I am starting to design a new chassis based on the basic dimensions of a Locost but I plan on installing a 425 hp 4 LB-FT LS1 A little underpowered, but I want reliability for the street. This chassis is about 4 times stronger in torsen than a standard Locost per grape FEA. Note that this chassis does not rely on stressed panels for strength, Any helpful thoughts would be great.



AW

[Edited on 7/9/06 by whittlebeast]

Hi and welcome,
Looks pretty substantial to me, but you dont show a transmision tunnel, I rigid tunnel will improve chassis stiffness agreat deal as well, also some well placed welded bulkheads will also improve stiffness.
Also, with using a large engine and the associated running gear I would suggest some strength in the mounting points.
Adam

Running a small block chevy in my Luego. When we were putting the chassis together extra triangulation went in everywhere to the extent where an extra 7 m of tubing went in.

The engine mounts were easy we got them from a hot rod place. They're rated to 1000 hp and as a result the car rocks not the engine !

Will be at Donnington if you want to cast your eye over the extra triangulation.

Are you making one with a roof??

What's happened to the side diagonal at the rear (can't remember tube name off hand, the one in red). On a standard locost chassis I reckon that would be a fairly important chassis member as a lot of the rear suspesnion forces would be directed down it.


Locost Chassis Design

Are you going to be able to get in and out ok with the roll cage design?

Best thing to do now is to build a balsawood model (available in thin square section from model shops, hot glue gun works well). Stick on some thin card (or plastic card if you're feeling flush) for any structural bulkheads.

Then you will find out how it actually performs, just grab it by the suspension mounts and twist gently, soon feel and see any weak points.

A good 3d model rather than a picture is also much better for deciding if it's practical and good looking - getting in and out, exaust headers, getting the engine in and out etc.etc.

iank, your sounding like my architecture lecturers! but very true, an old coat hanger, pairof pilers and a scale rule on the next shopping list!!!

whittlebeast, you have triangulation across your passenger compartment in the form of a roll cage(?). wont this make it hard to get in??

This car is going to be an autocross car and possibility licensed for ocational use on the street. Entry thru the roof or thru the side (and the small hole in the cage) is an option. Regarding motor mounts, I may just build a front and rear mounting plate and let that structure serve as the firewall bulkhead. I have built several cars and I always use Grape FEA to design the structure. see photos of my last car at

http://www.ncs-stl.com/racecar

FEA software is at

http://www.grapesoftware.mb.ca/

I have tube definition file for anyone that wants to do chassis design

AW

[Edited on 6/9/06 by whittlebeast]

As you as you know what your doing! hehe.You probably know far more than me by the sounds ofit. Personally I'd still mock up certain sections at scale. Id also probably be tempted to mockup the rollcake in cardboard tube just to see whether its quick to escape in an emergancy.
Sorry I didnt mean to sound like i was doubting you, just id prefer to ask and be told to shut up rather than not ask and you get stuck a few monthsinto the build or anything worse when running at speed.

JamJah
No problem. Generally I start with the basic triangulation of the chassis an prove that the basic concept is do-able. I next get the large components that are difficult to move like engine transmision and differential on the floor and mark the outline of the chassis and all major bulkheads to make shure I still fit in the big picture. In this cas I have the luxury of setting in existing cars and hold a peoce of tubing in about the location that the model FEA is showing to prove that I can fit my knees in the chassis and I can get thru the holes that I think will work. In sitting in an existing car I soon realized that my hips, knees and the center box structure were in conflict when I got the motor back far enough. This thing is slated for a powerglide transmission as I plan on it being fas enough that the last thing I want to think about is 5 or 6 gear changes. (I also own a shifter kart so...)
see http://www.ncs-stl.com/mike/Mike_Gateway_Spin.wmv to get an idea how fast these cars are.

AW

Here is and interesting option for a high HP install out of a 2000 Ford Cobra.



[Edited on 7/9/06 by whittlebeast]

I think I have desided to use the old racecar rear suspension. All I have to do is locate an old Nissan rear dif to mount in the center of it and redesign the mount. This way I own the wheels front and rear.



The front suspension will most likely be...




