I got bored and made a subframe.

[Calan]

Someone should get the dimensions and throw it into Catia or NX and see what happens when the load is put on it.
I would but I'm in my last semester. Lots of work going on.

As I said before...

Dynamic loading would be the most critical aspect though, and it would be next to impossible to define an accurate scenario for that without having access to a lab and $$$ worth of accelerometers and test equipment. Not to mention it would vary considerably over different cars.

It's not that the real world doesn't follow the laws of physics taught in school; it's just that there are usually countless factors that either aren't known when working out a problem on paper, or the interactions are so complex that it just isn't feasible to design something with 100% theory. Sometimes you just have to build it and see if it breaks, and experience will usually trump theory.

The problem is that it would be very difficult to properly constrain and load this type of model to get any definitive results. Even if you could come up with some numbers for the static forces and moments acting on it under many different conditions, those numbers would change with different suspensions, lateral grip, vehicle weight, etc. If you want to consider shock loading, modal frequencies, etc... then that's a whole other level of difficulty. It can be done, but I doubt that anyone here has the resources or knowledge to do it accurately.

So, you overload the shit out out of it in the analysis, and then design for a 50% safety margin. But, that won't accomplish much more than welding it up and seeing if it breaks.

Running a model through FEA is easy; accurately defining the constraints and forces you're trying to solve for is extremely difficult.

 

[dsmusmc]

I like how he just threw this together to kill some time. No it does look sturdy because its not finished he clearly said that it was just a prototype. He never stated anything about selling these or doing a mass production. I think he should just be left alone and let him do his thing before he gets jumped all over with bullshit. After all the shit id be surprised if he even continues this thread or the progress.

Hopefully he keeps up the good work. Good job and would like to see more. #### the hater.

 

[95g-s-x]

Give it a rest.....these cars have been around for 10-15 yrs. Things break! Christ, you'd think in that time, someone would design a lower balljoint that doesn't separate and "send you into a wall!"

OP, offer it as not for highway use/racing part with a diclaimer and you'll be fine. For that matter, send me one and I'll test the shit out of it. If I go into a wall, cool, I have something to do in my spare time. If i die going into the wall, my family is set from my life insurance!

 

[buhay6]

Um, harmonics and model analysis are WAY OVER THE HEAD of most people....

Even if it is man just don't word it like that, I'm not trying to be a dick I'm just looking at both sides. And I know a little about harmonics and I'm not an engineer at all.... just seen it first hand and watched a little blurp on it on discovery channel, so I know a little philosophy behind it. What I'm trying to say is don't just make assumptions that no one will understand because some people will, that's what everyone meant by arrogant.

 

[pauleyman]

I happen to have an engineering degree also. Good points from a lot of folks. There is alot to be said for both academics and real world experience. From a made for the public standpoint I'd be surprised if there wasn't some modeling taking place for anything that's structural. The factor of safety built into most consumer goods is also way out there. The only way things would break is typically some catastrophic event. Does that mean this subframe will or won't fail? No, I think it only means that it isn't modeled and we don't know the loads on any of the points. I would think at least a order of magnitude would need to be a factory so one would know what do and the manner in which to do it, engineering degree, real world experience or not.

 

[Boosted98gsx]

I've heard of a couple Magnus subframes fail, this is the only reason why I'm commenting in this thread.

 

[flyineagletsi]

I say Buck it and run it. The final version that is. For over 60 years, hot rodders drove what they built. There hasn't always been super high tech analysis programs or pencil neck mathematicians dissecting everything that one man wants to do to his own car. That is until the modern day computer and e dick measuring forums. Build the sh!t so that you feel safe enough because at the end of the day its your own ass. And IIRC, the OP never mentioned selling the product, which would completely eliminate the possibility to hurt anyone other than himself in a track car. We are not car manufactures and no one can expect or design around all of the variables which is why there are such things as recalls.

Experience is the best education.

