Going a little in depth with the front suspension

Another thing to keep in the back of your mind. Spring rates are different from wheel rates especially where there is significant leverage. No idea about the front, but at the rear , the rate at the wheel (lbf/in) can be calculated from the length of the semi-trailing arm and the distance the spring mount is from the rear end. eg if the spring rate is 260, and its mounted halfway along the arm, the wheel rate should be 130. Possibly not relevant here, but....
 
Yep, this is exactly why I’m doing this to accurately ascertain the wheel rates. Bear in mind that the arrangement with the front suspension means that the increase in load has a lower impact on creased spring rate under deflection. So, 160lbsf/in should scale to an extra 160 lbs of 1” spring compression. However, with the front arrangement that would equate to more like a variation of only 60lbs per 1” of deflection at the wheel with a load of 260lbs (not sure if I’m making sense here).

Another way of looking at it is you have a higher average wheel rate, but with a lower gradient of increased wheel rate by deflection.
 
Is this the steel that you're using Gargo, or did you just cite it as an example?
That is just an example. Truth be told, I searched for 4043 and didn't realise that the search engine had included results for 4340. Good call Ron as you should not be using this steel for ARBs, but be using a spring steel. I am using a modified Peugeot ARB.

Beryl already has a 1in Hex 4043
The only 4043 I can find is an aluminium alloy, 5% silicon. Sdibbers is what you are using? Or have you too made a typo?
For the Hex, you'll need to use a different equation for J the 'Polar moment of inertia' but you'll find it is very close the round bar equation.

Either way the spring steels I could find info for have a Shear Modulus of 10x10^6. Which is the base for the calculated number that appears in the spreadsheet. (My example used 11.5x10^6)
The other simple thing to do is measure what the ARB you have. Clamp one end, support the other and torque the bar, measure the deflection. Convert the result to the correct units.

In the real world cars are put on a rig that measures all these values to validate the designers numbers. The rig pulls and pushes, deflection and forces are measured etc.
In the image below, the rig is a post on each wheel, with the 4 golden jacks deflecting the chassis for roll and pitch. The weight transfer in roll/pitch can be measured. Sorry for the horrible car, but could find one with a Rover. These simple rigs can also measure the true CoGs, but overall and front and back CoG. Which I have done with my own car using bathroom scales, raising the front of the car and measuring the weight shift to rear. Simple but effective.
1669366710801.png
 
For the Hex, you'll need to use a different equation for J the 'Polar moment of inertia' but you'll find it is very close the round bar equation.

The polar second moment of area for a solid hexagonal cross-section ( for torsion about the centroidal axis) = (5(3)^0.5)a^4/16. If that looks a little confusing, in words it is 5 multiplied by the square root of 3 multiplied by the side length a raised to the fourth power, all divided by 8. The units are length to the fourth power.

Ron.
 
I wrote the equation out in Word, screen snip, and voila ;)
polar second moment of area for hexagon cross section about the centroidal axis.JPG
I prefer to refer to the resistance to torsional deformation as the Polar second moment of area as opposed to the Polar moment of Inertia. I do this to prevent the possibility of confusion as moment of intertia in a physics context refers to a torque to achieve a desired angular acceleration about an arbitrary axis. Second moment of area provides a measure of the resistance to deflection about a given axis. In an engineering context, the second moment of area also known as the second mass moment of area, and the moment of inertia all mean the same thing. Some mechanics of materials text books will use the terms interchangeably, while others frown on the use of moment of intertia. I think it stems from mathematicians that like to simplify wherever possible. The formulas for the second moment of area and Polar second moment of area are derived through double integration.

Ron.
 
Dyslexia got the better of me. 4340 was used on my car. I did get it flame hardened to (I think) 65 Brinnell to add to stiffness. It been in the car for around six years now. Certainly helps with body control. I would say that going stiffer might be a problem. I used loctite on grade 8 bolts for the mounting clamps, and even then they can back out after track sessions. It may indicate a point of weakness in the basic design.
 
Maybe it’s worth drilling bolt heads and lock wiring with uprated bars and usage. Have you also gone up a size on your ARB fasteners Sdibbers?
Jim
 
@corazon – wouldn't increasing the size of the ARB fastening bolts leave too little 'meat' in the clamps? Asking coz I had also wondered if that would be a smart move.
 
@corazon – wouldn't increasing the size of the ARB fastening bolts leave too little 'meat' in the clamps? Asking coz I had also wondered if that would be a smart move.

