Strut Brace test methodology
12-10-2002, 12:38 PM
#31
OK, I got to thinking about the discussion Danno and I had a while ago regarding Ti STB cross bars which eventually evolved into a discussion about bicycles. Then I got to thinking about various brace designs, including Neuspeed's old design for Volkswagens and the like, and the braces on cars like Shelby GT350s. My guess is their design is better, and all of the ones we either have or are talking about are flawed. Why? The bearing ends on the bar we are discussing allow rotation.
Think about it, you may be keeping the strut tops at the same distance apart, but you are not keeping the geometry the same. In fact the tower tops move, they will do so in and out of proper allignment in direct (or close to) correlation to one another.
BTW, it's about time you chimed in Danno!
So basically the brace needs to be rigidly fixed to the tower tops and ultimately triangulated or cross braced.
Think about it, you may be keeping the strut tops at the same distance apart, but you are not keeping the geometry the same. In fact the tower tops move, they will do so in and out of proper allignment in direct (or close to) correlation to one another.
BTW, it's about time you chimed in Danno!
So basically the brace needs to be rigidly fixed to the tower tops and ultimately triangulated or cross braced.
12-10-2002, 12:46 PM
#32
Oh, one more thing; our cars are very stiff. When I put mine in a ditch on Friday, it eventually wound up supported (basically) by only the right front tire and left rear tire. I literally could rock the car back and forth on this axis so that the other two tires were off the ground with one hand. Even still all the panel gaps looked even and I was able to open both door and the hood with out any resistance or sticking. When I had my VW I could barely get the doors open if I had one of the corners jacked up, that's how flexy it was
12-10-2002, 12:52 PM
#33
I was wondering about that one... looking (or as my SO say, "drooling") at pictures of a Turbo Cup car last night, the strut brace is an example of simplicity at its best. Basically just a flat bar with a slight curve in it, attached at each side by two bolts (15mm, I believe). If that isn't both effective AND cheap, nothing would be.
Ken, any chance you want to take a look at that, compare it to your design and comment on it? I am not taking anything away from your design, and I still stand by my original comments that it sure feels like the KLA STB made a difference in my 951, regardless of what the naysayers might offer as reasons why I shouldn't feel any difference.
I think I am still going to take a shot at measuring the effect, altho its gotten a WHOLE lot more involved than I originally anticipated (to quote my SO again... "you want to buy a WHAT? Strain gauge?? sheesh"). Funny how life works that way...
Regards, ...Scott
Ken, any chance you want to take a look at that, compare it to your design and comment on it? I am not taking anything away from your design, and I still stand by my original comments that it sure feels like the KLA STB made a difference in my 951, regardless of what the naysayers might offer as reasons why I shouldn't feel any difference.
I think I am still going to take a shot at measuring the effect, altho its gotten a WHOLE lot more involved than I originally anticipated (to quote my SO again... "you want to buy a WHAT? Strain gauge?? sheesh"). Funny how life works that way...
Regards, ...Scott
12-10-2002, 01:43 PM
#36
Advanced
Join Date: Aug 2002
Location: Midwest
Posts: 75
Likes: 0
Received 0 Likes
on
0 Posts
You’re right about simple. If Porsche built them you know they have to be good. We didn’t go with this design for a couple reasons. The first is aesthetics. We wanted to make a bar that was functional and looked good. The second reason is that I was not sure that the curved bar would hold the towers together without flexing. But, from these pictures it appears that Porsche made it work well
12-10-2002, 01:48 PM
#37
Hey, I'm not being a naysayer, well not a first. I have a strut tower brace too. I can say I do not recall noticing a difference when I added the bar.
What I am saying is that I think the design of the bar I have (Weltmeister POS) as well as the bar in question in this thread is flawed. A more effective bar would be on that is rigidly fastened to the chassis without heim joints (rod ends).
I think we are all guilty of being lured by shiny high tech looking bits and pieces, like the stainless braid cover hoses (with hose clamps) that folks have been hot for in the past, regardless of whether or not they items are appropriate for the application.
What I am saying is that I think the design of the bar I have (Weltmeister POS) as well as the bar in question in this thread is flawed. A more effective bar would be on that is rigidly fastened to the chassis without heim joints (rod ends).
