LocostUSA.com

Learning how to build Lotus Seven replicas...together!
It is currently March 29, 2024, 1:03 am

All times are UTC - 5 hours [ DST ]




Post new topic Reply to topic  [ 424 posts ]  Go to page Previous  1 ... 11, 12, 13, 14, 15, 16, 17 ... 29  Next
Author Message
PostPosted: December 26, 2011, 3:36 pm 
Offline
Automotive Encyclopedia
User avatar

Joined: December 22, 2006, 2:05 pm
Posts: 8037
Check out the pull down menu in the upper left hand corner where displacement magnitude is the solver default. There are various stresses in the pull down.

_________________
Miata UBJ: ES-2074R('70s maz pickup)
Ford IFS viewtopic.php?f=5&t=13225&p=134742
Simple Spring select viewtopic.php?f=5&t=11815
LxWxHt
360LA 442E: 134.5x46x15
Lotus7:115x39x7.25
Tiger Avon:114x40x13.3-12.6
Champion/Book:114x42x11
Gibbs/Haynes:122x42x14
VoDou:113x44x14
McSorley 442:122x46x14
Collins 241:127x46x12


Top
 Profile  
Reply with quote  
PostPosted: December 26, 2011, 3:45 pm 
Offline
Always Moore!
User avatar

Joined: November 9, 2007, 3:40 pm
Posts: 4074
Location: Pittsburgh, PA
I only had two stress options (I think total stress and total local stress) and both were showing 0. I'll have to try again. I must have done something wrong.

_________________
-Andrew
Build Log
Youtube


Top
 Profile  
Reply with quote  
PostPosted: December 26, 2011, 4:58 pm 
Offline
Automotive Encyclopedia
User avatar

Joined: December 22, 2006, 2:05 pm
Posts: 8037
Be sure to run the error checker next to the processor green equal sign button and rerun the processor after each change before selecting the post processor triangle.

FYI, on the twist angle (last selection on pull down), the value is in radians, not degrees, so multiply by 360 then divide by 2 pi to convert.

I've got tensile stress charts for 4 different points and one chart for user defined. I have values that are the same for the 4 different points. The user defined chart is present with totally different values. I have not attempted to decipher what that means or where the points are.

I can also display displacement and rotation for each axis separately, tensile force, shear force for V and W, bending about V and W, and twist angle. When it is working properly, each one will have it's own rainbow of colors. If you only see a solid color for the whole thing, rerun the solver, check the errors, and verify the load. You may also animate using default value and see if the whole thing slides back and forth without being constrained.

_________________
Miata UBJ: ES-2074R('70s maz pickup)
Ford IFS viewtopic.php?f=5&t=13225&p=134742
Simple Spring select viewtopic.php?f=5&t=11815
LxWxHt
360LA 442E: 134.5x46x15
Lotus7:115x39x7.25
Tiger Avon:114x40x13.3-12.6
Champion/Book:114x42x11
Gibbs/Haynes:122x42x14
VoDou:113x44x14
McSorley 442:122x46x14
Collins 241:127x46x12


Top
 Profile  
Reply with quote  
PostPosted: December 26, 2011, 10:55 pm 
Offline
Always Moore!
User avatar

Joined: November 9, 2007, 3:40 pm
Posts: 4074
Location: Pittsburgh, PA
Figured it out - I had truss selected. :oops:

_________________
-Andrew
Build Log
Youtube


Top
 Profile  
Reply with quote  
PostPosted: December 28, 2011, 12:50 am 
Offline
The voice of reason
User avatar

Joined: January 10, 2008, 4:47 pm
Posts: 7652
Location: Massachusetts
Thanks for the help on using the PrintScreen key etc. The result is below. This frame is about 5 times stiffer then the Locost model that Andrew put together and a bit better then half as stiff as his latest frame, if I remember the numbers correctly. It is nearly 4400 lbs. per inch in twisting at the rear roll bar. Surprisingly, Andrew's Locost model weighs about 111 lbs. and this model is close to 150 lbs.

