New Frontier (Part Six)

We dive beneath the skin.

Image: influx

Irrespective of whether Citroën’s Bureau d’Études was acting in concert or as alleged, in a contrary and fragmentary fashion, there were a number of engineering imperatives which for them would prove sacrosanct. The first of these and perhaps foremost was the mode through which drive forces would be transmitted.

The second and if anything, just as much a prerequisite would be the use of Citroën’s centralised engine-driven, high-pressure hydraulics for damping, steering, braking, levelling and attitude control. This highly innovative and technically ambitious oleo-pneumatic system was developed by Paul Magès and first employed for the rear suspension of the 1954 15 h model, prior to it being rolled out in fully fledged form in 1955’s DS 19.

Assisting Magès on Projet S was Hubert Alléra, who had amongst his other palmarès, designed the hydraulically actuated gearchange for the DS. Suspension-wise, the SM didn’t depart radically from existing practice, in fact a great deal of DS thinking (and hardware) was almost literally carried over; largely for cost reasons, but also because in the opinion of Jacques Né, not only were they strong enough to cope with the car’s additional performance potential, they were entirely appropriate to a car of the SM’s potential.

Né pointed out to author, Jan Norbye, that 100% of the SM’s roll stiffness was provided by the anti-roll bars. The main alteration from the DS saw the front suspension upper and lower arms mounted in a trailing orientation, primarily to help comply with US Federal impact standards due to come into practice in 1973.

However, Magès and Né had envisaged a mechanically actuated active suspension system, which by 1975 was running in experimental prototypes, incorporating hydraulic control of roll and pitch. However, this hydro-mechanical system was deemed too complex and expensive to productionise by Citroën’s masters, and was put on the shelf until such time as suitable electronic control units became available.[1]

The major departure from previous Citroën technical practice was Magès’ radical DIRAVI (DIrection a RAppel asser-VI) speed-sensitive, self-centring rack and pinion steering, introducing, in concert with the geometrically pure centre-point steering, an unprecedented level of control. By the time DIRAVI made its debut, it had already been fitted to a number of Projet-S prototypes, the first in 1962. Such was its directness, an unpowered system would have rendered the car undriveable.[2]

To alleviate this, Magès’ engineers developed a speed-sensitive variable servo, which provided maximum assist at very low speed (manoeuvring or parking) with a gradual easing of assistance as speeds rose, with zero assistance at high speeds when straight line stability was of greater importance. Coupled to this was powered self-centring, which came into effect once the handwheel was deflected from the straight ahead position, aiding high speed stability and reducing driver fatigue.

Heart of the matter. The SM’s steering regulator. Author’s collection

This was controlled by a gearbox-driven dual chamber hydraulic regulator, employing a centrifugal mechanism with bobweights, operating slide valves which increased or reduced hydraulic pressure in the larger chamber (the smaller chamber remaining constant) acting on the steering circuit. A cam regulated the amount of hydraulic pressure, set up to reduce the level of assistance at high speeds. The regulator was connected to a separate self-centring servo which applied hydraulic force upon the cam.

Citroën engineers determined from their exhaustive tests of the system that the smaller movements required upon the handwheel with the DIRAVI setup combined with the ultra-quick steering ratio, saved important time during direction changes, making for quicker and safer manoeuvring; similar perhaps to the siting of brake actuator and throttle pedal in the same plane, which led to often crucially quicker reactions from the driver. Another safety aspect of the DIRAVI system was that it was impervious to the otherwise potentially malign effects of a sudden tyre deflation or encountering a pool of standing water, for instance.

From an architectural point of view, Né and his team was somewhat hamstrung; first by the fact that he was not operating with an unlimited budget, and second, because there was a definitive way of approaching things at the Quai Andre Citroën, which Né, as an André Lefèbvre loyalist would have been bound to uphold. Furthermore, since the S-vehicle programme was intended to explore the upper reaches of front-wheel drive performance envelope, there was a compelling case to be made for continuity.

Image: Citroen Press via thecarhobby

Hence, the SM body structure and chassis layout closely reflected that of the existing DS saloons, with a large pressed steel ‘punt-shaped’ base unit, to which the structural elements of the SM bodyshell were welded. One notable advantage of the DS layout was that the engine’s fitment well inside the wheelbase not only aided weight distribution, but allowed for a more favourable dash to axle ratio, so beloved of exterior designers, which a fore axle FWD layout would have made impossible.

