Connect with us

Phone in use circuit not working

Discussion in 'Electronic Basics' started by steve, Feb 28, 2013.

Scroll to continue with content
  1. steve

    steve Guest

    I tested and built the following circuit.

    It all worked fine untill I tried to test it by calling into my phone.
    (the one test I didnt do before I soldered it all together)

    What happens is taht the phone rings for a about 1/2 a ring then stops ringing.
    To my surprise when I life the phone up the line is alive eg. Its like the device has answered the phone.

    I think that the circuit must be drawing too much power and then telling the phone that it has been answered, because the voltage has gone down.

    I have had to replace the Transistors with NTE199, which according to the book are the same. I have also used a rectifier that is a 2 amp 400v SIP 2KBP04M-1.
    Im wondering if I used a different rectifier if that would change the results?

    I recognize that your not suppse to take power from the phone co. but I would like to get this circuit working. Can anyone tell me why its not working.
  2. Peabody

    Peabody Guest

    steve says...
    That sounds right. The phone company senses an off-hook
    state based on DC current passing through the local loop.
    So your explanation makes sense.

    I think your part substitutions are correct. But I wonder
    if you have the bridge rectifier wired correctly - the phone
    lines connected to the ~ pins. Or, could that part be

    If everything is wired correctly, try removing R4 from the
    circuit and see if ringing works properly then. If it does,
    then you may just need a higher value resistor for R1.

    Actually, I don't much care for the circuit. During each
    cycle of the ring signal, the voltage on your circuit will
    go to zero. It won't go negative because of the rectifier,
    but it will go to zero. As it gets near zero, then again as
    it comes back up, Q2 will turn on for a bit, and current
    will flow, until the voltage gets high enough to turn on Q1,
    which will turn off Q2. But during that Q2-on period, maybe
    enough current will flow to signal an off-hook to the
    central office. Well, I'm clearly just guessing, but I find
    myself wanting to put a capacitor from the base of Q1 to

    Anyway, good luck.
  3. steve

    steve Guest

    Thank you for your comments.

    Well I have wired the phone line to the middle pins on the SIP [ + | | - ]
    And + to plus and - to Neg of the diagram.

    Thanks for the tips. I will try them.

    By the way one thing. The diagram calls for R2 to be 33k. But a comment on the bottom says that the value should be the same as R1 3.3. So I made R2 3..3. I asumed that the poster was right, because I know how scematics get screwed up some times. In your opinion should R2 be 3.3k or 33k. Maybe thats whats wrong?
  4. Peabody

    Peabody Guest

    steve says...
    Yes, I think that's right.
    Well, using 33k will make the circuit draw less current, but
    the green LED probably will not light up enough to be
    visible. But of course you really don't need the green LED,
    just the red one.

    My concern is that the circuit just draws too much current,
    and therefore shows as off-hook. I would try it at 33k, and
    take the green LED out (or just temporarily jumper across
    it, and see if it works. Beyond that, you could increase
    the resistance of R1, but then that will make the red LED

    There is another version of this circuit that uses MOSFETs
    instead of NPN transistors, and all but the LED resistor are
    in megohms, so it draws very little current when on-hook.
    That's because the MOSFETs switch based on voltage instead
    of current.
  5. Jamie

    Jamie Guest

    I think some one gave you some bad transistor numbers..

    THe phone system in the US rings at around 100 Volts AC and on hook
    is around 50Volts. Off hook may give you around 10 volts and it depends
    on how many phones you have at one time off hook.

    The 2N3392 can only handle 25Volts and it may work if you can insure
    that the unit will switch on in time to load it down. Further more, the
    other transistor will be sitting there with this 50v (on hook), leaking
    through the collector.

    The NTE199 replacement isn't much better, it has a 70V limit at best
    and most likely will work find when on hook, but when it rings, you can
    expect some leaking to be taking place.

    You first need to get HV transistors.

    Try getting some 2N5550 transistors, or the 2N5551 which is a little
    higher.. has 2N5551 for 0.78 ech and they have over 3k in stock.

  6. Jamie

    Jamie Guest

    That one would most likely work better, at least the transistors
    are rated at 200V, which is your biggest problem with the first

    Also this circuit has much less load since it seems to only work
    when off hook.

  7. amdx

    amdx Guest

    Here's one I used for years without problem.
    Similar to what Peabody posted, one less FET.
    I can't find the original URL.
  8. Jamie

    Jamie Guest

    I remember buying an off hook relay module for a phone line from radio
    shaft years ago, back in the old days, it never worked.

    would never switch on. Also, they stopped selling them. Must of been a
    reason for it :)

    Some times it is more sensible to buy ready made over making it.

  9. Peabody

    Peabody Guest

    amdx says...

