Steering angle sensor- best ideas!

All sensor specific discussions! Temperature, pressure, steering angle, brake and throttle, etc. Post adaptations of OEM-style sensors, and also your clever DIY hacks and custom designs here too!

Moderators: JeffC, rdoherty, stieg

brentp
Site Admin
Posts: 6282
Joined: Wed Jan 24, 2007 6:36 am

Steering angle sensor- best ideas!

Post by brentp »

Ok, Steering angle is one of the most valuable pieces of data to log on in the race car, so share your best idea! Anything goes.

- spring loaded lanyard connected to potentiometer shaft?
- belt driven multi-turn potentiometer?
- OEM steering angle sensor like from RX7?
- ???
Brent Picasso
CEO and Founder, Autosport Labs
Facebook | Twitter

Jaynen
Posts: 11
Joined: Mon Jan 07, 2013 9:02 pm
Location: United States

Post by Jaynen »

I've been looking for some options. There is a few spec miata threads about how they did theirs. It seems like the string potentiometer is the most popular.

It definitely tops my list along with RPM/TPS/Brake pressure

parid
Posts: 7
Joined: Mon Jan 07, 2013 8:55 pm
Location: Eugene, OR
Contact:

Post by parid »

Ironically, I even have an RX7 with a stock steering angle sensor... but I welded an aftermarket wheel to the splines on the steering column and the sensor no longer works or fits. If there was a solution easier that redoing our whole steering column that would be welcome.

GameboyRMH
Posts: 55
Joined: Mon Jan 07, 2013 10:45 pm

Post by GameboyRMH »

Hey Jaynen :D

We were discussing this very topic on GRM forums and a string potentiometer (lanyard idea) looks like the best bet. They're pretty expensive, well over $150 for the cheapest units, so building your own is worth it.

You can either measure at the rod end or have it wrap around the steering shaft. All the stock steering wheel position sensors I've found are just big encoder wheels, so a startup calibration procedure would be required to use those.

357
Posts: 3
Joined: Tue Jan 08, 2013 5:05 am
Location: Arlington, TX

Post by 357 »

What about using a linear distance sensor on the rack itself? Optical's the easiest, although an LVDT or linear potentiometer might give more precise high-rate data. An initial calibration of the steering angle-to-rack position curve would be required, but I would think you'd only need to do it once, as long as your data visualization/analysis tools had the ability to apply that calibration when using the raw data.
HCH-I/GMT800-HD/E34T

ebdavis
Posts: 4
Joined: Fri Nov 29, 2013 8:41 pm

Post by ebdavis »

I'm guessing the optical distance sensor from the "sensors" page that was last updated on 16 July 2013 http://www.autosportlabs.net/RaceCapturePro_Sensors
Is the current prefered method for measuring distance? Has anyone set these things up? Can the dumber people get a webinar or at least a flowchart on how to do this lol? They sure look a lot cheaper than string pots.

GameboyRMH
Posts: 55
Joined: Mon Jan 07, 2013 10:45 pm

Post by GameboyRMH »

A linear distance sensor on the rack would work fine, it's just that linear distance sensors are very expensive.

I'd say the optical distance sensor is still the best way to do it, but I've had an idea to use a BMW suspension travel sensor (looks kind of like a big RC servo) attached to the rack...that's more reliable and accurate but requires a lot more fab work.

ebdavis
Posts: 4
Joined: Fri Nov 29, 2013 8:41 pm

Post by ebdavis »

I am still setting up my rcp so I have no data to back this up but my current plan is to use throttle position sensors for steering angle and suspension movement and an EGR position sensor for brake pedal. The are super cheap and available at any auto-parts store.