[Edited on 8/9/06 by whittlebeast]

Just for fun, I did some anyalizing of the basic locost chassis this weekend Here is the results.

green panels as stressed members 938 ft-lbs/deg (note that the center tunnel is stressed on all four sides but is a little hard to present the lower picture showes the tunnel better)

green panels as stressed members plus added red braces 1780 ft-lbs/deg

green panels as stressed members plus added red braces plus blue panels 2000 ft-lbs/deg

my chassis with no center tunnel and no stressed panels at all but with a full cage 9320 ft-lbs/deg

As a general rule of thumb is you take the weight of the car with driver and fuel in lbs times 3 to get the minimum strenght required to get theshocks springs and anti-roll bars to do their job. so a 2000 lb car should have about 6000 ft-lbs/deg stiffness Just a little food for thought

Andy





[Edited on 10/9/06 by whittlebeast]

If you are going by the book there is still a diagonal missing between H and A1.

Previous analysis has shown that putting another R tube in on the other side of the engine bay, and triangulating the LA,LB,LC,LD stucture is very benficial. Would be interesting to see if your model supports this.

I tried adding the H - A1 tube but there was very little added strength. I was trying to add as little as possible to get the largest BANG for the added weight and complexity.

The front most bulkhead was not moving around that much.

AW

[Edited on 10/9/06 by whittlebeast]

MikeRJ

I added the blue tube and got up to 2400 ft-lbs/deg The tube is in addition to all the other mods, the 4 side tubes and the two roll cage stressed skins.

Remember that attaching these skins to support these sort of loads may take some effort. I try to stay away from stressed panels. They tend to be a fabrication and maintaince nightmare. I avoid drilling any holes into tubing as I like to vent every tube into another tube so that they are all innerconnected. That way I can pressurize a chassis and check for hidden cracks with age.

AW

It would seem that the driveshaft tunnel, firewall, bulkhead, and outer skin would add considerably to the overall strength turning the frame into a Monique. I agree with drilling the tubing from a structural point of view. I plan on using 16 ga steel for all sheet with 18 ga steel used for the outside panels. These are simple flat panels and not difficult to fab. The plan is to full weld the perimeters with skip weld and fill weld to interior tubes. The whole assembly to receive a light coat of linex to seal and provide additional bond. A cage such as the Westy will be welded in. I'm building for 500 hp.



[Edited on 10/9/06 by locostv8]

[Edited on 11/9/06 by locostv8]

LocostV8

Cool cage. The problem with seam welding in the panels is all the heat that gets applied to the chassis. I have had better luck sticking with tubes and for the most part square sections except where the actual roll structure is involved.

At some point I reach too much strength and start removing tubes to simplify the structure any tube that comes up in a blue color and is not there for some specific reason is removed?

What software are you using to model that cage?

Wish I could take credit for the cage but it is a Westfield competition cage. It would be too easy to use modeling software but it takes $$$ and I'm retired so I try to stick somewhat close to what is known to work.

I'm going a bit overboard with the first LC and using 1x1.5x11ga for the main rails and 1x1x14ga for the .75 x.75x16 ga. If you stitch weld in short sections and go back several times to fill in you minimize the heat applied. In doing this I'm going overboard but am turning the entire structure into a cage. This will be a street car using a roller 392(351) with efi, T5, and MN12/Cobra (MII geometry with longer A arms and Rorty rear with appropriate mods) suspension pieces and a budget of about 3k (with stock roller 351).

It appears you are into EFI, I would be greatly interested in talking offline about it.



Deman MN12 IRS


R/H 2000 Cobra R Center Explorer L/H MK VIII with explorer back cover
I will be using the Cobra R housing with Torsen T2(OEM) and Explorer cover giving 4 point mounting for the diff

[Edited on 11/9/06 by locostv8]

LocosttV8

Just go over to http://www.msefi.com and post a question. We can Squirt anything Just get a signon of locostv8 and I will realize it's you.

AW


[Edited on 11/9/06 by whittlebeast]

My FEA results have already been posted on this site.

The book chassis has 1180 ftlbs per degree and weighs 181 lbs.

Adding a V brace on the front face and diagonals across the front suspension mount tubes is a very worthwhile improvement. Another is to weld in the panels on the top of the tunnel and at the bottom of the front suspension region.

Some of the tubes in the tunnel do very little and can be left out.

Some more weight can be saved by making the panels in 18 gauge as opposed to 16 gauge.

The result is 2541 ftlbs and a weight of 171 lbs.

