Register now to get rid of these ads!

Technical Boxing a frame?

Discussion in 'The Hokey Ass Message Board' started by Hemi Joel, Apr 10, 2022.

  1. TRENDZ
    Joined: Oct 16, 2018
    Posts: 386

    TRENDZ

    I do a fair amount of structural welding/repair. Have had to make the best of using what was available to me. Im in the process of building a new work space/ garage. I decided to put some I beams in the floor for just this sort of thing. You never know when you might need it. 6” beams with 3/8 web. 16 ft long flush with the floor. Rebared into 10 inches of concrete. C67DFF55-4316-4B67-A34F-A20FF56E6352.jpeg A84C2BEA-1F17-47C4-A962-BE3BAF2FEEBB.jpeg 8658D109-848D-40C1-98B1-3BB9EB041D34.jpeg
     

    Attached Files:

  2. Was discussing this today. I’ve seen rail road track set up like this.
    I’ve got 18-20 floor pots for hooking chains for frame repair.
    Used to do lots of chassis repairs.
    We were talking about having a flush mounted structure to mount structural steel for chassis building or frame repair. Then store the chassis stands out of the way when not in use.
    Something like this but flush with the floor
    B1E98394-831E-4215-A3D1-156EE65718CD.jpeg D10516B5-423F-4CC8-9A60-76F43CB9B23C.jpeg
     
    Last edited: Aug 30, 2022
    Rickybop likes this.
  3. BFXJason
    Joined: Jan 2, 2018
    Posts: 87

    BFXJason

    Here was my solution to boxing....and I could still get to all the mounting holes. IMG_6293.JPG
     
    Budget36, Ned Ludd and AGELE55 like this.
  4. SS327
    Joined: Sep 11, 2017
    Posts: 2,463

    SS327

    I tried to box a frame once for a chevelle. It kicked my ass, knocked me out cold and I had a concussion. I’m more careful these days.
     
    Budget36, Ned Ludd and -Brent- like this.
  5. jimpopper
    Joined: Feb 3, 2013
    Posts: 321

    jimpopper
    Member

    I just finished boxing a replica model A frame. I boxed it from the front crossmember to the rear crossmember (Buggy Sprung). The rear ladder bars and heavy centered crossmember will end up moving a lot of resistance to the middle section of the frame. Since I was only doing one, no jig, controlled heat and it has a very slight twist. It is probably with in factory tolerances and a little pressure will level things out. Once the weight is back on it and depending what the body does, I doubt I'll lose any sleep over it.
     
    AGELE55 likes this.
  6. gimpyshotrods
    Joined: May 20, 2009
    Posts: 23,264

    gimpyshotrods
    ALLIANCE MEMBER

    That's unlikely to stop twist.
     
    clem likes this.
  7. BFXJason
    Joined: Jan 2, 2018
    Posts: 87

    BFXJason

    Possibly - I should probably rephrase. Not necessarily the solution to boxing, but an improvement over stock. There seems to be a debate whether it is needed or not? I'm only in the 375 - 400 HP range with an automatic transmission, so decided to go this route. Figured it would not hurt?
     
    ekimneirbo and AGELE55 like this.
  8. Hemi Joel
    Joined: May 4, 2007
    Posts: 1,528

    Hemi Joel
    Member
    from Minnesota

    That is very interesting, I've never seen it done that way. Rather than boxing, you are converting the inner part of the frame rail to a truss. Trusses are known to be pretty strong.
     
    SS327 likes this.
  9. NoelC
    Joined: Mar 21, 2018
    Posts: 668

    NoelC
    Member

    Yea. I agree with some of that!o_O
     
  10. gimpyshotrods
    Joined: May 20, 2009
    Posts: 23,264

    gimpyshotrods
    ALLIANCE MEMBER

    Trusses can hold an immense load, vertically.

    Take an entire house worth of trusses, lay them on their side, and leave 1/2 of the length hanging off of a flatbed, and they will curve to touch the earth.

    Trusses twist just fine.
     
  11. twenty8
    Joined: Apr 8, 2021
    Posts: 2,318

    twenty8
    Member

    What you have done adds strength to the frame rail mostly in the vertical plane. It will have very little benefit in resisting any twist of the rail (torsional load). Early frames (post model T) are already reasonably strong in a load carrying capacity, but respond well with a little help, such as boxing. Twist is another matter. Early unboxed frames twist far too much for what we expect them to do in a hot rod situation. We add heaps more power (torque twist), and we hang our modern suspension expectations off them. Boxing is a real improvement in this respect, and improves twist resistance immensely. Your truss design is not anywhere near as effective for twist as boxing.
     
