Thanks for waiting almost 4 years for part 2! Wow time flies... I hope you enjoy these results as load cell measurements aren't easy to begin with and even harder in a moving car. Let me know what you've experienced in your car. Does a strut brace help or no?
I suspect you are correct that the bar does little to nothing on street tires at the speeds driven in your test. A better test would involve “sticky” tires at autocross or race speeds. My Porsche got very good at Watkins Glen race track but the modifications really compromised the car for street driving. They are really two very different environments and you have to decide which is most important to you. Nice video and nice car! Regards.
Yes, stickier tires coming. This test was done at autocross speeds. Watch this video, it's almost all in second gear ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-S19riKPfgs0.htmlsi=G78djcOHoLrMUJS6
I give credit to Porsche as well. They continued to improve the chassis. For example the 964/993 chassis are much better. If you watch my torsional rigidity video you will see how easy it is to deflect the chassis. They aren't that strong, but they are light and they found a combo that obviously worked well for Porsche.
I’m not surprised by the results. That car is suspended by the torsion bars which react the spring forces in the lower pan where they attach, not into the strut towers. As you point out the strut towers are seeing the damping loads which should be low until they bottom out on their bump stops. Like you also state, front engined, coil over strut cars may seeing higher loading in strut tower bars and be somewhat beneficial. However, w/o triangulation (diagonal/s) and w/pivots at the ends (zero moment reaction) I wonder just how much.
Tom this is why I love your content. You come at it purely from an engineering/data driven point of view and offer the data to make decisions. I wish more people did this. On this topic I added a strut bar to my 993 and noticed nothing... but it was more of a vanity part than anything.
Thanks for doing this. While I had no proof, I have assumed for a long time that a strut bar in the torsion bar era 911's would not really do much. Your video is the first that I've seen that can back up that assumption. I'm not aware of a counter argument with empirical evidence. Keep up the good work!
Thanks for checking it out. I'm likely to fabricate a light aluminum brace that sized to handle 100-200lbs loads. But like you said, I don't expect it transform the car.
Thank you Tom for another good video thank you this is your subscriber on the West Coast Los Angeles, but I get to enjoy your videos over here on the big island of Hawaii and the sad part is there aren’t very many Porsches over here. There are but a lot of them are on the west side near Kona. That’s where you probably see a lot of the Porsches, but I might not East side so there aren’t very many. Thank you for another good video. Can’t wait for your next one.
Most strut tower braces I’ve seen are connected to the firewall also. Maybe that would make a difference? I’m not an engineer so I really wouldn’t know, but just looking at say a Mustang no matter the year they tie the strut towers together and also bolt to the firewall. Makes more of a cage. The 60’s Mustangs had “export braces” that went from the towers to the firewall. Then a “Monte Carlo” bar that went side to side. Both of those are hard mounted vs yours that can pivot. Totally different car I know. Maybe not a good comparison, but I would think the concept would be the same or similar.
Very good point. I can't say for sure but I think connecting the brace to the firewall helps with braking and overall chassis stiffening. I think those cars have heavy duty frame rails as a chassis, but I'm not familiar.
I'll speculate! I believe the thinking here is that the chassis is no longer flexing and soaking up the bumps, so the suspension will actually doing all the work. So the effect would be like fitting a stiffer front sway bar and/or springs?
Thanks for another interesting and informative video, Tom. I put a Rennline strut brace on the ‘72 Targa hot rod I used to own. This was the single bar that goes from tower to tower. My understanding is that the Targa tub is not as rigid the coupe and my car had no seam welding or any other extra stiffening as you’ve done with Mac. My car was lowered, the suspension was pretty stiff and I had sticky tires (wear rating of 70 I think). I never tracked the car and most of my driving was on curvy country roads below 70 mph. My completely subjective assessment is that it did give a slight improvement in steering responsiveness and feel which made for a marginally more enjoyable driving experience…the car just felt a little more planted. But again, this was a stock Targa tub.
