magnetorheological damper

[RacingManiac]

reading some old posts on the this forum and noticed this was mentioned a few times, in the context of Ferrari may have been using this in the past in F1. Working in this field now I wonder if it is possible since MR damper is really a form of controlled adaptive damping using current to locally changing the viscousity of the fluid to control damping. While not a completely active system I don't think system like this was allowed....

MR dampers nowadays have become so common place (a Delphi technology)a lot of US and European cars now have them. Ranging from Audi TT to Cadillac CTS to Ferrari 599.....

[scarbs]

MR dampers are banned in F1, there were rumours Ferrari were using them, but it seems this was a confusion with the inerters they were using from Monza 2007 onwards. I researched MR dampers a lot to see if they were an inerter-like solution, I found a Toyota patent on a MR Damper with a spinning generator coil inside. This coil generated power as is span around the damper rod and the power altered the fluids damping characteristics. In theory this could be an inerter like solution, as the damper needed no external power or control, and reacted only to the cars movement it could be considered legal. I don’t think solution has made it F1, as I was advised that any form of Mr fluid in F1 is a strict no-no.

[RacingManiac]

that's interesting.....its effectively then a "KERS" for suspension based on MR fluid...not entirely far fetch neither....

[jddh1]

This is what pisses me off with F1 sometimes. Why ban innovation like this? Pretty stupid.

[axle]

This is what pisses me off with F1 sometimes. Why ban innovation like this? Pretty stupid.

I expect it's a cost vs. return thing vs. sporting spectical impact.

If it doesn't enhance the sport and just leads to extra cost stick it in the bin.

However I'd have thought that in light of the push towards mechanical grip over aero for better overtaking the introduction of anything that can help (that's found on passenger cars) should be allowed.

[Conceptual]

Unless it increases the Drama on track, it has no place in a Theatrical Formula such as F1.

[Ciro Pabón]

Sachs rotational MR dampers are used in F1, at least by Ferrari, BMW and Toyota.

What I think is forbidden are active MR dampers. The ZF Sachs kind is passive, what I think scarbs calls inerters. Perhaps this can explain the difference (taken from this old thread, maybe the one that Maniac saw).

A rotational damper works by turning the vanes to push the fluid through a set of valves, as "clearly" shown in the right part of the picture.



Of course, the resistance you get when you turn the vanes inside the tube depicted depends on how viscous is (how sticky) the fluid inside.

There are several fluids that have diferent "rheological properties" (viscosity) when the conditions change. The first that comes to my mind is ketchup: if you slowly apply forces to ketchup, it has high viscosity; on the other hand, if you apply the force quickly, its viscosity lessens (that's why you hit the bottom of the bottle to "release" it)

Now, if you put magnets inside the damper, you can alter the "stickiness" of a magnetic fluid. This is called "magnetorheology" (viscosity altered by magnetism). The best I could quickly find is this linear mono-chamber damper, explained here: PDF
.

The light blue squares are the magnets and the lines on top of them are the "magnetic restricted zone", where the fluid becomes more "sticky" the faster it moves


Damping variation caused by permanent magnets (passive, allowed by regulations) is not too far from the one you get with active, electromagnetic ones (forbidden by regulations), as explained here: http://web.me.unr.edu/ciml/11.pdf. The important thing is that you have a more or less linear (variable) response, depending on the speed of movement of the damper, but it's not an active response. You can get active response if, instead of magnets you used coils and a controller to vary the current in them.

You can see that the force with "magnets-only" (gray lower line), even if lower than when electrically excited (pink upper line) it is still sizeable, which means the thingies work without clasifying as active suspension.



So, I don't know what happened to Scarbs, but I think those dampers are MR ones (they have a variable response that depends on the speed of the fluid passing the magnets).

BTW, the invention is old: 1940, by Jacob Rainbow, an american.

