MG-Cars.net

Welcome to our resource for MG Car Information.

Recommendations

Parts

MG parts spares and accessories are available for MG T Series (TA, MG TB, MG TC, MG TD, MG TF), Magnette, MGA, Twin cam, MGB, MGBGT, MGC, MGC GT, MG Midget, Sprite and other MG models from British car spares company LBCarCo.

MG Midget and Sprite Technical - Mumford - Roll Centre Height

Hi,

Anyone out there know the height of the roll centre they have / are using with their Mumford setups.

Cheers

Spencer
S Deakin

Just about the bottom of the diff.


Brad (Sprite IV 1380)

Hi Brad,

Thanks for replying, which intersections are you using to derive the RC height?

Cheers

Spencer
S Deakin

Its the imaginary line where the two rose jointed arms meet in the centre of the Mumford, thats the new RC, mines about an inch up from the bottom of the diff.
Brad (Sprite IV 1380)

Brad,

So just to confirm, I have added lines to your image

Cheers




S Deakin

Yup, that's it. When its fitted the angle of the arms is not quite so sharp.
Brad (Sprite IV 1380)

Cheers Brad,

I have been looking at the kinematic behaviour of various Watts configurations, the one shown in the image is a body mounted low roll centre version.

I am going to do the same with the Mumford for comparison to understand the differences in kinematic behaviour.

Cheers

Spencer




S Deakin

I sometimes read threads like this and realise that I havn't really got a clue....
John Collinson

Have you got Race car Vehicle Dynamics? doesn't say much about the Mumford but goes into depth on Watts, WOB links, etc..
Brad (Sprite IV 1380)

Yes I've got RCVD, Nixon, Gillespie and Matjinsky but none of them really touch on Mumford other than acknowledging its existance.

The Watts can be quite sensitive to link positioning as regards the lateral axle movement through bump and rebound, I've never been sure if the Mumford is a genuine straight line mechanism or like the Watts as a near straight line mechanism if everything is correctly chosen.

I guess I'll find out soon

Cheers

S Deakin

Good luck. Are you making your own Mumford or just playing around with computer simulations?
Brad (Sprite IV 1380)

I will make whichever looks best as regards it's behaviour. The Watts is simpler to make, but if the Mumford has enough advantages I will take that route.

The obvious area where the Mumford beats the Watts is the RC height as I can only get the RC down 40 - 50mm from the axle centre line (without compromising ground clearance) with Watts which is still approximately 200mm.

The Mumford is pretty well free in terms of RC height, my immediate concern is the axle attachment points if they are too high then it is difficult to keep them stiffly attached to the axle. But the simulations will show how high they need to be to get a RC in the region of 100 - 150mm.
S Deakin

I bent my first axle attachments, got a set made up out of EN40 & they seem pretty solid now. I basically went through the bump stop brackets, bit of a bugger to drill past the bump stop strap bolt unless you use a parallel peanut.
Brad (Sprite IV 1380)

Spencer,
I've also got a set of alloy spring pads rather than poly / rubber to stop the axle moving laterally on the mounts.
Brad (Sprite IV 1380)

I read recently in "Automotive Engineer" July/August 2009, that the new Astra is to be fitted with a Watts linkage. Centre rocker mounted to the chassis also.
David Billington

Brad,

I am planning on 4 linking the axle as far as longitudinal location goes as I want to be able to lower the car without using lowering blocks.

I have looked at the possibility of moving the forward leaf spring mount up in the body (to lower the car) as adjusting the shackle end will effect the wheel rate as well.

But 4 linking seems preferable (if I ignore the minor details of making them and the body mounts, not to mention where the coilovers will reside) :)

It all looks easy in CAD lets see if the enthusiasm is still there when I'm in a cold garage with an angle grinder for company

David,

It would be interesting to see GM's implementation of that, especially without losing boot space.

Cheers

Spencer
S Deakin

Spencer,

http://www.youtube.com/watch?v=OsU5SnWvqG8

http://www.richardaucock.com/astra-suspension-by-automotive-engineer/
David Billington

Plumbers vans just got faster.
Brad (Sprite IV 1380)

Brad,

What have you used for bearings in the two pivot points or are you just running on the bolt shank?

Spencer
S Deakin

I bushed them with bronze.
Brad (Sprite IV 1380)

Very interesting!

