Connect with us

Pulse frequency reduction circuit

Discussion in 'General Electronics Discussion' started by The_Biochemist, Apr 4, 2017.

Scroll to continue with content
  1. The_Biochemist

    The_Biochemist

    2
    0
    Apr 4, 2017
    Hi - as you can probably already tell I am not an electronics expert! i can just about hold a soldering iron.

    I have a problem and hope someone might be able to help me with a circuit design or at least give me pointers so I can research and learn myself.

    My problem:
    I need to remove the tachograph from my vehicle as I no longer need it. It is provided a pulse signal where the frequency is relative to speed from a hall effect sensor on the gearbox output shaft.

    I have identified from some minor wiring revision after removing the tachograph that once provided with a 12v power source the output from the hall sensor can be directly routed to the vehicles ECU and provides a suitable signal for the ECU to drive the speedometer, odometer and the like.

    So far so good - however the speed displayed is exactly 10x the speed the vehicle is moving at (based on GPS reported speed).

    What I think I need:
    I think I need a circuit that can take the pulse feed form the hall sensor and output a pulse feed that is 1/10th the frequency of the input feed but with the same wave form??

    Is this possible?
     
  2. AnalogKid

    AnalogKid

    2,427
    691
    Jun 10, 2015
    Depends on what you mean by "form". If the signal is a 12 V peak-to-peak square or rectangular wave, it can easily be divided by 10 with common ICs. The output waveform might be a symmetrical square wave (50% duty cycle, high 50% of the time and low 50% of the time), or a rectangular waveform with other duty cycles, such as 60% or 10%. If you want the exact input duty cycle at the output, that is much more difficult for a signal that varies in frequency.

    ak
     
  3. Arouse1973

    Arouse1973 Adam

    5,165
    1,087
    Dec 18, 2013
    Hello and welcome to EP.

    Yes it's possible.

    You need a decade counter. Something like this might do the trick. You will need a buffer to drive the ECU I am guessing the logic signal is positive going from the Tacho, you will have to make sure the signal level voltage is the same as the power supply for the ICs you are using .

    A CD4010B or a voltage follower constructed from transistor should work, you will need a pull up resistor from the divide by 10 pin to the supply rail. http://www.ti.com/lit/ds/symlink/cd4017b.pdf. See if you can work out what to do from here on. Come back if you have any questions.

    Thanks
    Adam
     
  4. The_Biochemist

    The_Biochemist

    2
    0
    Apr 4, 2017
    AK - thanks for the feedback. I have had a read in more detail on wave forms and it would seem to be a pulse-train/pulse waveform so my assumption is I would need the output to mimic the amplitude and mark but extend the space.

    Adam - also thanks for the feedback. I have looked at the CB4017b data sheet and think I get the gist; my assumptions thus far are:

    1) the IC can use the vehicle voltage between 12.3 and 14.4 - pin 8 being -ve/chassis earth and pin 16 being +ve
    2) this voltage range would give a 5mhz maximum input frequency which I think will be sufficient (i hope)
    3) the output would be connected to pin 11 thus trigering every 10th input pulse
    4) pin 11 would also cross to pin 15 to reset the counter on the 10th pulse
    5) the input from the hall sensor would be connected to pin 14

    From your comment I think I also need to check that the sensor pulse voltage is the same as the voltage between pin 8 and 14 and if it is not then compensate with some electrical trickery so that the IC is fed with the same voltage as the hall sensor output.

    Hopefully I am on the right track thus far but it seems entirely too simple as whenever I take apart even what I assume are the most basic circuits I seem to see resistors and capacitors everywhere so assume there must be more to it somewhere!!

    I am struggling with the voltage regulation aspect as I have read through information on voltage followers which seem to replicate the input voltage that is applied but just reduce the current draw; which doesnt seem to be my problem here??

    The vehicle voltage varies dependent on alternator output, battery charge level and utility draw from ~12.3v to ~14.4v. Given this is the power source for the hall sensor too I assume that the output pulse voltage must vary to some extent too. Is there a mechanism that can regulate voltage to the IC V(dd) pin from vehicle +ve relative to the sensor pulse voltage peak so as to synchronise the voltage??

    Hope this is not complete nonsense...

    Thanks

    Matthew
     
  5. AnalogKid

    AnalogKid

    2,427
    691
    Jun 10, 2015
    The 4017 is a decade counter, so if you need a divid-by-10 output, you do not need to connect any of the outputs to the reset input. Each output individually is the input frequency divided by 10 with a 10% mark/space ratio. The only difference among the outputs is that each one is shifted 36 degrees in phase. Ground the reset input.

    ak
     
  6. Arouse1973

    Arouse1973 Adam

    5,165
    1,087
    Dec 18, 2013
    Hi Matthew, no it's not nonsense, you have made a good start.
    Adam
     
  7. Arouse1973

    Arouse1973 Adam

    5,165
    1,087
    Dec 18, 2013
    If you take the output from the divide buy 10 pin as shown, you can see that the higher frequency (Green Trace) is 500 Hz is and the lower frequency (Blue trace) is 50 Hz. I'll leave you to work out the buffer :)
    Thanks
    Adam

    Counter Sch.PNG

    Counter wav.PNG
     
Ask a Question
Want to reply to this thread or ask your own question?
You'll need to choose a username for the site, which only take a couple of moments (here). After that, you can post your question and our members will help you out.
Electronics Point Logo
Continue to site
Quote of the day

-