Replacing tube R with a Y brace on each side of the engine bay gives 2683 ftlbs for 174 lbs weight.

This final modification results in a chassis that is ten pounds lighter than the book design and has just two extra tubes in total.

Look up kitcaranalysisv2.doc (version 2) for a full description of the mods.

If anyone would like a Nastran input deck I'll be happy to email it to them.

I can't see where you are going to fit the engine in with all the engine bay bracing. The support at the front of where the tunnel goes will obstruct your gearbox bellhousing unless I am looking at it wrong.

The tube over the bell housing has been raised a few inches to allow for the large motor. The final location of the four engine bay braces will be determined once the motor is in place. If I have to I can stress member the motor.

(yep americans interchange motor and engine. Nobody calls them enginecycles)

cymtriks

I was looking for 6000 ft-lbs / deg and wanted to avoid stressed panels if possible.

AW

[Edited on 16/9/06 by whittlebeast]

I have managed to get to 12000 ft-lbs/deg with a design that all of the lower frame rails are in book locations. The roll cage is situated in a way that the driver can get in and out without any real issues. The entire center tunnel structure is not required. It is all possible to do.

Now I feel confident that a suspension can be designed to realy handle as a new Lotus handles. I am designing to a 1850 lb minimum weight so I am hoping to have to add about 500 lbs of steel to the floor and get the weight distribution where I need it to accelerate as required. There should be room to add a turbo where the normal passengers feet go as there is no tubing there now if 400 HP ends up controlable.

Just a little food for thought.

AW

[Edited on 18/9/06 by whittlebeast]

1850 lbs? Crikey that's heavy. Why so much? Or does all the strenghening and engine of choice make up most of this?

see

http://www.scca.org/_FileLibrary/File/2006_solo_rules.pdf

starting at page 144 of 316

Its just the rules of the class. It is difficult to get anything to handle if it does not have about 60% rear weight and about 5 lbs/HP with driver.

AW

This would probably be a US only option but. Page 130 says EM, which includes Locost 7s, unlimited engine with weight of 1800. I know of a McSorley +4 car with a 302/T5/8.8 that weighs 1250 with 50/50 weight (but it is too flexible). That would say that you would have to do some rather serious balast. WHAT IF, the balast was in the form of a 460 (everything but block/crank/rods aluminum) with a big spline toploader, attach the toploader to an aluminum 8.8 diff then build a 7 clone around it. The 460 can easily be 513 and grow to as much as 600 infact a worked but mostly stock 460 can be gotten over 500 hp with ease. If you have to add balast anyhow why not in the form of unbreakable running gear?

Part of this entire project is to build a test ride for my web site. See http://www.v8efi.com

I want to develope a bolt on MAF controled Megasquirt kit for LS1 motors for streetrod and dragrace applications. I have friends that have done back to back testing with with removable 400 lb plates under the drivers set. His findings were that the car was just as fast with the added weight. The issue is it is impossible to get a RWD 50% rear weight car to accelerate off corners. 40-60 just works better. Point one F1 car with 50-50. Veiw http://www.ncs-stl.com/mike/Mike_Gateway_Spin.wmv to get a real feel for what these cars drive like. I am just trying to develope a road worthy, non-winged, thumper version of an A-Mod car with all the drivability of a super-bike.

The goals are under 11.5 in the 1/4 mile at about 135MPH, 1.6 geez in a corner and 1.5 geez on the brakes and quiet. A passenger seat is optional and my have to be left out.

AW


[Edited on 20/9/06 by whittlebeast]

quote:

---

Originally posted by locostv8
I know of a McSorley +4 car with a 302/T5/8.8 that weighs 1250 with 50/50 weight

---

The driver in a seven at CG? Only with a sever case of lead foot me thinks. With 50/50 ratio CG must be at half the wheel base, about the location of the pedal stand.

I thought about that later... okay, fair enough, though if the driver is same, and the wheelbase the same, the driver becomes unimportant.

Installing an engine literally twice as large, an axle twice as strong, and larger wheels and tires, and have it not only weigh the same, but not move the CG... is pretty amazing.

Maybe I'm wrong on this... just how much does a "typical" (automotive) 4-cyl Locost weigh?


[Edited on 9/20/06 by kb58]

This sounds like a familiar discussion.

