Hi, here's my problem,
I had a hall effect speed sensor on my car which, didn't work.
So I replaced it with an universal one because the original is out of production. There's no problem till this point, my old speed sensor has the following specification: 12v 16imp(pulse).
My new one is: 5-12v 4 pulse per revolution.
I red this before I bought the new sensor, but I wanted to see if my old one was calculated with 4 per rev, so it would be 4pulse per 4 rev= 16 pulse, but I was right thinking it was 16 pulse per rev ( a bit high :]]]])
I tested the new speed sensor and when the speedo was reading 20km/h
With the OBD2 the ECU was reading 4km/h ( I know I said the new speed sensor is 4 times lower but the speedometer needle remain a bit high, it starts about 3km/h instead of 0km/h so with the possibile calibration error of the speedometer it should be 16-17km/h which is right).
Made this premise to say that I need a speed sensor pulse multiplier which have to convert the 4 pulse from the new speed sensor to 16 for the ECU.
I forgot to say that I have a cable driven speed sensor, the speed sensor attaches on the gearbox on one side and on the other side there's the attachment for the speedometer cable, and in the middle the the hall effect sensor with 3 wires that are: Ground, Vcc+ and signal.
I need something that works, it's cheap, not too big cause I have not so large spaces, something easy to build ( well, that's not really a problem, I can try and I like build circuits I can), if it need programming plese link me a Premade software.
Thanks.

 

Hi,

Too bad it is not just a requirement to double the rate as that is easy to do.
So you have to go from 4ppr to 16ppr? That's double, then double again.

 

Hi,

Too bad it is not just a requirement to double the rate as that is easy to do.
So you have to go from 4ppr to 16ppr? That's double, then double again.

Yes, I have to go from 4ppr to 16 ppr.
I said cheap, I wanted to say not over 30 pounds.
Why "Too bad it is not just a requirement"?
Thanks

 

One way is to count both edges for x2.
How are the pulses being detected now?
Usually it is a gear sensor.
Max.

 

One way is to count both edges for x2.
How are the pulses being detected now?
Usually it is a gear sensor.
Max.

It is not gear, is a hall effect it has a magnetic wheel which pass in front of a pcb where are located th hall effect sensor and some resistors.
The photo below is my old speed sensor circuit.

 

Yes, I have to go from 4ppr to 16 ppr.
I said cheap, I wanted to say not over 30 pounds.
Why "Too bad it is not just a requirement"?
Thanks

Hi again,

Well if it just had to double the frequency there is a relatively simple circuit that does this.
Do you happen to know the duty cycle of the output of the new sensor, and the min and max frequency? That could help design something simpler.

 

Hi again,

Well if it just had to double the frequency there is a relatively simple circuit that does this.
Do you happen to know the duty cycle of the output of the new sensor, and the min and max frequency? That could help design something simpler.

Gimme a sec I'll do some screenshot.

 

If you want to search it is the brant gearbox speed sensor br4 for European car.

 

It is not gear, is a hall effect it has a magnetic wheel which pass in front of a pcb where are located th hall effect sensor and some resistors.
The photo below is my old speed sensor circuit.

I was meaning that many auto's that use a timing sensor, use the gear teeth on the flywheel with a hall sensor.
How is the x4 new version constructed?
Max.

 

If you want to search it is the brant gearbox speed sensor br4 for European car.

Hi,

We would actually need more data to be able to design something that will definitely work. With this limited data we could have many trial and error iterations to perform.
For example, each pulse could be 1us wide with 10ms spacing, or 1ms wide with 10ms spacing, for the same speed.
A scope pic would help a lot here. Any way you can get a scope shot of the output wave at a couple different speeds, or maybe ask for more data like this?

It may end up being a micro controller type solution, but to design such a circuit we would need more information about what the output pulses actually look like, such as the duty cycle and min and max frequency or spacing.

 

You could use two frequency doublers in series that gives four output pulses for one input pulse.
Below is a circuit using two XOR frequency doublers.
The values of R and C determine the output pulse widths.
The LTspice simulation shows an arbitrary input frequency of 100Hz since I don't know what your maximum frequency is.

 

One option I have used several times over the years to refit non standard control systems to older vehicles was to use a two stage conversion where I used a Frequency to voltage converter IC to generate a proportional voltage to whatever the input frequency range was then a second IC that did the reves and converted that voltage to the new frequency.

Option two is to do the leading/trailing edge double trigging concept then multiply that out by two.

Last common option I have used for doing even number frequency multiplication is to amplify the first signal with a common Op-Amp IC then run it through a capacitive coupled full wave bridge rectifier that goes into a resistive load. The waveform across the resistor will be double the frequency of the input. Then just do that circuit a second time to get the needed 4x frequency multiplication.

 

It is not gear, is a hall effect it has a magnetic wheel which pass in front of a pcb where are located th hall effect sensor and some resistors.
The photo below is my old speed sensor circuit.

What does the rotary/magnet side look like? Is the new one the same physically,
Any way to use the x16 wheel on the x4, do you have pic of the old magnet/wheel?
Max.

 

Would it not be possible to replace the hall device in the original faulty sensor as the magnet / toothed wheel on that would give 16 pulses. Can you post a picture of the inside of the original sensor to see if a repair is possible. ( Or show the picture of the toothed wheel part of the new sensor to see if it was possible to modify that.)

Les.

 

Thank you for all your reply, I think I will put the new sensor circuit in the old one with the old sensor magnet wheel.
I didn't think it was possible cause the new sensor circuit was all glued in place, but with a flat screwdriver slowly and with a lot patience I pried it off the sensor.
When my mobile phone will power up ( it's low in battery) I'll post some photos.