Board logo

Front Shocker top mount ?
Trollyjack - 30/5/11 at 09:41 PM

Hi
I have assembled the front end to day and now deciding where to put the top shocker mount
Do I put it on the side of the top rail

In both positions I am getting about 40mm of travel on the shocker





Or under side of top rail

On side of top rail
On side of top rail


Been told they are best stood up but please comment

I think under side would be better and more consealed

Comments welcome Please

There are more photos in my photo archive under front end


Chippy - 30/5/11 at 10:11 PM

I would mount it under the chassis rail for two reasons, 1/ it looks neater, and is better for IVA. 2/ you will get "slightly" more travel as its on more of an angle. HTH Ray


tilly819 - 30/5/11 at 10:22 PM

In the ideal world it should be placed were the load path passes through the center of the supporting chassis rail

tilly


blakep82 - 30/5/11 at 10:28 PM

quote:
Originally posted by tilly819
In the ideal world it should be placed were the load path passes through the center of the supporting chassis rail

tilly


personally i'd have brackets like this

Description
Description


with either of your options, you're never applying a sort of compression force, its always trying to push the brackets away from the tube


Jon Ison - 30/5/11 at 10:37 PM

Mine are as your lower photo but connected to each end of a rail that is welded to the underside of the upper chassis rails, also cross braced to the bottom rails. Well I know what I mean........


Trollyjack - 30/5/11 at 10:55 PM

Was thinking of lining the top eye around here and putting a channel across from the outer toprail the the cross support thar is 5" inside.



Peteff - 31/5/11 at 08:36 AM

Better on the outside of the rail to keep it more upright. More angle needs stronger springing.


Badger_McLetcher - 31/5/11 at 12:30 PM

IIRC if you place them on the inside of the chassis you'll get a more pronounced spring fall off rate- the more vertical the better.


Doctor Derek Doctors - 31/5/11 at 12:42 PM

Instead of just guessing you should sketch it up and check the installation ratio / suspension rate, really you want it to have a constant or 'rising rate'.

I.E for every 1mm of the wheel the shock compresses by Xmm, ideally with X being constant or increasing for each consequtive mm of wheel movement.

If you don't check and have X decreasing for each mm of wheel travel then you will end up with some very odd handling, especially in roll.

This is very basic input though.

Is this position not defined in 'the' book?

[Edited on 31/5/11 by Doctor Derek Doctors]


Trollyjack - 1/6/11 at 10:31 AM

Looked in Book and it does not say where to mount the top mounting brackets

some pictures show on side of top rail and some photos show it under side of top rail ??


Doctor Derek Doctors - 1/6/11 at 12:48 PM

quote:
Originally posted by Trollyjack
Looked in Book and it does not say where to mount the top mounting brackets

some pictures show on side of top rail and some photos show it under side of top rail ??


If you get me the following dims I can sketch it up for you in CAD:

Length of lower wishbone pick-up to lower shock mount
Angle between the line (between the above two points) and the horizontal Plane (floor)
The distance between the lower wishbone mount and the center of the chassis top rail.
Angle between the line (between the above two points) and the horizontal plane (floor)
Open and closed length of the shock
Overall length of the lower wishbone.

A simple sketch that I have knocked up will tell you the installation ratio in seconds and also if you have rising/falling/constant rate in various positions and then you can make an informed decision.


Trollyjack - 1/6/11 at 03:27 PM

Thats Brill
I'll measure next time i'm at my lock up

Kev


Bare - 1/6/11 at 04:56 PM

Simply Try to mount the shocker unit as vertically as absolutely possible. ..
Save yourself a maths exercise to tell you exactly that .
Wouldn't overly worry about structural 'niceities' as even the suspension pickup points on these Locost chassis are just daft/poor . The mount ears have been documented to fatigue off due to their poor placement and inadequate support.
Shock mount 'strength' is the least of your worries.