All I can talk to is what I'm doing. I keep refferencing Deman because his Wide 7 is closest to what I'm building.
http://www.deman-motorsport.com/widespecs.htm
with a Hyabusa 1300 his car weighs 985 lbs. He uses MN12 (89 to 92 Tbird) spindles and brakes and MN12 rear components with steel diff in what I believe is a +4 width 93" wheelbase car. I would imagine the eng/reverse box/driveshaft would weigh about 250 lb so you would have an approx rolling chassis weight of 735 lb

These are the same components I am using except. Front spindles/brakes are modded to remove caliper mount and rubbing blocks for pads and being replaced with PBR aluminum calipers for a weight savings of 20 lb per side. Rear diff is aluminum 2000 Cobra R with Torsen probably 20 lb lighter. The weight savings is nearly made up by heavier wheels and tires. The 351 weighs 525 lbs, I believe this is an early 351 with steel heads/intake/waterpump/exhaust manifold/flywheel, The later blocks as with the 5.0 are a bit lighter and with aluminum heads/intake/waterpump/flywheel and tube headers I would imagine that you could remove 75 lbs for a total of 450 lbs which is within a few lbs of a 302 or 2.3 turbo. A T5 weighs 85 lbs. I doubt I would add more than 75 lbs for steel floorpan and slightly bigger tubing. I will be adding power steering and brakes which probably add 50 lb. Total 1355 lbs.

The car example above is a +4 with MII knuckles/brakes and an 8.8 but smaller tires/wheels and a chunk of foam for seats so I believe it is concievable that he hit 1250 lbs and the engine is set as far back as possible.

http://www.strokerengine.com/EngineWeights.html

kb58
Sorry could not resist
It is easier to balance out a 7 than you might expect. Moving the engine back a few inches makes a big difference, increasing the wheel base is also effective. With a longer chassis like a +244 it should be possible to get 50/50 even with a quite heavy engine.

The problem is that 50-50 is not the goal, 40-60 is. If 50-50 appears to work good, then you need more power, a lot more power.

AW

[Edited on 21/9/06 by whittlebeast]

Back to the original point.

One of the current engine masters buildups, Livornious (SP) built a 509 ci 460 putting out 835 (within 1 or 2, going off memory) and the 460 can easily go 557 so without thier major masaging figure more ci a7 800 hp with an original weight of 720 which used HEAVY heads/manifold even with stroker should be able to get down to 650 (prolly a bit less). A car that over powered would be more easily handled with an automatic so throw in a C6 for 200 lb. Throw in another 100 lb for reinforcement still comes out about 1685 with a bit for balast over the axle. conservativly that comes out p/w of 2.3 lb/hp. An 800+ hp big block would be a challenge to megasquirt. How do you spell TIRES/TYRES.

[Edited on 21/9/06 by locostv8]

Night John boy.

OOPS!! Forgot myself. Them rubber thingys at each corner that keep wearing on the back corners.

And for that dirty great V8, he'll need some DAMN wide ones too!

Re my mods-
2500 to 2700 ftlbs per degree isn't the most that can be achieved with the locost but it is probably the best that you can get AND-
get the original bodywork to fit.
get commonly used engines to fit
not fit a cage
fit the book suspension
still save weight overall
not increase complication greatly
etc...

Fitting a roll cage will have a big impact on stiffness but will cost you on weight.

I can't understand the reluctance to use stressed panels unless you mean riveted panels which are a bit dodgy as the rivets can work loose over time. Welded thin gauge steel panels should be fine and may weigh less than a diagonal with an ally panel.

Re weight distribution-
I'm sceptical of the 40/60 weight distribution being optimum. Perhaps it was for the car tested with a 400lb weight bolted on (post earlier in this thread) but many mass manufacturers try hard to get get closer to 50/50. BMW made an advert boasting of their nearly "perfect" weight distribution of within 2% of 50/50 for every model. The advert stated how some of the cars had batteries in different position to get this figure. Why bother if it wasn't a good thing? Also some supercars are described as twitchy, or at least needing great respect, on the limit precisely because of their rear weight bias.

At the risk of hijacking the thread...

On a turn at constant speed, yes, I agree 50/50 is best, but how often are we in that situation? Coming into this turn we are on the brakes. Under braking we want equal load on all the tires for maximum braking ability, and a rearward weight bias helps accomplish this.

