Announcement

Collapse
No announcement yet.

welding a cage

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • #16
    Re: welding a cage

    Well it depends on what I can get over the winter. I am going to do a project car with my father, but he's oldschool and doesn't think anything with four cylinders can go fast.

    I've been looking at 89-91 bmw 325i or something similar.

    If left up to me, i'd get a miata or 240sx.

    I will certainly be putting more hours of research into this than hours of building. I'm sure that before the season is out, i will have checked out just about every cage I can get near.

    Comment


    • #17
      Re: welding a cage

      Here is some tips I snagged from another forum a while ago. I have plenty of pics of my car, but they are at home ... I'll post them tonight for you.



      -----------------------


      Well, I've been deeply interested in cars and following racing since I was 8 years old, some 37 years ago. Over the years I'd tried to learn all that I could. Jumping to the present, I spent a full year before building my cage analyzing other cages from amateur to professional. It helps if you have a rudementary understanding of physics and engineering. But, the basics are pretty simple and if you stick with them, you can't go too wrong.

      Perhaps this is a good time/place to discuss the basics of good cage building. Ignoring material/size selection (which is set by sanctioning bodies):

      1) Triangulate. And triangulate some more. No mult-sided polygons unless totally unavoidable.

      2) Where tubes connect to one another they should all connect at a single point. Not close, but a single point or node. This allows for multiple load paths and each stress on the cage will be distributed along those load paths reducing the load each tube takes/distributes. If you ties tubes close, but not to a single point, each tube will carry substantially greater percentage of the load.

      3) Whenver possible, tubes should not connect to the middle (anything but the ends) of a tube without other triangulated tubes connecting there also to support and distribute loads.

      4) Did I mention triangulate?

      5) Whenever possible, tubes should be straight, not bent. To make a straight tube fail, either some other portion of the cage must first fail or the tube must be ripped from the cage (assuming proper welds). Bent tubes may bend the other direction before loading any other portion of the cage.

      6) Cages are often have compromises. Be aware of what compromises are being made and especially why. I've posted photos of my cage before and I've made two compromises and have good reasons for them (it has two tubes that are bent that ideally should not be, but in the particular application the compromise is IMHO a good one).

      7) It's easy to build a strong cage. It's much harder to build a strong cage with a minimum of mass (tubes). Be efficient in your design for maximum strength with minimum material. There have been several cage analysis here that have been quite excellent. Keep in mind people will disagree over details. That doesn't necessarily make a cage good or bad, but the basics need to be right.

      8) Don't get too clever. I see a lot of poorly designed cages because someone tried to be too clever. You even see this in the pro ranks. I've seen a lot of cage designs come into favor and then fall from favor. But the basics always remain. And remember, the pros build a lot of cars and are looking for the smallest edge. If you have a one car team, don't get clever, but build a good basic cage.

      9) Cages should not only protect the driver, they should greatly stiffen the chassis. Whenever possible, tie the cage to suspension mounting points or over them or whatever is possible. You DON'T want the mounting points moving.

      10) Whenver possible, tie into bends with another tube (properly triangulated of course). This supports the bend (a weak point) and often, if done right, will crate more triangulation.

      Gosh, I could go on, but this is a good start. Now start looking at cages and try to evaluate what is good or bad in the design. Look for what you would do differently and why. Make sure the reasons you would do it differently are good ones.

      I'm sure (I hope) others will chime in. This could become a primer for designing a good basic cage.

      Another thing, if you do decide to build a cage yourselfI have three pieces of advice:

      1) Have a lot of patience. If you're prone to slap things together to get them done, building a cage is not for you. You MUST be honest with yourself here. In my case I had the patience to walk away when I got tired of frustrated. It was most important for me that it be right.

      2) Be demanding. Don't accept good enough unless it well and truly is. Most people accept good enough because they just want something finished. You'd better enjoy the process and the pride of accomplishment of doing something RIGHT.

      Note: if you can't do the two above, have someone build your cage, even if you take them a design to follow, find someone who will execute properly.

      3) If you are not a really good welder, get someone who is to do your welds. I weld like a blind monkey, so I had someone talented weld my cage. I just tacked the tubes in place until he could do a proper welding job on them.


      -------------------------------
      Arek Wojciechowski
      8legs Racing, SPC #77
      http://www.8legsRacing.com

      Comment


      • #18
        Re: welding a cage

        Originally posted by MazdaMatt
        You say that I should be using points, which makes perfect sense... Should I be mating braces (for example the ones that cross from the front loop to the main loop along the sides of the roof) right on the bend? or "as close as possible" (as the regs suggest)?
        As a general rule, try to have the main hoop diagonal, drivers side rear brace, roof diagonal and main hoop to front hoop tube all join into one point although it is often not possible in a smaller car as you also need to consider space for the driver.

