Connect with us

tachometer using 555 IC and LM3917

Discussion in 'Electronic Design' started by fever, Aug 21, 2006.

Scroll to continue with content
  1. fever

    fever Guest

    i want to build LED based tachometer for my motorcycle.my friend gave
    me the schematic of tachometer published in ELEKTOR in 10/1998
    http://fly.srk.fer.hr/~wgottwe/e98a034.pdf .it uses LM3914 as LED
    driver and 555 timer for frequency to voltage conversion.
    i bought every component and assembled on bread board.but its not
    working.
    1.i used another 555 IC insted of TLC555 is it ok?
    2.for induction pickup i did exactly what mentioned in the article.as
    my vehicle is CDI i can't get signal from ignition coil.
    3.when i measued the output as shown in the below pic.it was stable 5v
    without any effect to the RPM of vehicle.
    http://img97.imageshack.us/my.php?image=0mh0.gif

    i desperately need help to finish this project.
    pls help me
    thanks
     
  2. Can you exercise the input with a square wave generator?
     
  3. Donald

    Donald Guest

    Which manufacture ???

    The TLC555 is a low power CMOS, is the 555 you used a NMOS ??

    Do you have a o'scope available ??
    Did you wire this up yourself ??

    Did you wire up more then one unit yourself ??
    Desperate enough to pay money ??
    Help us help you, give enough information, please.

    PS: learn how to spell.
     
  4. fever

    fever Guest

    Which manufacture ???
    the 55 IC iam using is NE555N.do i need exactly TLC555 or 7555 IC?

    no i don't have osci'scope.but i tried blow ckt.and the LED present in
    the ckt start flashing when i rise the throttle of motorcycle.
    http://img90.imageshack.us/img90/3820/inductiveledix5.gif

    i don't understand where and what iam doing wong.is the elektor
    tachometer ckt right??
    or its not working bcz of using NE555N insted of TLC555???
     
  5. Donald

    Donald Guest

    You have no idea what you are doing.

    There is more than building a circuit then replace parts without
    reviewing the data sheets.


    If you used all the parts suggested in the parts list and it still did
    not work, would you blame the author for writing the project wrong.

    Good Luck.
     
  6. Depends. Some magazines employed illiterate hippos as printers, so you
    could never trust the circuits and parts lists to be error-free. Some
    still do; you see '47MW' resistors and '1000 F' capacitors called up.
     
  7. fever

    fever Guest

    If you used all the parts suggested in the parts list and it still did
    well iam not blaming any one.i just want to confirm.
    some guy just mailed me abt this elektor tachometer,that they printed
    wrong resistor values.and they corrected it in the next release.
    and iam from mechanical filed.i don't have much knowledge in
    electronics(just moderate).problem is i have to finish this project all
    by myself.
    so iam asking u ppl as u know more than me.
    any more views regading this project will be appriciated.
    thanks
     
  8. Fred Bloggs

    Fred Bloggs Guest

    No- you *must* use a CMOS 555 like the TLC555 for two reasons: C3,R3,
    and R4 are sized for a CMOS input current and 100ms time constant, and
    it is essential that the 555 output go all the way to 5V. The bipolar
    NE555 types will not work in this circuit.
    Sure-that's fine.
    Use a TLC555 and see if that doesn't improve things.
     
  9. Luhan

    Luhan Guest

    Measure the voltage on C1 with a digital voltmeter. If you have
    increasing voltage there with RPM, did you put that jumper in???

    Luhan
     
  10. fever

    fever Guest

    thanks fred(and other ppl too)
    i'll try cmos version of 555 first(can i use 7555 insted of TLC555 as
    it is also cmos version).and will post results.

    iam using breadboard for trials.
    i'll try 555 circuit first and then combined with LM3917.
     
  11. default

    default Guest

    No, or probably no
    With it outputing 5 volts - the thing is probably not working because
    the timing values are for a cmos 555 not a bipolar - recalculate or
    switch parts first.

    Beyond that. . . I had the same problem and ended up using a toroid -
    pass the cable through the core and wind a few turns around it for
    signal ground one end use the other to input the 555.

    You should be able to test your circuit by repetitively grounding pin
    2 on the 555 to simulate a signal - once you get it working.

    I wasn't too happy with the toriod on the spark plug because it took
    so much filtering to have only one led light at a time, that I was
    looking at the RPM ten seconds ago - I ended up using the signal from
    my 12 pole alternator with just a little filtering. One pulse per
    revolution isn't enough for a fast reaponse.
     
  12. fever

    fever Guest

    Beyond that. . . I had the same problem and ended up using a toroid -
    u mean like this.ofcourse with lesser dia. core. (sorry for such bad
    pic)
    http://img221.imageshack.us/img221/8909/ideaof4.gif
    first i want to trial on sparkplug cable only.if it failed i'll go for
    other method.as currently i don't have tools to messup with my
    motorcycle.
    and someone told me not to connect other end of signal wire to ground.i
    think i need to experement these 2(imean connecting to ground and
    leaving the end)
    **have u used 555 ic for ur tachometer?
    what shld be rough dia of wire that shld be wound on ferrite core?
     
  13. default

    default Guest

    Yeah, just like that. Or one of the "clam shell" type cores radio
    shack sells to put around power cords to limit RFI might work
    Wrapping the sense wire around the spark plug wire does not provide an
    "inductive" pickup. The wire is not oriented correctly to inductively
    couple much energy. Leaving it ungrounded gives you a capacitive pick
    up.

    When you use a core you're making a transformer with a single turn
    primary (the spark plug wire) and need to ground one end of the
    secondary to develop voltage input to your circuit.