Thanks for your report Mike! Stiffer shocks could certainly push the strut towers harder. The trunk areas are the same between targas and coupes so you probably got the same 10% improvement that I measured. The plane where the A pillars are is where the targa will flex a bit more than a coupe. I don't think a strut bar can help you there, in fact the door opening in a coupe is probably the weekest area too.
One of your best videos Tom. Maybe you could run with that set-up at an autocross someday. Also, I'm loving that HUGE tool chest in the background. Looks like your garage is seeing a bit of a makeover. Cool.
This would be an interesting test in a macphearson strut car. Most front engine cars are a wide open box unlike 911 which have enherentlu stiff front ends.
The track techs aren't going to allow a laptop and test equipment on the passenger seat. At least not for me. The G forces aren't that different honestly unless maybe I'm on a big track with bumps?
Another great video - It would be interesting to see the forces generated in Autocross/track conditions... I suspect these these forces would be far greater than those seen in even your spirited street driving test. Also, when you gonna tell us about the new garage hardware I saw lurking in the background???
I doubt the autocross track forces would be significantly more. My tires were squealing on the hard corner and slalom part. The highest loads were measured on bumps. I'm open to measuring at the track, but the race techs aren't going to allow my laptop and antique daq. A garage remodel is in the works, albiet slow going
Makes sense to me for that car. These are popular upgrades for sure. My question is are they popular because they work or are they a popular item to sell. You answered that for your BMW, thank you. 😀
I guess cars with McPherson Strut see the most benefit. Coil overs, per se, can be installed on a variety of suspension layouts. I suspect a four link set up with coil overs would perform similarly to the 911. An inherently stable system. As you note the 911 is pretty stiff around the frunk, unlike front engined McPherson strutted econo boxes turned into ‘performance’ wagons. Great stuff Tom. Always insightful 👍🏽
A future video suggestion, you made adjustable rear control arm mounting points. Would be very interesting to see how changing to mounting point height causes differences in handling. Thanks for sharing your efforts, very interesting.
Yes, good call. I'm waiting to sort out the bent suspension arm first and then will dive into anti squat and rear roll center stuff. I have the same roll center analysis underway for the front too by the way. Thanks for the reminder👍
I doubt it. Seam welding showed an 18% increase in stiffness. Perhaps the strut bar loads would be 18% more? It doesn't translate 1:1 like that, but I don't think seam welding is a magic bullet that transforms the car. Especially my car which is mostly a street car.
If your concern is torsional stiffness, then you need to measure the deformation of the body suspension mounting points under a cornering condition. Based on that data, you can determine how much twist results in the body as it flexes. To do that, you need to drive the vehicle on a controlled course(ie track), preferably through the same corner for several iterations. You will need to measure the deformation, which will be elastic, and will be very small. I have a late model C4 corvette coupe. When GM came out with the C4 convertible, the loss of the roof structure adversely affected the torsional stiffness of the chassis. To get the torsional stiffness of the convertible back to the levels of the coupe, GM added cross bracing between the frame rails under the floor. The problem with you 911 is the body is made of low carbon steel which is very formable but from a material point of view, does not help to improve the structural rigidity of the chassis. Nowadays, car companies use a lot of hsla in 340, and dual phase sheet steels out to 1000 yield to improve the structural rigidity and keep the first natural frequency of the body above the tire frequencies(50 hz). Basically, the body shell of your 911 is very soft in terms of structural rigidity because of the time frame at which the vehicle was designed. Things are much different today because the high strength steels are much stronger than the low carbon steels and help to produce a lighter vehicle.
I already measured the torsional stiffness as mentioned in this video. Watch this one ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-RfDPUD-RhMM.html I agree, that modern cars are much stiffer. Largely due to geometry, structural windows, structural foam, etc. The modulus of high strength steels is about the same as the steel in my car. The torsional tests I did are below the elastic limit, so if the steel had a higher yield strength it wouldn't increase the torsional rigidity. The higher strength means it could tolerate more twist without permanently twisting though. In the case of the stiffening rails under a corvette, that is adding more structural geometery to resist torsional loads. Makes sense to me.
