Frame strengthing gussets

[Radowick]

  I was talking with a friend last week about my Atom's frame. He pointed out that although the frame welds look nice and consistant, they look rather "cold" like they didn't penitrate very deep. I am not a welder so I have to ask others here whether this is accurate?  Mine looks like everyone elses.
  We got talking about the frame braces that DP offers to brace up the engine box part of the frame. I told that I am thinking of welding in some corner gussets in the bottom of the "box" to strengthen the corners. It would be similar to the welded in plate for the rear tow hook at the left corner. He noted that concidering all the loads from the suspension and all the loads of the engine that is constanly twisting back and forth which are working to twist this "box" by loading up the corner joint welds, it would be a good idea. It would be simple to do and would spread some of those loads around to avoid twisting the "box", or worse. I think it can be done at least to the two right side corners without removing the engine, and would still leave the required room to do the oil draining work.   

[DarthChicken]

The welds on our cars, that you can see, are NOT the welds that hold the cars together.  The frames are welded, and then bronze welded (not sure of the exact term) over top of of that, stictly for looks. 

[Radowick]

 I questioned the welders in our tooling shop today (all 3 of them with over 15 years of welding aircraft tooling) about bronze welding of steel. They explained that the term "bronze" welding means the steel filler rod has been alloyed with a small amount of bronze. This makes welding mild steel very easy and also reduces warping in thin sheet metal. This makes it perfect for body panel repairs. But they also pointed out that because of the soft alloy in it, it is only half as strong as traditional steel filler rod so it should not be used for structrual/critical work.

[Wmcmanus]

Not a bad idea.  I'd be interested in see pics of whatever you might come up with if you decide to go this route.

[Heywood-Yablowme]

Bad idea. The Atom frame is a brick shithouse. Why add unnecessary weight??

[positron]

I have to agree with Bruce's standby on this one... there hasn't been a failure. Rewelding a weld is bad medicine. Maybe Calscot will chime in, if he's in Country.

[DarthChicken]

You misunderstand - the bronze is just for looks, there is real mig (tig?  I can never keep them straight) welding under the bronze.

[Radowick]

[quote="Positron"]
..... there hasn't been a failure.
[/quote]
Are you sure?
[quote="Positron"]
Rewelding a weld is bad medicine.
[/quote]
Adding more weld to a joint is very typical for welds. Multiple passes of additional welds if fine provided all flux and scale is removed between passes..............provided using the same filler rod material. Grinding out cracks and voids of a weld, then rewelding is everday welding work.
But I am NOT suggesting building up the existing welds. I thinking of adding a gusset similar to the gussets the rollbar rest on into the lower corners of the box. A few ounces at most of extra weight at most, then some paint.  
[quote="DarthChicken"]
You misunderstand - the bronze is just for looks, there is real mig (tig?  I can never keep them straight) welding under the bronze.
[/quote]
I understood what you said.   Maybe this mixing of filler rod material types is what Positron is referring to as "bad medicine".

[CalScot]

I will chime in a bit.

IMHO, much of what Spanky says is true. Much of what the welders say is also true but as is the case with many things in life, it's what's not said that's probably more important. The welds on the Atom are basically excellent and fit for what the car was designed for. I do have some expertise here so I can say this. I worked in the welding & inspection business for 20 years. I graduated from the British Welding Institute and not only held a senior UK welding inspectors cert but I was an AWS CWI, ASNT level III in Ultrasonics, Radiography and more. I've inspected Nuke reactors to critical flight hardware to major oil pipelines 700+ feet under the North Sea. If it's been welded, I've seen it before welding, during welding and been at the court cases when they have failed.

There have been failures in the welds around the engine box structure. At least on my car. Two in fact. However, I would NOT go welding gussets without some genuine understanding of what you are doing. Its a complicated business. Even a seasoned structural engineer with the best FEA expertise will get it wrong often. Bottom line is if you start just adding gussets you simply shift the dynamic stress and loads someplace else. Possibly creating a far more catastrophic failure than simply allowing an existing weld to develop a crack (which will take time to propagate and fail)

Mines cracked because of 475WHP (thats over 500BHP), big slicks and being driven on track like a race car and some really silly driving for many many hours. It was never designed for that. The fact it managed it for so long is a great testament to how good it is.  I would not mess with it. I repaired the last crack with a simple reweld and a reduction in power. Well a small reduction

[Radowick]

.....Even a seasoned structural engineer with the best FEA expertise will get it wrong often. Bottom line is if you start just adding gussets you simply shift the dynamic stress and loads someplace else.