[RacingManiac]

I don't see the benefit of using passive magnet to actuate MR fluid over standard damper, since you don't get the "changing" viscousity effect of a electric current driven version, in effect it assumes a certain property when it approaches the magnet and resumes the normal property away from it. I can't see what addition control it would offer over just standard valving. Also how would the flux field form inside a rotational damper? I know that rotational damper are indeed in use in F1, but I don't know what would drive them to use MR fluid based damper if they can't make use of the control property of it, especially considering the extra fluid mass they carry over standard shock oil...

[Ciro Pabón]

Rotational dampers offer you a more compact case design. I guess the aerodynamic and weight savings you get, because of a smaller case, more than compensate for the excess liquid (if it is true that they use more liquid). Steel weighs 7 times more than oil, by volume, as all of you know. Aerodynamic influence is probably the main benefit: I think Ferrari can shave 5 cm of lateral bodywork at the rear of the car by using these dampers. Anyway, an image is worth a thousand words:

Sachs rotary damper


Williams front (telescopic) damper (picture by Scarbs)


The control you get comes from the fact that the faster the fluid circulates, the stiffer the damper gets, because the induced currents are larger.

Another benefit over normal valve-driven dampers comes from the linear response you get, exemplified in the graph previously shown. Actually, the response you get doesn't have to be linear: you can design it based on your needs. I guess you can do the same with a multi-stage cylinder damper, but instead of several "stages", that come into play one after another, you get a smooth response. Instead of valves that open or close at a predetermined load, by steps, you get a continously variable resistance.

Because the dampers are coupled by a conventional, telescopic damper, you get the "anti-roll bar" effect incorporated:



Finally, you "work" the liquid by shear forces, instead of compression. This means to me that there is a high probability that the problems from bubbles into the liquid (foam) are minimized, because you don't have to use a chamber filled with gas.

BTW, Sachs offered a previous solution to foam (and the need of compressed nitrogen) with its "Through-rod" damper. PDF on through-rod dampers.

What I have no idea is how they solved the vane seals "problem": I've read the pressure inside the case reachs 160 bars, so Sachs uses titanium cases. I also have read they might use the system already used in vane pumps: a small orifice behind the vane redirects some of the fluid, pressing the vane against the case.



LONG, UNNECESARY NOTE: Just in case, for those not familiarized with damper technology, the idea behind a damper is to get a different response when the damper moves slowly (like in rolling motions, for example) from the one you get when the damper moves faster (like in road bumps or in kerbs). This is called low-speed vs high-speed bump/rebound.

Most dampers have a "test" version, that you can graduate during Friday/Saturday tests and a "fixed", lighter (because it doesn't have a graduation mechanism) version that you use on Sunday. These test versions have two "dials", one for low speed and another one for high speed bump/rebound.

I think that an MR damper offers you a smoother and "graduated" response than a valve-driven damper.

I have to clarify that all of what I've written comes from my understanding of dampers. I might be wrong, I haven't read all this "somewhere else", so I hope somebody corrects me if I'm mistaken. However, I've been always able to understand how a piece of machinery works just by looking at it... and I don't think the people at Ferrari are dumb enough to use a worse damper version just because.

[Carlos]

Just a note - I have read that Toyota holds a patent on an active MR telescopic shock absorber that uses a coil (electric generator) on the shaft. The movement of the wheel, compresses the shock, rotating the coil, producing electricity which changes the viscosity of the MR fluid.

I remember scarbs posts and pictures of the inerter, and I think it was not electrical but mechanical. A circular flywheel activated by a screw on a shaft.

[RacingManiac]