I think we are all guilty of being lured by shiny high tech looking bits and pieces, like the stainless braid cover hoses (with hose clamps) that folks have been hot for in the past, regardless of whether or not they items are appropriate for the application.
12-10-2002, 05:05 PM
#39
I wasn't talking about anyone in particular, so no worries with the "naysayer" comment, Manning. I think your posts have been on target and constructive to the conversation, as have all the posts I have seen here and the other post on STB effectiveness.
I find it interesting that Porsche thought enough of the idea to include it on the Cup cars when they appear to have rubber mounted strut supports ala M030. I would think that would help to validate the claim that they must have some positive effect.
Just for grins, I went ahead and disconnected one side of my KLA brace to see what effect the STB has. This is all "seat of the pants" stuff, but I am more convinced than ever that the STB does indeed make a difference. The controlling factor is still the adhesion limits of my tires, not my car, but with the KLA brace, my car can push harder into a turn before getting to the point that the tires start to lose their grip (in one sense, it might be better for testing to have bad tires on the car, so that I have a lower threshold to hit to get the tires to break loose). The car steering also feels more precise with the brace.
It sounds like Ken is going to do his test, I think I will go ahead and do my original test as well. Depending on the results and the interest AND provided we can get some assistance from Sid, I may pursue doing a strain test as well.
Regards, ...Scott
I find it interesting that Porsche thought enough of the idea to include it on the Cup cars when they appear to have rubber mounted strut supports ala M030. I would think that would help to validate the claim that they must have some positive effect.
Just for grins, I went ahead and disconnected one side of my KLA brace to see what effect the STB has. This is all "seat of the pants" stuff, but I am more convinced than ever that the STB does indeed make a difference. The controlling factor is still the adhesion limits of my tires, not my car, but with the KLA brace, my car can push harder into a turn before getting to the point that the tires start to lose their grip (in one sense, it might be better for testing to have bad tires on the car, so that I have a lower threshold to hit to get the tires to break loose). The car steering also feels more precise with the brace.
It sounds like Ken is going to do his test, I think I will go ahead and do my original test as well. Depending on the results and the interest AND provided we can get some assistance from Sid, I may pursue doing a strain test as well.
Regards, ...Scott
12-10-2002, 05:26 PM
#40
Big thirst, Sore Thumbs
Rennlist Member
Napoleon
Rennlist Member
Napoleon
Haven't heard back from one source for a strain gauge. If that doesn't workout I have another but that may only be a vibrating wire stain gauge. Very accurate but it's sample rate is limited to like 2 hz.
I'll keep you posted.
I'll keep you posted.
12-10-2002, 11:51 PM
#41
Advanced
Join Date: Aug 2002
Location: Midwest
Posts: 75
Likes: 0
Received 0 Likes
on
0 Posts
Well I have completed my test on a 1986 951. I used a length of plastic pipe and a sharpie pin. I bolted the pipe to a bracket and slit the bar into the plastic pipe to make a slip joint. I wraped the bar with bubble wrap to keep it centered. I then drilled a hole in the plastic pipe the size of a sharpie pin. I used duct tape to hold the sharpie against the bar. When the car was driven, if the bar moved the pin would make a mark on the bar. We drove the car around the block. We did drive the car hard but it had street tires and was on a normal city street. I'll let the results speak for itself.
I would guess that we are + or - 1-mm. As you can see it moved 6.5mm. I think the strut will help this even on the street.
Ken
I would guess that we are + or - 1-mm. As you can see it moved 6.5mm. I think the strut will help this even on the street.
Ken
12-11-2002, 12:19 AM
#42
Unaffiliated
FABULOUS! Well done. Certainly not the most scientific, but proof plenty to me that the towers do move, and quite a bit I might ad. (I bet with track tires and a track suspension the movement is a LOT more than that) Kudos to the guy who guessed 6 to 8 MM!! Sure, the mass market braces don't triangulate, but most of us don't want to cut or drill, so boxing them in is a great choice. Thanks for your time and effort in laying to rest a long debated issue. Methinks the strut-brace guys may get a bunch of orders from naysayers..
Now let the mathemeticians calculate how much camber we are losing with that flex.
