Show Posts
Pages: 1 ... 12 13 [14] 15 16 ... 21
196  Forum 2005-2010 (read only) / Interfacing / Re: Car diagnostic (speed etc.) on: October 21, 2009, 06:31:52 am
It's certainly *possible* one way or the other. Does that car have electronic gauges or is there a speedo cable from the transmission to the speedometer? If you have electronic gauges then it'll be more straight forward. What you do is find the signal wire for the speedo and hook an oscilloscope to it (don't have an oscilloscope? Get one, I'll wait...) Now, you should see that the signal wire is likely a digital signal going from 0V to likely +12V. The trick is to figure out how many volts it is (because you cannot directly hook up +14V to an arduino pin!) and how many pulses per mile are being output. There is a remote possibility that this info is available somewhere. If not you'll have to sort of guess then calibrate your guess later. It could be in the ballpark of 16,000 to 64,000 pulses per mile. From there you should be able to integrate for time to come up with the current speed. Frequency to voltage chips are also available. Using one would make your job a *lot* easier. Then you could just hook the F2V chip output to an analog pin on the arduino and read that. You'd still need to figure out which voltage corresponds to which speed. That's easy enough. You guess then drive 60MPH on your gauge while someone else follows you. They tell you how fast you really went and you divide from there. Note that you should really be careful doing this as 120MPH and 60MPH are not the same to an officer of the law. smiley-wink If you are obviously off by a lot then fix it before you go out racing around.

If you've got a mechanical speedo then you would probably have to put a hall effect sensor somewhere (and don't forget the magnet to go with it) and read the output of that to determine speed.
197  Forum 2005-2010 (read only) / Interfacing / Re: Trouble reading with pulseIn with Infrared on: September 29, 2009, 07:43:26 am
Heh, yes, I suppose that's true! I don't do the LED backwards trick personally. Why not use the actual proper part? It's not as if photo transistors are a million bucks.
198  Forum 2005-2010 (read only) / Interfacing / Re: Trouble reading with pulseIn with Infrared on: September 29, 2009, 06:49:22 am
The part about the circuit being wrong and frying the LED is true.

However, it's not strictly true that one cannot detect pulses with a normal LED. If you reverse bias the LED it can be used as a photo diode. This will work for detecting light at the same wavelength that the LED would have put out. In this configuration, when the LED is in the presence of the proper wavelength light, it will start to conduct.

I don't know if you didn't mention that because it only complicates things for people new to the game. And, it's pretty confusing and easy to mess up. I'm not really saying that it's a good idea, but it WILL work. ;-)

So, assuming that the original poster still has some LED's around:

Wire it so that the LED is in reverse bias with the +5V rail going to the cathode of the LED. Then wire the 240 ohm resistor to the anode. Finally, wire the other side of the resistor to the digital input pin on the arduino. Then try your sketch.
199  Forum 2005-2010 (read only) / Interfacing / Re: Blinded by the light on: September 28, 2009, 11:04:30 am
It sounds like you are on the right track at least.

My suggestions
1. Shield it better
2. Go ahead and find a photodiode with a much smaller angle
3. If the photodiode is not actually being totally saturated then how about using two photo diodes? One that is much more directly in front of the IR LED and one next to it. Then they both would get approximately the same level of ambient IR but the one would get much more IR from the LED. So you could use your opamp to get the differential output. Depending on the noise level and the degree of saturation this might work fine.
200  Forum 2005-2010 (read only) / Interfacing / Re: Interfacing with car sensors. on: December 29, 2008, 03:34:20 pm
Although, isn't a internal combustion engine (Otto or Diesel) a closed system? It burns fuel, and fuel has a lot of energy inside. But it is a closed system in itself.

No, a close system is a system in which no energy enters or exits. As such it's quite a rare thing indeed. The whole of time and space is a closed system as it contains all the energy it ever will and no energy will ever leave it either. In an engine energy is entering as fuel and leaving as work, heat, and gases.

201  Forum 2005-2010 (read only) / Interfacing / Re: Interfacing with car sensors. on: December 28, 2008, 10:06:21 pm
It's good that the arduino doesn't seem to be getting too hot. It might be alright so long as you don't draw a bunch of power off of it's pins or the LM7805 voltage regulator. Even if you do you can cool it down with a heat sink.

As for interfacing with the gas gauge... Well, 13 volts is an aweful lot but you might be able to pick appropriate resistors to tone down the voltage a bit. Heck, even if you get carried away with resistance and there is only a 1V swing from empty to full you'll be fine. 1024 steps over 5 volts is still 204 steps a volt. That's half a percent resolution (though the arduino is likely to jitter at least 3-4 values)

And, yes, water has a lot of energy within it. Fusion reactions can liberate massive energy from hydrogen. But electrolysis is a different story. Fusion can be self sustaining (until the hydrogen is nearly gone) but electrolysis is lossy and thus not self sufficient. Electrolysis can be done chemically but it uses up the chemicals in the process. It tends to be done electrically. This requires more energy input in electricity than you get back by burning it. The only way for electrolysis to be a net gain is to turn around and fuse the hydrogen to helium. This releases a lot more energy than it took to split the water but it requires gargantuan power to split helium back into hydrogen. So you cannot get around paying the piper. No matter what you do reversing the process will take more energy than the original process. Have you tried turning exhaust into gasoline? That's essentially what you plan to do with water. In the proposed system water is both the fuel and the exhaust. You can't power a process with the process. That's perpetual motion. You can use something up (hydrogen, oxygen, gasoline, whatever) but when you are done you are left with a substance which is worth less than the input. (In the case of water you start with water plus electricity and end up with the same amount of water you started with. So all the electricity is waste) This is basic thermodynamics. Entropy will always win.

Gasoline works because someone has already added the energy for you. Namely the Earth and sun. You are taking a product of someone else's energy expenditure and using it up. When you are done with gasoline you are left with water and CO (well, mostly...) Plants take this water and CO and use solar energy to grow. Animals eat the plants and use heat from the sun to stay alive. Animals then die and get buried by processes driven by the spin of the Earth and the sun. Intense pressure and heat over time turn the animal (and plant) matter into oil. Eventually people drill for oil (using up energy) and pump it out (more energy). They then separate it into various fluids, one of which is gasoline. But you can see how much energy went into gasoline before you burnt it. And that energy was imparted upon it before you ever touched it.

The first two laws of thermodynamics in laymans terms:
1. You can't beat the house, you can only break even
2. You can't even do that

More technically:
1. You can't change the sum of the energy in a closed system.
2. Entropy will always increase systematically across a closed system
202  Forum 2005-2010 (read only) / Interfacing / Re: Interfacing with car sensors. on: December 28, 2008, 07:46:30 pm

A lot of people says the same than you, and a lot of people say the opposite. Some say it is impossible getting more energy than what you put in.

However, when I burn gasoline, when I burn coal, when I burn fuelwood, or when I burn Plutonium, I get a lot more energy out of them than the energy I input into the equation to split these fuels out to burn or whatever the chemical reaction is.

It IS impossible to get more energy out of a fuel than what was put in. The thing is, you don't have to be the one who put all of the energy in. In the case of fossil fuels most of the energy comes from decayed organic matter that has been processed in high heat and pressure. The organic matter supplies the initial energy. Everything there after is refinement until you finally put it into your gas tank. A lot of energy has been wasted in the process and you cannot get anywhere near the energy output from the fuel as what was put into it over thousands/millions of years.

So, instead, I want to make a real test.

The first hydrogen generator I burned used 30 A and generated about 1,5 lt/min of HHO.

The one I am testing now, uses 15 A and produces 3 times more HHO.

What I mean, is similarly to any other issue, you may have different models with different efficiency.

Yes, you will see difference efficiencies depending on the process.

Just consider this. For your scheme to work you'd need to get more power out of burning the HHO gas than you put in. You are using the alternator to provide power. Where does the alternator get it's power from? The engine. Let's say that the alternator is 90% efficient. Now, thermal engines (which fuel powered engines are) are around 10-20% efficient. Multiply .20 times 0.90 and the total energy efficiency of fuel into electricity in the alternator is 18%. This means that for every 100 units worth of energy you burn you are getting 18 back to put into electricity. Electrolysis isn't 100% efficient either. By the time you get down to it you are looking at maybe a maximum energy conversion efficiency of 10%. This means you'd have to burn 10 gallons worth of water in order to extract enough energy to convert 1 gallon to HHO. That is NOT efficiency. Now, you can argue that you aren't converting all of the engine's work into power to split water. And you aren't converting all of the energy in that way. But, what energy you do use will have the efficiencies outlined. So, once again, If you consider gasoline and water to have the same energy potential per gallon (which they don't. Gasoline is one of the most dense energy forms available) then it would take 10 gallons of gas to be able to release and burn the energy from one gallon of water.

