Strut Bar Benefit Testing Before and After

[tperazzo]

Originally Posted by aoncurly

Not trying to be critical, but I see a couple of questionable elements to your setup and test. First, I question why you cut and instrumented the top brace in the middle of the brace? By so doing, you create a "pin" in the middle. In tension, this is not an issue, but in compression it would significantly weaken the strut/brace. For the loads you are seeing, I suspect you are nowhere near the buckling force (again, in compression) of the brace, so you would not measure much difference. At more extreme loads, the strut or brace would deflect under compression, creating a weak system, and in theory, a collapse mode. A better (but not ideal) solution would have been to move the load cell to the end of the brace, where there is already a "pin".
Secondly, the addition of a diagonal for the loads you showed earlier in your video will not see much if any load. If the top strut/brace is in tension or compression, the diagonal sees no load. Only if there is a vertical load at the strut tower applied to the system (acting perpendicular to the top strut/brace) will the diagonal see any load. Also, if the diagonal sees any vertical load, it all ends up at the lower diagonal attachment point. If you are designing for some vertical applied load at the strut tower, you should have a diagonal in each direction or cross.
So, in summary, for the loads you have described in your video, the addition of the diagonal has no or little effect on the torsional rigidity. If the top strut/brace is significantly loaded, the strut/brace would collapse under compression.
I offer this commentary for your consideration.

Hi and thanks for you comments. I appreciate any constructive criticism and will gladly debate on technical grounds.

Regarding the buckling: I agree with your placement of the sensor recommendation that relocating on the side is a better choice for high loads. But as you say 70 lbs is no where near the buckling strength of the bar/sensor setup. So, if I did this test again with your consideration, I would get the exact same results.

Regarding the diagonal bar forces, I'm not clear why you think there is no load in the diagonal when the chassis is twisted. When the diagonal bar was connected the deflection dropped by 9%. The top bar did have some load as I showed on my computer screen, but I think the results speak for themselves: The diagonal bar improves torsional rigidity and a second diagonal bar or X may further improve it.

Your vertical load scenario in the strut only comes into play when I dropped the car on the right front tire. The strut doesn't have any springs in it and only transfers vertical load through the damper piston/fluid design.

I wish I had second load cell because several questions about that have come up. However, I feel like I've learned enough about this that I can make an informed decision on how to proceed.

I will size the top bar to be fairly lightweight but will avoid buckling and I will add a second diagonal bar in an X configuration.

Cheers,
Tom

[aoncurly]

The diagonals, much like building diagonals, only see load due to forces perpendicular to the strut/brace. (A building lateral system is your system, rotated 90-degrees, and diagonals are for lateral forces such as wind on a building.) I would agree that the addition of diagonals would increase torsional rigidity, but my comments were in reference to loads you described earlier in your video (you only showed loads in one direction). I think there is differential vertical displacement of the strut towers as the car flexes, thus creating a vertical load component at the strut towers. The actual load path is probably very circuitous so I think it's a bit more complicated than you originally described, but your point is well-taken: the braces will aid in torsional rigidity.

[tperazzo]

Originally Posted by aoncurly

The diagonals, much like building diagonals, only see load due to forces perpendicular to the strut/brace. (A building lateral system is your system, rotated 90-degrees, and diagonals are for lateral forces such as wind on a building.) I would agree that the addition of diagonals would increase torsional rigidity, but my comments were in reference to loads you described earlier in your video (you only showed loads in one direction). I think there is differential vertical displacement of the strut towers as the car flexes, thus creating a vertical load component at the strut towers. The actual load path is probably very circuitous so I think it's a bit more complicated than you originally described, but your point is well-taken: the braces will aid in torsional rigidity.

Yep, I thought in your first message you said that the diagonal bar doesn't contribute to torsional rigidity.

In terms of vertical loading due to the deflected shape isn't a big load compared to the reaction forces from the the strut. If you look at the cross section of the cars construction through the shock towers you see will this: Under the horizontal line is the deflected shape. Notice the double shear boxes.

Now even if the shock towers displace up or down, the strut does little to constrain it.

The same deflected shape is typical during torsional stress.

[Walt Fricke]

Curley
If one looks at the load paths, doesn't the top of the strut take very little force in a horizontal direction compared with down at the A arm? This is why I have always been suspicious of getting much benefit from strut bars. Of course, for a full on race car, you want even the small benefits. And there doesn't seem to be any significant disadvantage to go along with a strut bar.

A straight across bar has to benefit from a fixed diagonal in the vertical plane, so the forces on one side aren't just transferred to the other side, doesn't it? Again, maybe not much. An X or equivalent would be better (I did that on my track only car in a class with few restrictions), but as long as the loads are transferred to a stronger point, maybe a single fixed diagonal is enough.

It may be that the relatively low forces involved mean that the hinge pin effect of the strain gauge attachment doesn't matter?

