Project Overkill: Chassis

[carl]

Back to the chatter about roll cages, just a reminder they won't work for street use. I bought an ITC racer with cage for street use and my head was about an inch from the bars that run from the top of the rear roll hoop to the front hoop....without a helmet on you would be setting yourself up for severe head trauma.

 

[Steve Hoelscher]

Back to the chatter about roll cages, just a reminder they won't work for street use. I bought an ITC racer with cage for street use and my head was about an inch from the bars that run from the top of the rear roll hoop to the front hoop....without a helmet on you would be setting yourself up for severe head trauma.

^^^ This.

Driving a car with a cage without a helmet is far more dangerous than driving a car without a cage. Even a roll bar can be dangerous if the bar is anywhere near your head.

Remember that in a racecar, a cage is part of a system. The cage provides the structural integrity. The seat contains the movement of the driver. The harness keeps the driver located in the seat. The helmet protects the head from potential impacts. The HANS device limits the helmet's motion, etc... etc.... Remove any one element and you compromise the entire system.

So my idea of putting tubes in the rocker panels would strengthen the roofless chassis without having to add a cage, or roll bar, to a street car.

 

[Dr.Jeff]

So you're saying to wear a helmet, Hans device, and full harness when going to buy a quart of milk.

But true about bars, especially with the X; having such a small cockpit and relatively low Targa hoop leaves little to no room for a roll bar/cage to be located inside, such that the roof could be retained. For a street car anyway.

Similarly (also for a street driven vehicle) you might consider the additional weight that roll bars, rocker channel bars, or any significant chassis stiffening structures adds to the car. With a small, low power vehicle like these the overall performance will likely decrease with the weight gain. Kind of counter productive. That's what I like about the changes being made here (with Darin's project); they are not adding a lot of heavy materials onto the existing chassis, but either replacing things with stronger designs or adding relatively small amounts at key points.

 

[Steve Hoelscher]

So you're saying to wear a helmet, Hans device, and full harness when going to buy a quart of milk.

But true about bars, especially with the X; having such a small cockpit and relatively low Targa hoop leaves little to no room for a roll bar/cage to be located inside, such that the roof could be retained. For a street car anyway.

Similarly (also for a street driven vehicle) you might consider the additional weight that roll bars, rocker channel bars, or any significant chassis stiffening structures adds to the car. With a small, low power vehicle like these the overall performance will likely decrease with the weight gain. Kind of counter productive.

Well... to be clear, a roll bar or cage is NOT a good idea in a street car because you don't have the rest of the system's components in place. The roll bar becomes a major safety issue without a seat, harness and helmet.

You are also correct about added weight. Chassis stiffening is of little value in a typical street car. Sure, stiffer is better but the performance gains from chassis stiffening are really only realized on the track. On the street the extra weight only slows the car.

 

[carl]

I think most street bars and bolt together cages for sedans is more a vanity item for street cars. An X certainly does not need a roll bar for the street whereas maybe for a spider or Miata a street bar might help in a roll over event which I think is probably rare. And just to clarify, I did have a roll bar in my Miata and most of my spiders.....I'm that vain.

 

[Dr.Jeff]

Steve, I was joking about the helmet, etc for street use. But very true how it all ties together to create a safety system for track use.

Carl, good point; in general any sedan roof or targa hoop is sufficiently safe for regular street use, while a true convertible is more at risk. Can't help but think of those old photos with early race cars that had open cockpits and no roll-over protection, suffering a crash.

I recall my BMW and Mercedes convertibles having roll hoops that were designed to pop up in the event of a roll over. There was a occasion or two when driving hard generated enough G-loads to deploy it in absence of a roll-over. Fun times.

 

[AKimball92]

I spent a lot of time looking at the front-top of the tunnel where it attaches to the bulkhead as probably the primary weak point of the chassis - certainly the center tunnel.

I sure wish I had a jig to test chassis deflection so I could see how much it helps (didn't I see a post here of someone doing just that?). If it helped a lot, it'd be a super easy mod to any exxe without taking much off: just the cover under the heater, pull back the carpet a bit bolt it on with rivnuts. Just need someone to make a kit! (not me).

I might be mistaken but I think you can check this or at least see a delta with the use of 4 scales and one being elevated. Weigh each of the 4 corners flat, then elevate one say 6 inches (approaching suspension travel limits, record height). With the stiffer chassis, the raised weight should be a higher value than the original flat value given all other variables are consistent. Now that changes are made it might be too late. If you are into racing though, access to the scales should be easy or on the list of tools to buy. For future mods, the changes can be tested, or with the plate removed and then added.

 

[Pete Whitstone]

then elevate one say 6 inches (approaching suspension travel limits, record height).

6 inches! I doubt too many cars have 6 inches of suspension compression available to them. I think the X has something closer to 4-5 inches total travel - 2-2.5 in extension and about the same in compression.

