Hydraulic Actuated Rear Wing system

[AutoStudent25]

Hi, I am currently working on a project at Uni that involves coming up with an ECS for an adaptive rear wing system. I plan on using hydraulics due to their fast response time, I want to set up a pump attached to two accumulators which in turn are each connected to a solenoid valve and the linear hydraulic actuator at the start of one accumulator will be empty and the actuators top chamber full when the wing is required to tilt the full accumulator will empty into the bottom chamber forcing the fluid in the upper chamber back into the empty accumulator beginning a cycle in which the fluid is forced into accumulators from the actuator conserving energy. excess fluid from filling the actuator is rerouted with the solenoid valve to the pump where it is routed using a solenoid valve to whichever accumulator will be filled next cycle. W
Will This concept work and what are things I should tweak?

 

[BrianPetersen]

I don't understand your explanation. Flow-direction arrows and schematics indicating what is happening when, would help.

BUT. Three directional valves and two accumulators to control one cylinder? I understand it needs "forward", "stop", and "retract". If you have a single source of pressure (whether fitted with an accumulator or otherwise), a single directional valve, dual solenoid spring-centered with a closed-center spool, is all that is needed. Pressure to the center port, A and B to the ends of the cylinder, return-to-tank from the end ports (and the reservoir is the atmospheric-pressure reservoir for the pump). The pump needs to deal with being deadheaded appropriately. This layout is "fixed pressure variable displacement". This is a very common general-purpose hydraulic circuit design. Your average plain old ordinary industrial machine with a hydraulic power unit and a bank of directional valves to move various hydraulic functions, works like this.

Saving energy in that type of hydraulic circuit design generally means using a variable-displacement pump, so that you can turn it down to zero displacement when there is no demand. It still needs to maintain regulated pressure to achieve the maximum force on the actuator. This is not an inexpensive fixed-displacement gear pump. It's a variable-displacement vane or swash-plate pump. $$$, although lots of automotive engine oil and transmission oil pumps are variable-displacement vane pumps nowadays.

The other choice of hydraulic circuit design is something akin to old skool power steering systems, with a fixed-displacement engine driven pump (cheap, simple) delivering constant circulation through an open-center spool valve. Your input at the steering wheel shifts the valve which causes the pressure to build up in one side or the other, sufficient to overcome the load of the steering gear. "Fixed displacement and variable pressure". Doesn't work with an accumulator. If you insist on having a circuit design with momentary hydraulic flow into the cylinder that exceeds the instantaneous flow rate of the pump (thus needing an accumulator to store the pressure) then you're stuck with a fixed-pressure circuit design (with the pressure setpoint being whatever the storage pressure in your accumulator is).

If energy-conservation is a consideration, then hydraulics are not a good choice. Similar applications nowadays use a little stepper or servo motor to a gearbox to what amounts to a screw jack. No hydraulics. No standby power loss. Same reason power-steering systems have almost all gone to electric-servo nowadays.

 

[MintJulep]

Overly complicated and HEAVY.

Electrically driven ball screw or servo and bell crank will do the job better in every way.

 

[MintJulep]

Oh, and your scheme to recover
energy won't work and will only slow the response.

 

[AutoStudent25]

I don't understand your explanation. Flow-direction arrows and schematics indicating what is happening when, would help.

BUT. Three directional valves and two accumulators to control one cylinder? I understand it needs "forward", "stop", and "retract". If you have a single source of pressure (whether fitted with an accumulator or otherwise), a single directional valve, dual solenoid spring-centered with a closed-center spool, is all that is needed. Pressure to the center port, A and B to the ends of the cylinder, return-to-tank from the end ports (and the reservoir is the atmospheric-pressure reservoir for the pump). The pump needs to deal with being deadheaded appropriately. This layout is "fixed pressure variable displacement". This is a very common general-purpose hydraulic circuit design. Your average plain old ordinary industrial machine with a hydraulic power unit and a bank of directional valves to move various hydraulic functions, works like this.

Saving energy in that type of hydraulic circuit design generally means using a variable-displacement pump, so that you can turn it down to zero displacement when there is no demand. It still needs to maintain regulated pressure to achieve the maximum force on the actuator. This is not an inexpensive fixed-displacement gear pump. It's a variable-displacement vane or swash-plate pump. $$$, although lots of automotive engine oil and transmission oil pumps are variable-displacement vane pumps nowadays.