    Last edited: Sep 9, 2022
  12. Ned Ludd
    Joined: May 15, 2009
    Posts: 5,025

    Ned Ludd
    Member

    Flat plate, as used for boxing rails, also twists just fine — on its own.

    It isn't the additional torsional stiffness of a second flat plate which makes a boxed rail torsionally stiffer, but that the rails become closed boxes whose walls are stressed membranes which counteract torsional loads by their resistance to lozenging. A truss derives its beam strength from each bay's resistance to lozenging (whether the arrangement of bays is obvious or not) through compression or tension of the diagonal member crossing the bay. Stressed membranes and trusses are in principle structurally equivalent.

    The efficacy of a closed box in resisting torsion depends on the distance between opposite walls — or trusses, as the case may be. Being more about beam strength, early frames tended to be tall and narrow. When race car designers became aware of frame torsional effects (arguably as a result of lower roll centres associated with independent suspension) their first move was to shallower but wider rails, with circular-section rails becoming the norm in the years surrounding WWII. That went first to space frames, effectively much larger rails in which all the walls are replaced by trusses, allowing stuff to be packaged inside the closed boxes; and subsequently to monocoques, which went back to stressed membranes; and hence, with the addition of detailing to accommodate bending moments at the corners of openings, to the modern OEM unibody.

    @BFXJason's design has merit, especially as he appears to have taken care to place the truss's neutral axis in plan as close to the open side of the rail as he could. Smaller-diameter or oval tubing might be better, as that would enable an even greater distance from the vertical wall of the rail. I suspect that the tubing as arranged might be more resistant to lozenging than the alternative 3mm-odd-thick flat plate, so there might well be scope for reduction. Point loads on the rail flanges shouldn't be an issue, as alternate compressive and tensile inputs should cancel out.

    But as the aforementioned race car designers found, there is only so much you can do with a rail of reasonable dimensions anyway. Sooner or later you'll want more width and height — or smaller torsional loads, which is another story entirely.
     
    Hemi Joel likes this.
  13. ekimneirbo
    Joined: Apr 29, 2017
    Posts: 4,216

    ekimneirbo

    The thing here is that a lot of people are interested in one thing.......properly boxing a frame.

    Conversely, every one of the frames will be in a different state of condition and may even be warped or bent somewhat from 90 + years of use...........or brand new and in perfect shape.

    They will be attacked with every type of welder and people of varying degrees of proficiency in the use of those welders.

    Do we really expect that everyone is going to get the same result?

    Yep, some guys are gonna luck out and get it right with nothing to hold the frame rigidly......and some won't.

    Many years ago, as a project manager on a government weapons system, I was tasked with finding if Tig welding some pads on a flat aluminum part was viable and would produce a part that was still flat after being welded. It was about 8" wide and maybe 36" long and maybe .300 thick. Normally they were placed in a steel fixture to restrain them and the pads were furnace soldered in a heat treat oven. Basically the solder was put between the parts and heated till it melted and then cooled in the fixture.
    Concurrently our government station was being closed down and turned over to private industry. Many people were losing their jobs. There was a lot of people who were very unhappy ........

    The test part was sent to the weld shop and they were supposed to call me as soon as it was done. Well the welders in the shop were apparently among the most unhappy people. After several days and no call, I went to the weld shop..........and there sat my finished part. No semblence of flatness remained. All the pads were welded on one side. They had apparently "poured the coal" to it when welding and used no restraints.

    Instead of being even reasonably flat, it now resembled one of those trick water skis thats bowed on each end. What was originally flat now bowed about 2" on each end. I learned about the power of warpage from that. Granted it was an intentional strike back at being betrayed by elected officials, but the part was now junk. I took it up to the office of the top management official at the plant and laid it on his desk........and informed him that no more money would be available from my program. (Hey, I was losing my job too)

    The point here is that with so many variables in the frames being used as well as the equipment and skill level ...........and outright Luck, there is no way to say what result someone will get when welding long frame rails. Yes, its better to try and control the rails as well as the welds.....thats common sense.
     
    -Brent- likes this.
  14. gimpyshotrods
    Joined: May 20, 2009
    Posts: 23,264

    gimpyshotrods
    ALLIANCE MEMBER

    I would contend that as his diagonals are not in contact with each other, and this not welded to each other, that an area of focus stress will be produced in the rail flange in the gap between where they touch down. Cracking could originate there.