I think I understand the principle you are outlining, but either I'm mis-reading or you are over-estimating.

Pete

 

[darwoodious]

Here's a good list from 2 years back on some modern cars torsional rigidity: http://youwheel.com/home/2016/06/20/car-body-torsional-rigidity-a-comprehensive-list/

And here’s a link to what I’d build if I had more time:
https://peer.asee.org/development-o...nal-rigidity-of-a-formula-sae-space-frame.pdf

 

[carl]

What's truly needed for this forum (and many other old car forums) is one of those CAD programs that tests distortion of the chassis with mods via the computer rather than old school of putting the car on a torsion testing rig, but this would require someone imputing the data from an X into the computer. The more traditional DIY method is seeing if the door binds when jacking the car on one corner!

Similarly, access to a wind tunnel for aero aids would be nice but I imagine this is now done on computer simulations too.

Understand, that when I went to engineering school "the" computer was the entire ground floor of the engineering building and GWU actually had a small windtunnel which looked like something the Wright Brothers built.

 

[AKimball92]

6 inches! I doubt too many cars have 6 inches of suspension compression available to them. I think the X has something closer to 4-5 inches total travel - 2-2.5 in extension and about the same in compression.

I think I understand the principle you are outlining, but either I'm mis-reading or you are over-estimating.

Pete

Yes that 2 inches in compression on one corner should be enough. If you jack up the front right corner, a stiffer frame should apply more load to that corner than the front left i do believe, and probably more load to the rear left than the rear right. From this using simple sum of moments equals zero, calculate and overall torsional spring constant for the chassis as a whole. Think of a capital I (eye) with a wheel at each corner and the chassis torsional spring being the long vertical stem of the I.

 

[Dr.Jeff]

Understand, that when I went to engineering school "the" computer was the entire ground floor of the engineering building

We must have gone to college about the same time. As an undergrad our campus had one computer, also in a basement (filled the giant room), and it used punch cards. But no one was allowed to touch it except for a couple of grad students working on professor's projects. Even my first time through graduate school that campus only had one huge mainframe. But by then you could go to the library and pay a fee (by the hour) to have the specialist working in the library to do a data search on it for you. The results came back a few days later printed on the long scroll of paper with the little holes on the sides and perforated tear-off edges. By the time I went through graduate school again there were computer terminals available in the library connected to a mainframe. The last time I went through everyone had laptops on their desks connected by WiFi. Huge changes in a couple years time.

 

[dllubin]

We must have gone to college about the same time. As an undergrad our campus had one computer, also in a basement (filled the giant room), and it used punch cards. But no one was allowed to touch it except for a couple of grad students working on professor's projects. Even my first time through graduate school that campus only had one huge mainframe. But by then you could go to the library and pay a fee (by the hour) to have the specialist working in the library to do a data search on it for you. The results came back a few days later printed on the long scroll of paper with the little holes on the sides and perforated tear-off edges. By the time I went through graduate school again there were computer terminals available in the library connected to a mainframe. The last time I went through everyone had laptops on their desks connected by WiFi. Huge changes in a couple years time.

I used an IBM 360 for my undergraduate classes. You had to input your run with Hollerith (punch) cards and to do that you had to wait in line to use a punch card machine. Once you got the cards done, you would submit your deck and come back a few hours later to see if your job came out (on 133 column line printer paper). It was almost guaranteed that if it was an initial run there would be a typo somewhere or an error in the job control language that would prevent execution. I usually considered myself lucky if I could get a job to successfully run in less than a day what with all the wait time.

 

[Dr.Jeff]

you had to wait in line to use a punch card machine

My god you were lucky. You got to use the punch machine. Honestly I wasn't allowed anywhere near anything to do with the computers until my second round of graduate school. And even then it was indirect contact. I must have been a couple years before you, or your college was more advanced than mine (easy to do). Ah the good old days.

Sorry to have gotten so far off topic here Darin. Please get us back to your awesome build. I'm envious of what you are doing here.

 

[darwoodious]

Sorry to have gotten so far off topic here Darin. Please get us back to your awesome build. I'm envious of what you are doing here.

No worries... this is how conversations go - tangents abound if it is a good topic

Hint: next up on the chassis post will be related to those .065" SS tubes, how they're structural now, how I made them removable/serviceable and how I learned how to bend tubing!

Honestly, while I think the artifact I'll end up with at the end of this project will be super amazing, even more impressive will be the various skills I've learned along the way. I think that's pretty cool quite honestly. I've totally gotten on board of enjoying the journey while looking forward to the destination.

 

[Dr.Jeff]

That must be why my projects never seem to get finished, I love the journey so much.

Been looking forward to seeing the details on how you made the tubes structural, can't wait. Thanks.

 

[darwoodious]

As I've mentioned here before, the cooling tubes had holes and corrosion and clearly they'd had it for some time as the cover had been poorly cut to access and some crappy torch brazing done.

