I'm doing research for a speedometer project.
Most of the arduino based projects I've seen use a Hall Effect Sensor. but All the commercial units I have installed at work (I work on old pre-computerized cars) use a magnetic pick up like this one: Rostra Vehicle Speed Sensors
I had an opportunity to speak to a tech rep and he claimed that a magnetic pick up was far more rugged and reliable.
Rather expensive for what I am pretty sure is just a coil with a bolt through it. Same as the electromagnets we made as kids.
Does anyone know of a cheaper source for these?
If I either find one cheaper or I have to make one I will need to figure out how to read it.(could not find a data sheet for this)
The only test equipment I have is a DMM.
I am thinking that i could use diodes to clip the signal at 5v to protect the arduino but will likely need to amplify it first and possibly use a schmidt trigger to clean it up? or am i overthinking this?
Am I on the right track? any advice on how to proceed would be greatly appreciated.
EDIT: I am a god!! albeit not a very bright one. LOL see post count
If a magnet moves by a coil, a voltage and current is generated. The values depend on the strength of the magnet, distance from the coil, the speed it is moving and the size, wire resistance and number of turns in the coil.
If you give some details about what you are after, perhaps we could advise further. But, it is easy to wind coils and experiment.
In the commercial unit the magnet is mounted to the driveshaft. and the coil is mounted 1/2" to 3/4" away
speed would range from 0-10k RPM. ( not likely to be that high)
that coil is approx. 1" diameter x 1/2" thick. I do not know the no. of turns or wire gauge used. bolt is 5/16"
I can experiment just don't want to burn up an arduino if the pulse is to great for it. what would be a safe way to do this without a scope?
Pass an inductor through a magnetic field fast enough and you will get a voltage spike. I think you can make this yourself with a little trial and error with the number of windings (lots), strength of the magnet, and gap between them.
I don't think you will need the diodes on the arduino, as by using the schmitt trigger it would already be regulated. Also, using the diodes means diode drops, and you might not have enough voltage to afford that drop.
I just made a quick setup of a coil hooked up to my oscope. It is actually part of an electromagnet used for uncoupling HO scale model train cars, but it's a coil none the less with at least 1K windings, maybe more. For my magnets, I have some small rare earth magnets in a container.
With both wires of the coil hooked to the scope and passing the magnets over the coil at 1/2", I'm getting 5-10mV depending on how fast, and because I'm going back and forth, it looks more like a sinewave signal. The faster you go, the bigger the signal.
If you are determined to make your own, an oscope would be really handy to see what you are dealing with. You might be able to set up a motor with attached magnet (or bicycle tire, or fan with magnets on blade tips) to get a signal going and read it in mV on your DVM and build a circuit around that.
I fully understand the "make my own" thing, regardless of time and cost, but in the end, you are still using an analog signal and converting to digital outside the arduino, then using that input vs using a digital sensor straight to the arduino. A hall effect sensor can be had for a couple dollars.
Hutkikz:
In the commercial unit the magnet is mounted to the driveshaft. and the coil is mounted 1/2" to 3/4" away
speed would range from 0-10k RPM. ( not likely to be that high)
that coil is approx. 1" diameter x 1/2" thick. I do not know the no. of turns or wire gauge used. bolt is 5/16"
Only issues I see are slow rpm's not producing much signal and the driveshaft not staying the same distance from anything. (Unless you attach to the flange on trans or pinion)
The coil I used is approx the same dimensions as the one you describe.
You have a Variable Reluctance speed sensor. There are many automotive ones available and you should be able to find something inexpensive.
Here's a basic reference
As you have found, the output signal needs conditioning before use in the Arduino. You are not overthinking the issue.
I had an opportunity to speak to a tech rep and he claimed that a magnetic pick up was far more rugged and reliable.
I'm not sure that the VR sensor advantages of self powering and wide temperature range, offset the low voltage signal conditioning issues. Personally, I'd go with a Hall sensor.
@tinman13kup
Thanks for taking the time to experiment for me!
I would buy one if I knew where to find a workable coil at a more reasonable cost.
The only reason I wanted to go this way is because the tech rep. was adamant that using one was much more suitable for this application than a hall effect. maybe he was just defending his companies choice. Although the fact that every commercial unit I have ever seen over the past 20 yrs. uses the same does seem to support his claim.
An Oscope is on my wishlist, just not in my budget yet. Have been watching C/L but everyone in my area seems to think old broken 20Mhz analog scopes with no probes are worth $300+
the sensor is mount near the transmission where the driveshaft doesn't move much.
I am starting to think I'm swimming upstream here.
Whichever sensor I use will have to live outside under the car and survive our michigan winter weather conditions. this means it will be dragged through the snow a lot. This is the primary advantage mentioned by the tech rep.
The sensor I am referring to does not exactly match the description given for a VR sensor
" (VR) sensors are built with a wire wound around a permanent magnet"
instead the coil is wound around a ordinary bolt and the magnet is mounted to the driveshaft. Small difference I'm guessing.
I may just go ahead and try a hall effect first and see if it can survive this winter, I can always revisit this idea if it doesn't.
Hutkikz:
I may just go ahead and try a hall effect first and see if it can survive this winter, I can always revisit this idea if it doesn't.
You could always use a hall sensor from a car. My truck uses one for crank position and is totally encapsulated in plastic. It is a 5v sensor and can withstand the harsh temp and environmental conditions (oil, vibrations from diesel etc) It also has a handy (in your case) weatherproof connector. The cost new from the parts store is $15, magnet not included.
You could also look at ABS wheel sensors. Some are pretty cheap and thousands of different mounting configurations.
You would get better low speed resolution if you used a multi tooth wheel to trigger it.
If you want to use a hall sensor, there are threaded and fully encapsulated ones on E-Bay that are pretty cheap. Do a search for threaded hall proximity sensor.