    Also, it appears that they are only welded on the very edges of the rail flange. While this might be an early in the process photo, physicality will prevent perimeter welding. This could result in tear-out.
     
    twenty8 and Ned Ludd like this.
  15. ekimneirbo
    Joined: Apr 29, 2017
    Posts: 4,216

    ekimneirbo

    An interesting solution. It should help I would think. If I was doing that, I think I would have used some box tubing. Larger 1 1/2" could butt against the side and be tacked there as well............or more and smaller pieces, say 1/2" square in a lattice design. Then road debris would be easy to rinse out. Good to see some innovative thinking. Not every solution has to be something thats the same as everyone elses solution.;)
     
  16. alanp561
    Joined: Oct 1, 2017
    Posts: 4,610

    alanp561
    ALLIANCE MEMBER

    Round tubing would be stronger. That's why they make roll cages out of it. The welds on the round tube that the fellow made the truss bracing from need to be welded all the way around. Leaving one side open leaves the weld susceptible to cracking as the frame works.
     
  17. gimpyshotrods
    Joined: May 20, 2009
    Posts: 23,264

    gimpyshotrods
    ALLIANCE MEMBER

    Yup.
     
  18. BFXJason
    Joined: Jan 2, 2018
    Posts: 87

    BFXJason

    Thanks for all the feedback.

    BTW - the picture shown above is in the preliminary stages. I was able to complete the welds around the tube.

    With all the debate over boxing or not, I just decided to go with "something" better than stock. It also allowed access to running board bolts, and fluid lines.

    I did however box and brace the front and rear portion. Missing is the extra bracing in the middle of the "X" brace. IMG_8593.JPG
     
  19. ekimneirbo
    Joined: Apr 29, 2017
    Posts: 4,216

    ekimneirbo

    Looks pretty good. Just as an aside, if you think about it, thats how the old tube and fabric airplanes were built and they were strong and light.:)
     
  20. nochop
    Joined: Nov 13, 2005
    Posts: 3,818

    nochop
    Member
    from norcal

    I’m in the same camp, @67….how long am I gonna be here anyway camp
     
  21. Ned Ludd
    Joined: May 15, 2009
    Posts: 5,025

    Ned Ludd
    Member

    Would it?

    It's one of those things everybody knows, but nobody bothers to verify. But looking at the actual data for a few sizes of tubing commonly required by rulebooks for roll cages, square tubes of similar linear weight and wall thickness are about 15% stronger in bending — much more so about the diagonal axis, by definition identical in tension, and pretty much the same as regards compression/buckling. The round tubes are about 10-15% stronger in torsion, though.

    Roll cages are typically partially triangulated portal frame structures, so beam strength is very much an issue. Theoretically, square tube would make for a stronger roll cage, if it were detailed correctly. I submit that the reason rulebooks require round tubing for cages is that tubing junction details for square tubing are relatively complex to get right and extremely easy to get wrong. In a scrutineering situation it would not be practical to analyse each and every junction detail. With round tubing there is less to look out for.

    While I'm pretending, hopefully with a modicum of conviction, to be a structural engineer, it is perhaps time to ask what we're hoping to achieve by boxing frame rails. I've been following the off-again, on-again debate around this ever since that notorious TV episode, and I don't recall once encountering any figures in all these years about how much torsional stiffness boxing a rail adds. I've mentioned above that it isn't a mere case of adding the torsional stiffness of another flat plate, so it must be more than double. How much torsionally stiffer is a boxed rail than an unboxed rail? As much as 10 times? More than 10 times?

    Try around 400 times, depending on the exact dimensions.

    The torsional stiffness of a pair of boxed Model A rails on their own should be in the order of 500 lb.ft/deg. Unboxed they are barely 1 lb.ft/deg. I couldn't believe it either. Now a pair of rails isn't a frame: the overall torsional stiffness will depend on crossmembers and X-members and K-members, and especially the details where those join to the rails. Stiff rails are no good if they can simply rotate relative to one another in the pitch plane. I wonder if a good conventional hot rod frame will achieve 400 lb.ft/degree.

