Lancia Motor Club

Lancia APRILIA REAR SUSPENSION

Further to the postings back in Feb which I have just seen (being a new broadband subscriber) it is evident that the perception of how and why the Aprilia rear suspension works as well as it does is quite personal to the observer!! I would like to try to offer some unifying clarity.

Firstly, to be clear about the mechanics of the system, the trailing arms give excellent wheel location and without doubt account for the delightful steering precision,in conjunction with the sliding pillars at the front of course.
The springing comprises the torsion bars and transverse leaf working in parallel.
Considering each system separately the torsion bars are anchored to the body at their inboard ends and on their own could provide, on the face of it, adequate support for the full rear body weight. The Citroen of the period uses torsion bars at the rear in a very similar manner although with a beam axle. In the Lancia however the torsion bars are not wound up to carry the weight which is mainly supported by the transverse leaf. This spring is hung from the rear of the swinging arms, on its cables at either end (effectively hung from the wheel hubs) and the car sits on the middle of it,via the differential casing. The mounting (silent-block bushes) has no significant rotational stiffness so that if the torsion bars were not installed the body could rock freely. Thus the transverse leaf installation provides no roll stiffness at all
 Without the leaf spring, i.e. a la Citroen, the sizing of the torsion bars would be determined by the car weight and the required travel as is the case for the front springs. This would lead to an intrinsic degree of roll stiffness, again as is the case at the front. (This roll stiffness is not high because of the use of torsion bars. Any form of spring applied independently to the trailing arms to meet the same design criteria would give the same roll resistance.) If this intrinsic roll stiffness is considered to be too high the introduction of the leaf spring to carry some or all of the weight allows less stiff torsion bars to be used and the roll stiffness to be reduced. The presence and action of the transverse spring does not in itself lower the roll stiffness.

With reference to the concept of the suspension system influencing weight transfer during cornering this is only valid if the action of the front suspension is also considered. Overall if the car is cornering at a particular speed there will be a specific level of centrifugal force to be resisted and because the resistance is at ground level, at the tyre contacts, an overturning moment that has to be reacted. This reaction is evidenced as weight transfer from the inside to the outside wheels to a degree to exactly match the overturning moment.   
Whilst the design cannot alter the overall weight shift it can influence the amount of roll and the way its control is distributed between the front and rear wheels. If the suspension roll stiffness balance is not right the “soft” axle tried to allow more roll that the stiff one and the inside wheel on the stiff axle tends to lift off. If this is the back end it obviously leads to a loss of traction. If we start from a position where roll stiffness is equal at the front and the rear the rear inside wheel still lifts first because the turning radius is smaller for the rear axle so the centrifugal load, overturning moment, and hence weight shift is higher.  Hence the need for a reduction in roll stiffness at the back because as Neil says it is hard to  install an antiroll bar at the sliding pillar end (not impossible though) Comparing this situation to the Citroen I suspect they got away without these rear end complications because the front end is heavier so the suspension and anti roll characteristics would be relatively stiffer to start with.  Also with FWD it doesnt matter so much if the rear inside wheel does lift.

My personal view is that Lancia started out without the transverse spring (a la Citroen which preceeded it by a couple of years) but found the inside rear wheel hopping around on test and put in the transverse spring to get over the problem.  It also enabled them to use much softer torsion bars and run them at much lower stress.  Citroen are recorded as having trouble with torsion bars breaking on early cars.  The Lancia transverse spring certainly has a 'bolt-on extra' look about it which is not at all in keeping with the usual Lancia purity of form.


A

Great explanation which is perfectly correct in my view. I'm not sure I agree with the "bolt on extra comment" to be honest but it is a pain to work on in any case.If it was the intention to reduce the work required of the torsion bars and "down engineer" them, they went too far.The torsion bars break and more importantly the splines give up at the outside end and rotate in the arms.The torsion bars act against the direction of work of the spring and so define the ride height in effect . They don't really carry the vehicle and can only offer a relative torsional resistance depending on which spline you set them on.Without the transverse spring you would have to set the bars much more vertically and let the car sag on them in effect to create a sprung ride....If the splines dislocate you see the car move upwards under the action of the  transverse spring.On many cars they are welded in position.

I agree they overdid it. To reinstall the bars after my last axle overhaul (about 30 years ago!) I had to load up the boot to reach the neutral position. They probably break in fatigue because they see stress cycles from plus to minus as opposed to being loaded the same way all the time like most springs are. This + to - action, together with some corrosion as water can seep in, is probably what does for the splines.