The other choice of hydraulic circuit design is something akin to old skool power steering systems, with a fixed-displacement engine driven pump (cheap, simple) delivering constant circulation through an open-center spool valve. Your input at the steering wheel shifts the valve which causes the pressure to build up in one side or the other, sufficient to overcome the load of the steering gear. "Fixed displacement and variable pressure". Doesn't work with an accumulator. If you insist on having a circuit design with momentary hydraulic flow into the cylinder that exceeds the instantaneous flow rate of the pump (thus needing an accumulator to store the pressure) then you're stuck with a fixed-pressure circuit design (with the pressure setpoint being whatever the storage pressure in your accumulator is).

If energy-conservation is a consideration, then hydraulics are not a good choice. Similar applications nowadays use a little stepper or servo motor to a gearbox to what amounts to a screw jack. No hydraulics. No standby power loss. Same reason power-steering systems have almost all gone to electric-servo nowadays.

So it’s seeming like electric actuators are a better way to go? Would a geared servo motor be better for this application as it needs to respond to up to a 60 degree change in 0.2 seconds.

I had originally planned on using a fixed pressure hydraulic system with two accumulators one for redundancy similar to how you described however that seems a heavy solution now.

 

[BrianPetersen]

None of the rest of us knows the scale of your "adaptive rear wing". You gave us the time (questions below). How much force ... how much distance. That's the numbers you need.

Powered rear wings/spoilers exist in production. Take a look at a few.

Now ... Next question. Why on EARTH do you need such a mechanism to slam from one end of its stroke to the other in 0.2 seconds???

Powered wings on production vehicles are typically gimmicks. They extend when the vehicle goes over a certain road speed (perhaps 100 - 120 km/h?) for downforce or stability or what-not, and they retract when the road speed drops below some lower road speed for a slick appearance. If it takes a second, or even two seconds, for it to go through it's stroke, nobody is going to be troubled. Functionally ... Just ditch the actuator and mechanism and leave the spoiler up. It doesn't do anything for your aero at low road speed anyhow. There's no harm in leaving it up. The ONE thing that just maaaaybe might make a difference, is if it's a trap for snow or freezing rain or is a pedestrian-hazard of some sort if it's extended when the car is stopped - but if that's the case, I question your FMEA ...

There's an argument on racing vehicles that perhaps the spoiler could be in a low-drag position on straightaways and switch to a high-downforce (but higher drag) position during cornering or braking. If that's the situation then the number one first thing to consult, is the rulebook, to see whether that sort of thing is allowed at all. If that's the situation then I could also understand the motivation for having a fast changeover speed, but even then ... Doing stuff that abruptly changes the chassis dynamics (via abrupt changes to downforce) is a recipe for trouble, and I'd be more inclined to look for refining your control logic so as to eliminate the need for fast changeovers.

 

[IRstuff]

What vehicle on earth requires 0.2 second change in aerodynamic configuration? The fastest stock auto goes from 0 to 60 mph in around 3 seconds, and one would expect any aerodynamic configuration change to occur in the same time frame.

 

[gruntguru]

I think Brian described the scenario where that response is desirable. Vmax on a straight - transitioning to max braking.

 

[IRstuff]

Is the wing really going to make THAT much difference? Braking time from 60 mph should be something on the order of 6 seconds.

 

[GregLocock]

Bit less than half that on dry tarmac with decent tires for the actual braking bit. I was very surprised to see an SUV's front tires giving a mu of 1.2 (Fx/Fz) during a braking event.

 

[BrianPetersen]

We haven't heard from the original poster in a wee bit, but here's another thought. IF this is a roadracing vehicle, then there is some logic in having a rapid transition between the low-drag position and the max-downforce position, and there's also a desire to reduce weight. And with that ...

Is a "bang-bang" one end of travel to the other the expected mode of operation ... no modulating in between?

If so, then pneumatics may make sense. Air cylinders can move way faster than hydraulics, and air is lighter, and leaks don't matter much, and you don't need a reservoir on the low-pressure side, or even have a return path for fluid at all. You need a single 5-port 2-position solenoid valve fed from a compressed air supply. It can even be a single-solenoid spring-return valve. Your choice whether the default position is up or down, your FMEA should direct this choice, and that FMEA should address both loss of valve power and loss of compressed air supply.