[Edited on 1/6/11 by Bare]


Trollyjack - 1/6/11 at 09:08 PM

Length of lower wishbone pick-up to lower shock mount = 225mm

Angle between the line (between the above two points) and the horizontal Plane (floor) = 0°

The distance between the lower wishbone mount and the center of the chassis top rail. = 270mm

Angle between the line (between the above two points) and the horizontal plane (floor) = 74°

Open and closed length of the shock = 320mm open 255mm closed with rubber bump stop 235mm with out

Overall length of the lower wishbone. 340mm to centre od bottom ball joint.

I have drawn it up on a cad package but not ver good at it yet I can email you the DFX file if you wish


Doctor Derek Doctors - 2/6/11 at 09:32 AM

quote:
Originally posted by Trollyjack
Length of lower wishbone pick-up to lower shock mount = 225mm

Angle between the line (between the above two points) and the horizontal Plane (floor) = 0°

The distance between the lower wishbone mount and the center of the chassis top rail. = 270mm

Angle between the line (between the above two points) and the horizontal plane (floor) = 74°

Open and closed length of the shock = 320mm open 255mm closed with rubber bump stop 235mm with out

Overall length of the lower wishbone. 340mm to centre od bottom ball joint.

I have drawn it up on a cad package but not ver good at it yet I can email you the DFX file if you wish


I made a sketch yesterday, I'll stick the numbers in at lunchtime and test some positions.

I will test for Inside, Outside and Below the rail and tell you the rate, and rate change.


paulf - 2/6/11 at 09:48 AM

I had the same dilemma when i built mine and remember something about the mounting bolt being hard to fit when under the chassis rail as it was tight to the Fu tubes.I fitted the brackets to the outside of the chassis rail and made a wedge section that fitted between the rail and bracket to make it parallel to the lower mounting bracket .I thought the shock being more upright was the best way to have it and as mentioned the spring rate can be lower and wont tend to end up as a falling rate as could happen if mounted under the rail and to far in board.
Paul


Doctor Derek Doctors - 2/6/11 at 02:15 PM

Right then.

With those dims you cant put the shock directly inline with the chassis rail as you will have a very high ride height or your shock will be nearly fully compressed to keep that angle of the wishbone.

I have made an assumption that the centre of the shock top mount will be 40mm away from the centre of the chassis top rail (to give clearance) and that you will have 30-35mm of droop and 60-65mm of bump.

Inside mount:

-30 = 312.783
-20 = 307.713 = 5.07
-10 = 302.579 = 5.134 = 0.064
0 = 297.38 = 5.199 = 0.065
10 = 292.118 = 5.262 = 0.063
20 = 286.792 = 5.326 = 0.064
30 = 281.402 = 5.39 = 0.064
40 = 275.947 = 5.455 = 0.065
50 = 270.429 = 5.518 = 0.063
60 = 264.846 = 5.583 = 0.065


Outside Mount:

-30 = 315.486
-20 = 309.505 = 5.918
-10 = 303.469 = 6.036 = 0.118
0 = 297.38 = 6.089 = 0.053
10 = 291.328 = 6.052 = -0.037
20 = 285.044 = 6.284 = 0.232
30 = 278.798 = 6.246 = -0.038
40 = 272.502 = 6.296 = 0.05
50 = 266.157 = 6.345 = 0.049
60 = 259.762 = 6.395 = 0.05

The first number is the suspension travel (- = droop, + = Bump) The second number is the shock length, the third number is the change in shock length and the forth number is the change in shock rate.

In my conclusion mounting inside will give roughly a 5:10 installation ratio, mounting outside will give a 6:10 installation ratio thus outside would be stiffer with the same springs.

What's interesting as that mounting inside gives a lovely constant (amazingly constant) figure for the rate increase (0.063-0.065mm) over the entire travel of the shock which is ideal, whereas the outside mounted shock has a very erratic change of rate (-0.037-0.232mm) which will give odd handling.

If it was my car I would mount the shock inside to give more constant and progressive damper/spring travel with a constant rate increase, it will also allow you to have a longer wheel travel (also more ride height adjustment) and it will be a much neater finish to the car.

Hope that helps.