The way I understood how MR damper works the way it does on a car, is actually very simple and works very differently from a conventional valvestack. If you have have seen a pressure-flow curve of an orfice(a hole basically) on a flow bench, the curve would naturally assumes an almost exponential shape, where on low flow velocity(low shock speed), the pressure drop across the 2 side of the orfice is less, producing less damping force, where as in high speed the pressure drop is much greater because the same size orfice is less adequete at allowing fluid to pass through, thus creating a much greater differential to sustain flow, in damping terms, much more damping force. If you make a tubular shock with a piston with a hole in the middle, it would behave as such. And changing the size of the hole changes how steep that orfice curve gets. As you've said, there are different damping requirement for low speed(for handling event) and high speed(for bumps and uneven surface condition), thus usually a simple orfice curve would be really undesirable as you might not get the initial slope stiff enough for handling but gets good bump response at speed, or nice and stiff for handling but way too stiff for bumps. Regular dampers deals with this with valve stacks, either with steel shims stacked in such a way to for particular spring curve, or convental spring. What that does is you build a small pilot orfice in, to give you a stiff low speed response, but as the shock velocity increases(say hitting a curb), the fluid force pushs the valve stack open and essentially relieves the pressure. The knobs you turn on a adjustable damper can changes a number of things but what they mainly are doing is to control the size of the pilot orfice for low speed stiffness(usually a needle valve) and(this I am not so sure), the preload on the high speed valve stack or the amount of exposure the valve stack gets from the fluid.

On MR damper, at least on a production car, they only just have a single hole in the middle of the damper piston. The hole however is surrounded by electrical coil, and obviously the fluid is MR fluid. As you see the graph you posted before, it assumes the bottom curve with no current, because it basically is acting as an orfice curve at that point(btw, if that shock you shows goes to even higher test speed, although 2m/s is damn high, the curve would just keep going up). As you apply current to the coil, the flux field formed around the circular coil around the hole assumes a donut shape, now imagine the hole of the donut is where the flux field has no influence and where the flux field covers changes the viscosity of the MR fluid, where it assumes a much less viscous state, essentially shirnking the size of the orfice, and allows the damper to get stiffer. But you would write your control system to alter the stiffness(ie, the amount of current goes though the coil, and thus the size of the donut hole). Now if you see your stiff curve though, it is not an orfice curve, there is still tha "knee" to the curve(which in a conventional shock, its where the spring stack pops open), that essentially what happens in a MR shock too, but in slightly different fashion. As the shock velocity hits a high enough threshold, the pressure flow force will be high enough such that the "solid" MR fluid around the hole will shear off and you will get a bigger orfice then you did at lower speed, thus the "knee" happens.

Without control you can't do that, and you would need a conventional valve stack and/or different orfice to acheive the same effect, and at that point you might as well use regular shock oil...

[jddh1]

Very interesting read Ciro and RacingManiac.

Now, does anybody want to start working with me on Nuclear Powered Active Dampers (NPads)?

if we can just get a small nuclear reactor designed first...lol

[AeroGT3]

FIA is always touting road relevance, yet one of the biggest if not the biggest untapped road-relevant technologies is banned

[RacingManiac]

MR system has its drawback, it needs lots of current(GM system draws 4.5amp per corner on full stiff setting, can probably be lowered if you shrink the orfice and sacrifice a bit of ride), and its temperature sensitive at least from a subjective testing view. Where I work now we were trying to develop mechanical based valvling technology that can deliver similar performance for road application. As someone mentioned I think adaptive damping in general is more relavent for road vehicle than race car because of the compromise road cars need. Although I guess you can use it also to deal with the changing normal load that a race car experiences due to the aerodynamic downforce.....

[Ciro Pabón]

I think Racing Maniac hasn't got it, or I got it wrong (but I'm never wrong! ).

As he correctly points out, the liquid is magnetic. When this liquid passes close to a coil the result is an induced current in the coil. This current depends on the speed of the fluid. The induced current in the coils, in turn, influences the speed of the liquid, which induces more current. So, you do not need a controlling current for the MR damper to work. Actually, the reason why you cannot use a controlling current is because the device would fall under the category of "active suspension".

What separates magnetohydrodinamic theory from fluid theory is that: "... magnetic fields induce currents in a moving conductive fluid. This creates forces on the fluid, and also change the magnetic field itself." So, you have to solve simultaneously the Navier-Stokes and the Maxwell equations.

It would be funny for the thing to work otherwise: just imagine, Sachs using a magnetic fluid, magnets and coils that serve no useful purpose. If I understand Maniac, he thinks that the Ferrari damper works like a single valve one or they are risking to be denounced . I bet, if this were true, that Hamilton won the championship because Ferrari is managed by morons.