I have worked directly with stress engineers who did the finite element anlaysis work on my designs. It allowed me to better understand their work to make sure the concentrated loads are minimized. Instead of letting the load concentrate in the corner joint weld, a gusset would (is placed properly) help spread the load out & away from the weld and down the long adjacent tubes itself, which would handle the flexing better then the weld joint.    

But like I said, I am thinking about this. And I have a lot more thinking to do before I decide if, what, & how I do it.  

[CalScot]

[quote="Spanky"]

Instead of letting the load concentrate in the corner joint weld, a gusset would (is placed properly) help spread the load out & away from the weld and down the long adjacent tubes itself, which would handle the flexing better then the weld joint.    

[/quote]
Very true, however, the load would not be dissipated completely along the tube length and would transfer extra load to the weld at the other end of that particular tube. You would then need to gusset that joint, which in turn will shift more load to the next joint. You might ened up with a engine box frame with lots of ugly little gussets on every joint

Clearly, there are joints that are more streesed than others and selective strengthening could spread out the loads better. If you can get your engineers to do a take off on the Atom frame and run a FEA study of it to identify the load paths, I'd love to see the results. It would be interesting to see if they predicted the joints that failed on mine because they are not the joints I would have picked to fail by looking at it and trying to rationalize (guess) where the max stress is.

[Radowick]

[quote="CalScot"]
Very true, however, the load would not be dissipated completely along the tube length and would transfer extra load to the weld at the other end of that particular tube. You would then need to gusset that joint, which in turn will shift more load to the next joint. You might ened up with a engine box frame with lots of ugly little gussets on every joint
[/quote]

Good point.

[quote="CalScot"]
Clearly, there are joints that are more streesed than others and selective strengthening could spread out the loads better. If you can get your engineers to do a take off on the Atom frame and run a FEA study of it to identify the load paths, I'd love to see the results. It would be interesting to see if they predicted the joints that failed on mine because they are not the joints I would have picked to fail by looking at it and trying to rationalize (guess) where the max stress is.
[/quote]

I will see what I can do with this idea. I think we all would be interested to see what the results would be.

[AtomMadStew]

I know its even more weight but what about manufacturing some bolt on gussets that clamp around the frame? Would that be possible? Triangle gussets with 4 C shaped adjustable clamps with some thin rubber inserts as not to scratch the frame. Just a thought. Better than welding in something and then having to paint it to realise it doesn't work or causes stresses to move elsewhere. At least you could remove and evaluate a clamp's shape from use and see if it distorts over time.

[dingo]

If you're going to use a clamping arrangement like the Palatov "W" brace, then you should not put any rubber between the chassis and the brace.
All that does is reduce its effectiveness, by still permitting movement and it will hold dirt and wear away paint / powdercoat as fast as if there is none.   
I can attest from seat of the pants that the Palatov W brace redistributes the stresses Those vibes turn up elsewhere, places they didn't use to.
I believe the brace restricts movement of the front lower cross member of the engine bay, making it feel more of a solid unit. Yes really. Not that it was wobbly beforehand.
It might be a figment of my imagination, I admit.  His solution can only be fitted to a Brammo chassis that has the under seat U frame. 

[Radowick]

 I have 1/3 of the frame 3-d modeled. I am deliberated how to have the stress engr. treat the engine as a stressed member or not. Since it is mounted using rubber isolation bushings, it is not truely hard mounted.

 One of our stress engineers explained to me a long time ago that when designing a structure, the more direct and straight you can make the load path, the better. The more directional changes the load has to make to get to the members being evaluated, the less the load will be getting to it. Instead, the members prior to it will be end up carrying most of the load.
  In our Atom's frame case, this means that when the car is cornering hard, the loads coming up thru the suspension and into the frame have to travel around the front side of the engine box thru the smaller tubes before the loads can make it to the large frame tubes on the opposite side of the car and get their greater stiffening benefits. There is just one frame member on the front side of the engine going side to side, and it is way down low. A lot of load is passed thru it down low and then back up on the other side of the engine thru the one horizontal tube, and after a number of corners before connecting to the bigger side tubes. Not very direct.
  My guess is the highest loads are in the 2 joints way low in front of the engine and near the lower suspension A-arm pivot points. That horizontal tube is getting little help from any other cross tubes.  And the left end of that one tube also has the task of supporting the engine as that is where one of the engine mounts is.  I think the frame behind the engine appears very good, but the front looks like it will sway side to side under lots of hard driving. Granted, the amount of sway is probably very small, but when talking about metal fatigue, that is how it fails, from repeated small back and forth movements.