Moulton Hydragas and Creuat: coupled hydropneumatic springs

[autogyro46]

After Greg Locock mentioned it, I decided to go dig up Alex Moulton's 1979 SAE paper on the Hydragas system. I was startled by the similarites to Creuat's paper of 23 years later (attached), even though the coupled hydropneumatic units are "wired" differently. (In the Creuat, the hydropneumatic compliances are configured to induce common-mode motion in coupled wheel pairs, rather than differential mode motion in the Moulton).
What was compelling was that both address the global dynamic behavior of the vehicle. Moulton went so far as to use multi-modal frequency domain analysis (quite a trick in the late '70's) as well as time domain.
It would appear to naifs like myself to be an elegant and powerful solution (and if BMC and BL bought into it, cheap, too.)
My question is, why hasn't anyone else picked up the ball? It seems BMW let the MGF revert to coil springs, and brushed aside Moulton's proposal for Hydragas in the Mini revival. Is this simply a case of NIH? That sad thing is, that according to Moulton, all the Hydragas tooling has been destroyed. Good ideas can sometimes disappear for bad reasons.
It would seem to me that the diagonal coupling that Creat advocates has some advantages (roll stiffnesses can be 3-4 times higher than bounce), particularly if one is not particularly concerned with smoothing out very short wheelbase vehicles (aka Mini)
(I should add that, as a teenager, I had a pleasant summer in the '60's tooling around in a car equipped with the predecessor of Hydragas, the Hydrolastic MG1100)

 

[Peter7307]

autogyro46,
I suggest the main reason no other makers "picked up the ball" is simply the cost of paying Moulton to use their system when makers like BMW can engineer one of their own and not have to pay anyone to use it.

One practical aspect of the Moulton hydragas / hydrolastic set up is the very limited ability for adjustments and modifications which means the appeal for sports car owners is very limited.

The MGF Hydragas variant has a reputation for quality control issues and some of this related to the suspension. I am guessing when BMW took control the idea of a small , mid engined sports car appealed but little else did hence the reversion to conventional coil springs and gas shocks.

As an aside I owned a Morris 1100 with Hydrolastic suspension...bouncy , bouncy !

Pete.

 

[autogyro46]

Thanks Peter:
My main point was not to advocate that everyone should start paying Moulton royalties; there are other approaches to coupled suspension: Creuat, and perhaps most majestically, the Citroen DS19. (No "Bouncy-Bouncy" there, although having my brief encounter the the MG1100, I quite know what you mean.)

Hydraulics is simply an approach that facilitates coupling, and, by extension, an approach to global control of the vehicle's attitude in varying dynamic situations.

I would argue that the problems lie not in the concept, but in the execution.

 

[GregLocock]

Also check out Kinetik (I think they've been taken over by TRW), and of course any full authority active suspension offers as many cross coupling modes as you can be bothered to program.

Incidentally the 2CV originally had anti pitch bars, and the funniest rear shock absorbers.



Cheers

Greg Locock


New here? Try reading these, they might help FAQ731-376

http://eng-tips.com/market.cfm?

 

[Tmoose]

funny 2CV shox ? -

(reportedly) horizontal hydraulic

http://www.ecas2cvparts.co.uk/shock...v4mehariazu-original-lipmesa-notes-p-908.html

or friction type

http://www.ecas2cvparts.co.uk/friction-damper-assembly-frontrear-2cvdy-price-each-p-171.html

 

[Tmoose]

supposedly 2cv based Gothic aero

http://www.jalopyjournal.com/forum/attachment.php?attachmentid=114930&d=1125254056

 

[cibachrome]

Bose demonstrated a single very capable system using their voice coil technology to Cadillac a few years ago. GM was interested to the point of assigning a Vehicle Team to get it started towards production. However, Bose was unwilling to open up their software algorithms and manufacturing requirements to the scrutiny of Systems Engineering Staff to study 'fail safe', durability and response character tuning issues and concerns. There was also a concern about the ability to provide enough units for production and replacement parts. Much like the problems with the Lotus/Active Corvette, it is one thing to claim a certain set of response traits and algorithms, but when tested for specific claims and the claims don't hold up, Production Engineering people get cold feet...

 

[GregLocock]

That's the first I've heard of Lotus Active not meeting performance targets.



Cheers

Greg Locock


New here? Try reading these, they might help FAQ731-376

http://eng-tips.com/market.cfm?

 

[autogyro46]

The point I was driving at is, that full active systems may be wonderful, simple passive interconnection achieves much the same at a fraction of the cost. The mechanical 2CV and the hydragas/hydrolastic system proved that, and they were reliable over long prodution liftimes.
They were interconnected to deal with pitch, but further benefits may be obtained by cross connecting them to deal with all four modes, again passively.

 

[GregLocock]

OK, for a normal road car with a normal driver what benefit do you see from an interconnected suspension?



Cheers

Greg Locock


New here? Try reading these, they might help FAQ731-376

http://eng-tips.com/market.cfm?

 

[gt6racer2]

Hi, I've played with pitch coupling at length and extremes, and it is quite interesting. If you couple for in/out (Moulton), then two wheel input - such a sleeping policemen is great, but brake dive gets worse. If you couple in/in (others) then brake dive is improved ( nice for aero racecars or off road driving ) but two wheel bump becomes an issue. As a result of this conflict I don't see much future for a passive pitch interconnection.
Greg - I think you meant to say Kinetic with a "c" - and it was Tenneco that acquired them.
The key benefit of interconnected suspensions is the ability to reduce unecessary individual wheel rate on a given input by separating out the suspensions operating modes. Same reason you use stabilizer bars - just on the next level..

 

[autogyro46]

You've hit the nail on the head. Any suspension is a set of trade-offs and compromises, and coupled systems are no exception.
The in/in (which I prefer to call common-mode) situation you describe produces the same two-wheel bump problem, whether it's connected side-to-side as in an ARB, or back to front. In the former case, the one wheel bump becomes partly expressed as "pitch".
The really interesting case is in cross-coupled ARBs, which, due to their common-mode behaviour, provide increased compliant resistance to both roll and pitch (or anti dive/squat), for a given individual bump rate.
The downside would be that the two-wheel bump is espressed diagonally, and I confess I'm not sure what the implications of that are.
Any thoughts?

 

[BrianPetersen]

When you hit a two-wheel bump with the front wheels, increasing front wheel loading, it sounds like it'll try to pull the rear wheels off the ground while pulling the rear of the vehicle body down towards the ground, just in time for the rear wheels in turn to hit that bump with the vehicle body already heading the wrong direction (down, towards the bump) ... sounds interesting, to say the least.