 

[wishihadatalon]

What is the material? Please oh please say low carbon mild steel.

 

[Calan]

What is the material? Please oh please say low carbon mild steel.

How about CF? Now THAT would be interesting.

 

[nuclearr]

I have to agree with most people on experince vs. degree. My father is a big wig aerospace engineer. He was working on a project for NASA and the Air Force. They were trying to make a liquid hydrogen tank out of carbon fiber. Great Idea. Except the design called for the tank to be made in two pieces and be "stir welded" it told him when I was in high school it would not work. He told me it would. Turns out they could NOT make it work (with the tecnology available) and the program was so far over budget it was shut down.

To the op keep up the good work but do some testing before you say it will take "anything you can trow at it". I dd my car and have some of the crappiest roads in the nation where I live. I could destory a stock suspension in a day if I tried (not that I want to). Tubular sub frames are apealing to me becaue I hate the design of the front suspension on the 2g not a fan on the control arms on the rear but they work. But I am poor and can not aford to buy one so I may result in making my own.

 

[RipperXX]

I didn't even read the 2nd page after so much bull shit that was on the first.

Just wanted to say that I didn't see anywhere in the OP first post (nor any after for that matter) that he was going to start producing these for sale. Which is good since thats not what the custom fab portion of this forum is for. Also I wanted to say something to the OP. Ignore the haters, there are to many people on forums like this that want to be the popular guy/troll/Mr. Know it all etc. The WHOLE dsm community is not like this.

But I do have to say I can easily see why we tend to lose a vendor here and there, DSMer's in general are some cocky know it all's and im sure they get tired of dealing with a crowed like that, that doesn't tend to actually shell out a lot of $ when compared to say the Evo/Supra/RX-7/WRX etc. crowds.

 

[Morphius]

Just wanted to say that I didn't see anywhere in the OP first post (nor any after for that matter) that he was going to start producing these for sale.

^^This. I got the same out of this.

Lots of drama here. People need to chill.

OP, cool. Got any more pics to share?

 

[Morphius]

I've heard of a couple Magnus subframes fail, this is the only reason why I'm commenting in this thread.

I find that hard to believe. Magnus makes a quality subframe. I'd wager, they've actually done FEA analysis too. Marco doesn't take short cuts.

 

[dsmornothing]

Thank you to the supporters of this. I plan to run this on my GSX that should be in the 800 hp range. I do have more pictures that I will put up a little later tonight! Thanks for showing intrest in this! The other side is completed now. Now its just tying in my lower control arms and its ready to rock and roll. I am looking at doing a couple add ons to this one as well. Driveshaft loop and Axle loops so you dont take out any of that nice work!

You must be logged in to view this image or video.

 

[Alex7747]

Thank you to the supporters of this. I plan to run this on my GSX that should be in the 800 hp range. I do have more pictures that I will put up a little later tonight! Thanks for showing intrest in this! The other side is completed now. Now its just tying in my lower control arms and its ready to rock and roll. I am looking at doing a couple add ons to this one as well. Driveshaft loop and Axle loops so you dont take out any of that nice work!

You must be logged in to view this image or video.

Damn, that came out sick as f***. looks legit man,

 

[AK Maniak]

Looks good man. Be sure to let us know how it all works out.

I'll have my bachelors in petroleum engineering soon so hopefully i'll be able to understand what you guys are talking about since its way over my head

 

[Bud92gsx]

That super sweet. I can't wait to see this bolted up!!

 

[dsmdeliveryboy]

I think it looks awesome, its good to see people putting quality craftsmanship into things still and take pride in their own work instead of just going out and buying something.

 

[H@xtGSX]

..and the PE thing, come on man. Like two letters in front of your name convinces anyone you have a clue...

. So true!