Tperazzo, if he knows more about force paths than I do (which is not hard), reinforces my take about the front suspension mounts. The leverages with respect to those versus the forces at the axle are even more favorable to their not needing to be reinforced. Braking pushes the A arm backward. Vertical forces up front have to be inconsequential. If braking can try to pull the front mounts apart, well sheet metal does just fine resisting that, and the flat tubular sheet metal structure is plenty strong in compression as well.

I like the theories about the factory X brace. A race shop owner told me he favors it because it stiffens the front in case of collision, leading to somewhat less expensive repairs.

I have the strut brace on my SC cranked in as far as the hood/fender allows, just short of rubbing on one side. I wish I could get more negative camber.

If class rules allowed, I would widen the A arm in a flash. Easy enough to do if you can weld, and plenty strong. Because any shop could do this, there wouldn't really be a market for it, would there?

[tperazzo]

I agree with you Walt regarding everything you said about the upper shock tower loading and the front suspension pan loads.

Even though the X brace over the spare tire area may help save the car in a crash, I've also been warned that the X brace could be dangerous in a crash if its welded in the center and the car doesn't have a full roll cage. If I remember correctly the concern is that the bar could get pushed into the drivers area through a gauge hole. I hope that didn't happen to someone. Yikes.

[1979-930]

Well that settles it.
I’m ditching the STB so my Golf Clubs fit in the frunk again.


Sent from my iPhone using Tapatalk

[safe]

All I know about a "normal" strut brace is that I have never felt any difference with or without them. Pulling the body together for camber is another thing.

[aoncurly]

My comments were directed at the loads tperrazo described in his video, showing only horizontal loads at the shock towers. My belief is that the very nature of the shock towers (to stabilize the shocks) are loaded both vertically and horizontally. The shocks definitely see a vertical load component to bumps as the strut compresses and extends. The shocks are not vertical - they are cambered so there is also a horizontal component loading as well. Ideally, a bulkhead across the trunk would make that area torsionally very rigid. However, the impracticalities of this in a trunk negates this - no golf clubs! I am also not arguing that a strut and diagonal are necessary (unless you regularly track or race) - let's not forget, Porsche has the best engineers and they analyze the structure for loads that these cars experience and design accordingly. I would think that only in hard cornering while tracking and racing would any real benefits be felt. But I would agree that adding a strut/brace and diagonal braces would enhance torsional rigidity keeping the geometry of the struts and wheels as the designers intended.

[911pcars]

Good video to dispel the dreams of road race stardom.

My low buck, DIY 10% mod performed on a Weltmeister strut bar a few years ago:





I'm probably 10% poorer, but it's shiny and I'm still delusional.

I agree. There are probably better methods of increasing negative camber without using brute force.

Sherwood

[tperazzo]

Originally Posted by 911pcars

I agree. There are probably better methods of increasing negative camber without using brute force.

Sherwood

Its funny to me how many guys say a Cellette bench is needed for structural work and the tubs suspension points must be within tolerance and then other guys clamp the structure for more camber.

I'm not judging either way, its just that there is a wide range of thought on these things.


Nice work on your modified brace Sherwood!

[tperazzo]

Originally Posted by aoncurly

My comments were directed at the loads tperrazo described in his video, showing only horizontal loads at the shock towers. My belief is that the very nature of the shock towers (to stabilize the shocks) are loaded both vertically and horizontally. The shocks definitely see a vertical load component to bumps as the strut compresses and extends. The shocks are not vertical - they are cambered so there is also a horizontal component loading as well. Ideally, a bulkhead across the trunk would make that area torsionally very rigid. However, the impracticalities of this in a trunk negates this - no golf clubs! I am also not arguing that a strut and diagonal are necessary (unless you regularly track or race) - let's not forget, Porsche has the best engineers and they analyze the structure for loads that these cars experience and design accordingly. I would think that only in hard cornering while tracking and racing would any real benefits be felt. But I would agree that adding a strut/brace and diagonal braces would enhance torsional rigidity keeping the geometry of the struts and wheels as the designers intended.

Absolutely agree that there are forces in line with the struts when hitting bumps. That is why I measured the force in the strut bar when the right front strut was quickly cycled by dropping the car on a block. In fact, that is where I measured a peak force of 70 lbs in the horizontal strut bar (no diagonal connected).

The lateral strut tower forces were described when purely cornering. You are correct in the real world there are bumps in corners, however for the sake of experiment design these components were isolated into three different experiments as shown in the video.

[KTL]

Its funny to me how many guys say a Cellette bench is needed for structural work

A friend here on the forums had a minor front end accident with his 911. He bolted the chassis to a warehouse floor with some plate steel and then pulled it back into shape with a forklift. He said, "Who needs a Celette when you have a Crown?"

[tperazzo]

Originally Posted by KTL

A friend here on the forums had a minor front end accident with his 911. He bolted the chassis to a warehouse floor with some plate steel and then pulled it back into shape with a forklift. He said, "Who needs a Celette when you have a Crown?"