    If we're only worried about engine torque multiplied by the lowest gear ratio, we are looking at the part of the frame between the engine mountings and where the rear suspension attaches to the frame (or in the case of IRS or a DeDion axle, where the final drive unit attaches to the frame.) The salient thing is that the conventional tripod mounting with an engine mounting either side and a single gearbox mounting in the middle doesn't make a lot of sense for our present purpose. Doing it the other way around, with two mountings far apart at the back of the gearbox and a single mounting under the crank pulley would mean that we'd only need to box the 30"-40" between the back of the gearbox and the rear crossmember. That — or a four-point mounting — is completely doable, given a crossmember bolted rigidly to the back of the gearbox, with a rubber mounting at each end. It'd be a lot of stiffness for very little weight.

    In fact some gearboxes of the '20s and '30s were designed with "wings" cast into their cases for that exact purpose. Another way is the way Porsche tied the transaxle to the engine in the 924/944/968/etc. and 928 by means of a torque tube.

    With modern unibodies, dealing with multiplied engine torque is the least of the designer's concerns. It is all about suspension tuning, given a commitment to a dynamic paradigm which takes ample torsional stiffness for granted. I won't elaborate on my analysis of how that came about, as it is complex and necessarily deals with political economy. The upshot is mass-produced unibodies that do 24000 lb.ft/degree or stiffer. As we'd be lucky to reach 2% of that it would be stupid to buy into that dynamic paradigm. Alternatives are possible.

    A word on X-members: the way to understand them is to consider the left and right halves of them as part of the respective frame rails. Imagine the tops and bottoms of the resulting triangles boxed with horizontal flat plate, so that that part of each rail becomes a wide boxed rail, roughly triangular in plan, and plenty stiff in torsion. Now, replace the upper and lower plates with trusses in the horizontal plane — easy as the triangulation is right there — and we're back to an X-member as we know it.
     
    ekimneirbo, rod1 and Elcohaulic like this.
  22. Elcohaulic
    Joined: Dec 27, 2017
    Posts: 2,212

    Elcohaulic

    Isn't square or octagon tubing just as strong. I noticed in the seventies; they started putting octagon holes into I beams instead of round into commercial garage floor supports. This was only in the five floor and above structures. It would take the old look from the car though. I know my 69 el Camino has fully boxed frame, I think all the el Caminos from 68 to 72 are like that.
     
  23. Elcohaulic
    Joined: Dec 27, 2017
    Posts: 2,212

    Elcohaulic

    I noticed the 55-57 Thunderbirds had an X along with a boxed frame.

    This photo is off the internet, I do not know anything about this car.
    [​IMG]
     
    rod1 likes this.
  24. Ned Ludd
    Joined: May 15, 2009
    Posts: 5,025

    Ned Ludd
    Member

    Material near the neutral axis of a beam isn't doing much. Provided the beam is acting as a beam, i.e. the upper fibres are in compression and the lower fibres are in tension, the middle material is doing little more than maintaining the vertical distance between the upper and lower fibres. Hence the thin web which is characteristic of the I profile. Hence also the perfect viability of holes in the web, to save weight in the case of a front axle, or to save material cost in the case of a floor joist.

    The holes are actually hexagonal. The beam is made by making a crenellated cut along the web of a smaller beam, moving one half over by a pitch, and welding the halves back together.
     
  25. bangngears
    Joined: Aug 30, 2007
    Posts: 1,145

    bangngears
    Member
    from ofallon mo

    Boxed my avatar frame from one end to the other with no jig. Took my time and jumped around welding about 2 inches at a time. Worked well
     
  26. Flathead Freddie
    Joined: May 9, 2021
    Posts: 806

    Flathead Freddie
    Member

    No
     
  27. ekimneirbo
    Joined: Apr 29, 2017
    Posts: 4,216

    ekimneirbo

    I've warped enuff stuff to know the perils of welding................:p
     
    Budget36 likes this.
  28. sunbeam
    Joined: Oct 22, 2010
    Posts: 6,213

    sunbeam
    Member

    Better take out a loan when you go to buy the steel I just bought some 3/8 plate $2.80 a pound.
     
  29. error404
    Joined: Dec 11, 2012
    Posts: 383

    error404
    Member
    from CA

    yeah... not looking forward to the next time I need to buy DOM :-\ Fortunately I have a small stock pile.... but I use it more sparingly now, haha
     

Share This Page

Register now to get rid of these ads!

Archive

Copyright © 1995-2021 The Jalopy Journal: Steal our stuff, we'll kick your teeth in. Terms of Service. Privacy Policy.

Atomic Industry
Forum software by XenForo™ ©2010-2014 XenForo Ltd.