However, this layout, with the engine mounted aft of the gearbox and final drive, was not particularly space efficient, and despite the engine being tucked snugly against the bulkhead, there entailed a long frontal overhang – fortunately another Citroën tradition. A further advantage of this layout was a low polar moment of inertia, which aided stability and balance.

All of these factors would entail that the SM was a longer car than it might otherwise have been, more so even than the far more commodious DS saloon. Citroën engineers made use of the space ahead of the front wheels by siting most of the hydraulic units there, the car’s penetrating nose precluding the siting of the spare wheel in this section of the engine compartment.

The SM’s gearbox was a five-speed version of the unit fitted to contemporary DS models, with gear ratios optimised for the Modenese V6 power unit. In the autumn of 1973, a version of the 3-speed hydraulic torque converter-equipped Borg Warner Type 35 automatic gearbox was made available for the US market.

The SM also brought a new sophistication to something as prosaic as lighting, hydraulics being employed to provide automatic levelling for the six quartz halogen Cibie headlamps; the inner pair also turning in harmony with steering inputs. The outer pair of lamps provided main beam, while the middle pair supplied dipped beam.

Following DS practice, the SM also employed a pressure sensitive brake button, which operated the four wheel discs, the front brakes mounted inboard, alongside the final drive unit. The handbrake operated upon the front discs, employing separate calipers, mounted above the main units. As a failsafe, the hydraulic system accumulators stored sufficient fluid reserves to allow between 40 and 50 brake applications, should the system fail for any reason.

So while the technicalities of the SM were Quai de Javel distillate in its purest form, its engine was quite literally another country. It is here that we will turn our gaze next.

More on the SM

Sources and references:

Robert Opron L’Automobile et L’Art: Peter I Pijlman
Citroën SM : Jan P. Norbye
Citroën SM : Brian Long
Sa Majesté – Citroën SM : Peter I Pijlman/ Brian Cass
Citroën SM – Accidental Death of an Icon : Stuart Ager
André Citroen – Engineer, Explorer, Entrepreneur – John Reynolds
Maserati- The Citroën Years 1968-1975 : Marc Sonnery

[1] The Hydractive system made its debut in the 1989 XM, reaching its series production apogee with the 1997 Xantia Activa model. [Source: Jan Norbye]
[2] The DIRAVI steering fitted to the SM employed a ratio of 9:4:1 with a mere two turns lock to lock. [Source: Jan Norbye]

The SM gearbox and final drive was not only fitted to the Maserati Merak, but also the Lotus Esprit in conjunction with Hethel’s slant four unit until 1987, when it was replaced by a Renault-sourced unit.

Author: Eóin Doyle

Co-Founder. Editor. Content Provider.

21 thoughts on “New Frontier (Part Six)”

  1. Good morning Eóin. Fascinating technical detail, thank you. Given the huge effort that went to the development of the SM, I’ve always wondered why it was confined to a single model, the coupé, and not expanded to a wider range encompassing a luxury saloon and even an estate. Surely that would not have been prohibitively expensive and would have amortized the cost more effectively?

    The SM’s styling would certainly have lent itself to these formats, as coachbuilder HenriChapron demonstrated:

    The estate might have been a more questionable addition, but I think it looks just beautiful, if not especially practical.

    1. That’s an interesting thought, Daniel. Now I can’t get the image of the SM Plateau by Tissier out of my head. Or the SM pickup truck by Jerry Hathaway

    2. What fascinating pictures.;I had no idea such things had been done using the SM as a basis. The saloon works very well indeed.

    3. Hi Dave, the SM ‘landau’ is indeed a bit OTT but I think it’s extended tail looks better balanced with the front end. I’d be curious to Photoshop it onto the Opera version.

    4. I don’t think the Landaulet was meant to win any beauty awards, but was only intended for representative purposes of the Élysée Palace – hence the raised roll bar that the president could hold on to while standing.

    5. Many of Chapron’s creations are kind of remarkable. Some DS coupes are simply horrible and the open SM’s predecessor the DS présidentielle is no better

    6. Just sent an email to Santa to say I’ve changed my mind. I want an SM Estate for Christmas….

    7. At the risk of sound monomaniacal, the man who put the privacy glass in needs taking round the back of the bike sheds and shooting. Otherwise, utterly fabulous!