Trollyjack - 2/6/11 at 07:21 PM

So are you saying that the top mount is to be inside the top rail
where exactly is it to be, on the centre line of the top rail or under neath ?

could you you email me a sketch Please


Doctor Derek Doctors - 3/6/11 at 07:20 AM

quote:
Originally posted by Trollyjack
So are you saying that the top mount is to be inside the top rail
where exactly is it to be, on the centre line of the top rail or under neath ?

could you you email me a sketch Please


I have screen grabbed the sketch, here and in my archive

Basically though if you keep the wishbone inner pivot to - lower shock mount line at 0° (as you specified), extend to the shock to 295-300mm (297.38mm in the sketch) and then put the top mount so it sits 40mm from the centreline of the top rail you'll have what I used.

The two double white circles are the top and bottom shock mounts, the 25x25mm square box is obviously the chassis top rail and I I'm sure you can work out the rest.



Trollyjack - 3/6/11 at 08:50 AM

Thats brilliant thanks very much for you time, it has been much appreciated

By luck my idear was not too far away




now to manufacture the top bracket I intent to make this out of Stainless to give it better strength.


Doctor Derek Doctors - 3/6/11 at 09:30 AM

No Probs mate, I tried to base it on your initial projections to make something manufacturable.

Have fun building, I wish I could get some time to do some practical work rather than being in front of the desk.


Doctor Derek Doctors - 3/6/11 at 09:35 AM

quote:
Originally posted by Bare
Simply Try to mount the shocker unit as vertically as absolutely possible. ..
Save yourself a maths exercise to tell you exactly that


....or spend 30 mins doing some maths and find out that doing that would have given poor, inconsistent geometry.



[Edited on 3/6/11 by Doctor Derek Doctors]


Neville Jones - 3/6/11 at 10:00 AM

quote:
Originally posted by Bare

Simply Try to mount the shocker unit as vertically as absolutely possible. ..
Save yourself a maths exercise to tell you exactly that .
Wouldn't overly worry about structural 'niceities' as even the suspension pickup points on these Locost chassis are just daft/poor . The mount ears have been documented to fatigue off due to their poor placement and inadequate support.
Shock mount 'strength' is the least of your worries.

[Edited on 1/6/11 by Bare]


That statement above is about right on every count. What oriface DDD pulls his info from I don't know,( Wikipedia?or the same place as the ship's engineers?) but as the shock inclines from the vertical, once you get past about 10 degrees, things start to go way past non linear in the non desireable direction.

The locost setup puts some limits with the top wishbone design, and clearance for the coilover setup.

There are other ways of mounting the top shock bolt. I'll try and post a pic.

And yes, accepted practice among knowledgeable suspension designers would be to mount the shock as close to vertical as possible, with enough inward inclination to give a rising rate up to the point of max bump, where the shock would be 90 degrees to the wishbone.

Well, that's how the couple of designers I've met would do it.

Cheers,
Nev.


flak monkey - 3/6/11 at 10:11 AM

quote:
Originally posted by Neville Jones
quote:
Originally posted by Bare

Simply Try to mount the shocker unit as vertically as absolutely possible. ..
Save yourself a maths exercise to tell you exactly that .
Wouldn't overly worry about structural 'niceities' as even the suspension pickup points on these Locost chassis are just daft/poor . The mount ears have been documented to fatigue off due to their poor placement and inadequate support.
Shock mount 'strength' is the least of your worries.

[Edited on 1/6/11 by Bare]


That statement above is about right on every count. What oriface DDD pulls his info from I don't know,( Wikipedia?or the same place as the ship's engineers?) but as the shock inclines from the vertical, once you get past about 10 degrees, things start to go way past non linear in the non desireable direction.

The locost setup puts some limits with the top wishbone design, and clearance for the coilover setup.

There are other ways of mounting the top shock bolt. I'll try and post a pic.

And yes, accepted practice among knowledgeable suspension designers would be to mount the shock as close to vertical as possible, with enough inward inclination to give a rising rate up to the point of max bump, where the shock would be 90 degrees to the wishbone.

