Vehicle speed sensor signal

[fastback86]

I have a vehicle speed sensor (vss) that produces an ac sine wave whose frequency is directly related to the input speed of the shaft on the sensor and the maximum frequency would probably be 350 Hz. While the frequency is what is used to determine the speed, the peak to peak voltage also varies greatly from 2V peak to peak to about 60V. I want to use this signal to drive a microcontroller input. What type of signal conditioning circuit should I use to get the most usable range from the sensor? I was thinking a transistor based opto isolator but how should I drive the LED?

 

[crutschow]

You can put a resistor in series with the opto isolator LED input to limit the current to its maximum rating with a 60V input. If you have a particular opto in mind I can give more details.

 

[fastback86]

I thought of that too but i was concerned at low speeds where the peak to peak voltage is 2v it would not work, but say a Lite On LTV-817 optocoupler...
https://media.digikey.com/pdf/Data Sheets/Lite-On PDFs/LTV-817_827_847(M,S,S_TA1).pdf

 

[crutschow]

To accommodate the low voltage I added a transistor buffer on the input as shown below. It will operate down to 2V but will not be damaged by the 60V input. The output ground can be isolated from the input (vehicle) ground if you want.

Note that the opto isolator model is just one I had in my simulation library. The one you selected should work the same.

 

[fastback86]

Thanks for the help. I hadn't even considered using a transistor to by-pass the LED. I'll give that a shot and see what happens. BTW, what's the purpose of D1?

 

[MikeMl]

Look at post #6 in this old thread.

This would work if you access to both ends of the speed sensor.

 

[crutschow]

Thanks for the help. I hadn't even considered using a transistor to by-pass the LED. I'll give that a shot and see what happens. BTW, what's the purpose of D1?

D1 conducts current during the negative half of the cycle to avoid excessive reverse bias voltage on the transistor base-emitter junction (which normally can only tolerate about 5V or so).

 

[fastback86]

Look at post #6 in this old thread.

This would work if you access to both ends of the speed sensor.

Thanks for the link. I can get to both sensor wires. That sender also sends a signal to the cruise control and the ecm. Hopefully I don't interfere with it too much.

 

[fastback86]

D1 conducts current during the negative half of the cycle to avoid excessive reverse bias voltage on the transistor base-emitter junction (which normally can only tolerate about 5V or so).

Thanks!

 

[MikeMl]

Thanks for the link. I can get to both sensor wires. That sender also sends a signal to the cruise control and the ecm. Hopefully I don't interfere with it too much.

Then neither the circuit I linked to, nor Crut's circuit is suitable. You will have to use something like an opamp with a high input impedance (an AC-coupled) so as not to prevent the sensor from working with the existing circuits.

 

[crutschow]

Thanks for the link. I can get to both sensor wires. That sender also sends a signal to the cruise control and the ecm. Hopefully I don't interfere with it too much.

Where do the two wires go?

 

[fastback86]

Where do the two wires go?

That depends on the year of the car. It's a Ford sensor that they used from about the mid 80's until about the year 2000. On the early cars it ran to the cruise control module because that the only reason it was there. On the later cars it runs also to the instrument cluster also because the factory speedos are electric. I'm 99% sure neither side of the sensor is grounded but I would have to consult the vehicle schematics to be sure. right now I'm looking at some instrumentation amplifier circuits as possible solutions.