After we make our turn we step on the gas. For maximum traction we want more weight on the driven wheels and a rear weight bias accomplishes this too.

One other thing, it is very common to have larger rear tires. In that same corner, our 50/50 car will not have evenly loaded tires; the fronts having a higher load/area than the rears. I'd expect understeer from this "perfectly balanced" car.

I know in the big picture all sports cars do pretty well, but I do feel 40/60 is a more optimum target than 50/50.

As far as manufacturers saying what's best, I think there's a fair bit of marketing BS thrown in. That is, with a typical front-engine car, I don't think it's physically possible to get weight distribution to 40/60, so it's in their best interest to say 50/50 is "best." About moving the battery, yes, it improves things, by moving the weight toward the rear, which is where the 40/60 already is.

[Edited on 9/24/06 by kb58]

From my experience weight bias front/rear is much less important than centering the mass to make a car stable and predictable. The polar moment of inertia (PMOI) is far more important to me.

This is why when you are adding ballast to get up to minimum weight you add it to the center of the car, not the back.

Think of a stock, front engine rwd sedan. Weight bias of 60/40 or so. Can expect some understeer with easy/smooth throttle oversteer. Pretty predictable and easy to drive at steady state cornering right? So lets try and make it better and get 50/50 bias by moving the engine back further. So now you have the heaviest lump almost in the middle of the CG, no weight up front, and the same weight in back. Add a little throttle oversteer and the mass in the back is going to pivot on the CG with no mass up front to counter balance it and it will try to come around much faster and be much harder to get back. Think of a dumbbell on a pivot VS a bowling ball of the same weight. The dumbbell is harder to get turning, but once it gets going its hard to stop.

Will this car put down faster lap times? Maybe due to the faster PMOI up front on entry but maybe not because the driver is too scared to push it to the limits.

The point I am trying to make is that centralizing the mass is a better thing to worry about than weight bias. PMOI is what gets you through tight corners faster and lets you feel more comfortable pushing the car to the edge. Yes, it makes the car easier to slip out, but it also makes it a lot easier to get back and less likely to turn around on you.

I feel like I haven't explained that very well but Ive run out of time for now.
Maybe I can clarify later if it doesn't make sense.

Cheers.

quote:

---

Originally posted by cymtriks
Re my mods-
2500 to 2700 ftlbs per degree isn't the most that can be achieved with the locost but it is probably the best that you can get AND-
get the original bodywork to fit.
get commonly used engines to fit
not fit a cage
fit the book suspension
still save weight overall
not increase complication greatly
etc...

Fitting a roll cage will have a big impact on stiffness but will cost you on weight.

I can't understand the reluctance to use stressed panels unless you mean riveted panels which are a bit dodgy as the rivets can work loose over time. Welded thin gauge steel panels should be fine and may weigh less than a diagonal with an ally panel.

Re weight distribution-
I'm sceptical of the 40/60 weight distribution being optimum. Perhaps it was for the car tested with a 400lb weight bolted on (post earlier in this thread) but many mass manufacturers try hard to get get closer to 50/50. BMW made an advert boasting of their nearly "perfect" weight distribution of within 2% of 50/50 for every model. The advert stated how some of the cars had batteries in different position to get this figure. Why bother if it wasn't a good thing? Also some supercars are described as twitchy, or at least needing great respect, on the limit precisely because of their rear weight bias.

---

“the front center of the chassis in the z direction”
This one sounds a little suspicious, do you get any reaction here?
Try locking one of the front spindles in z and let the reaction force be one part of the force pair.

EDIT:
Sorry. Early Sunday morning z is “sideways” I guess, so let's forget about the force pair.
Still, if you get a reaction in that point something is fishy.

Cheers!

[Edited on 06-10-8 by leto]

Restraining of the front center realy is in the Z (up/down) direction. Without that even with matching but opposite forces making a torque couple without the vertical movement restrained you get a division by zero in the calcs. If I constrain one front spindle and apply the torque load to just one front spindle I get the same chassis stiffness.

The build begins

Any progress and how did you do the bearings on your AM cart?

Hey Andy,

I talked to you a little while back about the sucker car, thanks again for the info. I Didn't know you were on this forum, just wanted to say the locost build looks good, should be fun... (I'm still designing and finding sponsors for the project.)





[Edited on 1/1/07 by turbo time]