Controlling Linear Actuator via Accelerometer

[wingguy14]

Hi All,

I am new here, and am looking for some help with a new project, that goes into the realm of simple electrical engineering, which is outside of my comfort zone. I have spent quite a while trying to figure out a solution, but I can’t come up with any sort of elegant solution… I am hoping I can find some help here, or someone who is willing to take one a very small side job and design this for me, I can offer a fee.

Project Overview: Automotive race application, need to control a pair of linear actuators based on either a rotary dial or accelerometer outputs.

Part 1
I need to be able to move a linear actuator to 5-7 different remembered positions. The linear actuator I am using has a built in potentiometer. I want these 5-7 positions to be controlled via a multi-position rotary switch. This would all be off of a car’s 12V system. I have been told that this is pretty simple micro controller work, but again its outside of my scope of knowledge.
Part 2
I also would like to have the ability to have 3 of these positions to be selected via outputs from a 2 axis accelerometer. Position 1: This would be the default position (Less than .5G from either axis). Position 2: Needs to be selected when the Y axis sees >.5G(car is braking) Position 3: Needs to be selected when the x axis sees >.5G(car is turning), this needs to override position 2 if both axis are experiencing G loads. It would also be good to have some ability to adjust the G thresholds.

That’s really the quickest overview I can give. I have found an off the shelf actuator control that can remember 3 positions, and I was going to use that, but I keep getting hung up on what type of switch you used that is activated at a certain voltage output from the accelerometer. So, at this point I am really looking for a custom solution as opposed to cobbling together off the shelf parts. Thank you ahead of time to anyone who may be able to provide some ideas, or to anyone who would want to design this for me. Its much appreciated!

 

[BrianE22]

It would be a simple job for a microcontroller. Microchip has a link to outside vendors that will design a circuit for you:

http://www.microchip.com/stellent/idcplg?IdcService=SS_GET_PAGE&nodeId=2502&param=en534075

 

[BrianE22]

Or a PLC with analog inputs.

One thing to consider are the failure modes. If the electronics or software messes up you don't want bad things to happen.

 

[wingguy14]

Brian - thank you. I will check these both out!

 

[MacGyverS2000]

Sounds like someone is attempting to build an automatic downdraft/wing system? Though I'm not sure how an accelerometer would be useful in such a case...

Dan - Owner

http://www.Hi-TecDesigns.com

 

[wingguy14]

Dan,

That's very close to what I am trying to do. Perhaps an accelerometer will not work how I thought, but I simply want the actuator to change position when it sees certain g loads... Is there something fundamentally wrong with this idea? Thanks much

 

[MacGyverS2000]

Wing position should be based upon desired load (downforce)... this is tied (most directly) to speed, not acceleration. An accelerometer is not the appropriate sensor for this.

You need an EE to design this for you.

Dan - Owner

http://www.Hi-TecDesigns.com

 

[wingguy14]

Dan,
I see your point, but I don't want the wing to move based upon vehicle speed. I want it to move based on what the car is doing (braking, turning, steady speed, etc.)

I have sent a few inquireies out looking for an EE, but have not found anyone yet. Anyone who is looking for a side job, please let me know! Thanks

 

[MacGyverS2000]

I assume you are trying to vary the downforce based upon action of the driver (e.g., increase downforce during hard braking). Have you considered how this will affect the handling characteristics of the vehicle at speed?

Dan - Owner

http://www.Hi-TecDesigns.com

 

[wingguy14]

Thanks eveyrone fotr your replies. Does anyone have a direct contact to someone who may be interested in this project?

 

[MacGyverS2000]

I think you need to define your problem with a lot more detail before an EE would want to step in... I asked some very top-level questions that didn't appear to have a quick answer from you. This tells me you have a difficult time defining the problem, and I'll charge extra if I have to not only solve the problem but also determine what the problem is first. I don't think you'll find many EEs who think differently...