Respectfully, the issue with higher strength steels is that when the parts are stamped, they are work hardened to a much higher stress state than the initial yield, such that the the stress state at forming becomes the new yield stress. This work hardening through stamping is what gives the part it’s stiffness. In fact, today we use hot stamped parts which phase change from austenite at 900 degrees to martensite through a quenching process. These parts can have yields upwards of 1700 MPa. Hence, a work hardened part could tolerate much higher force per degree of twist, hence we know that from mechanics stiffness times deformation is equal to force. The greater the stiffness, the greater the force per unit deformation.@@GarageTimeAutoResto
Well, I suggest you review your statics and material properties then. Google the equation for torsional stiffness for a round tube or square beam, it doesn't matter. The stiffness does not depend on the yield strength of the material. It depends on the materials elastic modulus and the part geometry. Stamped parts are stronger than a flat sheet due to the increase in second moment of inertia. Our cars are not designed to go anywhere near the elastic limit. What you are saying is not relevant to this discussion.
Isn't most of the forces going through dampening instead of spring? Speing has a consistent rate, but dampener is dynamic and can transfer huge loads when driven hard.
The springs or torsion bars are supporting the weight of the car plus any dynamic jounce forces. Figure on 900 lbs plus 200 lbs per inch of deflection. The dampers can also create force with rapid jounce. Maybe 100lbs depending on the rate of jounce (in/sec) Driving hard on a smooth track doesn't create rapid jounce. Weight transfers from one side to the other relatively slow. The rapid jounce is what showed the most force in my testing. For example the speed bumps and water dips.
Wow great content. I expected typical youtube hyperbole. I really liked the discussion of torsion bar feame forces at the start. I asked greg fordahl (legendary suspension engineer (rip)) about strut braces and he said the only time hes seen it help was on a race car with a cracked frame. Ive seen many porsche strut bars tie the shock tops together. I was thinking this might help since stock shock tops are held with big rubber doughnuts so camber is more controlled. Handling on my 930 improved when i installed monoball strut tops, felt more stable and controlled especially in rough corners. Wrt ‘porsche always right’: modern street performance tires are crazy sticky compared to 1974. Maybe it can make a difference, gotta test and measure to see! Good job. A few question: is it possible all those stitched seams are making your car stiffer? Could you do test again tying strut tops together? That would show how much they want to move relative to each other.
Thank you for commenting and watching. I loved what your expert friend said about stut braces and cracked cars😀 Very good point about modern tires versus when these cars were designed. The tires currently in my car are much wider than stock and do have good grip. You mentioned the compliance in the upper bushings versus solid mounts. You are absolutely right that most of the deformation occurs in the rubber bushings versus the shock tower sheet metal bending or moving. The question is what direction do they deflect and how much? Going back to stickier tires means the strut tops may actually move outward upon cornering. The spindle location is placed such that cornering loads counteract the torque when the tire tries to fold under the car. Therefore sticker tires might actually move the strut tops outward. If you look at the diameter of the rubber, they can't move too far, but mono balls certainly eliminate that. I would love to test. Just need to find the time, ha ha. Always good discussion and thanks for generating more ideas😀
I would really like to see HPDE testing of this, but understand the difficulties. I plan to add a Racing Dynamics strut brace to my 944S2, but I have a lot more loads up front, so it's definitely beneficial. When I swap to coilovers I want to add a lower strut brace, tying my caster blocks (rear control arm mounts) together.
Yes, I would love to rent a track and do more testing. It's just outside my current scope, but will continue to push this car on the track and learn more. And share more. Ha ha, it's taken me a few years just make part 2 😀
@@GarageTimeAutoResto Lol I wish I could afford the time to try and R&D and document the gains. There's no room in the hood of my car to put a load cell with a strut brace. I can barely fit a strut brace to begin with on an S2.
Yeah, my equipment is very old. Load cell measurements are not easy, but my rig could be miniaturized with some expense. You can't just run it with a voltmeter 🤣
Thank you. I agree the Metric system is better and easier. For me it's hard to relate to units that I haven't grown up with. For example I know 100lbs is 45kgs, but that doesn't register in my brain because it's just a number. I'll do my best to convert or post both.