  2. The truly interesting feature of DIRAVI is that it is a fully hydraulic system with mechanical backup that as long as the hydraulic system is working correctly doesn’t need the steering column to transfer the movement of the control unit immediately below the steering wheel under the instrument panel to the steering rack.
    DIRAVI also works exactly opposite to a conventional PAS. A standard PAS is pressure-less as long as the steering is in the straight ahead position and hydraulic pressure is built up when the wheel is turned. DIRAVI is under full pressure in the straight ahead position and hydraulic forces are equal on both sides of the servo cylinder, effectively locking the steering in the straight ahead position and preventing any forces from reaching the steering. As soon as the wheel is turned the hydraulic pressure on the ‘target’ side is reduced and the hydraulic ram is pushed in the desired direction and then locked again – that’s why they gave DIRAVI artificial feel by hydraulic pressure on a kind of heart shaped feedback disc in the control unit. This self centring mechanism in the control unit makes for very funny driving when the system is maladjusted. Setting up the steering geometry of an SM of a CX is not for the faint hearted – there are even special jigs for adjusting the horizontal alignment of the steering rack relative to the steering linkage.

  3. Fun fact: the SM’s gearbox that was derived from the DS’ was strong enough to be pressed into service in the Peugeot 205T16 group B rally cars…
    Just proof of the inceredible built in sturdiness of DS and SM (and CX), something completely amiss from the Peugeot based Citroens like XM and BX.

    1. Citroen gearboxes were also used in the early rear-engined Cooper open-wheelers including F1 cars. In 1957 Jack Brabham, who was working at Cooper then, visited the ESRA factory where the gearboxes were made and found the foundry foreman willing to cast some strengthened cases if Jack modified the cores himself, which he did, adding ribs and increasing thickness where needed before they cast a run of 25.

  4. Here’s a film showing the SM undergoing crash and handling tests. Looks like they tried fitting air bags, at one point, no doubt with US regs in mind.

  5. These cars badly needed the active roll and pitch control function they were originally intended to possess. Without it,…..well…..

    The later Xantia Activa system is not based on the same principles as the semi-active system referred to in the article. The arrangement for the Activa works well, but it is not as elegant, nor as “pure”, as the original.

  6. The delicate lady beside the SM in the opening photo, S-ra I.Paci, used to have a dark green, mint
    XM V6 24v (the PRV, 200hp one).

    They had a reputation for premature cam wear, and engine-out for spark plug changes (almost!).

    But what an utter gem of
    a car it was!

    It unhinged a certain dynamic lunacy in the XM, that all its other iterations failed to disclose.

  7. Love these pieces about my second favorite Citroën, which is quickly becoming my favorite Citroën.
    I was lucky enough to be in Normandy last year when they held the Citroën centenary. It was amazing seeing so many beautiful cars in one place.

    In my opinion, the Maserati Quattroporte 2 was the best reliazation of the SM. Hopefully you guys talk a little bit about that.

    1. Pleased you are enjoying the series Neil. We have covered the QP-II, albeit not in depth, more in stylistic terms. That piece you will find in the archive. However, it is my intention to return to this subject in the fullness of time – it sits in the latter portion of the SM tale. The Chapron cars should also gain a mention along the way. Thanks for stopping by.

  8. Pedantry time!

    Unless I’m gravely mistaken, the DIRAVI steering system really shouldn’t be described as variable assistance, but rather fully powered with variable feel forces; the hydraulics are always providing 100% of the actual force to steer the wheels, and the driver 0%. This is in contrast to conventional “power steering” systems (which should really be described as boosted rather than powered, since they respond in proportion to the force applied by the driver, with the driver always providing some proportion of the actual force to steer the wheels). In DIRAVI, the hydraulics respond not to applied force, but to the position of the steering wheel, always positioning the road wheels to match. Since the driver is providing 0% of the needed force, an artificial feel mechanism is needed, and the sensation of increased steering effort at higher speeds comes from varying this mechanism. (I could even describe the famous “self centering” as a mere byproduct of this artificial feel mechanism, but that wouldn’t be entirely honest; the fact that the feel mechanism centers the steering wheel, and the steering rack responds purely to the position of the steering wheel is why you can drive through standing water and not feel the steering wheel jerk to the side–no doubt a deliberate design decision–and also why the powered flight controls in large airplanes work in precisely the same way.)

    Of less consequence, but no less pedantic: The SM’s dipped beams are provided by the outboard pair of headlamps (the ones that are larger than the other two pairs).

  9. Very accurate, and I can even add: In normal operation there is no real mechanical connection between the steering wheel and the wheels. The connection is purely hydraulic … like on airplanes.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.