Electronics question

[Bill Musarra]

I have a problem that I am hoping someone can help with. I am trying to measure the pulses of a magnet such as those in the 930 transmission. I am told that it can be measured with a multimeter such as the ones from Harbor Freight etc. I can not figure this one out. Have tried most of the combinations without any success. I will try to measure the sensor that Stewart Warner uses for thier speedometer. Mine is at the front wheel and measures the bolt heads from the disc brake. This is a 2 wire setup. If there is a waay for this signal to be shared, then I can use it for a source
What I am really looking for is the number of pulses that the transaxel has per revolution, or for the front wheel. This number will be used in the programing of the electronics in my power steering. But I can't get anything to come up. The only sensors that I can find that will count the pulses is a "go-no go" which I assume is what the speedometer people use to get a signal. The wire from my transaxel is a single wire. I don't know if I have to supply a voltage source to the trans and read the sensor or what. Any help would be appreciated.

Bill

 

[David]

Presumably a VR sensor ( Variable reluctance sensor - Wikipedia, the free encyclopedia ) earthed to the case.

I'm not sure exactly how you'd read it, I'd do it with an arduino but I've not done it before so would have to look it up myself.

 

[John Fitzpatrick]

It sounds as though you are trying to measure frequency with a voltage meter. Multimeters are typically analog devices and your pulses are digital. For that you need a digital pulse counter that would count the pulses produced by one revolution of your wheel. Pulse counting can usually be selected on a digital frequency meter which measures the time between a signal's leading (or falling) edges and converts that to its period in seconds or its frequency in pulses per second.
An easier method would be to buy a cheap USB oscilloscope and use it with your laptop. That way you could see exactly what's going on. Then you can adjust it to measure exactly what you want.

If you do have a hand held pulse counter, it could be seeing a high DC voltage that swamps the small pulses from the magnet. Makers usually put capacitor is series with the input to prevent that, but the lower the input frequency, the bigger it has to be.
You trying to measure a very low frequency pulse waveform--what's the low range of your meter?

 

[Mike Pass]

I suspect that the voltage of the pulses is very small. If they are in the thousandths of a volt range then a basic multimeter will not detect something that small.
An oscilloscope will detect very small voltages as it has an amplifier built in. The oscilloscope will also give a display of rapid pulses and from the time base markings on the screen you can work out the frequency of the pulses.
Cheers
Mike

 

[Bill B]

Bill,

The typical variable reluctance probe is going to generate very low voltage pulses, especially at slow passing speeds. As the gap between the ferrous material you are sensing decreases and the relative passing speed increases the voltage will increase. With a DMM set to a DC milivolt range you might be able to see some numbers flicker on the meter as you rotate your bolt heads past your sensor.

For my speedometer application on my GT40 I needed 10 pulses per reveloution with my wheel and tire package. I machined 10 spot faced locations (about .040 deep) around the outside of one of my CV joints and made a bracket to hold the VR sensor about .020 off the OD of the CV joint. Seems to work fine and speedo reads from about 4-5 MPH on up. I attached a couple of pictures of when I was testing it on a lathe.

You can "share" the output of a VR sensor with multiple electronic devices, however, they must all be very high input impedance (would suggest at least 1M ohm or better). The very small magnetic flux change results in this type of sensor generating a very small amount of voltage / currrent.

 

[Keith Stafford]

Hi Bill, just saw your postings, if the sensors are unpowered, then the output could be small, you may indeed need a scope to fully quantify the pulse as Bill B suggests; in electronics it is sometimes essential to know the magnitude, duration, frequency and other characteristics of the input signal.

Is there any way you can come at this from the published specs of the parts used? failing that you may have to look at scoping it, do you have or have access to a scope?

 

[John Fitzpatrick]

Bill,​

with a two line sensor what you've got is the magnetic equivalent of a microphone. The signals it generates as a bolt heads passed by are probably amplified at some point before they're usable by your speedometer. If you can open the speedometer and locate the circuit that does that, you can share its output with another device.​

​

As has been said, an oscilloscope is a tool of choice, but if you don't have any electronic equipment, you can connect a probe to a pair of high impedance headphones. Clip the ground lead to negative and listen to various points on the circuit board while someone spins the wheel. What you're listening for is a series of clicks that will speed up as the wheel turns faster.​

