How to make a magnetic speed sensor similar to those used to sense wheel speed in ABS systems

Step 1: A magnetic speed sensor is...

The sensor works using a iron cored coil with a magnet attached to one end. When a peice of ferrous metal is moved towards the end of the sensor it changes the shape of the magnetic field in the coil, this changing magnetic field then induces current to flow in the windings of the coil resulting in a small amount of electricity being generated.
This sensor only detects movement of ferrus objects near the sensor so is typically used for speed sensing (for example a wheel sensor in an ABS system) usually in combination with a toothed steel wheel.
They are very rugged sensors and are not affected by dirt and have a very high signal output making them less sensitive to noise (ideal for automotive applications)
For example the sensor I have made will generate a sine wave of 50Vp-p into a 1Kohm load (that is a peak of 50mV)
<p>can i see the circuits for real life use</p>
Nice project. <br> I'd like to see a basic circuit to check speed without needing to tote an oscilliscope around. <br> Is there a simple, cheap, robust, mobile, lightweight circuit that would ride on my bike, for instance? Would this sensor be sensitive enough to detect a bicycle spoke?
what would i do if i wanted it to detect a magnet. i lost the sensor for my bicycle speedo
What sort of landy you have? Mine is an old range rover with a faulty speedo, and I'd be very glad to install an alternative to know what's my current speed. <br><br>First I thought of fixing up a hall sensor somewhere close to the transmission brake drum and putting a small magnet just on top of that drum, but that might cause some disbalance and it would show me just the speed of vehicle, while setting up four bolts like yours would allow me to track the speed of each wheel - I could even vizualize the traction on a pda or an ipod :-) <br><br>Right now I am investigating how the ABS/TC sensors are installed on the newer landies. I'd be glad to see continuation of your post - have you installed your sensor?
Sadly I had to part with my Landrover before I got any further with the project. (My first child turned up - now just turning 4 on Monday!)<br>You will have to take on the mantle and post an instructable showing the sensor in action.<br>Pete
i would love to pick your brain over how you are going to mount this on your engine because i need a speed sensor on my 73 v8 dodge honey bee so i can add cruse control
Hi, I'm not Peter the ungreat but I have a suggestion. take it or leave it. AutoMeter makes a digital speedo kit which comes with a modern sensor that you could &quot;piggyback&quot; to get your cruise input from. (you would lose your stock speedo functionality though.) You could use a modern rear diff/axle assy with an ABS sensor. This could be the input for cruise, and you would still have the cable op speedo. You could modify the oem gear to fit a modern speed sensor that might fit the original location on the extension housing of your trans, again this would make the oem speedo inop. Last but not least.... Old Cruise control units were vacuum operated, find one of your vintage and make it work. I hope this helps.
I actually found a old dodge motor home in the junk yard while harvesting a carb i pulled the cruse control and its running much better now except the cab over bed needs work
Nice!&quot; well, except for the fact that bodywork isn't any fun. &quot; best of luck to you.
ya well it turns out that there was water damage in the frame and a crunch with a branch smashed it in. to bad but6 its tarped off waiting for the snow to go
Fantastic! I love it, some nice bit of magic there.<br><br>I have a tricky question. I want the sensor to produce a signal, which goes to a guitar bass amplifier, and produces a thump, when metal comes close to it. <br><br>How do I make the sensor produce a signal around the 33-100 hz range ?
is there a limit to the maximum signal frequency ?<br><br>I'd like to make a sensor like this to measure the six sided nut from a turbocharger shaft with a top speed of about 120 Krpm<br><br>not sure if I should go with an optical sensor instead (also not perfect because of condensation in impeller inlet blocking the view)
Hi,<br><br>I have never tried to find the upper frequency limit, but I guess it is limited by the iron core - I know that when you start making high frequency transformers you need to use dust cores rather than iron. although having said that it is the change in flux through the windings that matters.<br>My best advice is give it a go. To test I suggest using a large diameter toothed wheel running at a lower speed.<br>This sort of sensor is very good in a dirty environment - however I know that rare earth magnets are limited by temperature (about 90 degrees Celsius) and they start to loose their magnetism.<br>Hope that helps<br>Pete <br>P.S. could you use an audio sensor with a high pass filter as nothing else in the engine will emit the frequency that the turbo will?
