• Forum membership has its advantages....

Computer Aided Design tools used in the industry

Bob_Sheaves

Well-Known Member
Ohayo goziamasu! (that means "Good Morning" in Japanese):

I have followed the link-coil discussion vs. leaf (hotchkiss) suspension design with great interest.

I have a question of some personal relevance (I am a vehicle dynamics engineer by trade) directed at the industry reps on this board, "What CAD tools (if any) do you use and why or why not?"

The second half of this question: "Is there any interest in utilizing the same design tools the 'Big 3' use for your projects?"

Best regards,

Bob Sheaves
 

Donahoe

Well-Known Member
DRE uses Autocad...Turbo cad .......... Solid works....... But mostly I use chalk & concreate...

NEVER LIFT!!!!!
 

Bob_Sheaves

Well-Known Member
I appreciate your answers to the first part of the question. Judging from the tools, it seems that cost is an issue. Is this true? Being in the industry for a number of years, I know what I have invested in my personal CATIA and DADS software and hardware. Which leads me back to the original post...

How do you do your preliminary analysis to ensure you haven't made an unforseen "oops" in the design of the vehicle? A/Cad, nor the others has even the slightest analysis capability (by this I am NOT referring to FEA- Finite Element Analysis- but rather dynamic analysis of the handling characteristics you have designed in). Experience can cover a great multitude of sins, but not performing some dynamics on a design seems to be "unusual". Could you please explain in more detail?

Arigato goziamasu! (formal "Thank you"- accompanied by a bow)

Bob Sheaves
 

Attachments

Donahoe

Well-Known Member
The problem comes down to veriables... And LOTS OF THEM... In and off road car you have so meny unforseen things that can throw a wrench into the best laid plans... As far as stress.. Most of the time I will consult with outside engineers. We are actually working on the rear arms for a new truck we are building now and we went over and over what we thought how thick the arms should be at the shock piviot points... Its a 51" long arm with 2 shocks riding about 30" from the piviot point... We went over it and over it.. Then called our engineers to get the tech answer as to what the loads will be and how thick to build the arms... Before we called I came up with 8" thick at the shocks.. We called them and after a few hours they called back with the answer... 8" thick.. Now keep in mind they started not knowing what I thought ...Just the facts... So they will take all the info we give them and start running analisis programs on the computer... My caculation came from looking at the span and thinking about it for a second and saying to myself... HMMM? I think about 8" ...Total guess..with a little bit of past knowlage... But whats funny is... Its happens 9 times out of 10... We go hmmmmmmmmmm?? I would stick that hiem about there... They come back and say it goes right there... We say oh yeah right where we wanted it. Cool. But what they can never figure out is the load a by-pass shock will take. We do most everything in G FORCE.. Well how meny G's does a by-pass shock see? Any one??? Any one??? Who knows...?? Thats just one veriable in a huge slew... Mother nature... Polar moment....Scrub??? WHat effects does scrub have in the dirt?? Figure that one out..... Plus the other thing we have going for us(off-road trucks now-a-days) is HUGE amounts of wheel travel... So how much does caster and camber matter when your pushing 20+ inches of wheel travel in the dirt... The problem most engineers come across when working with off-road trucks in OVER ENGINEERING... Thinking like its a road car... Its not... Lots of the came priciples apply but its not the same... So basically I build the thing send the weight and piviots to the engeneers and they come back with...YEAH IT SHOULD WORK!! What is the Right way to build an off-road truck or car??? You tell me??? Is there a RIGHT WAY??? Is there the RIGHT WAY to build a-arms?? Or there is? Well what about all the wrong ones out there that win races??? TO MENY VERIABLES.....To be totally honest most of the fab guys out there are not to cad freindly.. We are just learning ourselves... We have been playing with the cad for a couple of years... But we still like to use the old methods of desighning a truck... String....Chalk..... Floor...... I know of 4 trophy trucks that were built from complete computer drawings the rest started with a flat concrete floor and a pile of tube... Not to say that one is better then the other... I just know that sometimes what looks great on paper looks like ca ca in the real world. But we are really getting into the whole cad thing... Having a good time.... Learing more and more everyday... Its alot easier to copmlete piviot points with a cad thats for sure.... But I learned how to build these things on the floor.. Its hard to teach and old dog new tricks.