Some people argue that burning a little hydrogen along with the gasoline actually increases the efficiency of the engine. That could be true. But, as you can see above, it would have to increase the efficiency of the engine A LOT in order to even break even. Also, beware: burning hydrogen will increase the head temperature of your engine. If you burn too much you will warp your head and ruin the engine.

In summary: I will test it myself to be sure to say: "from the best of my experience, it does / doesn´t improve yield".


Sure, I don't mean to tell you NOT to do it. It's your choice. I'm just trying to let you know that what you really want to have happen is almost certainly physically impossible. The best way for you to learn that is to try it yourself. That will give you a much better understanding than anything I could ever say to you. In the end you'll end up more experienced in a variety of areas. That's not a bad thing at all. In fact, if I'm somehow wrong then you'll have done the world a favor by conducting detailed analysis.

Sorry for this long post. I don't mean to p$@% in your cornflakes but I do think that you should be aware of the potential roadblocks you are about to face.
203  Forum 2005-2010 (read only) / Interfacing / Re: Interfacing with car sensors. on: December 28, 2008, 06:45:41 pm
I'll tell you this much. It takes more energy to split water than you get from burning it. It's just physics. So, it depends on whether you are doing electrolysis onboard or ahead of time and compressing the hydrogen. Either way it's very unlikely that you'll see favorable results. This is actually an area I have done a lot of research on. You are really better just going with either hybrid electric/fuel or straight electric.

Oh, and a shameless plug. If you need parts for your 1989 Mercedes please try
204  Forum 2005-2010 (read only) / Interfacing / Re: Interfacing with car sensors. on: December 28, 2008, 05:54:47 pm

The arduino has an LM7805 voltage regulator (it's the IC) and so you can power it with anything from 7 to about 35v. At anything 12 or up you'll be generating a lot of heat on that regulator. The solution is to bolt a heat sink to the regulator or otherwise screw the regulator to the chassis of the car so that there is more surface area to dissipate the heat.
205  Forum 2005-2010 (read only) / Interfacing / Re: Getting sound into arduino. on: March 02, 2009, 11:53:32 am
This is harder than it would first appear. A guitar does not output a perfect sine wave at the primary frequency. Instead it's a complex waveform with a primary frequency and several harmonics all with attack, sustain, etc. There are ways to find the principle frequency: fourier transforms.

Be sure you are sitting down before you visit that wiki page... It's math soup.
206  Forum 2005-2010 (read only) / Interfacing / Re: Safe to connect analogOutput(PWM) to analogInput? on: February 16, 2009, 10:48:02 am
Ooops... Yes, I was being dumb. You're right, that's output pins.
207  Forum 2005-2010 (read only) / Interfacing / Re: Safe to connect analogOutput(PWM) to analogInput? on: February 15, 2009, 09:52:29 pm
I think that you would want to put a resistor in between the output and input to limit the current. The input pins can't take more than 40ma. That's a resistance of at least 125 ohms (at the 5V that the output pins output).
208  Forum 2005-2010 (read only) / Interfacing / Re: vexplorer hack on: February 14, 2009, 11:30:04 pm
If you really just need a simple answer and don't want the added commentary then the best I can do is this: probably. Take that as a yes if you want. But, don't say we didn't warn you.
209  Forum 2005-2010 (read only) / Interfacing / Re: vexplorer hack on: February 14, 2009, 11:14:35 pm
Heres a link to a grumpy mike tutorial on driving DC motors from an embedded system:

It pretty well covers everything.
210  Forum 2005-2010 (read only) / Interfacing / Re: vexplorer hack on: February 14, 2009, 10:52:34 pm
it will all work with what i have?

I don't know... Do you plan to power the H bridges off of the arduino's power? If so you at least need a heatsink attached to the LM7805 voltage reg. How much power do the motors on the robot need? You should have some decent sized caps for decoupling the power supply to those h bridges. Motors are inductive loads so hopefully your h bridge setup has backfeed protection diodes. Be sure to use fast acting diodes such as schottky diodes.
Pages: 1 ... 12 13 [14] 15 16 ... 21