How do i put the magnet in between the coil and the iron...there is no gap there...any workaround??
The magnet is just stuck to the end of the nail using it's own magnetism. Stick the magnet on the non-sensing end. (See step 5)
Do we have to fit the sensor inside the bolt???If yes..why??<br><br>Does the sensor work directly without enclosing it in the bolt...Please reply urgently...
Yes, the sensor works fine un-enclosed. The bold is just to make the whole sensor rugged and easy to mount securely.<br>This type of sensor is often fitted in very hostile environments (I.e. near the road on a car - hence needing to be very rugged)
Do we have to fit the sensor inside the bolt???If yes..why??<br><br>Does the sensor work directly without enclosing it in the bolt...Please reply urgently...
I might be missing something, but that looks AWFULLY similar to a Guitar pickup to me. I have a feeling that this is more or less the way Hammond make their &quot;tone wheel&quot; organs. The profile of the wheel obviously reflects the wave form of the desired sound....<br><br>Maybe I'm wrong, but it DOES look like one could feed the signal into an audio output device and yield some VERY interesting effects.
This is actually just the same that Hammond organs have: tonewheel.
Well done.<br /> This looks to be what I'm after, the&nbsp;general&nbsp;design looks about right.<br /> I'd assume that if&nbsp;the&nbsp;sensor were looking for magnets passing past it on a rotating wheel it could work over a longer&nbsp;distance?<br /> <br /> Does anyone know if this is available as a pre-made sensor by any name?<br /> <br />
Thanks! Really nice job. A great alternative for rigging up a quick custom sensor. Excellent.
I think I'm going to end up doing something similar to this in order to detect the speed of a moving (ferrous) object. The amplitude of the signal should be proportional (?) to the speed of the object, right? Any ideas as to how I can figure out what the proportionality constant is? Do you think it's going to be linear? Awesome project, btw!
I would like to rotate the magnet and see the voltage rise and fall as it passes the coil. Would I just need a coil? Is there a simple way to get an iron core to wind the wire on? How does the coil winding orient to the magnet? If I need a 10 volt output do I put on more turns or get a stronger magnet? Both?
A steel bolt will do. Radio Shack sels (or used to; it's been a while) magnet wire in small spools. Magnet wire has thin enamel insulation that lets you pack more turns into a given space, but any insulated wire will work. You're not trying to generate significant current, so smaller wire will let you use more turns. Note that small coin-shaped magnets are usually have their poles on the flats, so you need to spin the magnet like a flipped coin. The voltage you get is proportioal to the product of the strength of the magnetic field, the number of coil turns and the speed of the coil through the field. Depending on your application, you may be able to save some time by driving a small permanent-magnet motor as a generator.
Sorry but I don't get the "(that is a peak of 50mV)" part. I thought 50Vp-p meant 50 volts peak to peak.
i think it needs the hall effect wheel gear sensor to connect,and it is very cheap, about USD18, and if anyone want to buy, please contact me. my MSN: vicky.hu_123@hotmail.com
Ya its true and it will be useful for all if hall effect whell gear sensor to connect. How much prize it cost?<br/><hr/>williamgeorge<br/><a rel="nofollow" href="http://www.drivenwide.com">Buzz Marketing</a><br/>
would this work in a similar way with a coil that detects a magnet fixed to the moving wheel? I want to use a reed switch like a bike speedometer, but I don't know if it can switch on and off at 10k RPM...
i'm lovin the old school game boy in the back of this picture
<h2>relays use cadmumium contacts and cadminum is a toxin eeeek!=</h2>
Reference please.
This is almost the same concept as the Tonewheel in the old hammond electric organs, instead of a magnet like you have Hammond used a guitar pickup and amplified it.
I really love your project but I wasnt really minded doing it cause i dont have any oscilloscope and it would be too hard to find an old computer and not compact enough..So i decicded to check in my workshop, i found a kellogs pedometer. Easy to work with, why not use this? incredibly easy, kellogs pedometer rules!
