145Views19Replies

Author Options:

I need help! Answered

Greetings! I have never built an electronics project like this, however, I do understand the basics of what needs to be done.

We are building a car that will run on the Bonneville Salt Flats in 2010. What I am looking for is to build a device to show the difference in wheel speed between the front wheels and the drive wheels. The reasons for this is to give the pilot information on how much traction he is getting at the rear wheels. Since the engine and drive wheels are located 15 feet behind the driver, you cannot do this by ear (especially at 400+mph). The second reason is that the tires used at these speeds cost $800 each, there will be 4 drive wheels, that's $3200 worth of tires. Because of the speed, the tire treads are very, VERY thin, any spinning will destroy the tires in short order. It's not just a money thing, at 400+ getting a flat tire could be fatal.

My current thought is to place wheel speed sensors at one front wheel and one drive wheel. That data will be collected and one value subtracted from the other. That difference will be presented to the pilot. I am not opposed to an LCD display or a series of lights that will indicate 5, 10,15, 20 and over 25mph difference.

I am a computer programmer by trade and we are planning to integrate a PC into the car for data logging and tuning. I will have a complete CPU available managing the input and output. If we can use the PC to create an application to do this, it should not be an issue. If it is easier/better to build a chip-set to do this, we can go that route as well.

I am open to any and all ideas, I have a rough idea of what needs to be done, just no knowledge of parts or processes to complete it.

Thanks,
Brandon.

19 Replies

user
lemonie (author)2009-11-26

Looks fun - what's the power-plant and what are the tyres made of?

L

Select as Best AnswerUndo Best Answer

user
bbarn (author)lemonie2009-11-27

The tires are a Kevlar chorded tire with about .030 of rubber over them for the tread. The Mickey Thompson tires that are on our lakester are speed rated to 590 mph, that's probably what we will use on the streamliner as well.

The engine in the lakester is a 540 cubic inch twin turbo-charged V8. Each of the turbos came from a large truck application (tractor/trailer size). They are 88mm Garret turbos, each one moves about 1350 cfm of air and charge the system with up to 38 pounds of boost. Your average production turbo engine usually puts out somewhere between 5 and 15 pounds, you can see we are WAY above average.

The new streamliner project we are working on will have a 496 cid Chevy with twin turbos as well.

You can see the current project here www.spiritofstlouisracing.com , there is also a really good video of the lakester on the dyno search youtube for "Lakester Test", I can't get the post to link correctly here. The donor engine we are using for the new project can be seen on the related videos page "Rob's 63 LeMans on the OST Dyno" as well as at www.onebadpontiac.com. We will be doubling the size of the turbos and using a modified cam to increase the HP by about 600. The other gains in HP to the rear wheels will come from removing the torque converter an making it a direct drive.


Here are some specifics I need help with.
High level goals:
Collect input from multiple sources (2).
Aggregate the input by subtracting non-powered input from powered input sensors.
Output difference in sensor input to a gauge, bar graph or series of LED lights.

Specific questions:
Which PIC is best suited for accomplishing the goals?
Which software(s) can you recommend for coding the logic?
What equipment is needed to transfer programming logic onto the PIC?
Will I need to design other items into the circut prior to or after the PIC? (Basically, do I need to breadboard a PIC and the sensors with resistors, capacitors or other elec equipment or can I just directly connect two sensors and an LED bar graph to the PIC? If you can't tell by my question, I have no idea how to design the circut...)
Is there an electrical circuts for dummies book or some reference material that you can recommend? I'm not afraid to do some research and reading, however, I'm not looking to earn a doctorate in electrical engineering either. If I am going to be spending 6 weeks reading how to do it, I'd just as leave pay someone to do it for me. However, where is the challenge and fun of having someone do it for me?

Thanks for you postings, I appreciate the help.

Select as Best AnswerUndo Best Answer

user
lemonie (author)bbarn2009-11-27

I'm more interested in big engines myself. I looked around and found that Malcolm Campbell hit ~300 (Bonneville) with a 2,800 HP Rolls Royce R, and George Eyston got a bit over 350 (Bonneville) with two. That was the 30s, but your machine still seems a little underpowered for 400+
?

L

Select as Best AnswerUndo Best Answer

user
bbarn (author)lemonie2009-11-27

They are making great strides. I watched the "Spirit of Rett" run a 392 with an exit speed for 403 this October at B'ville. They are running a naturally aspirated 500+ cid engine. At B'ville elevations, they can't be making much more than 800-1000 hp, we will be making almost double that. If we can get the power to the ground, we should be able to get it there. This is why we are looking for a slip indicator, it will help make sure we are putting the max power to the ground without wasting any on spinning the rears.

Select as Best AnswerUndo Best Answer

user
lemonie (author)bbarn2009-11-28

How things advance.... I see the need, but wonder whether if you asked nicely one of the F1 teams might give you one?

L

Select as Best AnswerUndo Best Answer

user
bbarn (author)lemonie2009-11-28

Where's the fun? I love to learn new things by doing them myself, at least once! Heck, before September 2009, I didn't even know that B'ville was still running. Now I have crewed on a car, am building a car, and am PM for a serious effort at an alternative fueled (Ethanol) LSR vehicle. It's fun to learn new things, that's why I picked this project. It should be fun to learn how to do and may teach me some other new tricks I can apply elsewhere.

