Slimy38
|
| posted on 21/9/14 at 09:39 AM |
|
|
Engineering help needed (rear upper wishbone design)
I wonder if I could get some input into the fixing for my rear upper wishbones? I've got the wishbone, upright and coilover all intersecting on
one 8.8 M10 bolt, and I want to ensure I get a strong enough fixing.
I have got three different designs. In each, the view is from above, the top blue box is the coilover bush and crush tubes, the lower one is the top
mount for the MX5 upright, and the green is 5mm plate steel (the rest of the wishbone would extend off to the left);
The first (left) is what I've seen from some Westfields, with the coilover just attached to the end of the bolt. Obviously it's a design
that works for Westfields, but I'm not sure. Especially as I'm only using an M10 bolt which might be smaller than Westfield use. Will the
coilover just bend the bolt over time?
The second (middle) is a little bit better, but again will the combination of the upright and the coilover result in a bent bolt in the middle?
The third (right) was my original design, with both the upright and the coilover within a u bracket. Definitely no bending forces here. But is it
overkill?
I'd appreciate any thoughts, mainly so I can stop overthinking things and just get on with it!
[Edited on 21/9/14 by Slimy38]
|
|
|
|
|
Slimy38
|
| posted on 21/9/14 at 10:07 AM |
|
|
This is a pic of a Westfield coilover mounting being on the end of a bolt;
How does this type of joint retain it's strength and not simply bend in two/three?
[Edited on 21/9/14 by Slimy38]
|
|
|
Sam_68
|
| posted on 21/9/14 at 11:25 AM |
|
|
The crush tube of the bush (or the ball of the spherical bearing) effectively adds to the cross-sectional area of the bolt in bending, if it's a
reasonably close fit, so the loading will approximate more closely to single shear.
IIRC an M10 8.8 grade bolt in single shear can take over 2 tonnes of load (ie. in excess of an 8g bump on a 250kg corner weight)... so if you design
it so that the offset is kept as small as possible, to minimise any bending effect, you should be fine. Though I'd still probably be happier
with an M12 bolt (which can take over 3 tonnes in single shear). It's Sunday morning after a long night, though, so don't quote me on
those figures or sue me if the bolt shears. 
Structures purists will tend to suggest that single-shear loading is a heinous crime, whatever the circumstances.
Pragmatists will tell you that, in practice, the bolt size will be dictated by the size of the bushes on the components you're using and will
normally be much larger than it needs to be for the loads it's taking, in which case you might as well save yourself a bit of weight and
complexity on the bracketry by accepting single shear.
The Westfield design does look pretty dreadful, though, and isn't something I'd want to put my name against - what with the big, squidgy
polybush creating a large offset for the coilover load path, and the wishbone Rose joint possibly supported partly on the threaded portion of the bolt
- but I'm sure it'll have been calculated as safe.
Of course, if you do use single shear, you should always ensure that the bolt is arranged so that the threaded portion is NOT taking the shear.
|
|
|
Slimy38
|
| posted on 21/9/14 at 11:53 AM |
|
|
Thanks for the very comprehensive reply, it's appreciated. The M10 is due to the upright upper crush tube being M10, I'd rather keep that
as it is. Some people have changed it, but it's not an easy job.
Interestingly, I have also found this other Westfield design, which is similar to what I was planning as option three;
Which to me looks infinitely better.
|
|
|
Sam_68
|
| posted on 21/9/14 at 12:03 PM |
|
|
^^^ That design presumably uses a bush set in the upright?
... Much more sensible and elegant solution, I agree - and everything in double shear for minimal increase in weight and complexity!
Camber/toe adjustment is a bit more of a faff, appears to be the only downside.
|
|
|
Slimy38
|
| posted on 21/9/14 at 12:10 PM |
|
|
Yep, there's a beefy metalastic type bush in the top of the upright that I'd like to retain.
Sounds like everything is just pushing me right back to the original 'three plate' design. I'd have had them finished by now if I
had just stuck with my first idea and not tried to over-complicate it!
|
|
|
Lew The Machine
|
| posted on 21/9/14 at 08:34 PM |
|
|
I'm only mid way though my mech eng degree so take this with a large pinch of salt... but I'd say that because there is no measurable
moment or distance from the "pivot" where a bend would occur, the only forces that you need to work out your FOS for would be the shear
force across the bolt, which shouldn't be too hard to calculate if you know the car weight etc!
ITS FINE! i welded it!
|
|
|
Sam_68
|
| posted on 22/9/14 at 06:27 AM |
|
|
quote:
Originally posted by Lew The Machine
I'm only mid way though my mech eng degree so take this with a large pinch of salt... but I'd say that because there is no measurable
moment or distance from the "pivot" where a bend would occur, the only forces that you need to work out your FOS for would be the shear
force across the bolt, which shouldn't be too hard to calculate if you know the car weight etc!
I don't think that's quite the case, because the polybush will allow a small amount of bending deflection within it's
walls, but yes, that's what I was getting at with my post above, when I said that it would approximate to single shear.
Then, as you suggest, it's just a matter of multiplying the corner weight of the car by the anticipated instantaneous deflection, in
'g'. You can calculate the strength of the bolt in single shear quite easily, but it's even easier to simply look it up in
manufacturer's tables. 
The tricky bit is deciding what factor of 'g' to apply. Costin and Phipps (from memory) suggests a 4g bump for calculation, I think, but
that was back in the days of relatively soft suspension and tall section tyres with large amounts of sidewall deflection. I'd be inclined to use
at least 6g, possibly 8g, these days.
But whatever, I still think that the double shear design shown immediately above would be the better one.
|
|
|
skov
|
| posted on 22/9/14 at 12:33 PM |
|
|
Slimy, if you did want to go to an M12 bolt you just need a different crush tube.
A company called strongflex supplied me with polybushes and M12 crushtubes for the MX5 uprights for the same price as their standard M10 ones.
Came to £23 posted for both sides (was a couple of years ago now though).
|
|
|
prawnabie
|
| posted on 22/9/14 at 12:43 PM |
|
|
That pic is the same as my westfield MX5 chassis setup. I have changed the lower mounts for rose joints for ease of setup though.
SHaun
|
|
|