For those of you out there who want an <strong>Oscilloscope</strong>: build one! Seriously. Find an old Pentium 1 or 2 machine, put a teeny hard drive in it, install a small OS (linux, or windoze 95-8) and get a couple oscilloscope programs. There is a ton if info on the web about said <strong>FREE</strong> programs, and how to interface them with computer and the test leads. They may not be as pretty as having a packaged oscilloscope, but they get the job done. In the US you can usually find these old computers abandoned with an operating system still on the HD. Do some basic cleaning, and you have a scope on the cheap!<br/>
where to buy a cheap oscilloscope?I want to build this machine
Check you nearest local thrift/recycling store. My bro-in-law found an oscilloscope for $3US that just needed a fuse. His local store's name is Deseret Industries. E-bay probably has a few at near-giveaway pricing too.
Question. Now i know very little about this but had an idea. Theoretically since this just sends out a current proportional to the amount of ferrous material near the sensor could it not be used as a rudimentry metal detector? Thanks
Yes, fundimentally it is a kind of metal detector, however it does not produce an output when there is no movement of ferrus material, it's output is proportional to the ammount of ferrus material moving through the field. If you wanted to use it as a metal detector you would have to 'sweep' the sensor over the area to be detected. It would also have a very small range since the field at the end of the sensor seems to be quite tightly contained. Pete
Nice- although there are easier sources of magnets. You could probably sell the phone for $20 and buy a ton of small neodymium magnets!
How long is your coil? Does it have more than one layer of windings? You're right about a hall-effect sensor being less rugged. Our robot team used a couple laster year to count sprocket teeth. They had to be so very close to the teeth that the tiniest amount of deflection would, er, did destroy them...
The coil is just salvaged from a relay (it was a 12V 1A relay - so quite small) The coil does have many layers, I am guessing about 10 or so. If you want to wind your own coil I suggest either experimenting with different numbers of turns and measuring the output or calculating the no of turns from the desired output (don't ask me to help with any calculations!) These sensors also work best if they are close to the teeth but they will give a useable output with a gap of 0.5 to 1mm. If they do touch the teeth however they tend to just make a nasty graunchy noise and then carry on working Pete
A hall effect sensor is a semiconductor device and relies on a current being deflected inside a peice of silicon in the presence of a magnetic field. A hall effect sensor will detect a static magnetic field, the one above will only detect motion. I think a hall efect sensor is less rugged in terms of noise immunity and heat immuntity, it will also need a power supply to it (I.e. more wires or a bit of circuitry in the sensor) Pete
is this the same as a hall effect sensor?
My motto in life is 'laziness is the mother of invention' why make a coil when someone else has done the hard work for you! Plus relay coils usually have lots of turns which is what you need to get a decent output. Pete
Smart move using the relay for the coil! Good instructable!
Cool project.. I have two questions though: 1- what would the signal be like without using the magnet (which is attached to the end of the coil core)? 2- in the case where i am monitoring a plastic disk instead of metal and i attach that magnet to the disk, would i get the same results? Thanks!
1) Without the magnet you will get no signal at all unless there is some residual magnetism in the toothed wheel. The sensor relies on the deformation of the magnetic field crossing the wires of the coil. No magnetic field - no current generated. 2) Yes a magnet on a plastic wheel should generate some output in the coil, it will be higher if the coil has an iron core (I.e. a nail or something). Have a play and see what works. If you have a plastic disc and a relatively clean and noise free environment an opto sensor with a slot in the disc would probably be better. Pete
hey man, just took a look on your profile, you training to be an architect? well, i did walk that path long ago buddy.. am an architect! i live in japan now where i continue my studies
Yup - I am 6.5 years through the 7.5 year process! Pete

About This Instructable


99 favorites


Bio: I am Married to Beth, I am an Architect and have four wonderful children
More by PeterTheUnGreat: Remove a stuck alloy rim without damage Find an exhaust leak without burning your hands! USB pimp my Landy
Add instructable to: