) ID (in) Cross Sect. Area (in^2) Tensile Strength (lbs) Weight (lb/ft)
1.75 0.120 1.51 0.61 55,303 2.089 lb
1.75 0.095 1.56 0.49 44,453 1.679 lb
1.75 0.083 1.58 0.43 39,119 1.478 lb
1.50 0.120 1.26 0.52 46,821 1.769 lb
1.50 0.095 1.31 0.42 37,738 1.426 lb
1.50 0.083 1.33 0.37 33,253 1.256 lb
1.25 0.120 1.01 0.43 38,339 1.448 lb
1.25 0.095 1.06 0.34 31,023 1.172 lb
1.25 0.083 1.08 0.30 27,386 1.034 lb


DOM Tensile Stength = 70,000 psi
Diameter (in) Thickness (in) ID (in) Cross Sect. Area (in^2) Tensile Strength (lbs) Weight (lb/ft)
1.75 0.125 1.50 0.64 44,668 2.169 lb
1.75 0.095 1.56 0.49 34,575 1.596 lb
1.75 0.083 1.58 0.43 30,426 1.478 lb
1.50 0.120 1.26 0.52 36,416 1.769 lb
1.50 0.095 1.31 0.42 29,352 1.426 lb
1.50 0.083 1.33 0.37 25,863 1.256 lb
1.25 0.120 1.01 0.43 29,819 1.448 lb
1.25 0.095 1.06 0.34 24,129 1.172 lb
1.25 0.083 1.08 0.30 21,300 1.034 lb

HREW Tensile Stength = 40,000 psi
Diameter (in) Thickness (in) ID (in) Cross Sect. Area (in^2) Tensile Strength (lbs) Weight (lb/ft)
1.75 0.120 1.51 0.61 24,579 2.089 lb
1.75 0.095 1.56 0.49 19,757 1.596 lb
1.75 0.083 1.58 0.43 17,386 1.478 lb
1.50 0.120 1.26 0.52 20,809 1.769 lb
1.50 0.095 1.31 0.42 16,772 1.426 lb
1.50 0.083 1.33 0.37 14,779 1.256 lb
1.25 0.120 1.01 0.43 17,039 1.448 lb
1.25 0.095 1.06 0.34 13,788 1.172 lb
1.25 0.083 1.08 0.30 12,172 1.034 lb


If 2 chassis were identical, the Moly car would weigh less and be stronger over all but flex about the same.
If you go to mix matching tube sizes and types, there's a lot more math involved.

If you look at 1.50 x .120, there is no weight difference. Which, really is more reflective of dimensions being the same. The other ones have weight difference because cross section is different due to tolerances allowed for the tubing type.

There is no value here to determine flex. That is determined by modulus of elasticity when you are below yield. You better be about 45% of yield at most, on any chassis members, with maximum loading during a lap, or that thing will be junk very quickly.

[MasterSbilt_Racer]

Tensile stress = load to this value, tube is in 2 pieces

Yield stress = load to this value, tube is one piece, but permanently deformed

Modulus of elasticity = slope of the stress/strain curve until you load to the yield point. Think of this as the spring rate. All materials deflect with load. If you don't reach yield, the dimensions remain unchanged when you unload it. Operations below yield mean you use this to calculate deflection, just like a spring. For example, 300# on a 100#/in spring compresses it 3 inches.

I'm taking a lot of liberty for simplicity as axial and bending loads are handled differently.

[Lizardracing]

I think I wrote that wrong. I was trying to agree that according to those numbers, if they are legit, flex is in the mechanical design structure of the chassis and not because of the tube being DOM or Moly. Assuming identical chassis, identical tubing sizes used.

To put this another way, 2 10’ sticks, hanging off the edge of a table 5’ with 200 pounds hanging from each will bend downward the same amount and spring back when the weight is removed. The Moly tube, with more elasticity, will be able to hold more weight before it bends permanently.