    Yes, and that has the capacitor I kept wanting to put into
    the first circuit. That should delay the turn-on of the FET
    enough to keep the LED off during ringing.
  10. Daniel Pitts

    Daniel Pitts Guest

    I remember as a teenager, I was playing around with the phone line. I
    learned two things. One, if you hook the right resistance across the
    line, incoming calls get a "busy" signal, but you can still make
    outgoing calls. Two, when someone is calling you, you can get a jolt if
    you're holding both ends of the line.

    So, when the phone rings, you get a 60V AC signal. Perhaps you need
    slightly more resistance in the line. Unless D2 is already too dim, I'd
    try changing R1 to a higher resistance.
  11. Sticking your fingers across a voltage source was never a precise means of
    measuring voltage.

  12. Jasen Betts

    Jasen Betts Guest

    33K looks right to me, if perhaps a little too low.

    the green LED illuminates when the circuit is off-hook which
    is 40V or more. 33K only gets you ballpark 1.3mA, but much more
    than that and it'll start looking like off-hook.

    If you want it to be brighter get a better LED or connect several
    LEDs in series.
  13. Jon Kirwan

    Jon Kirwan Guest

    There is quite a long specification for telephone lines in
    the US. I have a copy, thanks to Don Bowey, when he dropped a
    copy over at my home back around 2003 or so. It represents an
    industry composite of various systems and includes envelopes
    of operation. The gist of it is, as Don wrote before coming
    over back then,

    "The FCC R&R require a minimum of 5M of customer premises
    equipment loop resistance in the on-hook state. Also.
    at 220uA some Central Offices will declare a line fault
    and remove the line from service."

    The basic idea of presenting 5MOhm total customer premises
    load (that is the TOTAL of all phones and connected
    equipment) when on-hook is incredibly hard to meet with an
    unpowered attachment that shows "in use" and "not in use" led

    I've also read that, "if you are in USA or canada Bellcore
    specs allow a max of 5micro amps at 48VDC in an on-hook
    state." That's 10MOhm, in effect, though I believe the 5MOhm
    for the entire premises still fits the spec.

    The upshot here is that you aren't going to be seeing an LED
    lit continuously and still meet on-hook specifications. You
    can design a system to blink periodically, by drawing very
    little current onto a capacitor, which charges slowly up and
    is then discharged into an LED when the voltage rises to the
    trigger voltage. You can design that to meet specifications.
    But that's not what you've been given to build.

  14. ehsjr

    ehsjr Guest

    There's a lot wrong with that circuit, but as an experiment
    add a small (10uf or less) electrolytic from the base of
    Q2 to ground, then ring your phone to see if it works.

    Then, when you've got it working or not, remove it from the
    phone line and use the parts for something else. It is not
    "phone company friendly" nor are the parts properly V rated.

  15. Jon Kirwan

    Jon Kirwan Guest

    Yeah, that is true enough I suppose. So examine: 2.png

    That should keep the draw under 1 microamp, while pulsing an
    LED for about 10 microsecond pulses. I didn't include a
    resistor limiter in series with the LED, as that is something
    for the end user to insert as appropriate. I also didn't add
    a bridge rectifier, either. Again... up to the user. I don't
    think that circuit will do much at 10-12V or less for the off
    hook voltages. But if it does, it will blink ... very slowly,
    I think.

  16. Or surely a modem, just uinsg the "DAA" part. Even the "winmodems".
    THere has to be an off-hook indicator to tell the modem the line's not

    I recall back when this sort of circuit seemed to come up here over a
    decade ago, John whatshisname suggested something so that when his modem
    went online, it disconnected the other phones in the house. Obviously he
    had to arrange things right he could control the rest of the phone wiring
    in the house.

  17. P E Schoen

    P E Schoen Guest

    "Ian Field" wrote in message
    You can also scrounge some HV transistors in CFL bulbs. And some fairly good
    capacitors and a transformer, and a diac, which can be used to make a
    relaxation oscillator to blink an LED without overloading the phone line.

    I have wondered how much power can be drawn from the phone line without
    causing an error notification to the TELCO. The maximum power is probably
    the point at which the line voltage drops to 1/2 the 48 VDC, which may be in
    the order of 10 mA, so at 24 VDC it's about 1/4W. You might get away with it
    if you go off-hook every 10 seconds or so.

    A lot more power could be extracted if you could cause an incoming ring
    signal, which is about 100 VAC at 25 Hz and possibly 25-100 mA, maybe 40%
    duty cycle, so perhaps a couple of watts.

  18. Jasen Betts

    Jasen Betts Guest

    You draw 10mA and they'll think your phone lines ahva a fault (perhaps
  19. Jasen Betts

    Jasen Betts Guest

    some modems had a switched socket that would be disconnected from
    the line when the modem went off-hook, some had the pads but just
    had wire links in place of the relay.

    A second relay was needed as the main relay was also used for pulse
  20. tuinkabouter

    tuinkabouter Guest

    For off hook indication you can put a led in series with the telephone
    line. Across tte led a diode.

    Instead of the diode you can use an optocouple and then control whatever
    you like.

    Old fashion method, a 12 V relais in series.
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