Well, that's how the couple of designers I've met would do it.

Cheers,
Nev.


Absolutely correct


Fred W B - 3/6/11 at 11:11 AM

quote:

to manufacture the top bracket I intent to make this out of Stainless to give it better strength.



Please don't make the brackets from stainless unless you intend to use the correct consumable to weld it to carbon steel.

Stainless is alloyed to give certain properties, if you dilute the alloys (by welding to carbon steel) you end up with the wrong structure and form a brittle rather than ductile weld.

When welding with 316 use 316 electrodes, when welding with 304 use 309 (best) or 316 electrodes.

You would be much better off to make the bracket in thicker carbon steel if you want it stronger.

Cheers

Fred W B.

[Edited on 3/6/11 by Fred W B]


paulf - 3/6/11 at 12:21 PM

quote:
Originally posted by Doctor Derek Doctors

In my conclusion mounting inside will give roughly a 5:10 installation ratio, mounting outside will give a 6:10 installation ratio thus outside would be stiffer with the same springs.


Hope that helps.

Are you sure that you got this the correct way around? surely the nearer vertical the damper is then the stiffer the effective spring rate.
Paul


mrwibble - 3/6/11 at 12:29 PM

i'm confused, as a leyman, DDD's proposal and numbers seem very convincing... but then i can't claim to understand what the numbers mean...

what was your job title again DDD?


Doctor Derek Doctors - 3/6/11 at 12:31 PM

quote:
Originally posted by paulf
quote:
Originally posted by Doctor Derek Doctors

In my conclusion mounting inside will give roughly a 5:10 installation ratio, mounting outside will give a 6:10 installation ratio thus outside would be stiffer with the same springs.


Hope that helps.

Are you sure that you got this the correct way around? surely the nearer vertical the damper is then the stiffer the effective spring rate.
Paul


Yes thats what it says. 6:10 (outside) is stiffer than 5:10 (inside).

Not the clearest sentence I have ever written I know.


Doctor Derek Doctors - 3/6/11 at 12:39 PM

quote:
Originally posted by Neville Jones
quote:
Originally posted by Bare

Simply Try to mount the shocker unit as vertically as absolutely possible. ..
Save yourself a maths exercise to tell you exactly that .
Wouldn't overly worry about structural 'niceities' as even the suspension pickup points on these Locost chassis are just daft/poor . The mount ears have been documented to fatigue off due to their poor placement and inadequate support.
Shock mount 'strength' is the least of your worries.

[Edited on 1/6/11 by Bare]


That statement above is about right on every count. What oriface DDD pulls his info from I don't know,( Wikipedia?or the same place as the ship's engineers?)



The numbers are pulled from TrolleyJacks stated suspension layout, put into a sketch of his supsension and the the wheel moved 10mm at a time in CATIA V5, no opinion, no expectation, just numbers from an unbiased CAD box.

The numbers are clearly listed for all to see, yes mounting on the outside gives a higher installation ratio but mounting on the inside gives a more constant rising rate and rate of rising rate. As clearly stated

I don't know what your problem is, you have provided no figures, no facts, just the same old BS heresay, you have added nothing positive and made no effort to even read the numbers. I on the other hand tried out the two options that TrolleyJack had open and simply provided a list of numbers back and said what I would do if it were my car. What TrolleyJack reads into those numbers and does is upto him/her (It's a neutral name)

Once again you just bring pointless negativity and zero help.

[Edited on 3/6/11 by Doctor Derek Doctors]


Doctor Derek Doctors - 3/6/11 at 12:45 PM

quote:
Originally posted by mrwibble
i'm confused, as a leyman, DDD's proposal and numbers seem very convincing... but then i can't claim to understand what the numbers mean...

what was your job title again DDD?