>>Which Harder duly did, despite nobody doing it on a production car before

There have been Watt links on all sorts of cars, including Scimitars... or it it a unique combination of beam axle and Watt? Then I'd say what's a live axle if not a rather strong beam...(!)

Anyway, very interested in the simulation Watt vs Mumford.

A
Anthony Cutler

Anthony,

I was wondering about how to interpret that sentence also, I was thinking Rover SD1, but assumed in the end he meant the combination of torsion beam and Watts Linkage.
David Billington

Is that linkage the one available from MOSS?
Ron Koenig

Ron,
it's one i made.
Brad (Sprite IV 1380)

Impressive computer machining. Looks suitable for mass production. You should sell these Brad.
Ron Koenig

Thanks Ron,
i'm pleased with it, i designed it on Google Sketchup, then made it by hand the old fashioned way. It would be easy enough to water cut.
Brad (Sprite IV 1380)

Spencer,
can you drop me a mail.
brad_r@fastmail.fm
Brad (Sprite IV 1380)

Hi Brad,

You have Mail

Cheers

Spencer
S Deakin

so looking at that astra vid got me thinking.
if you want to build your own watts link just get some astra parts from the junk yard (they always seem to chrash soon) and adapt to fit!
Onno Könemann

So have you Spencer, enjoy.
Brad (Sprite IV 1380)

Brad,

I have replied

Cheers

S Deakin

beginning to sound like "Round the Horne" in here lately

bill sdgpm

Bill,
i was sending Spencer some diagrams of my Mk3 version, if it works out then i will post them on here for everybody, but at the moment its just a 3D design.


Brad (Sprite IV 1380)

very cool Brad

looks nice

but

:)




maybe you are too young



(however you are scarily good with Sketchit/SketchUp, I just cant get past making tubes and swimming pools and long long walls)
bill sdgpm

Bill,
i'm neigh young, scary maybe, but neigh young anymore.
Brad (Sprite IV 1380)

Spencer, Brad,

V. interesting thread.

Some thing not yet discussed (unless I missed it) is that my spacial awareness tells me, the lower you set the roll centre the more wheel arch lateral clearance you are going to need, as the greater the height from the roll centre to the top of the tire, then the greater the lateral movement of the wheel arch adjacent to the top of the tyre. (Can sketch if need be).

Also the lower the roll centre the more more roll torque is generated as this is C of G above roll centre, and hence stronger spring rates are required for equivalent body roll angles for given lateral G.

Also I thought I had read that a slight declining roll centre axis from rear roll centre to front roll centre was preferable. Hence if you set the rear low, you would also have the set the front lower. Any thoughts on this?

James
James Eastwood

Hi James,

My reason for creating the kinematic 'models' was specifically to look at rear axle lateral displacement, roll centre migration etc...

The axle lateral displacement of the body mounted Watts linkage (the example I modelled lowered the roll centre about 40mm)was not affected by the roll centre (RC) height, assuming the behaviour of the Mumford is as good if not superior to the Watts then I would say no there is no discernible lateral movement (due to the kinematics anyway)as they are used because they are (near) straight line mechanisms specifically to stop / reduce lateral axle movement.

Yes if the the rear RC is lowered and everything else left the same then yes more body roll will result as the overall vehicle roll resistance is less. Stiffer springs at the rear could restore the level of roll resistance or a stiffer FARB / springs would do the same but also change the roll moment distribution.

The vehicle will be lowered at the front as well which has the effect of lowering the front roll centre as well, yes I agree that the rear RC should be higher than the front RC (and this will be part of my considerations on how low I take the rear RC.) Whichever rear lateral control (Watts or Mumford)I ultimately use will be adjustable for RC height in or around the achieveable front RC height.

Cheers
S Deakin

Funny you should mention that James, i ended up lowering the front by 3/4". I also have an uprated ARB with poly mounts (saying that it might even be solid mounts, i haven't seen the car for that long, i can't remember)& rose jointed linkages.
Brad (Sprite IV 1380)

Some simulations carried out so far:

Did some work on optimizing a body mounted Watts setup, firstly wanted to look at the effects of the centre to centre length of the centre pivot, baseline was 100mm but I also tried 50mm.