    I started with the 555 charge pump then tried a LM2917 (I think that's
    the number - frequency to voltage convertor) Better smoothing with a
    smaller cap, but still not what I wanted. Leds have the ability to
    outperform a mechanical meter for response time - seemed a shame to
    hobble them with a large filter cap.
    Use whatever you want - the current and voltage are low it won't
    matter too much. Better to wind the wire around the entire core
    rather than have all the turns bunched at one side - so space them
    out.

    When I started thinking about what I was doing . . . do I really need
    to know when it is turning 500 RPM? My motorcycle idle speed is
    supposed to be 1100 RPM and it won't run slower than 800 RPM, so one
    LED is virtually useless unless the starter motor is spinning the
    engine.

    Then the idea of only 20 LEDs wasn't too my liking either. Red-line
    is 12,000 RPM not plus or minus 500 RPM. The final version of my tach
    used a set of four comparators to change the scale of the 20 led
    display. 1000-4000, 4000-7000, 7000-10000, 10000-13000.

    Each of the comparators has a reference voltage that they compare to
    the signal from the charge pump, as each comparator's range is
    exceeded it lights an LED to tell me the range it is on, and connects
    its reference voltage to the display (via an analog switch IC).

    So I've got 20 LEDS doing the work of 80 and have 150 rpm resolution -
    and it took me a month to get working. but the display goes around
    four times now.

    I used a 4" PVC plumbing fitting for a housing and put a thin ring of
    4" pipe in it to give me a ledge to seal the display board to - I
    didn't use a PC board and made it in two assemblies.

    Most of the time and fun was on the breadboarding of it and developing
    the idea. I think I have something like a total of 7 IC's in it if I
    count the two regulators (one is for the power supply and another for
    the reference).

    I even incorporated a high/low switch for display brightness but found
    that 6 ma per led was more than enough for high efficiency leds.
     
  14. fever

    fever Guest

    sure i'll keep in mind abt grounding the other end of pickup coil.
    wow thats a quite detail explanation.
    thanks a lot.
    at first i want to use that F-V converter only(LM2917) and i found
    circuit in this german page.
    http://www.digitaler-drehzahlmesser.de/ this is completely dedicated to
    LED based tachometer which uses LM2907 with 8-pin.but the problem i
    faced is with the F-V chip.in my area LM2907 with 14-pin are
    available.no LM2917 and no LM2907 with 8-pin.so i stuck at this point.
    and is it possible to see ur tachometer pics.
    thanks for ur help once again
    fever
     
  15. default

    default Guest


    Well I'm an American don't speak much but English and a smattering of
    Spanish - enough to get me in trouble most days.

    The LED digital tach in the German picture - is there a link to the
    schematic? Sorry for my ignorance.

    I have an absolutely - "damn wish I'd thought of it" schematic of a 64
    LED display from a guy that makes use of a 8 bit A/D converter and
    decodes the display into a 8X9 matrix using a few common CMOS parts.
    That is 64 LED's with only three IC's ... minus the regulator, and
    only 20 wires to the display board.

    At the time I built mine. the LED's were costing 29 cents each and
    using 20 to represent 80 was a cost consideration

    I built my tach four years ago and it has been on the MC ever since.
    I have a shematic that I put in my notebook and can scan and send and
    it will give you an idea of what I did.

    Should I send it to ?

    Take care
    bob
     
  16. default

    default Guest

    I was looking at your web space - what does the alternator file
    contain? I just rewound the rotor coil on my bike and wondered if
    there was something I might use.

    I'd rather replace the field coil with permanent magnets and then just
    design a new regulator.

    The program for the USR modem? what does it do?. I have a USR
    external.

    Thains
    bob
     
  17. fever

    fever Guest

    The LED digital tach in the German picture - is there a link to the
    yes in download section u'll find the circuit details.if ur facing
    problem with the langauge then use altavista's-bable fish translator to
    read the site.this site contain 40,30,20 LED tachometer circuits with
    PCB layout.

    if possible do mail me the schematic of your tachometer.u said it uses
    LM2917 is it 8-pin or 14-pin.
    thanks
     
  18. Fred Bloggs

    Fred Bloggs Guest

    Right- the author of the Elektor article mentions "induction" and the
    "high voltage" in the spark plug cable. He doesn't translate into
    English very well because his sensor is capacitive and works on the
    principle of rapid voltage collapse in the wire when the spark fires.
    The 20 wraps of the sensing wire around the spark plug wire form a small
    capacitance, and the rapid high voltage collapse in that wire couples
    charge into the sensing wire and into the 555 circuit, triggering it.
    There is no connection to GND. The sensor is just a single wire going
    from the 555 circuit to the plug cable, wrapped about 20 times, and ends
    there. This is better than an inductive type pickup using a core in some
    ways because it is immune to stray magnetic fields.
    You seem to be having a problem following directions. Elektor doesn't
    publish circuits that have not been built and tested, which means it
    will work as advertized if you follow the directions. But when you go
    off on your own changing part numbers, component values, and sensing
    methods, when you don't know what you're doing, you will end up with a
    pile of malfunctioning crap.
     
  19. fever

    fever Guest

    ways because it is immune to stray magnetic fields.
    could be.
    because the article is quite old.and its kinda hard to get the exact
    components in the market.still iam trying my best to get them.(7555
    insted of TLC555 as both r cmos versions)
    and till now i experemented the signal wire by leaving the other
    end,and was using less dia wire.this time i'll try different dia wires
    with 7555 IC.
     
  20. default

    default Guest

    Done
     
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

-