NEVER LIFT!!!!!
 

vwguy

Well-Known Member
as an autocad user i thinks its gocd for some things to sketch out but im just starting to design 4 links etc and to me like kreg said its seems easier to do it all by head, experience, and concrete. but most its TRIAL AND ERROR
 

FABRICATOR

Well-Known Member
Bob,
We all use CAD+R, and that stands for Cactus And Dust, and we must never forget Rocks. CAD as you may know it, is good in our sport for checking geometry, strength of parts, some ratios and spring rates, chassis design (most of it) and that's about it. No CAD program at any price will figure in all the variables, which the dezert is full of. Even if you had the greatest CAD program, you would have to "know" what safety factor to throw in at each step of the way. Otherwise your computer would design an Army tank. CAD is very useful in the design of each part, but only if you know ultimate loads, cycles, and the old problem of abuse. There is no doubt copious use of CAD could improve nearly any off-road race car. The most sophisticated programs might collectively be able to somewhat design a car for the rough, but how rough and at what speed? I don't think even NASA is there yet. Unfortunately, there are few if any Off-road race cars that are at a level where this is would make enough difference. Most off-road cars are loaded with parts, brackets, and mounts, that are mismatched as far as ultimate strength goes, and that's where CAD would help the most. It's sort of like putting an off-road race car through wind tunnel testing (and this has been done). Does it help? Yes. Does it win races? No. Most other forms of racing and also aircraft are made to do one thing and not be abused to any great extent. Ask yourself how an IRL car would be different if it were designed to survive anything short of hitting the wall at a steep angle. By the time CAD is truly a major force in this sport, all the cars will be virtually the same. We don't even know if they (the fastest) will be 2 or 4 wheel drive! That's what makes it so interesting and elusive. Craig is right about using chalk on the concrete. We also use pieces of welding rod, string, carpenters levels, tape measures and most of all the eyeball. So far, for the most part, it's all been a matter of improving what has already been done and borrowing ideas from wherever we may see them.
 

Greg

Well-Known Member
Kreg, has anyone that you know of taken the time to run real time tests on shocks? Like to use an accelerometer and presure monitors (on both ends of the shock) and gathered data? I have access to equipment like this. I have not run tests yet but even when i do i wont be shure what to do with the data. If someone has or could come up with basline parameters then this info could be used to set up other vehicles with much less trial and error.

Greg
 

Bob_Sheaves

Well-Known Member
Hi Greg,

This is a very common thing to do in the automotive industry-both on a
"shock dyno" and on development vehicles. As Kreg stated, it inputs are
what must be know for the data to have value for development of vehicle
specific valving.

FYI- these parameters are some of the most highly guarded secrets among
the automotive manufacturers.

Best regards,

Bob Sheaves
 

jeff

Moderator
Greg-

I can't remember the team names but... About six or seven years ago I remember seeing some teams that were testing shocks in the manner you describe... I remember watching some pretty interesting footage of all the gizmos and gadgets that they had hooked to the shocks. They ran a few vehicles, both street and off-road through their paces - in some cases they had an Indy car hitting a "bump" at 200mph - the off-road guys had footage of a Dakar type rally vehicle nailing a whoop at 100mph... the instruments recorded just about every damn parameter you could think of. All sorts of interesting stuff came out of it... ESPECIALLY THE EXTRA SLOW MOTION footage - it's amazing to watch springs do weird things. There was good footage of certain winds in the spring actually collapsing under extremely fast compression - to the point where the coil would no longer absorb energy - it just locked there before your eyes. It's cool to hear about it - but to see it was amazing. I can't remember which shops I was at - one was off-road and the other was a Indy type roundy round chassis analysis lab. I'll put a post-it on the dash of my truck to remind me to try and remember the names.

Oh yeah - on the CAD discussion... history has proven that even the best designs fail when poorly constructed. I know that's not the point of this thread - but if the fabricator screws up or the materials aren't up to spec - all the CAD in the world won't save you.
 

Dylan

Well-Known Member
A note about "CAD"
people think CAD is much more powerfull than it is. Programs such as autocad simply draw lines, one step above pencil and paper. Solid modleing (solid works, catia) works like a lathe and a mill in real life but with no chips on the floor and quick easy changes. These programs with the added FEA package are very usefull and will simulate testing like bending and braking parts, but in the wrong hands will give totaly worthless results " garbage in garbage out". I would like to take a look at at DADS software from what i've heard its very usefull. you can also do dynamic analisis on math software like " mat lab or Maple" but you need to know a lot about dynamics to use it. I got maple for $100 but i havent used it lately.

I use a solidworks sketch and excel to figure out caster, camber, scrub, scub radius, roll center,anti squat, dive, motion ratio, ect. but I could do it all with paper and pencil too, "and have done it with chalk and string on the shop floor at 2:00AM when need be". As for strength of parts its good to do some rough calc's on vulnerable parts then multiply by 2 just to be safe because you never know how big a rock your going to hit.

My questions:
I can calculate pitch moment, anti squat, ect.... but what do you want for these values? If DADS can tell me that then PLEASE send me a copy. (I've got my own guesses)

how well does FEA simulate a tube's impact with a rock?
what about the second impact when the tube is already dented in??
with a lot of work you could do that on FEA but it would be a lot more fun and understandable for the dezert-people to just go try it.
 

Bob_Sheaves

Well-Known Member
Hi Dylan!

It sounds as if you have spent some time working with "real" CAD/CAE/CAM systems so I will briefly describe the interactions of CATIA and DADS before getting to your specific questions.

The necessary modules of CATIA are ADD, KIN, COM (with the KIN kinematics module the key to the interaction with CATDADS-the graphic display pre- and post-processor for the DADS solver). When a vehicle is designed, as you surmized, "blocks of solid material" are located in 3 dimentional space to represent all the components of the vehicle. The more accurate this information, the more reliable the model (and the less guessing the system has to do). In the SOLIDE/ANALYSE function, you apply the material mass properties and CATIA will generate the CG (Center of Gravity) for each component. Once this is done, you built the kinematic motion sets defining the joints, travel, and all uniform/linear motion components. This sets your basic handling geometry in space (anti-dive, anti-squat, toe in, toe change, linkage ratios, etc.)

At this point, you would switch to the CATDADS/SETUP function where you now would define all NON-linear/NON-uniform deflections (i.e. tire spring rate, tire slip angle, durometer of any bushings that are urethane or rubber- like engine mounts is used- and CATDADS will add them to the kinematic set. The biggest issue with using MatLab or other analysis programs is that you need to be very conversent in the math behind motion and kinematics to be able to get a good, reliable result, as you pointed out. DADS and CATIA get around this limitation by utilizing expert system knowledge built into the programs. To illustrate, the systems "knows" that 2 solid bodies cannot pass through each other, therefore, if the design has interference when the motion is studied, the system will halt and highlight the interference zone and ask you if you wand to return or modify the parts at that point.

Now, to your specific questions. The values you are referring to are all vehicle specific and driver specific. When I design a vehicle and do not know the driver, I will design it to drive in the "neutral" or fully balanced zone, then add adjustment to allow the car to be setup for the individual driver's preferences. There are some rules of thumb, but the various builders use different settings for the rules. DADS is used for exactly that-determining what operates at the lowest "damage level" for a given set of circumstances. I obtained from TAACOM (Tank, Automotive, Army COMmand) a set of course profiles used by the military for implementation in DADS. several of the courses are a good representation of the Baja 1000 race, in both speeds and terrain quality. To make a long story short- the US Army measured and graded the soil samples taken from a couple of courses (2 at Aberdeen PG and 4 at Yuma PG) every 6 inches along the course. The input data to DADS is profile, soil penetration, moisture content, tractive shear, and a few other more esoteric items.

I had to chuckle about sending you the program...do you run an IBM RS/6000 S70 with 32 processors and 2GB RAM per processor? Or run AIX 4.3.3 for the operating system? Or utilize the GXT500 graphics card? If not, I'm afraid it wouldn't do you any good to have the software (it is also node-locked licenced). You can obtain a copy from CADSI in Iowa, but be prepared to shell out $37,000.00 per year for the licencing fees. Oh yeah, that's on TOP of the fees for CATIA (another $26,000 per year) and the hardware to run it (used, like mine, about $70,000 more). That is why I thought there may be some interest in the off road industry- I already have this "stuff" for other reasons, and can offer it for other's projects.

For you final question, the answer is "if you model the correct level of detail for both the loading and material properties, yes FEA will work".

Hope I have cleared up some of the questions....

Best regards,

Bob Sheaves
 

Dylan

Well-Known Member
My request for a copy of the software was not serious. I was aware that it does not run on your atari 800 home computer. The fact is if you are looking to drum up business the desert people aren't ready for $37,000,000 software. It would be interesting to see what you could do for the off road racing industry, so build a car and bring it to the baja 1000. If you would like to work with someone try calling Light Racing he may be interested in doing that level of analysis. But I don't think he speaks Japanese
 

Jerry Zaiden

Well-Known Member
I agree with Dylan. The desert is way to hard to simulate. We can build a fast truck on a flat surface with string, chalk etc... or in a cad program. This has been done both ways and both ways are proven. Baldwin / MacPherson trucks CAD by Brian Cudela those trucks are awesome. The Walker, Mike Smith, trucks built with real world experience seem to also do the job very well.
We are looking at the program called Solid Edge. It is easy to use and can simulate in motion simple things like a 4 link. This will show us where something might bind or the changes in parts. But I do not feel there is any way to simulate the desert, or the mistake the driver made at 100+MPH.
Look at Jim Baldwin in his fully CAD built truck. He hit a tree stump and that was the end of it. The rear end of the truck fell out.
I would love to learn more about you program. It sounds very powerful! But I do not think we are ready to buy a copy any time soon. :)
 

rdc

- users no longer part of the rdc family -
I am engineering student and for the past 4 years I have been designing and building off-road race vehicles with help of CAD/CAM software. The first vehicle I built was done the the old way, Trial and error. It took much time to get every thing the way it need to be because with out a plan you don't know what the next step is. The next vehicle I built I progressed to using AutoCAD. This was an improvement over the "string and chalk" method, but still was lacking solid direction during the build up process. Now I have come to my final year of school and have evolved to using solid modeling to design with. I use SDRC Ideas to build a vehicle in virtual space. It allows you to see the final product in 3d space before you every put the saw to the chrome-moly. I don't know if I could go back to using the old methods now that I have seen the power of this high end software. The fabrication process goes so much faster when you have a design sitting on the computer that you can look at anytime there is question on how the next stage of construction is to be done. One of the big advantages is that this particular software has a generative machining module built right in so it is very easy to take most any thing you can draw and create CNC toolpaths to produce the parts. The use of the software doesn't mean you have to stop using your 'gut instincts', it just allow you to tryout new ideas and and see a design before it is ever built. I say everyone that is building pro class vehicles should be using some form of CAD/CAM software, heck even the rednecks in NASCAR are using this stuff. A trial version of SDRC Ideas can be had for around $150 and it will do most everything most of use would need and will run on a standard PC. Well those are my thoughts on the issue, I attached a screenshot of a partially built solid model of a superlite buggy, take a look. Oh yeah if any of you buliders are looking for someone to do this kind of design work I just graduated and am available.

Later,
Karl

<P ID="edit"><FONT SIZE=-1>Edited by KustomKarl on 05/13/01 08:59 PM (server time).</FONT></P>
 

Attachments

rdc

- users no longer part of the rdc family -
Hey Karl, what engineering school are you in your fourth year at? Is that a Cad design in your attachment? Did you design the rear cage for Collins tt?
 

rdc

- users no longer part of the rdc family -
I am at the University of Wisconsin-Stout. It is a very hands on school, we not only get to design these vehicles but we also fabricate and race them on school time!! The attached file is a solid model of superlite design I am working on. The solid model is basicly a 3D mathematical model that is exactly the same as the real car(assuming we can build it exactly to specs). I wish I could design anything on a TT, any offers welcome.
 

Attachments

Jerry Zaiden

Well-Known Member
Just a sample of the program SOLID EDGE.

We are able to check the fit of parts. This is our F-150 pivot bracket. This is one great thing about CAD. If you spend the time you will have great parts!

 

vwguy

Well-Known Member
you dont need solid edge to drag a bracket i can draw a shackle in autocad 2000 in 10 minutes what im trying to say jerry is that solid edge puts a grey tint to the "metal" and a background thats it autocad works great for brackets
now as far as these programs go drawing a bracket or something thats going to be milld or cnc machined is great on computer but the 4 links and tubed frames and what not need to be chalk and string for now

now this is pretty easy to draw a shackle in autocad2000 and im only 15
 

Attachments

Curtis Guise

Administrator
I built this entire front suspension on my class 7 Toyota in about a weekend. I have allways wanted to learn Autocad, or similar modeliing programs that some of you have mentioned but don't have the time. I spend about 10 (or more) hours a day in front of a computer and haven't used it once to help build anything offroad related. But don't get me wrong, I wish I had the time to do it that way.
This is just more proof that you can definitely build something the old fashoned way, and have it work for you. I built this suspension once and it has worked great since. And I didn't even use chalk and concrete. Mostly just my brain, and maybe a pencil to write down some measurements so I wouldn't forget them. I built the sub-frame, arms, steering, etc.... 17-18" of travel.....


<font color=red>BTW it's for sale....<font color=red>
Curtis Guise
Juice Designs jdfab.com
 
Top