If it has to modulate position at all, forget pneumatics completely. Servo/stepper motor through a gearbox is the way to go, but 0.2 seconds for it to go full stroke is unlikely unless this stroke is quite short, which the rest of us don't know ...

 

[MintJulep]

The Chaparral 2E wing only used 2 positions.

Petroleum Museum | Chaparral 2E

petroleummuseum.org

 

[AutoStudent25]

We haven't heard from the original poster in a wee bit, but here's another thought. IF this is a roadracing vehicle, then there is some logic in having a rapid transition between the low-drag position and the max-downforce position, and there's also a desire to reduce weight. And with that ...

Is a "bang-bang" one end of travel to the other the expected mode of operation ... no modulating in between?

If so, then pneumatics may make sense. Air cylinders can move way faster than hydraulics, and air is lighter, and leaks don't matter much, and you don't need a reservoir on the low-pressure side, or even have a return path for fluid at all. You need a single 5-port 2-position solenoid valve fed from a compressed air supply. It can even be a single-solenoid spring-return valve. Your choice whether the default position is up or down, your FMEA should direct this choice, and that FMEA should address both loss of valve power and loss of compressed air supply.

If it has to modulate position at all, forget pneumatics completely. Servo/stepper motor through a gearbox is the way to go, but 0.2 seconds for it to go full stroke is unlikely unless this stroke is quite short, which the rest of us don't know ...

The Idea is a high-performance racing morgan that will need to go from full speed to cornering on tracks so I thought a fast response was desirable how quick could servos go?

 

[AutoStudent25]

The Idea is a high-performance racing morgan that will need to go from full speed to cornering on tracks so I thought a fast response was desirable how quick could servos go?

The regs state 4 degrees of motion with a 60 fegree range we are being pragmatic making sure it can go from full Drag reduction on the straight to full downforce should a hazard arise

 

[BrianPetersen]

So. How much stroke on your actuator, and how much force??

 

[AutoStudent25]

So. How much stroke on your actuator, and how much force??

Its Stroke should be between 25 and 51 cm and the depth should be between 30 and 60 cm (I have a reasonably useless design team so I do not know this yet) i just want to know what generally would be a good direction to go in as exact values will need to wait until my group catches up

To calculate power when I have more exact values I plan on using the formula
demand torque = (load torque + acceleration torque + friction torque) * design safety factor of 1.25 .

Load= (Force of air resistance at max angle and speed + weight) acting at 90 degrees to the wing surface
air resistance (1/2Air density*velocity2*Coefficient Drag*Wing Area) Cos Wing Angle
weight((Mass*Gravity) Cos opposing angle to wing)


Acceleration = angular momentum/angular velocity
Angular velocity will be the change in wing angle/Time(t)
Angular momentum weight will be the Length of the wing* Mass of wing*linear velocity.

Friction = the sum of the friction posed by all the mechanical components in the control structure such as hinges

will this be the correct way of doing it?

 

[BrianPetersen]

"51 cm" as in half a metre of actuator stroke?????? Usually automotive-scale drawings are in millimetres ... and the entire tailgate actuator on my car doesn't have half a metre of stroke!

 

[IRstuff]

The Idea is a high-performance racing morgan that will need to go from full speed to cornering on tracks so I thought a fast response was desirable how quick could servos go?

This really depends on HOW you think your control system is even going to know WHEN it's cornering. You can't depend on braking alone, since there are other times you might be braking, and it might be disastrous if you suddenly deployed air braking while you were maneuvering for an advantage.

I would guess that it would if you have a high performance inertial navigation system, you might get an answer in under 0.5 seconds; may be less if it's augmented with a detailed mapping system

 

[3DDave]

Button on the steering wheel.

 

[AutoStudent25]

This really depends on HOW you think your control system is even going to know WHEN it's cornering. You can't depend on braking alone, since there are other times you might be braking, and it might be disastrous if you suddenly deployed air braking while you were maneuvering for an advantage.

I would guess that it would if you have a high performance inertial navigation system, you might get an answer in under 0.5 seconds; may be less if it's augmented with a detailed mapping system

There will be a full logic system developed by another member of the group it makes use of sensors for wheel speed steering angle gps and EMS and a few more to decide when to corner