<img src="graemlins/beerchug.gif" border="0" alt="[cheers]" />
This also goes a long way in explaining why Porsche gave us significant hood/fender gaps! DOH!
Nice Job!
Now let the mathemeticians calculate how much camber we are losing with that flex.
<img src="graemlins/beerchug.gif" border="0" alt="[cheers]" />
This also goes a long way in explaining why Porsche gave us significant hood/fender gaps! DOH!
Nice Job!
12-11-2002, 12:50 AM
#43
Big thirst, Sore Thumbs
Rennlist Member
Napoleon
Rennlist Member
Napoleon
While I don't discount the movement we still don't know which dirction it was. To be perfectly honest it could be allot of things even the engine, well maybe not that much, which doesn't discount the need for a strut brace regardless of what caused it.
Okay packed with some financial info. I think I should pass you through my train of thought this afternoon. WARNING if you don't like booring technical info skip to the last paragraph which is lighter and has Marvin in it.
Okay I have a line on some electro-resistive strain gauges. Unfortunally we would have to make a Rosette bridge. i.e 4 gauges attached end to end in a closed loop. Needed to cancel out the noise (which would be considerable under the hood) So were talking almost $400 in gauges. Not incluiding all the signal conditioning. And the gauges would be ruined. Non-starter.
Okay Vibrating wire which we could easily measure accurately with one gauge but at no more than 2 hz. Readout wouldn't be to much to rent. still. Non-starter. Unless we have a corner we can hold for like 3/4 of a second.
Then we come to a fiber optic gauge. Excellent dynamics, up to 40k hz. low noise so one gauge would be sufficient. only about 100 and it too would be ruined. Problem would be the readout, about 5k to buy but I could rent one for low dough. $100 or so. Still a bit exspensive.
So I start thinking about low dough alternatives. One would be similar to the sharpie experiment. with a hole drilled through the dowel and pipe. Wooden dowel pieces could be installed and see what size dowel stops snapping. That would give us the amount of force induced. Kind of wierd but would work, if time consuming.
So how to do it more real time like. In my head pops a robber packet of water with a pressure transducer. As it's squeezed pressure comes up. That would give us a displacement and energy. but a bit complicated.
So were going back to the scratched paint/sharpie experiment. But suitably modified for real time. We replace the shaprpie with a linear potentiometer. We can read with standard multimeter, one with peak capture would be best, not sure if mine has that, but not a big deal. So driver throwing the car around. Passenger holding the multimeter. This would not only give us displacement and if we get a mechanical type geek we can extropolate energy (still not happy, energy would be a better measurement to make). Best thing about this is we can then drive over potholes and the like and see what kind of deflection we get then dig out the trigonometry books. So not only do we measure flex but when. Suitably made this could kept after a strut brace install and we could see what the improvement is. If neccesary we could make a 2d or 3d version. For your useless info file this is called an extensometer. Total cost 30 bucks plus the manufacture of the bar bit. And that's for the 3d version. 1 d would be like 6 bucks plau bar.
Now we know I have no life. Anyone else with the fab equipment to make the bar bit? I think the intersting thing is the various ways that this measurement can be made. pressure, strain, displacement. I thought of temprature (The increase in temprature of the bar under presure) but doubt we could surpass the ambient noise. There are also stick on strain gauges that change color we could pick up cheap, but I have no idea of a source. Best part of that is it could be attached to the strut itself.
Okay packed with some financial info. I think I should pass you through my train of thought this afternoon. WARNING if you don't like booring technical info skip to the last paragraph which is lighter and has Marvin in it.
Okay I have a line on some electro-resistive strain gauges. Unfortunally we would have to make a Rosette bridge. i.e 4 gauges attached end to end in a closed loop. Needed to cancel out the noise (which would be considerable under the hood) So were talking almost $400 in gauges. Not incluiding all the signal conditioning. And the gauges would be ruined. Non-starter.
Okay Vibrating wire which we could easily measure accurately with one gauge but at no more than 2 hz. Readout wouldn't be to much to rent. still. Non-starter. Unless we have a corner we can hold for like 3/4 of a second.
Then we come to a fiber optic gauge. Excellent dynamics, up to 40k hz. low noise so one gauge would be sufficient. only about 100 and it too would be ruined. Problem would be the readout, about 5k to buy but I could rent one for low dough. $100 or so. Still a bit exspensive.