They aren't my numbers really, I just put Trolleyjacks suspension into a standard 2D layout of front suspension and then moved the wheel by 10mm at a time and wrote down the numbers that came out. they would have been the same had anyone done them

Job title? Well technically I'm listed as Mechanical Engineer / CATIA V5 Designer but at the moment there is stuff from Fluid and Thermo dynamics, Man' Eng and Material involved as I'm working on a major project for a future Formula 1 car.


mrwibble - 3/6/11 at 02:00 PM

quote:
Originally posted by Doctor Derek Doctors
quote:
Originally posted by mrwibble
i'm confused, as a leyman, DDD's proposal and numbers seem very convincing... but then i can't claim to understand what the numbers mean...

what was your job title again DDD?


They aren't my numbers really, I just put Trolleyjacks suspension into a standard 2D layout of front suspension and then moved the wheel by 10mm at a time and wrote down the numbers that came out. they would have been the same had anyone done them

Job title? Well technically I'm listed as Mechanical Engineer / CATIA V5 Designer but at the moment there is stuff from Fluid and Thermo dynamics, Man' Eng and Material involved as I'm working on a major project for a future Formula 1 car.


well i'm sure your a very clever man. i don't understand why someone has been so rude to you for trying to help someone with their build...


britishtrident - 3/6/11 at 04:13 PM

quote:
Originally posted by mrwibble
quote:
Originally posted by Doctor Derek Doctors
quote:
Originally posted by mrwibble
i'm confused, as a leyman, DDD's proposal and numbers seem very convincing... but then i can't claim to understand what the numbers mean...

what was your job title again DDD?


They aren't my numbers really, I just put Trolleyjacks suspension into a standard 2D layout of front suspension and then moved the wheel by 10mm at a time and wrote down the numbers that came out. they would have been the same had anyone done them

Job title? Well technically I'm listed as Mechanical Engineer / CATIA V5 Designer but at the moment there is stuff from Fluid and Thermo dynamics, Man' Eng and Material involved as I'm working on a major project for a future Formula 1 car.


well i'm sure your a very clever man. i don't understand why someone has been so rude to you for trying to help someone with their build...



+1


Doctor Derek Doctors - 3/6/11 at 04:57 PM

quote:
Originally posted by britishtrident
quote:
Originally posted by mrwibble
quote:
Originally posted by Doctor Derek Doctors
quote:
Originally posted by mrwibble
i'm confused, as a leyman, DDD's proposal and numbers seem very convincing... but then i can't claim to understand what the numbers mean...

what was your job title again DDD?


They aren't my numbers really, I just put Trolleyjacks suspension into a standard 2D layout of front suspension and then moved the wheel by 10mm at a time and wrote down the numbers that came out. they would have been the same had anyone done them

Job title? Well technically I'm listed as Mechanical Engineer / CATIA V5 Designer but at the moment there is stuff from Fluid and Thermo dynamics, Man' Eng and Material involved as I'm working on a major project for a future Formula 1 car.


well i'm sure your a very clever man. i don't understand why someone has been so rude to you for trying to help someone with their build...



+1


Cheers guys

Clever man? I wish! Just lucky when it comes to getting jobs I think (sshhhhhhhhh..... don't tell anyone I work with)


Neville Jones - 3/6/11 at 05:46 PM

DDD,
Considering that my work has involved motorsport design, and I have many years of experience, and my statements have been corroborated by at least two other bona fide engineers, then I have every right to question the extremely poor solution you have put forward to the gent asking the question.

To lay the coilover at the angle you suggest, will impart high loads into the top mount and wishbone, and also give undesireable behaviour to the system.

My explanation was quite simple, and states what is considered to be the accepted normal practice when designing a suspension system of this type. As I have done far too many times past.

Working in F1 doesn't make anyone a genius or expert, sometimes quite the opposite, and maybe more than a little lucky, as you pointed out.

And as I've stated many times, there is far too much wrong information on this forum, and the internet in general, and if I can stop just one person from coming to grief because of these misleading statements, then I can say I've done good.

Cheers,
Nev.