Love it! Although he should have used a Toyota forklift since Toyota apparently use German engines now...

[VWLoosenuts]

Great video and work, thanks. I just wish you did this testing before the seem welding was done. I believe results may be slightly different if you had?

[Walt Fricke]

Jeremy - this may be a good point, but as I recall the results are consistent with what Burgermeister did? And his car wasn't seam welded (though neither was the wall of his garage).

Once this car is ready to drive, a day at a DE - or even an autocross - with the strain gauge bar in place and hooked up to a data recorder, should be even more informative.

Long ago - perhaps before the PC? - the makers of a strut brace (Automotion) instrumented a car with dial indicators with a telltale. Their conclusion was that the change in location of the shock towers was outward. Based on this, the brace they sold was just a bar taking up tension forces.

A competitor (Stable Energies), who made (and makes) a strut bar system which has a diagonal (and is carefully designed to fit rules which prohibit welding in or otherwise messing with the sheet metal to create attachments)disputed this, but did not offer data.

With the introduction of the Boxster tub, there is no need to do any of this, at least in the side to side direction, as the tub comes up pretty much to the top of the tower.

[tperazzo]

Originally Posted by VWLoosenuts

Great video and work, thanks. I just wish you did this testing before the seem welding was done. I believe results may be slightly different if you had?

Thank you and I hear ya. The reason I did the seam welding first is because there are seams under the strut tower brace.


Seam welding isn't a magic silver bullet to be honest. I think the seam welding and strut bracing yield comparable results (18% and 10% respectively in torsion).

Believe me, the car is still plenty flexible despite the seam welding. If it was a full race cage tied into the suspension, then I would say, "Yeah, not fair!"

[tperazzo]

Originally Posted by Walt Fricke

Long ago - perhaps before the PC? - the makers of a strut brace (Automotion) instrumented a car with dial indicators with a telltale. Their conclusion was that the change in location of the shock towers was outward. Based on this, the brace they sold was just a bar taking up tension forces.

Hi Walt,
That is consistent with what I measured when dragging my car across the garage floor. The strut bar was in tension meaning at least one tower was trying to move outward. I only got 5-10 lbs in the bar with this test, but I hope to try again by driving the car on a track.
Thank you,
Tom

[ClickClickBoom]

Wouldn’t the load path be from the axle to the a-arm mounts, with any multiplication factoring coming from the distance the axle is relative to the a-arm pick up points? The strut towers are for shock absorption mounting and some small lateral deflections.
Side note, I have the x brace aluminum mounted, and while driving in the rain the windshield gasket woul weep, took it out and dry as a bone. Putting it back in now that the fuse panel update is finished, will go slip sliding in the rain to test.
Moving stiffness around means forces are focused in a way that the factory never engineered for.

[tperazzo]

Originally Posted by ClickClickBoom

Wouldn’t the load path be from the axle to the a-arm mounts, with any multiplication factoring coming from the distance the axle is relative to the a-arm pick up points? The strut towers are for shock absorption mounting and some small lateral deflections.
Side note, I have the x brace aluminum mounted, and while driving in the rain the windshield gasket woul weep, took it out and dry as a bone. Putting it back in now that the fuse panel update is finished, will go slip sliding in the rain to test.
Moving stiffness around means forces are focused in a way that the factory never engineered for.

Yes most of the load is directed to the heavy cross member due its proximity to the front axle. However, since there is a large tire mounted to the spindle the friction force from the road tries to fold the tire under the car. The only thing that prevents this is the shock tower attachment.

As far as the up and down loads go, the strut tower is very strong in this direction because the metal is in tension instead of bending as with the lateral forces.

Thanks for sharing that experience with your windshield! Wow, that's pretty crazy. Did your X brace have heim joints on it to tension the tub?

Thanks
Tom

[ClickClickBoom]

Originally Posted by tperazzo

Yes most of the load is directed to the heavy cross member due its proximity to the front axle. However, since there is a large tire mounted to the spindle the friction force from the road tries to fold the tire under the car. The only thing that prevents this is the shock tower attachment.

As far as the up and down loads go, the strut tower is very strong in this direction because the metal is in tension instead of bending as with the lateral forces.

Thanks for sharing that experience with your windshield! Wow, that's pretty crazy. Did your X brace have heim joints on it to tension the tub?

Thanks
Tom

This is the unit I have installed. D-Zug brace install
Prior efforts were modified Weltmeister units. I have 2 sets of Weltmeister brackets that were torn apart while using rubber bushings. There is plenty of force being transferred while driving. I guess my point was/is there many forces present and moving them from one point will just transfer the forces to the next weakest point in the load path. Everything from the front suspension to the rear, is in the load path, and I would imagine anything outside those points could be contributing to stiffness via boxing/triangulation.