So I start thinking about low dough alternatives. One would be similar to the sharpie experiment. with a hole drilled through the dowel and pipe. Wooden dowel pieces could be installed and see what size dowel stops snapping. That would give us the amount of force induced. Kind of wierd but would work, if time consuming.
So how to do it more real time like. In my head pops a robber packet of water with a pressure transducer. As it's squeezed pressure comes up. That would give us a displacement and energy. but a bit complicated.
So were going back to the scratched paint/sharpie experiment. But suitably modified for real time. We replace the shaprpie with a linear potentiometer. We can read with standard multimeter, one with peak capture would be best, not sure if mine has that, but not a big deal. So driver throwing the car around. Passenger holding the multimeter. This would not only give us displacement and if we get a mechanical type geek we can extropolate energy (still not happy, energy would be a better measurement to make). Best thing about this is we can then drive over potholes and the like and see what kind of deflection we get then dig out the trigonometry books. So not only do we measure flex but when. Suitably made this could kept after a strut brace install and we could see what the improvement is. If neccesary we could make a 2d or 3d version. For your useless info file this is called an extensometer. Total cost 30 bucks plus the manufacture of the bar bit. And that's for the 3d version. 1 d would be like 6 bucks plau bar.
Now we know I have no life. Anyone else with the fab equipment to make the bar bit? I think the intersting thing is the various ways that this measurement can be made. pressure, strain, displacement. I thought of temprature (The increase in temprature of the bar under presure) but doubt we could surpass the ambient noise. There are also stick on strain gauges that change color we could pick up cheap, but I have no idea of a source. Best part of that is it could be attached to the strut itself.
12-11-2002, 01:46 AM
#44
Addict
Rennlist
Lifetime Member
Rennlist
Lifetime Member
I find it interesting what you guys are doing. I can also appreciate it as I am somewhat of a numbers guy. But in this particular case, I'm leaning to more of the following...
Fork over the $125, slap that puppy on, take your baby for a drive, and know you're probably getting the most bang-for-performance out of any benjamin you've spent on your p-car ever.
Merry Christmas!
-Matt
Fork over the $125, slap that puppy on, take your baby for a drive, and know you're probably getting the most bang-for-performance out of any benjamin you've spent on your p-car ever.
Merry Christmas!
-Matt
12-11-2002, 12:01 PM
#45
Big thirst, Sore Thumbs
Rennlist Member
Napoleon
Rennlist Member
Napoleon
From e-mail to an automotive engineer I know.
> From: "Jiro" >
> To: "Sid">
> Sent: Wednesday, December 11, 2002 10:46 AM
> Subject: Re: picz
>
>
> > Let me save you a lot of trouble.
> >
> > Jack up one corner of the car. Measure before and after with a
> good
> > tape measure.
--- Sidv wrote:
> It's lateral load they are looking at.
> ----- Original Message -----
That's what they think.
A nice even lateral load shouldn't make the strut tower move very
much - relative to each other; one should tend to move inboard, the
other outboard. Depending on geometry, they probably will come
together a bit, just not as much as worst case. But if you take a
bump while in a turn (on the outside wheel), expect that to be the
worst case closing in of the shock tower. The guy with the Sharpie
did his test on surface streets; I bet the peak motion there was over
one-wheel bumps.
Try it. Jack up two wheels equally. Then do just one or the other.
> From: "Jiro" >
> To: "Sid">
> Sent: Wednesday, December 11, 2002 10:46 AM
> Subject: Re: picz
>
>
> > Let me save you a lot of trouble.
> >
> > Jack up one corner of the car. Measure before and after with a
> good
> > tape measure.
--- Sidv wrote:
> It's lateral load they are looking at.
> ----- Original Message -----
That's what they think.
A nice even lateral load shouldn't make the strut tower move very
much - relative to each other; one should tend to move inboard, the
other outboard. Depending on geometry, they probably will come
together a bit, just not as much as worst case. But if you take a
bump while in a turn (on the outside wheel), expect that to be the
worst case closing in of the shock tower. The guy with the Sharpie
did his test on surface streets; I bet the peak motion there was over
one-wheel bumps.
Try it. Jack up two wheels equally. Then do just one or the other.