Maybe, I'll repeat the test with coilovers someday and I could add another load cell in the diagonal bar. In the meantime, I should fabricate something with diagonals and no sensors to finish what I started.
The parallelogram analogy isn't very useful when only 2 of the 4 points are actual pivots. Since there is obviously some stiffness without the strut bar adding the strut bar would allow both sides to fight against deflection in all but one of the 3 possible loads. Those 3 being; both sides in tension, both sides in compression, and one side in tension and one in compression. If both sides are in tension or compression, even if not equally, the Brace is going to help. If there is equal tension and compression but on opposite sides, then the brace won't be doing much if any good, but if there is a lot of compression on one side and very little tension on the other side, then the brace should give you twice as much stiffness as you had before. I'm not sure how there would ever be much tension in either side, the most would be if one tire was off the ground just hanging and that doesn't seem like much compared to the compression the outside wheel would have in hard cornering.
The empirical test data suggests there isn't much force in the bar under hard driving conditions largely because the way the McPherson strut design places cornering loads on the chassis on a torsion bar car. Your rationale is not supporting what happens while driving the car and of course the strut towers are rigid by design.
@@GarageTimeAutoResto I wasn't trying to speak to this particular use case, but the idea in general. It does seem in this instant the car is plenty rigid from the factory.
Whew! I could write a doctoral thesis on this, how do I put it in a comment. Making one singular change may or may not improve the 'mechanical' performance but it may well change your perception of the car while driving it. A better test would to take it to The Streets of Willow (Willow springs Raceway) and test the car with and without the bar. I did EXTENSIVE testing and changing things on my Lotus 61 formula Ford over the years (yes, I know its not a Porsche), and what worked best for me and my car was different from what others did with similar cars. (My performance gauge was lap times). As an example, I ran less negative camber on the rear than most racers, but far more front negative camber than most. I ran toe-in on the rear and toe-out on the front. I also corner balanced the car with my weight in the seat for every race. Other things, like ride height will affect handling. (I set the ride height 3/4" higher for California Speedway, or else G-force caused the bottom of the car to drag on the track on the turn 1-2 banking). Then there's shock dampening settings... and the list goes on. I wish we lived down the street form one another. We would have LOTS to talk about....
For sure we could talk for hours. I totally agree lap times trumps all theory and indirect testing. My issue is that it would take me 10 years of driving to acquire the skill to know the difference. I'm a beginner driver, but eager to learn and get better. My personal feeling is that tires, alignment, and shock settings are the ones to play with first. If I was a crew chief, I could get to the bottom of this in just one $100k weekend 😀 But, I'm just not there yet. I will do willow springs though. I will be trying to keep it on the track.
Not sure if you have answered this before but where did you get all the plugs for the holes in the fronk compartment? I know rennline and other companies sell various aluminum covers for the AC and HVAC holes but the oval holes in the fender arch and other small holes near the smugglers box I cant seem to find any kits or generic plugs for them. Thanks in advance!
So cool! This is a topic I’ve been super interested in for all cars I’ve ever owner. I’ve always wondered how to gather data on one of these bars be it in the front or rear. Would it make sense to run without the diagonal bar (if it’s able to be separated) to get a better baseline of how much movement there is, then do an identical run with the diagonal bar?
Thanks, the best way to test would be with another load cell in the diagonal bar. But I'm afraid the forces won't be high enough to measure?? When I put coilovers on the car, I will try this. Another idea is to use linear transducers to measure deflection instead of force. I have to think about that one. I ran out of time, but I was going to remove the shoulder bolt on one side and replace it with a pencil. (Or similar wooden dowel). If the pencil doesn't break then the bar isn't doing much.
I appreciate the efforts you went through to obtain actual data! Do you know if these have jounce bumpers on top of the shock as a travel limiter? If so and if there were a bump big enough and at enough speed I suspect you would see much higher loads (ie say a water runoff type dip mid-corner at medium/high speed).