​

​

It's crude and low-tech but it's quick and cheap, and you can't hurt anything as long as you don't short out two pads on the circuit board with your probe. If you can post a photo, that would be helpful.
​

Good luck!​

 

[John Fitzpatrick]

JEGs sels a 6 pulse/ revolution hall effect speedo sender with a pass-through for a speed control that might work, or you could check for an on line manual to get an idea. It's about $80

 

[Bill Musarra]

First let me thank each of you for your input on this question. I am one of those people that understands electronics, but has a difficult time with its application. I have limited funds due to some recent health problems with my wife, so I am trying to keep it simple. The source I am trying to work up is for the power steering unit I have installed in my car. The hookup of the unit was fairly straight forward. Now to get the steering unit to actually work. Keith Stafford and I have been kicking this around for some time now. I believe Keith is about to work out all of the bugs in a simple arrangement. His is to use a timer and limit the time for the high assist /low speed setting for a period of time and then have it switch back to the high speed /low assist mode. I was shooting for a totally automatic arrangement. If I were an electronics kind of guy, I think it could have been done and completed by now.
This started out as a, tail end of the winter, project and it still goes on. Have only made a few trips with the car this summer. I am now working on a gremlin in the electronic ignition area. A simple problem that I think has been diagnosed and is awaiting a remedy.
I think I will await the progress with Keith's application as it makes sense to me and is probably the best way to handle it. Keith will let us know when he feels his work is ready for prime time. So thanks again. If any of you work on this idea, come back and post what you have come up with.

Bill

 

[Paul Thompson 'Hooligan']

Hi Bill
Just came across thia whilst having a browse and i have a simple cct i've used on occasion to convert a simple magnetic ABS sensor output to a 'square-ish' wave that i've then used as inputs to an ECU for traction control. It works well up to 50khz inputs and only requires a few readily available components.

I've sketched it below, sorry for poor diagram but i do not have my laptop to hand so did it on my iPad which is not overly user friendly for 'finger drawing'..

The 4k7 pull up resistor was only required on ECUs that did not have an internal pull up. I am guessing your speedo input may or may not require the resistor depending upon its input configuration.

Out of interest, it was also fitted to all 4 wheels of a car that still had it's ABS function working and it did not throw any fault codes to the ABS system either (BONUS) although i did inform the owner that i did not guarantee this would be the case.

Anyway, if you try it, i hope it helps...

Oops!! Checked my hand written notes and the diodes i had actually used were 1N4001 or 1N4004 although it might still work with the 1N916 / 1N4148.

 

[John Fitzpatrick]

Bill –

You're probably sick of advicd and I hope this isn't along the lines of "What time is it?", – "Here's how you build a watch." But a little bit of technical background might help you as you go along on this project.

The Hall effect transistor is simply a semiconductor that’s sensitive to magnetic fields. It’s biased the same way as an ordinary transistor, and in the absence of a magnetic field, is in the "off" state.. The actual voltage change it develops at its output is small and has to be amplified to be of much use, and by itself it's not interesting to anyone except second year physics majors. However, they do make good sensors when incorporated with other components in an integrated circuit.

Those ICs usually are comprised of a voltage regulator (any noise in the bias voltage shows up at the output), the Hall effect transistor itself, an amplifier to bring the output level swing well above the noise level, and a buffer, called a Schmitt trigger, to clean the amplified signal up and make it useful as a digital output. Ordinarily, the IC will have three leads: positive, ground, and output, and the circuit is packaged in a way that makes it easy to properly orient (very important) to the object being sensed. By putting all the components in one package with a high level output, the low level signal from the Hall transistor itself doesn't get corrupted by electrical noise on the way to whatever component is using its output. If the case is the ground, you just have an output lead and the power lead to connect – that's probably the situation in your case.

I changed out the AC motor on my l0” lathe to a DC servo and used a little Hall effect switch from Radio Shack to sense the RPM and keep it constant with varying load. They're cheap and really easy to use – about the only thing you have to be concerned with is keeping the supply voltage within its limit. I'm a klutz with a soldering iron, but I didn't have any trouble.

I suggest you just put a small 12 volt bulb between your sensor output and ground, spin the wheel and see if you don't get flashes – even if the output is an open collector. Once you see how it works, hooking it up should be pretty straightforward. If you want to play around with this stuff, $20, a sensor from RadioShack, an LED, a 9 V battery and a refrigerator magnet are about all you need, and your kids will have fun playing with it when you're done!
Good luck