Select as Best AnswerUndo Best Answer

user
lemonie (author)bbarn2009-11-28
user
Weissensteinburg (author)lemonie2009-11-26

Good question, and I wonder what the tires are made of, too.

:P

Select as Best AnswerUndo Best Answer

user

u would think they were made of dimonds or something.......

Select as Best AnswerUndo Best Answer

user
NachoMahma (author)2009-11-24

.  I'd go with a unitless bargraph type display. Logarithmic so that just a little slip make a big indication.
.  -  [][][][][][][][][][] 0 [][][][][][][][][][] +
.
.  Logrithmic so that just a little slip make a big indication.

Select as Best AnswerUndo Best Answer

user
bbarn (author)NachoMahma2009-11-24

I like it, however the graph would only need to be  presented this way
-0 [][][][][][][][][][][][][][][][][][][][] + The only case for the front wheels to be going slower than the rear would be if you were on the brakes and locked up the rear wheels, it is not necessary to alert the pilot of this. (Actually, with both parachutes out, the brakes are not even necessary.)

For the display this is a GREAT idea. What I am really in need of is HOW to get it to display. I am a complete noob to the world of electronics. What I would love to get is how to capture the output of some speed sensors and how to convert that information to the graph. Part number and/or examples of devices and how they do/do not integrate with a pc would be great.

I do not have an issue with building a standalone system (ie. not integrated with a pc), I just have no technical experience with how to accomplish this. I am sure there are microprocessors that can be built into a breadboard that will accomplish my goal.

Thanks for the post, I greatly appreciate the input. BTW, I would be happy to document the project here an let everyone see the final product.

Select as Best AnswerUndo Best Answer

user
Goodhart (author)bbarn2009-11-25

Well,  just throwing out an idea here (as it will be a wee bit more complicated then this example) but if the spinning can be attached to a small DC motor or generator (very small, but high current) then a voltage by means of rotation speed is generated:  attaching something like this circuit to it will accomplish what you want to know.

Select as Best AnswerUndo Best Answer

user
bbarn (author)Goodhart2009-11-26

That design looks like it will work for a single wheel speed, but I actually need to compare two wheel speeds to each other and display the difference. I was looking at using a hall effect sensor and sprockets to collect the data on both wheels, then subtract the front (non-powered) wheel from the rear (powered) wheel speed and graph or indicate the difference. Where I am really struggling at is what goes in the middle. Is a PIC the way to go or is it easier to capture via PC? I have no experience with PIC, and way too much experience with PC.

Select as Best AnswerUndo Best Answer

user
Goodhart (author)bbarn2009-11-26
but I actually need to compare two wheel speeds to each other and display the difference.

Would the use of two of those circuits and a comparator IC work? 

Well, the PIC or another up like  the Aduino which has plenty of sources out there...software-wise, would be the smallest solution; for the way you are looking at it.
 

Select as Best AnswerUndo Best Answer

user
NachoMahma (author)bbarn2009-11-26

.  First thing I'd do is figure out max RPM of each tire/wheel.
.  Then I'd see if a reluctor system similar to an automotive distributor would work at that RPM. Should be able to feed those signals to a microprocessor with little or no signal conditioning.
.  Since the wheel/tires are different sizes, you will need to multiply one set of inputs by a fudge factor so the numbers are the same when no slippage present).
.  Compare.
.  Send signal to indicator in proportion to difference

Select as Best AnswerUndo Best Answer

user
bbarn (author)NachoMahma2009-11-24

By the way, you can check out our current car running here:
www.spiritofstlouisracing.com

This is a "lakester", it makes 1,800 HP and should run 340mph in it's current configuration. We are just the builders for this particular car, not the drivers. We are currently building our own to drive in the Streamliner class. It's about the same length, just way more aerodynamic and should make a bit more power, somewhere in the neighborhood of 2,000 hp!

Select as Best AnswerUndo Best Answer

user
Marsh (author)2009-11-26

 You really only need a single white bar on a field of black to get speed. Shine an LED on it and detect the white spot with a photo transistor. You could pick up the signal directly using a standard PC game port (it has like 4 debounced switch inputs). This way you could detect the speed of each wheel.
At high speeds like this, there's no telling what could happen. A bearing could seize or begin to chatter. Who knows what else could happen. 

Select as Best AnswerUndo Best Answer

user
bbarn (author)Marsh2009-11-26

The biggest concern that I have is that the rear wheels loose traction for an extended period of time, they will shred, but will not fully deflate. While safety is a huge concern, getting up to speed is the goal. Loosing traction means we are expending engery without gains. Since the pilot can't effectivly hear what the engine is doing, some means of indication is called for. What I really need is guidance on the best / easiest method of aggregating two inputs and displaying the difference. I think a hall effect sensor is probably how we are going to go, one of the other issues we have is that the front wheels are a different diameter than the rear. Using a toothed wheel and a HES, we can alter the teeth count between front and rear to "calibrate" to the difference. My struggle is how to best perform the collection and display - PC or PIC. I have no experience with a PIC and would appreciate some hand holding through the startup process.

Select as Best AnswerUndo Best Answer