This almost made me choke on a bite of pizza, it made me laugh on reflex... I work with a PE I wouldn't trust to tie my shoes. hahaha

As for harmonics and FEA... Well, my degree is actually in critical care nursing, and I know what they are... I work as a designer, draftsman, and am an "engineer" because of the nature of my job. I engineer both mechanical and structural things for a working mine. You don't need a degree to run AutoCAD, ReSys or Solidworks, but it seems every engineer sure thinks you do... 99% of engineering is common sense, and the other 1% has computer programs written that do the math for you. The degree is nothing more than a piece of paper proving you could pass those classes... In the real world only engineering stuff I see engineers doing is running the exact same programs I use. The computers do all the work... I would put $20 on me being able to walk into quite a few engineering departments and ask someone "Quick! What's the moment of inertia of this beam sitting like this?" and most engineers would say "Hang on, let me put that in Solidworks real quick...". Now, don't get me wrong, an engineering degree gives you more of a working knowledge of the "why" behind the way things work, and that's very useful. I have to ask or read up in engineering text books fairly often to get that kind of info, and that can slow me down in comparison to someone with a degree in it.

 

[dsmdeliveryboy]

I'd be interested in something like this for a 1G, I do road-racing and this would help out a lot. I wish there was a way to x-brace the front and rear subframes like evo owners do.

 

[Timechg]

Give me a CAD document (IGES preferred) and I'll throw it on ANSYS for some Modal analysis.

Boosted98gsx, you are a ridiculous man. I have an engineering degree too, but I would never talk the way you do. People who do FEA know that the model is never perfect and can never simulate 100% actual conditions. For example, you keep on throwing out harmonics (amplitude blah blah blah), what values would you assign to this loading condition? What value you would you assign to the torsion, bending etc?

Not to mention your basic premise that you must anneal the welds first.....How the eff are you going to make a model that accounts for the micro-structural composition of the metal/weld. What would your boundary conditions be? You are throwing words around and claiming people don't know what you are talking about thinking that your better than everyone else. If you were to create a model like this (which I know 100% you could never do) you would be major author in computational journals. Ansys wouldn't be able to handle what you are proposing.

If you really want to find out how much it can handle, break it. A lot of R & D engineers spend their time breaking their prototypes to see where the failure occurs, a FE model would be a complete waste of time.

Let the man make his part. It's better than seeing a fellow engineer result to trolling.

 

[TSimage]

Great job OP and get some safe testing in on that thing as I am sure you are aware. If this actually does end up in an 800whp DSM then make sure you try it out at half title first and work into it. Hell put it into a beater and back into a wall just for sh*ts and giggles, see what kind of harmonics it makes ..

 

[wishihadatalon]

FWIW I asked about the material because I too am an engineer (in training) with a primary focus in welding. If it is made from from low carbon steel, annealing would be pointless. If the carbon content is low enough, you couldn't quench the part in oil and build martensite if you wanted to. Simply put, if the carbon content of a given steel is low enough, there is absolutely no possible way to form martensite (brittle grain structure) and normalizing would do nothing but waste money.

 

[Raph]

What are you lot on about?..race parts are not designed for road use,race parts fitted to cars that requierem are constantly to be inspected by the owner /driver etc..

Or do race engines last 100k miles on the street ..no.

And to the expert saying that the OP will face a lawsuit if the parts fail injuring people..

no he wont, the user is responsible for his equipment,and that equipment isnt for road use as it isnt approved..so whoever bus it and uses it on the road is at fault..
And who ever needs these parts,definatley doesnt have a streetable car anymore anyway.

So whats the point of argueing over a bit that the OP made for race/off road purpose?

A F1 engine does two weekends max..

A Rally cars suspension,subframe etc gets rebuilrt every season..

Diffs and gearboxes even more than once a season..

See what im getting at?

You cant use car with a Modified frame/sub frame etc on Public roads/highways etc...so why suggest the part will fail killing people.

a Driver using these parts does it on a rack and wears/has appropriate equipment.

 

[hubz91talontsi]

I'll take one for a 1g please.