Back to the flex and handling effects.
Billet says in his tests, cutting out bars and all that, didn’t appreciably effect the handling of the car or the driver feel. Can we infer that flex changes the sensitivity of the chassis adjustments? IE, a flexable car needs bigger changes to effect the car handling while a stiff car might be so sensitive the car is tough to keep on point?
What I mean by that is dirt tracks that change a lot all night, a stiff car might be good for a lap or two but begin falling off at a faster rate while a more flexable car will be good over a broader range of track conditions. If that is indeed true, one might see why an asphalt car or a drag car can be stiffer, while a dirt track car appreciates some flex.

I can tell ya on my modified, made with 1.5 and 1.25 .083 wall DOM tube, is very insensitive to changes. I know this, and just make bigger swings as the night goes on. It also doesn’t sit flat on four jack stands so maybe I’m ignoring something I shouldn’t be! Hahaha!

[talclipse]

Glad to hear bonehead is doing well Billet. Sure miss his input around here.

I think cars today are more consistent than they ever have been. Most chassis builders have went to some form of CNC tube notching and bending so their fitment is much better than they were 10 years ago. (I can remember picking up a modified chassis and it having 3/8" weave welds where their fitment had gotten sloppy.)

With that said I will die on the hill that cars built from 4130 need some form of stress relief/ normalization. I know rocket uses a mechanical stress relief device that mounts under the jig and uses ultrasonics to vibrate the chassis at high frequency to achieve this. I've also been told it was down at one point and they built a few cars without it and there was no difference.

I know a local "CHASSIS BRAND by DRIVER" built a fixture where they would secure the chassis, then use a jack with a load cell to raise the RF corner of the car a given measurement. They had found that their stuff worked better in a specific range. They went as far as to having different numbers for SLM, crate, steel block, etc.. Its probably worth noting that at one point they were replacing a chassis after only a handful of races. ¯\_(ツ)_/¯

[billetbirdcage]

talclipse: With that said I will die on the hill that cars built from 4130 need some form of stress relief/ normalization

They are when they get powder-coated, least I've had people tell me that. Look up normalization and see if you think 400 degrees an oven might get is gonna get the job done. . .shrugs/smiles

[Lizardracing]

No.
Annealing temp are around 1500' for several hours and a control slow temp decrease and normalization/stress is around 1200' for 24 hours.
450' for 30min ain't doing nothing......Measurable anyway.
Maybe part of the process is a 12-1500' to bake off impurities first, the coated, then baked at 400 for 30minutes and you have an argument.

[Austin34471]

"We here at * ______ Race cars* perform a special stress relief and annealing process in an large oven that is big enough for an entire chassis while using a special coating to prevent oxidation and further material degradation over time."
Oh so you powdercoated it. Nice.

[Ghopper]

It is hard to "like" a comment here. The forum needs to adapt to 2022 and mimic some disruptive social media trends.

You guys are really getting into the weeds of component details. Tubing thickness and gusset designs are all over the map in this market. Now everyone is guessing on residual stresses like they are talking about dark matter (https://en.wikipedia.org/wiki/Dark_m...ough%20gravity.) Don't most people vibrate at time of weld?

The answer is to measure the current spring rate of the car, and tune this to be in a window that is determined from objective/subjective feedback of multiple track test sessions. Billet was already doing some of this.

No dirt chassis manufacturer has contacted me about measuring their vehicle stiffness. I will just end up knowing it for their cars. Cup teams have chassis-only twist rigs to check variability in cars. I was lucky to have a friend that got this rig off FB Marketplace to make it easier find stiffness of the whole car assembly.

[grt74]

alot of over thinking going on here, just saying, its more important to just and some bars in a particular area,

[ZERO25]

So the real question is.....can it be measured accurately and predictably to enable us to adjust our wheel loads effectively?


Side note.......Hoffman says he buys 15 chassis at a time. They come in pieces, in a box, pre-bent from a cnc company. His results speak for themselves!

[MasterSbilt_Racer]

So the real question is.....can it be measured accurately and predictably to enable us to adjust our wheel loads effectively?


Side note.......Hoffman says he buys 15 chassis at a time. They come in pieces, in a box, pre-bent from a cnc company. His results speak for themselves!

You can easily build a test setup. It's not about adjustment. The car is either in a range you know is good or you don't use it. That's what one should do with the data.

We live in a world though where the chassis is blamed when a guy has something else screwed up and thinks the chassis is wore out. It's expensive, but allows egos to remain intact.