        Originally posted by MazdaMatt
        It is suggested that all connections to sheet metal be made on verticle sheet. The diagrams in the CASC book show all six mounting plates on the floor... Is this incorrect? Would it be better to 'widen' the angle of the legs so that I must cut the tube at an angle for an eliptical intersection to plates on the walls? Wouldn't more strength be found on the floor near the rails?
        The main hoop mounting plates in most of my cages have been welded to the floor but whenever possible I extend additional plate up the vertical wall of the car a couple inches and weld it to both the car and the main mounting plate for maximum strength.
        Jeff Graves

        Ongoing Miata project

        Comment


        • #19
          Re: welding a cage

          Originally posted by MazdaMatt
          Well it depends on what I can get over the winter. I am going to do a project car with my father, but he's oldschool and doesn't think anything with four cylinders can go fast.

          I've been looking at 89-91 bmw 325i or something similar.

          If left up to me, i'd get a miata or 240sx.

          I will certainly be putting more hours of research into this than hours of building. I'm sure that before the season is out, i will have checked out just about every cage I can get near.
          The Miata is a great car for track days, Solo 1 and Solo 2 but the interior is very tight for a cage. I've "caged" (no door bars yet) my roommates Miata and be doing my car this winter if I can find time between customer cars. Take your time and pay attention to the driving position and driver clearances with a helmet.

          I've heard Adam may be partial to the 240
          Jeff Graves

          Ongoing Miata project

          Comment


          • #20
            Re: welding a cage

            check out the CARS rule book for cage disigns that are ally legal.
            I honestly think the rally regs are far more stringent then the race regs
            http://209.250.151.132/CARSRally/

            take a look at 4 stars web site and take a look at the picture of the Sprongls car at STPR. That was a very very heavy impact directly on the pass door and it took over 30 minutes to extract dan Sprongl
            http://www.fourstarmotorsports.com/C..._STPR__250.htm

            If you are going to build a cage put some door bars in place especially if you gut the doors

            Last year at Targa we had a very very heavy hit on the passenger door and the door bars probably saved Ryan from being more seveerly injured then he was.

            You have the right idea, do your research and build it strong. This is one area you do not want to cut corners. I speak from experience ( 1 rollover, 1 end over end and 1 sideways into a tree)

            Regards
            Tony Kloosterman
            www.neverquitracing.com
            Co-Founder of http://spda-online.ca
            Subaru Performance Drivers Association
            neverquitracing.com

            2007 Targa Newfoundland 1st Churchhill Award ( Nations Cup)
            2007 Targa Newfoundland 2nd Open Class
            2005 Spirit of Targa Award Winner
            2005 Ontario Champion ESP
            2004 Canadian Champion ESP
            2004 Ontario Champion ESP

            Comment


            • #21
              Re: welding a cage

              Originally posted by ADAM
              yeah if you are building a 1 off cage then $300 is a bit low, I buy the DOM tubing in 500ft sections so I get a better price..


              9-11 hrs with 2 people should be more than enough time if they are both working.. 1 welding 1 cutting and bending..for the basic 6 point..

              now if you start to go tube crazy like Arek's cage thats a different story
              2 people is a different story. For a minute I thought you were a cage building machine, able to replace a CNC bender and robotic welder without breaking a sweat.

              Although I did Arek's (current) cage as 2 stages, I think there is about 150' of tubing and 75 hours in it. He keeps asking when I'll have time to do stage 3. Once that's done, he'll be able to drive the cage around without the car.
              Jeff Graves

              Ongoing Miata project

              Comment


              • #22
                Re: welding a cage

                Originally posted by JGraves
                2 people is a different story. For a minute I thought you were a cage building machine, able to replace a CNC bender and robotic welder without breaking a sweat.

                Although I did Arek's (current) cage as 2 stages, I think there is about 150' of tubing and 75 hours in it. He keeps asking when I'll have time to do stage 3. Once that's done, he'll be able to drive the cage around without the car.
                Lol ... yeah, the car is just too light right now ... needs more tubes!
                Arek Wojciechowski
                8legs Racing, SPC #77
                http://www.8legsRacing.com

                Comment


                • #23
                  Re: welding a cage

                  With enough tubes you'll be able to run T1 with that bimmer.

                  Thanks for the good snippets, links and advice.

                  Does anyone know of a 3D modeling program that has a tube steel module built in? I talked to the drafters here because I knew that SolidWorks has a sheetmetal module, but unfortunately no tube steel. Even better would be a frame force distribution simulator program so i can see how much force the design could take.

                  Comment


                  • #24
                    Re: welding a cage

                    Originally posted by MazdaMatt
                    With enough tubes you'll be able to run T1 with that bimmer.

                    Thanks for the good snippets, links and advice.