So I needed new tubes and a new cover. I didn't know about the SS tubes available here that others have used (hint: learn how to ask others first!) but my brother owns a tube bender, so I figured I'd just do it myself. I picked up a couple of 10 foot .065 walled SS tubes in proper 1 3/8" OD and a couple of shorter 5/8" OD for the heater tubes.

Also picked up some long 14g (.060) steel for the lower vertical spars of the tunnel and a flat sheet of similar guage SS for the cover. First step was to get the cut out and welded up. The important part here was to match the corresponding vertical tunnel in the cabin to continue the vertical part of the tunnel profile. As I'd need to attach the SS cover to that and make it removable, I needed to add some weld-nuts to the inside.

Next step was to build some brackets to keep the tubes in place. Since I'm not just holding the tubes from moving but attempting to make them structural, they needed to be quite strong. After some initial failed attempts, I settled on some 3/4" solid bar to span the 5" of tunnel. The upper part (mild steel) is welded to the vertical spars picking up any vertical load. The lower part will be welded to the SS tubes. The whole thing is sandwiched together with 3 load points (aft, mid and fore). Probably the dumbest part of this build is that I wanted to engineer the tubes to be removable and servicable - so the tunnel had to have a cover that was removable, the tubes had to be able to unbolt AND I wanted it structural. These are all in odds with each other, but not an insurmountable task. I did need to acquire a new tool tho...

Enter tiny little mini-mill from LittleMachineShop.com. Hey, I needed a new drill press anyways, so might as well go overboard. With this tool I drilled and tapped the mild steel together with the SS and with it bolted together, drilled out the areas for the 2 coolant tubes and the single heater tube that runs below. Here's the start of one. Don't think I took photos of them finished but not welded in.

(continues...)

 

[darwoodious]

So now I have the basis for the tubes to sit in. Once the project is all finished, the SS tubes will be TIG welded to the SS clamps.

Time to bend these tubes! I've never done any tube bending but as I mentioned, my younger brother had a pneumatic one and I borrowed it, tho I did have to get some dies for 1 3/8" and 5/8" (1 1/4" and 1 1/2" are the popular sizes). There's lots of instructions on YouTube on how to bend and those came in quite handy. The worst part is that there are 2 axis bends in the front part where the tubes exit the tunnel, so lots of measuring. Bending tubes for the X could be an entire thread on it's own and it takes a lot of time and you're bending some pretty expensive SS tubing! The heater tubes were easy tho.

I needed some forward attachment point to replace the welded straps and they also needed to be bolted so I can remove both tubes. Found some very strange "box section" metal that was perfect. TIG'd on some extra metal, threaded and fab'ed up an aluminum cap

(continues...)

 

[darwoodious]

Finally I needed to get the bends at the tail end of the tubes. These are really tight bends - far to tight for the tube bender since it needs like 4" of straight tail, so my best option was to bend up a the tubes to the angle I wanted, cut them and weld them back together to keep it compact. I also knew that since I'm doing a K20 swap which takes 1 1/4" coolant tubes, I might as well choke down just after the tunnel.

These need to be leak-free, so in addition to pressure testing which I did later, I picked up one of those cheap USB cameras you hook to your phone and investigated my welds on the backside. Looks good.

Now I needed to cover the whole lot. I picked up a piece of SS sheet and had a local metal fab shop bend it with their giant bender. This was drilled for the specific locations on the tunnel spars and fastened with some M6 SS bolts (yeah, went stainless crazy where I could).

Finally, to keep it relatively clean (and mostly dry) in the tunnel, I needed a way to cover the front. While the rear structural support already works to keep water debris out as it is at the very end of the tunnel, the front doesn't have that since the spar and tunnel cover moves forward past where the return line for heater drops thru the tunnel floor:

I fabricated a simple 3/8" aluminum clamp/cover that bolts over the exiting tubes and is perfect size to stay with the tubes and block the tunnel entrance. This is not structural but looks nice and complements the SS.

That front plate will get some kind of grommet to go around it to further keep the tunnel area clean of as much water as possible. The structural members or anywhere else on the tunnel cover won't get any kind of gasket treatment so any water that makes its way in the tunnel can flow out towards the back.


I did ALL of this because I wanted the chassis to be stronger. Tubes are by nature very very strong torsionally and since I've got two of them, why not make use of it. Any twist the tunnel wants to do will force the clamps to twist (due to the strong vertical spars) and since the tubes are welded to half of the clamps, it will transfer that torsional load forward to the next clamp (midway and vice-versa). It'd be even stronger to use steel tubes and just weld them in place, but then they wouldn't be removable.

This was fun, but not practical.

 

[Dr.Jeff]

Very cool. I am in complete agreement of your design, and I can appreciate the work it took to create it. It is pretty much as I imagined from your earlier comments, although I wasn't sure how you planned to tie the tubes to the tunnel rigidly and still be removable. Excellent execution, admire your work. Thanks for sharing the details.