The problems to overcome are:
1. Mounting.
This is true of any add on sensor short of tapping into factory ones which Ive read is a bad idea. Some aluminum angle or bar has been helpful so far in my application.
2. Actuation
The problem with making a sensor detect a linear movement when it was designed to measure a 90* movement can be overcome by attaching an arm to the sensor. A better way (due to having a true linear relation) would be to attach a cam to the sensor such as whats attached to most throttles. The size (diameter or radius) of said cam would be determined with some math to give the full range of motion that the application requires. So the bigger the cam attached to the tps the longer a linear motion it can translate. Another thing to decide is whether or not to actuate it by attaching the wire or string that will twist the cam to the steering arm or to have it wrap around the steering shaft. If wrapping around the shaft more math (which I feel is just as bad as the worst four letter words) will be needed obviously to decide the size of the sensor cam.

My current plan is to use one off an 06 kia optima (google a pic) because it has a metal tab that is relatively easy to rig a cam/arm to. The egr sensor is more straight forward in that all it needs is to be placed in a position close to the pedal that allows full movement of each, so closer to pivot point for less movement and further for more. I'm not able to put the time into this that I might have years ago so I have no hard tech to post but I thought there may be others that scoff at the prices of such things as string pots etc. as I have.

brentp
Site Admin
Posts: 6282
Joined: Wed Jan 24, 2007 6:36 am

Post by brentp »

Thanks for sharing!

If you do use an arm you can re-map it to be linear again by creating a custom mapping in RaceAnalyzer.

For brake, your best results will be an actual brake pressure sensor, but a linear sensor would give some approximation, as you know. Please post pics as your project progresses!
Brent Picasso
CEO and Founder, Autosport Labs
Facebook | Twitter

jakekooser
Posts: 64
Joined: Wed Apr 16, 2014 11:03 am
Location: Florida
Contact:

Post by jakekooser »

For my Porsche 911, I was thinking of putting a pulley around the steering shaft where it passes through the 'smugglers box' and using a 10-turn potentiometer. Where I work we use these potentiometers for monitoring the position of a rotating assembly and it's reliable in my experience.

Been doing some googling and something like Motec has is what I had in mind: http://www.motec.com/sensors/position/

brentp
Site Admin
Posts: 6282
Joined: Wed Jan 24, 2007 6:36 am

Post by brentp »

If you use a pulley, you definitely want to use a cogged belt - looks like a miniature timing belt.

Yes - those motec sensors look very expensive.. :-/
Brent Picasso
CEO and Founder, Autosport Labs
Facebook | Twitter

jakekooser
Posts: 64
Joined: Wed Apr 16, 2014 11:03 am
Location: Florida
Contact:

Post by jakekooser »

yeah, they typically use 1/4 inch wide timing belts. I 'may or may not' be mooching some parts from the office to test this idea out with. Jameco has 10-turn pots for under $20, who knows what the luxury of a Motec label would cost you. :)

brentp
Site Admin
Posts: 6282
Joined: Wed Jan 24, 2007 6:36 am

Post by brentp »

Take pics of your work please! We've found the biggest challenge is to find a large bore pulley to fit the steering shaft. Any part numbers you find will be appreciated!
Brent Picasso
CEO and Founder, Autosport Labs
Facebook | Twitter

jakekooser
Posts: 64
Joined: Wed Apr 16, 2014 11:03 am
Location: Florida
Contact:

Post by jakekooser »

will do. Looks like there is quite a variety of pulleys available. I'm thinking I may have to take a large pulley and split it to go around the existing steering shaft.

I'll search throughout the day on and off as time allows.

http://www.amazonsupply.com/search?keyw ... g%20pulley

smig
Posts: 31
Joined: Mon Feb 11, 2013 7:39 pm

Post by smig »

Here's what I came up with.

I started with a couple of gears that I got off amazon and a single turn precision potentiometer and did some machining/fab'ing and came up with these.

Image
The aluminum bracket will be trimmed to size and fix/clamp length wise down the steering column housing.

Image
The drive gear mounts to the column, siding over the spline and will mount on the actual steering column shaft with 3 set screws.

Both original gears were machined for a press fit.

The precision pot is a single turn with no stop. The gear ratio is 1.33:1

Time will tell if it's effective.

Post Reply