                    Does anyone know of a 3D modeling program that has a tube steel module built in? I talked to the drafters here because I knew that SolidWorks has a sheetmetal module, but unfortunately no tube steel. Even better would be a frame force distribution simulator program so i can see how much force the design could take.
                    Unfortunately, the motor is too big for T classes (2.8L or less), and besides ... the M3 for some reason is automatically in GT ... even the e30 M3, which only had a 2.3L motor You think there is a bias at work there?

                    I didn't use a modelling program, but if you do get one going, I'd love to put my cage design into it, just to see how effective the planning on my cage was. Something like that sounds like it would be pretty complicated to get up and running though (I played a bit with 3d software years ago).
                    Arek Wojciechowski
                    8legs Racing, SPC #77
                    http://www.8legsRacing.com

                    Comment


                    • #25
                      Re: welding a cage

                      Originally posted by ScotcH
                      Unfortunately, the motor is too big for T classes (2.8L or less), and besides ... the M3 for some reason is automatically in GT ... even the e30 M3, which only had a 2.3L motor You think there is a bias at work there? .
                      I think he was referring to Solo Sprints T1.
                      -=Brian=-

                      Comment


                      • #26
                        Re: welding a cage

                        Originally posted by MazdaMatt
                        Does anyone know of a 3D modeling program that has a tube steel module built in? I talked to the drafters here because I knew that SolidWorks has a sheetmetal module, but unfortunately no tube steel. Even better would be a frame force distribution simulator program so i can see how much force the design could take.
                        I use SolidEdge at my day job and have used CADKey and Ideas in the past. I've never used SolidWorks but I can't imagine it would be very difficult to model a cage in it. Just protrude a pair of circles along a path.

                        I use a basic version of Algor (linear material and static loads) for FEA. It won't tell you much about crash survivability but it will predict stiffness, static strength and tell you something about load paths.
                        Jeff Graves

                        Ongoing Miata project

                        Comment


                        • #27
                          Re: welding a cage

                          Originally posted by MazdaMatt
                          Does anyone know of a 3D modeling program that has a tube steel module built in? I talked to the drafters here because I knew that SolidWorks has a sheetmetal module, but unfortunately no tube steel. Even better would be a frame force distribution simulator program so i can see how much force the design could take.
                          I have exactly zero experience on this topic... but if you're up for a VERY lengthy but interesting read, here is a thread from a Porsche forum from a while back which discusses in particular the effectiveness of different door bar designs:
                          http://forums.rennlist.com/rennforum...d.php?t=259343

                          A few pages in, an areospace engineer gets into some basic, linear FEA modelling of different designs to compare the deformation and load distribution characteristics. Interesting stuff, but from the discussions that happen towards the end of the thread, I gather that getting any meaningful information from such basic modelling of force distribution (particularily when you are not modelling the rest of the car) is very limited due to the many complexities of simulating a crash... unless you have solid engineering knowledge, tons of free time on your hands, and a supercomputer at your disposal...

                          As an added benefit though, there are numerous photos of various cage designs littering the thread from both the professional motorsports level and the common club racer level to get some ideas from.
                          Erik Z.
                          1987 Porsche 944 Turbo (SoloSprint #944 GT1)
                          1998 BMW 328is

                          Comment


                          • #28
                            Re: welding a cage

                            shuttles still crash

                            I am not sure there is any data you will gleam from getting overly complex using computer modeling...

                            even if you stop the ingress of foreign metal from penetrating the driver, at some point the G's will be to high and the driver will die regardless if the cage has protected the driver from ingress of metal..

                            its such a dynamic thing..
                            Former CCC member, and shit disturber

                            winner of a bunch of different classes over many years

                            breaker of parts, burner of engines

                            CREATOR OF "ULTIMATE DOOM 570SX"

                            Comment


                            • #29
                              Re: welding a cage

                              neat Porsche thread..they really went nuts with what ifs
                              Former CCC member, and shit disturber

                              winner of a bunch of different classes over many years

                              breaker of parts, burner of engines

                              CREATOR OF "ULTIMATE DOOM 570SX"

                              Comment


                              • #30
                                Re: welding a cage

                                Well i wouldn't expect to be able to simulate crash conditions. I just wanted to use a 3d modeling program for design purposes, and then maybe a stress test program to see if its a pathetic design or reasonable good.

                                Yeah, i could use solidworks to extrude bars manually, then attach gussets manually, but i wanted to do it like the sheetmetal module that lets you pick a shape and it designs the sheet shape and where the bends and notches go. If i could draw simple lines into the form of a cage with the measurements of a car and have it poof into a design includeing gussets and everything, i'd be sailing.

                                I know that I have seen my school's Forumla SAE chassis in a force distro program. They were checking for particularly vulnerable spots (one bar that turns red when the others around it are unstressed as opposed to forces being properly distributed). I've seen the same program in use for a suspension bridge design. Its all final-year or masters mechanical engineering stuff.

                                This is all speculation, though. I'll probably end up with a set of paper drawings and a fair bit of during-build inginuity.

                                Comment

                                Working...
                                X