[grt74]

You can easily build a test setup. It's not about adjustment. The car is either in a range you know is good or you don't use it. That's what one should do with the data.

We live in a world though where the chassis is blamed when a guy has something else screwed up and thinks the chassis is wore out. It's expensive, but allows egos to remain intact.

completely agree with this one !!!! ill say this and leave it alone, all chassis take a set per say, thats why you can build 2 cars the same but one is going to run better, put that one on a pull down ,get the numbers the same and there its is (its amazing that we are looking at 1/16th of an inch and less these days) ,now you have the same car (notes are the big key here), of coarse one will always feel a tick better than the other, just the way it is, now the issue with tubing isnt racing the car that is the issue, its when you stuff her in the wall at over 150mph, at that point its becomes a saftey issue, thats why you get mandates from nhra,cup, and others in the racing world, we are getting closer and closer that point, over 150 is a factor and every 25mph after is, thats why youll always see cup under 200 for the most part
just my 2 cents
consistency is going to be the biggest factor, on all parts, tubing,welding, testing, tweaking, then great notes,
hoffman is winning because of his complete package and consistency, not the chassis by itself

[Ghopper]

You can easily build a test setup. It's not about adjustment. The car is either in a range you know is good or you don't use it.

I reject this. Change the stiffness matrix of the existing car (which is an assembly of many parts). There are many parts on a racecar that changed after they leave the factory; Tires, shocks, motors are matched to the changing needs of the consumer.

How: Maybe you are changing motor mounts, number of bolts in components, adding a brace, cutting a brace.

[MasterSbilt_Racer]

I reject this. Change the stiffness matrix of the existing car (which is an assembly of many parts). There are many parts on a racecar that changed after they leave the factory; Tires, shocks, motors are matched to the changing needs of the consumer.

How: Maybe you are changing motor mounts, number of bolts in components, adding a brace, cutting a brace.

I didn't say you couldn't bring it in.

[ZERO25]

I was hinting at making it easily adjustable........at the track if need be! Theres plenty of super stiff bump springs that could be hidden in the right places.

[TheJet-09]

Somewhat off topic, but I was curious...how many man hours do you think go into building a chassis (on average) from the first piece of tubing on the jig to ready for powder coat? And having asked that, where do you think the largest percentage of cost in building that new chassis comes from...material (but I doubt that)? The design itself? Labor to build it?

And I don't ask because I think I have an answer, I'm truly just curious.

[billetbirdcage]

Somewhat off topic, but I was curious...how many man hours do you think go into building a chassis (on average) from the first piece of tubing on the jig to ready for powder coat? And having asked that, where do you think the largest percentage of cost in building that new chassis comes from...material (but I doubt that)? The design itself? Labor to build it?

And I don't ask because I think I have an answer, I'm truly just curious.

Designing car, making jigs, and labor to do that stuff is a lot to recover so that cost has to be spread out into the number of cars your gonna build. The number of hours you have in just drawing tabs, brackets and etc is overwhelming. I know on one of my cars I have over 90 different tabs/brackets and lasered parts just for that one style of car.

Material (before stuff got crazy) cost just for the main frame (no bumpers and etc) is gonna run close to 2K if you buy enough to do multiple cars (say 5) and then you have lazered parts, brackets, tabs, and etc (that can easily run 700.00 per car).

Time to build a car from raw tubing to finish ready to powder coat: That's gonna very a ton, things like how often you do it and your equipment change that a lot. For example, I've had a guy weld for me before and he's really good and fast as hell. A car all tacked together on jig took him 6.5 hours and then 1.5 hours out of jig to weld underside that you can't get to on jig. I did the same car and didn't keep real good time but I was around 14 hours just in jig part. Course he put his helmet down and didn't lift it until he was done.

2 guys working can usually do one in around 40 to 50 hours, if you done a couple before. Again that's no bumpers, fuel cell cages, etc just a bare frame.

Honestly frames should cost way more then they do, If a guy was honest.

#Note: this is bending the tubing and notching it, nothing CNC'ed prebent and notched

[ZERO25]

Do you think buying cars in a box, CNC'ed, will lead to more consistency in rigidity?

[fastford]

do they have CNC,s that can weld an entire LM chassis ?