And also the length of the lateral Watts links, my baseline setup has Watts links pretty well as long as they can be - they reach up to the leaf spring positions, then I halved the length of them.

The optimized Watts gives around 0.1mm of kinematic lateral displacement for 50mm of rear axle travel, however it's very easy to end up with a lot of lateral movement if the geometry of the links isn't right.




S Deakin

Spencer,
what program are you using for this? Can you run my Mumford MK3 design thorugh it?

cheers
Brad
Brad (Sprite IV 1380)

Hi Brad,

It's ADAMS Multibody Body System

http://www.mscsoftware.com/Contents/Products/CAE-Tools/Adams.aspx

No problem as soon as the Mumford system is fully modelled I will put the 'MK3' through it. I will just need the X,Y,Z position of the points on the axle that you connect to, the static laden radius (slr) of the tyres you use the height of some datum on your Mumford system to the ground.

The Watts and the Panhard are pretty straight forward to model so got done first, the Mumford just requires a little more effort for things like the roll centre change with ride height motion etc...

I have also modelled a Panhard system, the following graph was the absolute best I could get from the Panhard, shorter link lengths or getting the pickups in the wrong place can see 7 - 8mm of lateral movement for 50mm of bump / rebound and can be very assymetric.


50mm of vertical axle travel gives about 1.2 - 1.3mm of lateral displacement.

Cheers



S Deakin

Spencer,
i'm changing the rear springs at the weekend, so will take some measurements then.
cheers
Brad
Brad (Sprite IV 1380)

Also looked at 4 link (4 links not including the lateral restraints)

The midget will only allow the use of fairly short longitudinal links which can effect several things, wheel recession (fore / aft movement of the axle) and the anti squat behaviour changes more with ride height than longer links would.

Just looked at wheel recession so far:

Based on a lower link which goes to the front leaf spring mount and the rear to the axle centre line (could be longer but not much more)there is quite a lot of wheel recession 3.5mm for 50mm of bump travel.

This graph shows the effect of making the longitudinal links 50mm longer (black curve)it reduces it by 0.5 - 1.0mm

Sorry I've switched the axes to further confuse!!

Cheers




S Deakin

Must admit you've lost me on wheel recession, anything to do with Gordon brown?
Brad (Sprite IV 1380)

Gordon Brown? Who's he?

Didn't he design the Mclaren F1 GTR or was that his brother Murray?

Recession is how far the rear axle (in this case) move backwards / forwards with bump / rebound.

In a bad recession the axle doesn't move at all coz you can't afford fuel ;)

S Deakin

Hi Spencer,
the dimensions for the axle pick ups are:

Height 60mm above the centre of the axle.
Lateral 100mm behind the axle centre line.
Inboard 200mm from the face of the half shaft.

I make the imaginary roll centre 1 inch above the bottom of the diff.
cheers
Brad
Brad (Sprite IV 1380)

to go back to the original post

The ideal "daily use" placement of the imaginary roll centre is about 1 to 1.5" below the centreline of the halfshafts when the axle is in a neutral position with someone driving.

so when you have her up on stands and level get a mate to sit in the drivers seat then offer all the stuff into place.

You can (and i have been doing so for days) do as many calculations as are available to try and work it out, but nothing really beats experience and trial and error.
PJ Moore

you know what. iv not got a bloody clue what you lot are on about. im gona go away and feel inadequate and unmanly now. thanks alot!!!
roy judd

Roy,
I like you Psychedelic number plate, sh*gtastic dude as Austin Powers would say.

http://www.imdb.com/title/tt0118655/quotes

it's all about making the car roll less, have a look at this site.

http://www.billzilla.org/newrcar.htm
Brad (Sprite IV 1380)

Brad,

I will get back to you if I need any more dimensions.

I have measured the hard points on the standard car now so just need to sanity check them before doing any more with the kineamtic model.

Cheers,

Spencer
S Deakin

Ok Spencer,
it will be interesting to see how the Mumford fairs on the kinematic model,

cheers
Brad
Brad (Sprite IV 1380)

Spencer,

After speaking at great length with a well known chassis builder (many Class A, B and C Midgets under his belt now) he has in short told me to stay away from the RTL systems available and stick with a well sorted Watts linkage.