I wasn’t the first, I think Mike did it years before?
No issues, there’s plenty of material in the mounts to go from 5/16 to 3/8 unf.

519D9172-D542-41DF-AAE7-F259251708DA.jpegA0B633B5-71A0-45F9-9832-1DED006EE587.jpeg
 
@corazon – Looks good! Why did you opt for Allen head bolts? Are they sufficiently strong? Asking coz I can't see any markings indicating their strength. Apologiez for perhaps diverting this thread – not pretending to be an engineer, don't worry chaps!
 
@corazon – Looks good! Why did you opt for Allen head bolts? Are they sufficiently strong? Asking coz I can't see any markings indicating their strength. Apologiez for perhaps diverting this thread – not pretending to be an engineer, don't worry chaps!

Personal preference more than anything, they fit the component nicely, they’re 12.9 grade zinc plated from memory

Jim
 
I would say that going stiffer might be a problem. I used loctite on grade 8 bolts for the mounting clamps, and even then they can back out after track sessions. It may indicate a point of weakness in the basic design

I would say that it is due to the dynamic loading the bar is placing on the bolts. The induced torsion in the bar is only resisted at the clamps. Repeated displacements, however small at these locations, places cyclic loading on the bolts, which due to vibration, can now loosen. The loss of bolt tension leads to larger bar displacements, increased vibration, and an even more significant loss of tension.

Given that is phenomenon does not occur in a road-going car suggests that the design is perfectly satisfactory for that application. What is now happening though is outside of the original design specification with larger diameter bars delivering an increase in bar stiffness. The clamps and the bolts are therefore compromised, being required to perform in a situation they were not designed for. In a perfect world, larger bolts delivered at a higher tension would be the answer, with a redesigned clamp that contained the bar with greater security.

Ron.
 
That sounds about right. Next step will be stiffer springs on the car. This will add support which will most likely reduce torsional loading by reducing suspension travel relative to both sides. I like the idea of stepping up the thread size next. I’ll most likely do that at the same time as I strip the front suspension.
 
Personal preference more than anything, they fit the component nicely, they’re 12.9 grade zinc plated from memory

Jim
Class 12.9 is a metric spec. These are UNF bolts. Grade 8 are good, but can crack in a failure mode due to hardness. Grade 5 actually might be better for this application as the have a small amount of give in them that allow for a small amount of preloading when torqued to spec.
 
Yes I went 1 size larger years ago and lengthened the thread in the arm at the same time. I had to buy a whole box of Grade eight bolts to get four....Imperial cap screws should be significantly harder (HRC45 to HRC39,) and have a higher yield strength (162k vs 130k) than Grade eight and you can normally buy them in small numbers. I found it was very important to ensure both top arms are in exact alignment with the stiffer bar and the cap is tightened up evenly so everything is in full contact and square on both ends. The numbers above are also an indication of the damage they can do if they do fail.
 
Class 12.9 is a metric spec. These are UNF bolts. Grade 8 are good, but can crack in a failure mode due to hardness. Grade 5 actually might be better for this application as the have a small amount of give in them that allow for a small amount of preloading when torqued to spec.

The original bolts are grade 5, so your point Sdibbers is on the money.

I found this article that details why higher-strength bolts are not always the right answer.
the-stronger-the-better-is-not-necessarily-the-case-for-fasteners.pdf (boltscience.com)

Ron.
 
I know it’s a metric spec but Margnors sell these UNF offerings as such, I don’t know why they don’t say Grade 8. Listed as BS2470. They are also zinc plated which is unusual for higher strength cap screws, they’re usually that black oxide that rusts quickly..

I wanted a plated capscrew in this location and this was all I could find at the time.

I’ve read that article before Ron, and do generally have that in mind when selecting fasteners but thought I’d trial these and deal with it if there ever is a breakage. The higher clamping power will be beneficial if they can handle the forces don’t you think?

I do think for hard use that cross drilling and lock wiring whichever fastener you have would be a good idea with knowledge they can start loosening in spite of loctite

3/8 UNF x 1.1/2 H/T Skt Cap Screw, Zinc Plated

Jim
 
Last edited:
Talking of the uprated ARB clamps & bolts, l'm still suspecting that it would be best to space these gaps between the clamps out rather than leaving the centre of the bolts free to take all of the twisting force. Is that a reasonable assumption?

Screenshot_20221127-121714_Chrome.jpg
 
Back
Top