I am helping a guy in the Midwest who restores vintage tractors to restore a 1969 Versatile (Canadian) D100 tractor. It has a Ford diesel engine and its tachometer was made by Smiths Instruments in England. The input to the tachometer comes from a reluctor style sensor that reads a 6 tooth rotor that spins at 1/2 the rate of the engine. That means that there are 3 pulses per rev, the same as a 6 cylinder, 4 stroke gas engine.

The system was not working when it was removed from the tractor. The tachometer and sensor were sent to me in California to see if I could fix it. Here is how the sensor looks:



I opened the sensor by removing the connector screws to expose the internal wiring and I saw that one of the wires had broken, presumably doe to vibration over the years. I repaired the broken wire, put it back together and moved on to the tachometer itself.

I connected a gear driven reluctor style vehicle speed sensor from a 1994-98 Mustang to the tachometer, powered it up and spun the sensor gear with my fingers and saw that the tachometer was working. I connected my function generator to the tachometer using a 10uF cap to block the DC and am able to calibrate the tachometer on the bench.

Now I want to be able to test the Smiths sensor with the tachometer, so my task is to fabricate a rotor that can cause the sensor to trigger the tachometer.

My first attempt was to put a carriage bolt through a hose bib handle and put it in my drill to create a small rotor:



I tried it using a Output Shaft Speed reluctor sensor from a 1999-2004 Mustang and was able to get the tachometer needle to move. But when I switched to the Smiths sensor, while I could see a small sine wave on the scope, it didn't have the amplitude needed to trigger the tachometer. I am pretty sure that I need larger teeth on the rotor to be able to trigger the tachometer.

But that got me to start wondering about the construction of these reluctor sensors.

The Mustang OSS sensor has one bit of metal protruding from the face of the sensor, while the Smiths sensor has two bits of metal protruding from the face.

My theory was that the OSS sensor has a bar magnet in it and the Smiths sensor had a U-shaped magnet in it with the 2 metal bits being the ends of the U.

I know for a fact that there are magnets in the sensors and that the teeth in the tractor rotor are not magnetized. As an experiment, I held a compass next to the OSS face with the face to the east of the compass and the compass needle aligned with the metal bit on the face of the sensor.

I then ran the same experiment with each of the metal bits of the Smiths sensor with the following results:

OSS:
Smiths top:

Smiths bottom:

I was not shocked to see the compass needle point away from the top metal bit in the Smiths sensor, but I was shocked to see the compass move to 90 degrees away from the bottom metal bit. I rotated the whole experiment 90 degrees to make sure that the earth's magnetic field was not the culprit and it was not.

Can anyone explain to me how this sensor could be constructed to cause the 90 degree field near the bottom metal bit? I am baffled and have not been able to find much about the construction of sensors like this on the internet.

 

Whilst I’m also confused by the movement of the compass I’m wondering if the separation between the two protrusions is the same as the separation between the teeth on the rotor? This would allow the rotor to successively open and close the magnetic circuit of your U shaped magnet which would create a much stronger signal than the usual single pole sensors.

To test, perhaps you could use a bent nail or a piece of flat steel in your drill such that it passes the two poles simultaneously once per revolution?

 

Whilst I’m also confused by the movement of the compass I’m wondering if the separation between the two protrusions is the same as the separation between the teeth on the rotor? This would allow the rotor to successively open and close the magnetic circuit of your U shaped magnet which would create a much stronger signal than the usual single pole sensors.

To test, perhaps you could use a bent nail or a piece of flat steel in your drill such that it passes the two poles simultaneously once per revolution?

Jerry, thanks for the reply. This photo I just received from the restorer seems to agree with you:



I am planning on bending a piece of steel wide enough to cover both metal bits on the front. I am 99% sure it will work, I just want to verify it.

But I am really curious as to how there is a transverse magnetic field near the bottom bit of metal on the face. That baffles me...

 

Actually not quite as I envisioned from the original photo, it’s the height of the castellations which complete the magnetic circuit between the two protrusions in the sensor, but I was nearly right. Why the transverse field? Baffles me too..,,

 

While I still don't understand how the magnetics work with this sensor, I was able to create a test that could trigger the tachometer. The wife wanted to browse some antique stores, so I looked in the tool bins and saw a beefy 2" Forstner bit that looked like it might do the trick. I talked them out of it for $5:



I built a little jig to hold the tip in place and prevent the bit from hitting the sensor:



By spinning the Forstner bit, I was able to get the tachometer needle to move, proving that the system still works together.

But to add to the mystery, it appears that the polarity of the sensor mattered. It would trigger the tachometer when connected one way, but not the other.

I am wondering if the orthogonal magnetic field on the bottom has something to do with the apparent differences in sensor polarity...

 

Nice solution
I'd want to open it up and explore inside but since it's working, probably not a good idea!

 

I am with you. But they tell me that you can occasionally find one of these tachometers so they can be found. The sensors are often strapped with the engines, so the sensors are even more rare than the tachometers. It is about to go into a brown truck for its trip home.

And the magnetic fields remain a mystery.

 

I wonder if it's like a Wiegand effect sensor (1969 might be too early for a real Wiegand sensor). You can use the Wiegand effect in several interesting ways with the most popular today being access cards but rotary sensors is also an application.

 

I wonder if it's like a Wiegand effect sensor (1969 might be too early for a real Wiegand sensor). You can use the Wiegand effect in several interesting ways with the most popular today being access cards but rotary sensors is also an application.

I had never heard of Wiegand wire. Thanks for sharing. I looked it up in Wikipedia.

It was patented in 1974, so I am pretty sure the Smiths sensor is not made with Wiegand wire. Wigand wire also generates a higher voltage signal than regular copper wire does. I actually suspect that my Ford sensors are actually made with Wiegand wire. I believe that may be why I get a stronger signal from them than I do from the larger Smiths sensor.

Very interesting...

 

Wigand wire also generates a higher voltage signal than regular copper wire does.

Both, ordinary sensor and Wiegand sensor have pickup coil made from copper wire.
Difference is in core of coil.
In ordinary sensor core of coil is piece of steel wire.
In Wiegand sensor core of coil is piece of Wiegand wire.

 

Thank you for the clarification!