This is the version of my frame for the transaxle, the front transmission one is being worked on. I think the numbers will be similar, but it will weigh more because it will have more of a drive tunnel.

This frame has about 60 tubes last time I counted. They are not highly stressed, in the torsion case anyway. I will check it for suspension and other loads soon. It should be reasonably build-able and repairable. Many cars suffer various indignities at the track so being repairable is a good thing.

I am uncertain about the X brace under the engine. It may only work with some oil sumps or engine and it isn't clear yet how to do the engine mounts. I will also draw up the alternate ( again... ) and give the numbers for that too.

This would be easier to see on a black background. Is that something you can set your window to or can it be done in MS Paint?


You do not have the required permissions to view the files attached to this post.

_________________
Marcus Barrow - Car9 an open design community supported sports car for home builders!
SketchUp collection for LocostUSA: "Dream it, Build it, Drive it!"
Car9 Roadster information - models, drawings, resources etc.


Top
 Profile  
Reply with quote  
PostPosted: December 28, 2011, 1:24 am 
Offline
Always Moore!
User avatar

Joined: November 9, 2007, 3:40 pm
Posts: 4074
Location: Pittsburgh, PA
Its coming along nicely Marcus. So it takes 4,400 lbs to move node 31 up 1 inch/node 35 down 1 inch? What is the distance between node 31 and 35?

Engine bays are such a royal pains in the butt. I may revisit it one of these days - a bolt in brace always seemed appealing. The kicker will be getting the bolted joint stiff enough so the brace is effective and not just ballast. I have to say I do like your geometry though. It seems very simple and to the point.

If you remove those lower engine bay diagonals and/or make them something super stiff like 2" diameter solid stock, how much does the stiffness change (mine just had to be there - if they were there, there was a nice increase: missing, a terrible stiffness decrease)? Ultimate stiffness comes from gaining half a percent or a percent here and there and taking advantage of the total gain but if it only reduces it by a percent or two but makes chassis design easier (or saves you from having to assemble the engine in the frame ;) ), it may be worth going for a "less than optimal for stiffness" geometry.

Just for curiosity's sake, what happens if you toss a diagonal in the bulkheads formed by nodes 9-13-14-16 and 17-19-22-23?

_________________
-Andrew
Build Log
Youtube


Top
 Profile  
Reply with quote  
PostPosted: December 28, 2011, 2:54 am 
Offline
The voice of reason
User avatar

Joined: January 10, 2008, 4:47 pm
Posts: 7652
Location: Massachusetts
Quote:
Its coming along nicely Marcus.


Thanks Andrew! I am just going to admit up front that I sweated bullets to get this anywhere near the stiffness of your frame. I am trying to do basically the same test that you did. I am applying 1000 lbs. up at node 31 and 1000 lbs. down at node 35. It is constrained in X,Y,Z at both front top corners. Interestingly if I left off the X on one corner the model moved around a lot. It was funny in the animation. Almost the same result for vertical but it moved a foot or two sideways.

The frame is 40" wide and the hoops are 1 3/8" diameter. So Node 31 and 35 are about 38 5/8" apart. The vertical displacement is .228" at node 31.

Quote:
toss a diagonal in the bulkheads formed by nodes 9-13-14-16 and 17-19-22-23?


There is a diagonal in that first bulkhead. I tried to rotate the model to make things visible, but that one seems to have sneaked by. Oddly the other bulkhead does not seem to need one. I even removed the tube between nodes 20 and 21. It was only carrying about 50 lbs. of force. It turns out the tubes with purple and yellow colors have opposing forces that are canceling out there at the node. Wish that could always happen.

The X brace under the engine replaces the more normal lengthwise tubes. Those lengthwise tubes were tending to be the highest stressed in the frame. Especially as other places improved. The X brace is enough to change the color on the tubes behind the main roll bar. I will do those tests tomorrow and post a picture.

I really have had a hard time keeping track of changes. It doesn't seem on this that you can change tubes in isolation for very long. It sort of becomes trees of choices. Then yesterday we had a blackout in the morning and I guess I didn't hit save the night before. The frame wasn't nearly matching the numbers I wrote down and I can't remember those changes :oops: I did make a good find today though, moving the front end of the upper engine bay braces back one bulkhead made a pretty big difference.

I'll put up the files tomorrow. There might be a little more improvement, but I'm getting the feeling this is sort of where it wants to be.

_________________
Marcus Barrow - Car9 an open design community supported sports car for home builders!
SketchUp collection for LocostUSA: "Dream it, Build it, Drive it!"
Car9 Roadster information - models, drawings, resources etc.


Top
 Profile  
Reply with quote  
PostPosted: December 29, 2011, 11:47 am 
Offline
User avatar

Joined: October 24, 2008, 2:13 pm
Posts: 5326
Location: Carlsbad, California, USA
I am such a wonk. I think this FEA analysis is really cool. It is also surprising. Based on what Andrew and Marcus have posted, it also appears that sometimes "less is more" and a lighter thing can also be a stronger thing.

When I modified the Haynes Roadster for my 3.8L V6 donor and the 300 HP I expect to eventually have, I really had to use some common sense knowledge (to increase strength go up in section size or gauge or both) plus applying lessons learned in the Linton analysis, the Aussie mods plus some other improvements done by others. It's fundamentally the same design, but more substantial and consequently heavier (image attached). I know I could have done a better job and have a lighter design if I'd had a good FEA tool to use when doing the modifications.
Attachment:
SN95-Mustang-Locost-Roadster-1.jpg


After seeing the results here and poking around the GRAPE and LISA sites, I found LISA accepts STEP files. I had a STEP file for my final chassis as a single part, so I downloaded LISA and loaded my STEP model in not really knowing what I'd get. Unfortunately, in STEP, it does not appear as a single part and there are discontinuities and it fails to mesh. It's a huge model, so I'm not too surprised. A simple node-based model as has been posted here would probably be much better anyway, but I no longer have a license for the software used to design the chassis and can't recover all the X,Y,Z points of the nodes. That's too bad because all the junctions of members are at the intersection of the section centroids for the most part and would be pretty simple.

However, LISA does load and mesh simpler components and does a really cool job of meshing some complex parts. Here is the original Haynes Roadster pedal box meshed by LISA from a STEP file and a separate rendering of it so people can get clear on the geometry, which is hard to see just from the LISA mesh file. This is a 3D, volume mesh.
Attachment:
LISA-Pedal-Box.jpg


Part Rendered.

Attachment:
Pedal-Box-Image.jpg


I know enough to know the mesh above is much more complicated than it needs to be for this simple part. But the point is you can get an immediate model for use and analyzing by loading STEP files in LISA. Almost every 3D modeling program can output to STEP, so you get something to analyze for almost nothing when you're designing. I would definitely like to use LISA for analyzing many Locost components that I'm designing subsequently. Besides the chassis itself, there are many elements like the control arms, mounting brackets, pedal assemblies and so on that could benefit from basic FEA analysis.

Cheers,


You do not have the required permissions to view the files attached to this post.

_________________
Damn! That front slip angle is way too large and the Ackerman is just a muddle.

Build Log: viewtopic.php?f=35&t=5886


Top
 Profile  
Reply with quote  
PostPosted: December 29, 2011, 2:49 pm 
Offline
The voice of reason
User avatar

Joined: January 10, 2008, 4:47 pm
Posts: 7652
Location: Massachusetts
Here are a couple more pictures from Grape of my frame work. They both show simple changes to the frame shown above. The first one shows the X brace under the engine replaced with elements like a normal Locost frame. the other shows the frame with just a diagonal across the upper engine bay.

The frame with the standard arrangement under the engine is about %50 less stiff and about 3 lbs. heavier to prevent more stress in those bottom frame rails and engine supports.

Attachment:
Car9AxialstressB.jpg


This one with a diagonal across the upper engine bay shows what improvements in this area might bring. The car is about %25 stiffer and might be lighter because tubes could be removed under the scuttle that support the current braces.

I think we are stiff enough, at least for the moment, and time is better spent now on the differential / transaxle mounts, engine mounts, and suspension supports.


You do not have the required permissions to view the files attached to this post.

_________________
Marcus Barrow - Car9 an open design community supported sports car for home builders!
SketchUp collection for LocostUSA: "Dream it, Build it, Drive it!"
Car9 Roadster information - models, drawings, resources etc.


Top
 Profile  
Reply with quote  
PostPosted: December 29, 2011, 2:59 pm 
Offline
The voice of reason
User avatar

Joined: January 10, 2008, 4:47 pm
Posts: 7652
Location: Massachusetts
There is some good news. I found a mistake in the description of the tubes I was using in the Grape model. Somehow I specified the same size hole in my 1"x2" rectangle tube as the 1"x1" square tube. Steel weighs a lot and the 4 pieces fo rectangle tube I used cost about 25 lbs. more then they should have.

So the weight of the frame is now coming in at 123 lbs. Yeah! There will be a few pounds gained for practical details.

_________________
Marcus Barrow - Car9 an open design community supported sports car for home builders!
SketchUp collection for LocostUSA: "Dream it, Build it, Drive it!"
Car9 Roadster information - models, drawings, resources etc.


Top
 Profile  
Reply with quote  
PostPosted: December 29, 2011, 8:02 pm 
Offline
User avatar

Joined: October 24, 2008, 2:13 pm
Posts: 5326
Location: Carlsbad, California, USA
Today is a sick day for me but, I wanted to make good use of my downtime and see what I could turn up that's useful in exploring FEA in general and LISA in particular. Here are two useful things on the Internet for the true FEA novice that I thought others might benefit from:

1) Explanation of basic FEA theory ==> http://www.sv.vt.edu/classes/MSE2094_No ... heory.html

You'll need to recall some facts from basic mechanics, trig and at least understand that there is a relationship between forces, displacement, stress and strain and maybe know about Young's Modulus - at least what it describes. A bare bones understanding of matrices and their multiplication is required. This is a very simple mechanics case in a Virginia Tech tutorial, but it defines all the basics and shows how things relate and what assumptions made to make the FEA system work.

2) After the above, there is a super-simple, online FEA program, tutorials and lesson assignments at Carnegie-Mellon University all available for free;

CMU FEA stuff here ==> http://engineering-education.com/miniFEA/index.htm

It may even be a better place to start since there is an intro tutorial that fills in the various values for you and all you do is watch the caned interaction and results from the Java-based FEA program. You don't have to download anything, but it does require your browser to support Java.

It sets up meshes, forces and displacements in a very visual way and then show the results they yield. The mini-FEA program is limited, but it's the educational content that's important here.

Cheers,

_________________
Damn! That front slip angle is way too large and the Ackerman is just a muddle.

Build Log: viewtopic.php?f=35&t=5886


Top
 Profile  
Reply with quote  
PostPosted: December 30, 2011, 2:25 pm 
Offline
The voice of reason
User avatar

Joined: January 10, 2008, 4:47 pm
Posts: 7652
Location: Massachusetts
I have made one more improvement to the engine bay bracing at the top of the frame. I don't think it will work for the V8 crowd, but it appears to be very good for an I4. The previous braces and the support tubes behind them under the scuttle area are replaced with a simple V with the apex at the middle of the front bulkhead. This is more rigid and would appear to be a good step easier to build, it requires 4 less tubes. It does raise the stress a bit in the frame because these tubes take a good amount of the load.

Here's a picture and hopefully a zip of the Grape files. On to other things to test in Grape.


You do not have the required permissions to view the files attached to this post.

_________________
Marcus Barrow - Car9 an open design community supported sports car for home builders!
SketchUp collection for LocostUSA: "Dream it, Build it, Drive it!"
Car9 Roadster information - models, drawings, resources etc.


Top
 Profile  
Reply with quote  
PostPosted: December 30, 2011, 2:40 pm 
Offline
User avatar

Joined: October 24, 2008, 2:13 pm
Posts: 5326
Location: Carlsbad, California, USA
Marcus,

It looks like the nodes labeled 20, 21, 22, 23(?) are your hoop at the dashboard. Is that right?

If so, what happens if you add a tube directly from 20 to 21? Sometimes the connections are white in these graphics, so something may be there already. If not, I'm guessing you'll get more stiffness from joining the two complex joints together laterally rather than just having 22-23 connected.

Cheers,

_________________
Damn! That front slip angle is way too large and the Ackerman is just a muddle.

Build Log: viewtopic.php?f=35&t=5886


Top
 Profile  
Reply with quote  
PostPosted: December 30, 2011, 3:36 pm 
Offline
The voice of reason
User avatar

Joined: January 10, 2008, 4:47 pm
Posts: 7652
Location: Massachusetts
Yes, that's the dashboard hoop. When I started this I didn't know how to model the bend properly so I just made them joints. Grape does support the right way to do this, I believe, but I haven't made the effort yet. Just thinking I need to keep making progress and do things like engine mounts, drivetrain and suspension loads.

So adding that tube does make a difference, but only a small one. I 'm thinking it's not worth it. When you put it in it carries a load of something like 150 lbs. and improves stiffness by about 1%. That comes to something like 2 thousands of an inch less flex where I am measuring it ( node 31 at the rear roll bar ). That tube adds 2 lbs. and reduces room a little under the dash.

In comparison the basic change to the layout of the upper engine bay bracing was more like %20 more stiff plus a weight and complexity reduction on top of that.

I think we are in an area now where this frame is stiff enough. Not certain on that, but we are around or under 1/4" of flex to what I think is pretty hefty load. That should be enough to make the suspension move about 2". What are folks using for wheel rates on their cars? Then we can put this in context. The point of this test is to see if one could use spring rates and anti-roll bars to affect the cars handling. If most of the springiness is in the frame, you can't do that very well.

In this design I am placing very high value on a couple of things: simplicity for ease of construction, lightness because race car, strength for safety. Well three things really, add stiffness for tuning. Oh wait, plus being general or adaptable enough to be used with a variety of bodies or layouts. So five things: simple, light, strong, stiff, adaptable and really, really fast.

_________________
Marcus Barrow - Car9 an open design community supported sports car for home builders!
SketchUp collection for LocostUSA: "Dream it, Build it, Drive it!"
Car9 Roadster information - models, drawings, resources etc.


Top
 Profile  
Reply with quote  
PostPosted: December 30, 2011, 11:01 pm 
Offline
User avatar

Joined: October 24, 2008, 2:13 pm
Posts: 5326
Location: Carlsbad, California, USA
I think you definitely have the sense of it, Marcus. Nothing is ever going to be perfect in every way and there is a lot to say for having simplicity and practicality in a design, which in real life can definitely trump "perfect."

Sailors say, "A simple boat is a happy boat" and it's true. I think we can extend that sentiment to Locost sports cars too.

Do you think you'll actually build your design?

Cheers,

_________________
Damn! That front slip angle is way too large and the Ackerman is just a muddle.

Build Log: viewtopic.php?f=35&t=5886


Top
 Profile  
Reply with quote  
Display posts from previous:  Sort by  
Post new topic Reply to topic  [ 424 posts ]  Go to page Previous  1 ... 11, 12, 13, 14, 15, 16, 17 ... 29  Next

All times are UTC - 5 hours [ DST ]


Who is online

Users browsing this forum: No registered users and 16 guests


You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot post attachments in this forum

Jump to:  
POWERED_BY