It may not be as gucci looking, but its definitely a function over fashion item in both of our opinions. With the amount of vertical axle movement in a midget is miniscule in comparison to that of the Mk2 Escorts, and they happily and successfully run watts linkages in high horsepower applications, on all sorts of terrain without problem.

I guess i feel its a case of keep it simple when it comes to these things. Less to go wrong and an easy life maintaining it.

Each to their own though, if its purely an engineering experiment to show your ability, then i am definitely up for watching that. If i had a choice i would too :D
PJ Moore

Pete,

Motivation is purely lower rear roll centre height whilst still having sensible ground clearance - not possible with a body mounted Watts.

Problem with Mumford is knowing what to do with it - plenty to get wrong and make the car worse - still more than possible to do that what a Watts as well.

I am also looking a 4 link Satchell configuration - doesn't get much simpler or less 'flash'

I want what is lightest (usually cheapest - fewer bits) and does what I want.

Mumford - Fashion over function! Arthur Mallock was never known for using bits for the bling factor!

Cheers

Spencer
S Deakin

Gotcha,

Well, i will be keeping a close eye anyway for personal interest.

Im personally doing the body mounted watts linkage on an adjustable pivot to begin with. Then shift it about until it feels right when driving. Once i have determined where i want it, to suit me, ill finish fabricating the final mount.

I dont mind all the math, the sketching, etc etc, but as i mentioned either above in this thread or in another thread, all the math in the world wont tell you how it will feel in a car that was never built straight from the factory anyway

I'll get a link up to my site when i get the new axle to start playing with ;)
PJ Moore

Get a link up anyway - we're all nosey!

Out of interest are you coilover or leaf sprung
S Deakin

my current design caters for coilovers and 4-link

but in the mean time i will be running the coilover shock bodies without the springs and maintaining the leaf springs.

I need to finish my doodles and final designs then get a friend who is rather handy with the old CAD programs to make me the drawings.

Then its a case of welding everything together and cutting roads to length then get everything plated so it looks pretty and doesnt rust.

This will be the first time I will be doing anything to my own car, so there is a lack of fear in delving into certain unknown areas. There is always doubt when you are just "trying" something for a customer or a friend.

PJ Moore

Pete,

I would suggest making the pickup points on the axle adjustable as well if you plan to experiment with the height of the Watts pivot on the body.

If you get the relationship of the two wrong the straight line behaviour of the Watts dissapears rapidly and the axle moves laterally sooner than you think which will start to bind the system.

As far as there being an "ideal" placement for the kinematic roll centre (1 1/2" below the axle centre)this gives a roll centre height of approx 215mm static. My target is someway below this (with a range of adjustment,) the springs will need to be revised to suit the reduction in roll moment distribution but can still easily give sensible wheel rates.

Cheers

Spencer

S Deakin

Spencer

Thanks for that, i am still lacking in underpinning knowledge when it comes to individual items within the setup of the rear axle. All my experience prior to midgets has been FWD race applications where modifications didnt go much further than setting up of "off the shelf" parts that were already proven to work.

This is handy though and does at least give me grounds to build up from.

Coming from an engineering background I will pick 90% of it up immediately, but as with anything there will always be that 5 or 10% there to bite you in the ass.

Im off for some more study time now, although i do intend to sleep tonight as last night was spent looking for axles, diffs, brakes and nice trick bits that i dont need :D

Pete
PJ Moore

oh spencer

http://www.shedfive.com

im still working on it and getting stuff up in some sort of logical order ;)

PeterJMoore

Spencer,
when you have a moment can you run my Mumford through your program, it will be interesting to see if i got it right or totally wrong.

cheers
Brad
Brad (Sprite IV 1380)

Looks like the ex-DC Midget (GFA) - very nicely modded car; before DC went FWD!

A
Anthony Cutler

Anthony, it was a car with competition history in England, I believe through sprints, hillclimbs and similar events. unfortunately I cannot remember the owners name, but when Will returns i will ask him to clarify.

The spec of the car was lovely on paper, but almost every single aspect of the car has now undergone change and modification as a lot of the parts and add-ons just were not suitable.


Pete
PeterJMoore

This thread was discussed between 14/02/2010 and 27/03/2010

MG Midget and Sprite Technical index

This thread is from the archive. The Live MG Midget and Sprite Technical BBS is active now.