Dan - Owner

http://www.Hi-TecDesigns.com

 

[wingguy14]

Dan,

Thanks for th input. I really thought it was a pretty deinfed set of ideas, do you think it would help if I write up a very clear and detailed problem, and post it up?

Perhaps I am not articulating this well, but I think I have a good idea what I want.

My background is in static aero componets, as well as being a race driver, so I have a pretty good idea of how I want the device to act on the wing, and how that action will impact the handling of the car, certainly I would want the failure mode to be the higher downforce level. Thanks again for the feedback.

 

[BrianE22]

Can you afford the lag between body accels and wing adjustment? A more anticipatory system might involve feedback from the driver's inputs and maybe vehicle speed as well as body accels.

You might want to run the question in the Automotive Engineers forums on this web site. Seems like some of those people have raced before.

 

[MacGyverS2000]

If all you want is a knob to change the foil, any EE worth his salt can throw together a circuit and code in a day or two without thinking (in fact, such a circuit is typical of a home hobbyist project to learn about simple micros). If you want real-time reactive to external sensors, complications increase very quickly.

You would be better off to fully describe how you want aero surfaces to react based upon very specific inputs... don't just say I want a wing to tilt downwards when the brake is pressed. Specify the range of angles the wing will travel, how it will react to varying pressures on the brake pedal (preferably via a function of some form, such as pedal pressure versus current speed, etc.), and so on. If you cannot do this, there will be a bunch of back and forth between you and the designer, and costs will rise very quickly as you waste time. Even if the equation you initially come up with is not perfect, it will help the designer give you a more complete system right from the beginning.

You will need to be prepared for a number of back and forth sessions with the designer. For example, you initially suggested an accelerometer... on a drag strip this might be fine, but take it to an autocross or an oval and a single sensor will not provide enough info (brake lightly on a tight curve while on an embankment and your aero component is going to dig your rear into the pavement so hard your handling is out of the window... because the single sensor can't take off-axis angles into account). If you can't describe the problem fully, the designer is only going to design for the easy path.

Dan - Owner

http://www.Hi-TecDesigns.com

 

[wingguy14]

Brian - It would need to be fairly instantaneous, and I was thinking the packaging would be easier if it could rely on external sensors. I have been considering using the brake pedal switch to activate the braking position, and just using the accelerometer on an x-axis and just have the accelerometer defeat the brake switch if they are both triggered. Again, I suppose I am going to need to physically test some accelerometers to see how well they discriminate between x and y. ( I have order 2 types, and am trying to hook it up to a multimeter, and go to the track, hopefully that helps me make some decisions).

Dan,
Firstly, this will be for road course racing.

I think what I am looking for is simpler than you are imagining. Only 3 seperate automated positions, I will do a full brain spew of what I am thinking tonight.

To your point, I think overlapping signals (braking and turning( are the biggest issue, but I have thought hard about this and if I am on the brakes hard enough to need that wing angle I will not be turning, so the turn setting will have to defeat braking. I think that I will learn alot if I can get an accelerometer rigged up to a read out, and record the data from a few laps.

I will keep thiniking and post up later, I really appreciate the ideas and food for thought.

 

[PNachtwey]

Changing the wing/spoiler orientation as a function of g force and velocity should be easy. Picking the right actuators for the spoiler or wing might be a challenge. I see mounting, shock resistance, weight,heat, and power consumptions as being problems.
Do you have a formula for spoiler angle as a function of g force and velocity? If so then you have completed one of the harder obstacles.
If you have resistive feed back you really don't need to limit the spoiler position to just three positions. It will be just a easy to make it continuous. The inputs are continuous so the outputs will naturally be too. If you plan to have steps then you need to worry about when and how fast the spoiler angle will transition between the steps.
If you have even a small micro you should be able to get updates within a few milliseconds. If you want to implement the quickly I would get a small starter kit or development board and not start from scratch.


Peter Nachtwey
Delta Computer Systems

http://www.deltamotion.com