[Edited on 3/6/11 by Neville Jones]


procomp - 3/6/11 at 05:50 PM

Hi

One quick glance at that lot on the pics shows that worrying about a few mm of travel on the damper is the last of your worries. Id be way way more concerned about the camber gain problems your going to have with that layout on the pickup points of the arms. But as it stands simply move the base damper mount out as close to the bottom ball joint as is practical and mount the top mount out on the side of the chassis to gain as much damper travel per wheel travel as possible. You can use a basic ruler for that.

But then again assuming this is just a ROAD use car leave as is and stop worrying. It is still far better than whats available from some of the kit manufacturers with the top bracket in either position.

Cheers Matt


Trollyjack - 3/6/11 at 06:11 PM

Just read all the comments above and my head is now spinning

What to Do now ??????


Doctor Derek Doctors - 3/6/11 at 09:17 PM

quote:
Originally posted by Neville Jones
DDD,
Considering that my work has involved motorsport design, and I have many years of experience, and my statements have been corroborated by at least two other bona fide engineers, then I have every right to question the extremely poor solution you have put forward to the gent asking the question.

To lay the coilover at the angle you suggest, will impart high loads into the top mount and wishbone, and also give undesireable behaviour to the system.

My explanation was quite simple, and states what is considered to be the accepted normal practice when designing a suspension system of this type. As I have done far too many times past.

Working in F1 doesn't make anyone a genius or expert, sometimes quite the opposite, and maybe more than a little lucky, as you pointed out.

And as I've stated many times, there is far too much wrong information on this forum, and the internet in general, and if I can stop just one person from coming to grief because of these misleading statements, then I can say I've done good.

Cheers,
Nev.

[Edited on 3/6/11 by Neville Jones]


We're all waiting with bated breath for you to come out with anything other than petty negativity, rambling arguments and personal insults.... but then again that seems to be all you can come out with.

Seriously, get a life.


flak monkey - 3/6/11 at 09:34 PM

While Nev is not particularly tactful in his reply, his information is as well founded as it comes I am afraid. Coupled with Matts comment above.

With outboard shocks on a seven type car its nearly impossible to get the ideal set up as they sit at such a shallow angle. If you are that determined to get it as good as you can then start looking at inboard shocks and at better suspension geometry overall.


mark chandler - 3/6/11 at 10:16 PM

Shock on outside point, get the bone mount as far out as possible as recommended by Matt, personally I do not like forcing the wishbone to act as a lever, you just load up both ends anyway.

You do need to try and get more travel on that shock, its nearly closed in droop, this achieves this.

My qualifications are none BTW

This is the front of a winning Brabham for reference, okay its 30 years old so a load of old tat!




nb/ do not agree with all the nasty comments, pointless and rude on both sides.......

[Edited on 3/6/11 by mark chandler]


Neville Jones - 4/6/11 at 09:56 AM

quote:
Originally posted by Doctor Derek Doctors

Seriously, get a life.


I've got a sense of humour instead. You might get the same.

If yourself, and one or two other wannabe's who want to be engineers, stopped posting wildly incorrect and misleading, and sometimes downright dangerous information, then I wouldn't be wanting to correct same.

There's nothing wrong with being a cad jockey, or a ships engineer, be happy in that, and stop trying to be something you're not.

Research your subjects properly, and post from a base of knowledge and experience.

Cheers,
Nev.


JF - 4/6/11 at 11:08 AM

Nev, with all respect to all your wisdom and experience. Can you back your story up with numbers and proven examples? Like you say yourself... there is a lot of false information out there. Why should anyone assume your information is indeed right.

What went wrong with DDD's calculations? Instead of burning him to the ground, maybe you could give him some pointers where to look out for or some secret parameter we meager mortal humans don't know about.

And yeah maybe DDD doesn't present himself in the smartest way possible, and yeah he might often give wrong advice. But instead of just burning him to the ground, maybe you could help him a little so he might get it right next time. And help many people to the right information.

Because not only are you burning DDD down to the ground. Trolleyjack still doesn't know how he should do it.

Oh please allmighty Neville Jones, please share some of your infite wisdom with us lower humans.