Years ago (in about 1963 or 4) I had a very tidy second series Aprilia which went exptemely well, EXCEPT that the torsion bars had been welded into their sockets - and not into the correct splines. So the car had a built-in lean to the right, which was great for going round roundabouts, but not so good for exiting said roundabout! There were few cars that could keep up on right hand bends. I didn't get around to mending it, as the Maclagan Augusta came along in the nick of time, so the Aprilia was sold. I haven't owned another Aprilia since.
 Regards, John

I was driving my Aprilia to the MOT centre when there was a bang from the rear.  "Big pothole" I thought but, a few hundred yards further on I decided to get out and have a look.  There was a very slight list to port but looking underneath disclosed that the left hand wire harness linking the spring end to the suspension arm had snapped.   The attitude of the car would seem to bely the idea that the weight of the car is taken principally by the spring!

Don't want to worry you but is it a combination of circumstances perhaps? the bang sounds more like the torsion bar breaking and the steel wire then broke having to work alone? Let's hope not.

I have always believed that if the cable broke, the car turned turtle! Evidently not!

No, torsion bars are fine and I have removed both the broken and the intact wires and tied the spring ends with straps to prevent them escaping.  I wonder if the shock absorber lever had broken as well, the spring end may have contacted the road with some violence.  Not enough to turn the car over but enough to have graunched.  Omicron are currently fitting new wires and repairing the bearings as well, they say it is "one of the things they do" so maybe not so uncommon.

David,

It sounds like you might have been lucky. When something like this happened to me a thousand years ago, the spring-end dug into the road and ripped the whole spring off from the diff, leaving only the long-leaf attached to the cable at the other end and all the other bits strewn down the road… The rear wheels sank up into the rear wings, as you might expect and the wings had to be removed to enable me to limp the car on to a place of sanctuary.

Good luck with repair. More on the genius design of the Aprilia rear-suspension to follow.

Colin

I have now replaced the cables with the recon units from Omicron.  Photos show how I did it.  Note the belt and braces nylon webbing - the power of the spring is intimidating to say the least.  Although I took off the unbroken original cable on one side using only tha jack and webbing, by the time I came to replace it with the new the webbing had stretched and the jack plus the weight of the car was insufficient, hence the spring compressor.

When this thread was active back in February I held back hoping that I would be able to find an old article by Paul Frere, which I had referred to earlier. This was a technical appraisal by a knowledgeable writer and had I found it, it would I think have substantiated the claim for the Aprilia rear suspension being the work of genius. Ironically about the same time there were reports of Paul’s death at the age of 91 – see this link to an obituary: http://www.guardian.co.uk/sport/2008/apr/02/motorsports.france 

Paul was one of the Lancia D50 drivers and a Le Mans winner - I am pleased to see Jack Romano is to publish an article about Frere in the November issue of Viva Lancia!

Anyway, the article never turned up, but all is not lost because I have found another appraisal by one Ted Eves, which was published in the Autocar in April 1984. This is a bit too long to post here on the Forum and I will send it to Jack with the thought that it should be published in Viva Lancia! I am indebted to Grant Gibson for guiding me to this article and for his most interesting thoughts on the subject, which were prompted by me telling him that there were folks on the Forum who were getting critical about what they saw as the ‘delicate’ nature of the Aprilia irs. Grant’s comments are a delight and I will try to get his agreement to publish them.

Meanwhile, let me add a few more comments of my own:

First, I simply cannot accept the idea that Lancia was trying for full springing by torsion bars and had to put in the transverse spring because of shortcomings in that approach. It is inconceivable that such short bars could have coped with such loads - they would have been so rigid as to be too inflexible. Other fully torsion bar systems, such as VW, Renault and possible Citroen use much longer bars, which are frequently longitudinal (fore and aft), or in the case of the Renault 14 run the full width of the car. Had Lancia been intent on fully torsional suspension, they would have used coil springs, not short bars. Remember, the springing action of coil springs is torsional in nature across the cross section of the spring wire, i.e. they act just like very long torsion bars, unlike leaf springs, which have a conventional bending action.

Then, the idea that some torsion bars had to be welded in place does not invalidate the approach. By the 1960s these cars were well beyond their design life and will, at least in the UK, most likely have been submitted to some bodge maintenance. To find some of them that have been welded up after 60 years should be no surprise.