For sure there are rubber bumpers on these strut inserts. I know for a fact that the jounce travel is very short at about 1". So I'm pretty certain that the front suspension bottomed out during some of those bumps. I will be fixing this with raised spindles and shortened struts soon. I also did a driving video with suspension travel and shock speeds. That was a good one too.
I don’t think driving the car at 30mph or even 40mph will allow one to notice the effect of a strut brace in the car, I also don’t feel a lot of performance parts benefit would be felt doing urban driving at legal speeds anyways. I only felt the difference on my car when doing 3digit numbers in the Pyrenees or Alpine roads in EU not cruising around the city at “see me” urban pace. So their is benefits to be had during spirited driving on country lanes or closed track else Porsche or even BMW Motorsport won’t have been utilising performance parts like strut braces etc in their race cars
Not surprised and I welcome skeptics. The data showed low speed dips and speed bumps generated the most force. High speed stuff might make a small difference, but there is always something to poke holes at. If you think it helps with your 911 then go for it. Porsche didn't always get it right by the way.
@@GarageTimeAutoResto I think it depends what one wants to do with the car. If one plans on tracking or indeed high performance driving probably a good idea, but if not really pointless doing those types of mods. A pal sold his 993rs tribute cause it was simply too hardcore(had strut brace as well amongst other full on OEM RS bits). Drives a standard supercar at the min. Point being you are not wrong, just a question of personal requirements from car
That full on bolted in x brace setup you showed, any actual info on its benefits? Side Note: I bought one for my 997 cause I thought it looked cool and had extra project loot laying around 😊
Yep, this video shows the triangulated strut brace improves the chassis rigidity. I measured almost 10% improvement. ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-xTlj2XLEOF8.htmlsi=Fgkpm3_QgQZtiDKG
Interesting video and results. A few thoughts... The RSRs of the 70s had already changed to coil overs, so that explains (at least in part) their use of front and rear shock tower bracing. I recall reading somewhere that there is quite a bit of flex in 911's chassis up front when cornering hard (perhaps on Pelican's forums?) - so I added a front strut tower brace on my build with the vague hope it would add some stiffness, but I have no data to back that up. Is 50-70lbs substantial? Would it get even higher if you were on the track, with semi-slicks? Maybe there is a tipping point where it's worth it...
Is 50-70 lbs substantial? The front of a 911 weighs around 900lbs when not moving. Dynamic road going forces in the ball joints and suspension pivots can easily be 5X that. My assessment is 70 lbs is very small compared to what the torsion bar area sees. Downforce, higher grip, combined with stiffer suspension is where the strut bar becomes more important in my opinion. 935 coil over suspension had lots of strengthening on these strut towers for good reason. Can someone give me a 935 to test. I'd be happy to do it.
Makes sense! With that said, I'm not ready to remove the bar now, lol. BTW, I had never heard of the strut bar being used to adjust camber before... Brutal stuff.
Just noticed the front of your car all stich welded too. How much does that help over the stock car? Would the numbers be much different? Edit: Never mind, I saw your answer to someone else in the comments. :-)
Watch this one. I explain how to stress the chassis by twisting it and measuring it at the same time. ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-RfDPUD-RhMM.html Another one: ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-xTlj2XLEOF8.html
@@GarageTimeAutoResto 🤣🤣🤣 I'll never rat out a fellow Porsche brother. It looks similar to my old stomping grounds. Costa Mesa, Newport, Balboa, Laguna Beach 🏖
Frankly your driving in this video is not what I call spirited, very mild. Your conclusion is no surprise. If you add racing slicks or even cup2 tires and drive at 90% plus of the car's ability you will come to a different conclusion.
Fair point, but when tires squeel you don't get much more lateral force when driven harder. Call this drive .8G at either the hard right or the slalom section. If I had cup 2 tires, then we are at 1.2-1.3G? Go ahead and gross the measured 40lbs strut bar force to 60lbs. It's still insignificant. Guys want to see smoke from the tires to prove a point, ha ha.