PS: I for one would be interested to learn a bit more about this subject as I'm trying to work out a chassis design. Wich might get built someday, but till I get the chance to do so I enjoy doing some design work.

PSS: Yes you should read this post with your sense of humor turned on.....


britishtrident - 4/6/11 at 11:44 AM

When somebody shouts about how bright they are and denigrates others we know exactly what they are.

About 20 years back I used to share a lunch table with a guy who was later nominated for a Nobel prize he was very quiet and unassuming but he was the real deal.


Neville Jones - 4/6/11 at 04:25 PM

OK, sense of humour turned up and sarcasm alarm turned down..... And aware that a couple of wannabe's are gunning for me... ( If they had any sort of engineering education, then they would not be saying what they are, and would understand what I've said. Flakmonkey IS an engineer, thus his comments.)

Short of taking a lot of time to do a bundle of sketches, I'll try and explain in words. (A picture is worth a thousand words, I know, but anyway....)

Draw a bottom wishbone, inner pivots and coilover mount at the outer end.

Now draw a coilover at the outer pivot, at 90degrees to the wishbone.
Any load put upwards into the coilover, from the wishbone, at this point will be directly along the axis of the coilover, so the full rate of the spring can react to it.


Now draw the coilover at an inclination inwards towards the chassis, with the bottom pivot still in the same place on the wishbone.


A vertical load at the coilover pivot is reacted only by the vertical component of the coilover. As it is inclined, the coilover will be pushing outwards on the wishbone as well. This should be fairly easy to see. As the top of the coilover is now inclined even more, the outward push increases in relation to the sine of the inclination angle, and the vertical decreases with the cosine. Thus, a heavier spring is needed to give the same vertical force at the pivot, as would be required if the coilover was vertical or close to it.

At 10 degrees inclination, the cosine is 0.9848, the sine is 0.1736. So, if the spring in the coilover is pushing along its axis with 100 lb force, then 17.36 lb force wqould be acting outwards, and 98.48 lb force would be acting vertically at the wishbone pivot. At 15 degrees, these figures go to 96.59, and 25,88. 20 degrees then gives 93.96 and 34.2. If you graphed these, you would find that they would follow a sinusoidal graph. Up to about 20 degrees, things are fairly linear, but most would prefer to keep the angles as low as possible. Once you get past 20 degrees, the vertical component decreases at an increasing rate, and the horizontal increases.

That about sums it up.

So, if you put the coilover at 90 degrees to the wishbone when at the max travel up of the wishbone, the force seen by the wishbone will be increasing as the cosine decreases. Rising rate. Quite simple.

With a Locost book situation, it is difficult to achieve what I've just outlined. However, the top of the coilover should be as near to vertical as is possible geometrically. This will probably put the top mount some distance from the chassis, and needs to be dealt with by adding an extra tube laterally, attached to the chassis and picking up the top coilover mounts. This would do the same job as what is commonly known as a 'strut brace' on tin tops.


Clear as mud!

Cheers,
Nev.







[Edited on 4/6/11 by Neville Jones]


JF - 4/6/11 at 09:26 PM

Thank you for this constructive post Nev. I'd offer you a drink if we were in a bar. But sadly I don't feel like swimming across the pont tonight

It's hard to find good sollid information on the net. And well I'm no engineer, nor do I have any experience designing cars or even parts for cars. Although I'm working on both. I'm a mechanic who works a lot with metal, hydraulics and electronics but not an expert on any field. And won't claim that I am either.

But I believe in doing things right, certainly when it comes to suspension etc. And therefor it's always nice to obtain some knowledge from someone who has been there and actually knows a thing or 2 about it.

So might I ask you, and others offcourse, to atleast supply the TS with the info he needs. Preferably with some sort of fundation, as there is way to much bogus info out there. If besides that you feel the need to burn anyone who gives a wrong answer... well have fun...


mrwibble - 6/6/11 at 06:41 AM

out of interest, for a coil over mounted inboard with rocker arms, what would be the optimum, 45 degrees?

[Edited on 6/6/11 by mrwibble]