Finally, the issue of what constitutes a ‘neutral position’ for the torsion bars also needs thinking about. Clearly the neutral position will vary with the static load that is applied to the rear end of the car – it can’t be neutral in all circumstances. The Aprilia was designed to take up to three people in the back and it should be no surprise that the rear of the car needs to be loaded when re-fitting the transverse spring – that is after the torsion bars have been fitted to their correct datum marks. And remember the very clever vernier adjustment that made it possible to do this with some precision.

When I first faced this loading problem I contemplated asking the then mother-in-law and wife to climb into the back, but I recall stacking paving slabs in the open boot to be a better alternative. Thinking about it now I remember the bigger problem was that I didn’t have a trolley-jack then (nobody did!) and the job required some deft work with a selection of screw jacks.

More anon, but probably in Viva Lancia! where the subject should get a wider readership, which it deserves.

If anyone wants to see the Ted Eves article, let me know and I’ll email you a scan.

Colin

And here below is a John Maltby photo from 1964 of Aprilias in the snow. This is here because it’s lovely and also because it enables me to track just how few (or how many!) members read these posts and bother to open images!

Hi Colin,
Would you email me a copy of the article please.

Regards
Brian
8227 8)

I wonder of the tail on the left belongs to NS 1692 who came from JM in 1970 to my tender care

David,

I am fairly certain the front Aprilia is that of the late Rodney Wilkinson, with registration number EGH 335. It was at the Covent Garden event a year ago fully restored by its new owner, but still in the same colour that Rodney had it. See below.

Colin

I assume thats not an original colour Colin ? Does anyone know in which colours the aprilia was available when new, I think I have seen every possible variation over time but I'm guessing the original choice was limited.

 William, from the excellent publication by Lancia to celebrate the Aprilia's 50th anniversary the colours are listed as follows:-
Model 238 (1st series) Grigio, Blu, Nero, Granada, Verde, also for Lusso model colours Metallic (although not stated what colours!)
Model 239 (Chassis) Blu Lancia, Grigio chiaro, Beige chiaro.
Model 438 (2nd series) Grigio metallic, Blu, Nero, Beige, Verde, Bruno chiaro, Metallic scuro, Bruno, Blu metallic, Grigio acciaio, Grigio chiaro, Nocciola, Beige, Amaranto.
Model 439 (2nd series chassis) grigio metallic, Blu.
There is also a full list of which interior colour matches the paint colour, but if I include all of these I will be here all night!!
Regards Ade.

gosh! (does anybody still say gosh?), i didn't expect such a range of choice and perhaps shortsightedly, I always thought the metallic ones I saw were not original colours...

what form of publication is this that you refer to Ade, leaflet, book, magazine?

It is a soft bound Book/magazine produced by the Lancia Club with help from Lancia with a collection of articals and technical data, 60 pages, silver cover with 'Aprilia' and Lancia logo on the front. Mine was sent to me from Lancia when I enquired about details of my Aprilia in about 1988, I've not seen another..........and this one is NOT for sale!!!!!!

And I was going to offer 25000 pounds....... ;)

   The book Ade has sounds intregueing. I have the same data in a 44 page booklet called the LMC Journal for spring 1969.There is a John Maltby photo of a standad Aprilia on the front.Editor at that time was Nigel Trow.Text is in english!
   Going back to rear suspension matters I previously suggested that employing the transverse leaf spring enabled the aprilia designers to use much softer torsion bars.I have now measured up the various components and made some first order estimates, using very basic theory, of the actual spring rates.
   I found that the leaf spring has a rate of about 145 lb/in (loaded in the centre and supported at the ends).Because of the leverage effect due to  the connection points, via the cables, being about half way along the trailing arms,when installed in the car this gives an overall stiffness for the back end (bounce stiffness) of only 43 lb/in.
   My calcs. show that the torsion bars give a much higher rate of bounce stiffness, viz. 317 lb/in, i.e. 7.5 times stiffer than the leaf. Total rear end bounce stiffness is thus 43+317=  360 lb/in.---(It is a good job I am not a green-grocer or I could be arrested for using these Imperial units)----If 360 lb/in was the design target the torsion bars on their own would only need to be 14% stiffer, i.e. 4.5% thicker,which is a somewhat smaller difference than I expected.
   Checking with the drivers handbook confirms that the system is set up "by the factory" with the bars in a neutral state, although it is not specified what weight the car should be carrying. Experience seems to show it should be well loaded. Thus when travelling light the bars and the leaf are working in opposition but when load is increased the torsion bars go through the neutral position and then start assisting with the load carrying as previously described. It is certainly true that with this system the stress level in the torsion bars is very much lower than would be the case if they were carrying all the weight.
It would be an interesting experiament to set up the rear end of the car without the transverse spring and see how it behaved. Because they are so stiff it would only be necessary to set the bars around by 17 degrees (two splines at the inner ends.There are 42 and 48 at the outboard end to give the vernier ajustment capability referred to by Colin M. They can be set to within approx. one degree which gives a variation in ride height of about 1/6" or 4 mm.)
   Incidentally Colin I am allmost certain that the Light 15 Citroen rear torsion bars are anchored in the middle. I know the fronts were longer and ran longitudinally and as you say  Renaults ran the full width at the back---giving rise to unequal wheelbase distances between the two sides on the R16 and R4 and maybe others-----something I am sure Vincenzo would never have countenanced! With short bars it is easy enough to get the required spring rate but at the penalty of high stress levels.
   Overall however the excellent handling of the Aprilia is I am sure down to the well engineered wheel location rather than the choice of springing.
   Finally (if you are still awake!) --with reference to D.W.'s observation of a slight list to port when his nearside cable broke---in this state most of the near-side weight of the car is still supported by the spring which is now bearing down directly onto the road.It has to move down about 8" before striking the road but even after this degree of "relaxation" it is still loaded enough to support the car. Acting directly the force onto the road is roughly double what it is when acting through the road wheel via the leverage system of the swing arm.

...except that the spring was not resting on the road, it was hanging on the end of the friction shock absorber and so still attached to the car but the lower end displaced by about 6 inches.  When I replaced the cables I had the springs tied to the torsion bars by nylon tapes and the attitude of the car was much as usual despite the tapes stretching to, again, some 4 to 6 inches displacement.  All is well now!

   I had not picked up on that mod which I am sure saved you quite a bit of grief. The standard attachment point for the shock-absorber arm is directly onto the suspension trailing arm just below the bump stop platform.
  It is interesting that we now have two instances of rear suspension cable breakage reported on this forum and in neither case did the legendary barrel-roll occur. I wonder if your car was deiberately modified to avoid this.
  Glad all is well now.

   Me again!   Two further points of clarification.

                1.   Re-reading Colins latest thoughts the re-assembly procedure that requires the boot to be loaded to reach the neutral position of the torsion bars involves installing the leaf spring first. The load is to compress the leaf, with the wheels on the ground, and then the bars are rotated to find the position where they will just slide in.
                     If the bars were inserted first one would need to know precisely what angle to set the trailing arms to in their free state before inserting the bars. The leaf spring could then be fitted (using the special clamp) without having to load up the car.

               2.   With the leaf spring trapped by the shock-absorber as I now understand was the case for DW's cable failure the spring load is "locked in" and that side of the car is fully supported by the torsion bar alone. The modest list he observed shows the capability of the torsion bar if it is put to the task.

I actually had a choice of two possible mountings.  As my hydraulic mediated shockers which mounted on the side were perished I fitted the dry ones which mount to the end of the spring.  Lucky!!

Ben,

I think I explained it badly. The assembly operation I was recalling was with a fully relaxed transverse spring – like it was a semicircle. This required me to raise the rear of the car high off the ground with blocks under the sills. Before offering up the spring the angle of the trailing arms were set by reference to datum marks. I forget what these were, but there was a measure whereby the arms (without the bars) were set to some set position so that the distance from the top of the hub flange to some point under the rear wing ( I half remember a ‘blue plate’ with a simple horizontal line on it!) was set and equal both sides. Then the bars were fitted, using the vernier options to get the best fit.

Then it was a matter of jacking up one end of the leaf spring at a time to get it to a point where the end fittings, with the cables could be bolted in. This was the point at which I contemplated getting mother-in-law to climb into the rear seat, which was of course very high off the ground – hence the better option of paving slabs in the boot. And this is why the absence of trolley jacks, which we all take for granted now, made it all a bit more problematic.

Happy and innocent days!

Colin       

with the car perched precariously in the air on blocks and jacks and using the spring compressor tool referred to as Colin describes .With the bars inserted you need to place the trailing arms at an angle so the distance from the top of the hub disk to a projected line from the damper mounting bolts is 165 mm. This requires no family ballast or other relative contributions of course...
i'll post a drawing if this is unclear.