Alarm clock time advance speeding-up project help

Discussion in 'The Projects Forum' started by tomoimrt, Dec 2, 2011.

  1. tomoimrt

    Thread Starter New Member

    Dec 1, 2011
    5
    0
    Hello,
    I am an absolute newbie trying to self learn. I have a commercial alarm clock based on the LM8560 IC. I need to get up at different time almost on different day of the week and I need to set the alarm every night before I go to bed. The problem is to set the time for the alarm, I would push and hold the 'hour' or the 'minute' button one at a time and they only advance forward, at a 2Hz rate. So if I need to change my alarm time from 7:45am to 6:30am, my finger will have to hold down the buttons a combined time of over 30s. The buttons are very stiff and hard on the finger too.
    I opened it up and saw the circuit is very close to this reference circuit http://www.datasheetcatalog.org/datasheet/sanyo/LM8560B.pdf
    Holding down Min-set(pin 21) or Hour-set (pin 22) will short the individual pins to Vss and the chip will automatically advance the respect digit at 2Hz. It doesn't look like the 2Hz rate can be made faster from outside the chip. However, I am thinking if I construct a switch using semi-conductors that will do the on-off switching on those time-set pins at 10-12Hz, it should serve my purpose.

    I took some measurement using a multimeter as follow:
    Vss to ground = 16.4V
    Vss to pin 21 or pin 22 = 12.8V
    short circuit current flowing from Vss to pin 21 or 22 = 6uA.

    Initially I was thinking of using a 555 timer to supply the 10Hz Vcc to pin 21/22, but was not sure the "0" phase (ground) output of the 555 pulse will short the pin 21/22 to ground causing any damage to the ic.

    Alternately can this be done with simple transistors-only circuits?

    Any help or suggestion will be greatly appreciated.
    Thanks!

    Thank you.
     
    Last edited: Dec 2, 2011
  2. SgtWookie

    Expert

    Jul 17, 2007
    22,182
    1,728
    Hi Tomoimrt,

    Have a look at the attached schematic and simulation.
    I elected to use a CMOS 555 timer and a PNP transistor as the main components.

    Radio Shack carries the TLC555 for a couple of bucks; they probably carry most of the rest of it as well. You don't have your location in your profile; if you are not in the USA, you probably don't have a Radio Shack near you.

    S1 is used to cause the hour or minutes to advance rapidly.

    S2 is used to select which will be advanced; hours or minutes.

    VR1 provides a means of adjusting the speed of the advance; from about 5Hz to about 17Hz. If you don't want the speed to be adjustable, replacing VR1 and R1 with a single 100k resistor will give you about 8Hz. If you go below about 50k, the numbers will advance much more quickly than you will be able to see. Don't go below about 20k, or it may stop functioning.

    D1 limits the maximum voltage across the 555 timer to 12v. R3 limits the current through the Zener, and consequently the remainder of the circuit.

    Rsim and Csim are simply there for simulation purposes. Rsim is representative of the resistors that are internal to the clock IC. Don't include it in the actual circuit, as that could cause problems. Csim is there just to speed up the simulation; the simulation takes a very long time to run without it.

    If for some reason you must use a regular bjt 555, then replace R4 with a 330 Ohm resistor; otherwise your clock will be "stuck" with the hours or minutes advancing constantly. I really do suggest that you use the CMOS timer instead.

    This addition will not use any power unless S1 is closed. This is important if your power goes out and your clock is running on battery power.

    Just FYI: for some odd reason, the datasheet shows the Vdd and Vss labels used opposite to modern conventions. Vss is usually either at 0v or a negative voltage, and Vdd is normally more positive than Vss. I am uncertain why they did that, but I have also seen that with other manufacturers' datasheets for this same IC.
     
    Last edited: Dec 2, 2011
  3. tomoimrt

    Thread Starter New Member

    Dec 1, 2011
    5
    0
    Hi SgtWookie,
    Thanks for your response. I have a few NE555 kicking around which I guess are not CMOS. So I will change R4 to 330ohm. What does R4 do anyway?

    1. Can u point out which Resistor, if present at all, is related to the biasing of the transistor and why such a value is chosen?

    2. With regard to using 10k for Rsim, does it matter if the actual resistance of the hr/min trigger pins to ground resistance is 2M ohm (short circuit current at pin 21/22 is only 6uA)? Will it change the value of any of the resistors in your circuit?

    3. Am I correct to that the transistor is turned on on the "0" phase of the clock output from 555 and registers a connect-to-Vcc action on the time-set pin? Would it work if I changed the PNP transistor to a NPN and connect its base to the 555 output through a diode and drove the transistor at the "1" phase of the cycle? The diode would have its positive end connecting to the 555 output, and negative end to the NPN base to protect the pin 21/22 from shorting to ground during the '0' phase of the 555 output. The base would need a resistor to limit the base current of the transistor because the emitter to clock ic portion has high resistance already... does it make sense?

    Thanks again for your expertise.
     
  4. SgtWookie

    Expert

    Jul 17, 2007
    22,182
    1,728
    R4 keeps the transistor Q1 turned off when S1 opens.
    If you are not going to be using a CMOS 555 timer, then perhaps you can use a couple of diodes (1N4148/1N914 suggested) in series with the transistor's emitter, instead of decreasing R4.

    R2 is simply being used as a saturated switch. It's not going to take very much current to pull the inputs up. Admittedly, I just gave R2 a "ballpark" number; it didn't have to be terribly accurate. R2 will allow (12v-0.7v)/10000 = 1.13mA current to flow through the base. You said 6uA; I figured you could have actually meant 6mA, or really did mean 6uA. In either case, 10k is actually a bit low resistance. If there was 6mA current to the set hours or minutes, then (12v-0.7v)/(6mA/10) = 18,883.333... Ohms.
    Here is a decade table of standard resistor values:
    http://www.logwell.com/tech/components/resistor_values.html
    Use the green (E24) or yellow (E12) columns.
    As you can see from the table, 180 (18k) and 200 (20k) are the closest values to 18,883.333... Ohms. In this case you could use either.

    If the current you saw was actually 6uA, then you could actually use around 19MEG for the base resistor - but the standard table only goes to 10MEG.

    So, it's up to you. 10k will work. 20k will work.

    You won't use Rsim at all. You shouldn't have to change any values, but I really suggest that you use a CMOS 555.

    Yes. When the output of the 555 goes low, it pulls the base of the PNP low, which turns on the transistor.
    Why complicate it like that?

    Just use the circuit I gave you, and it will work. There is no point that I can see to make a bunch of changes; and you would be using an NPN as a voltage follower.

    When S1 opens, the output of the 555 is pulled to Vss. The NPN transistor would be wired as a voltage follower, and would keep the set pin high for whichever is selected by S2. If you want your clock to advance at the same rate it does now when you hold the button down ALL THE TIME, then go ahead. You would not be able to stop it from advancing with your suggested modification.

    If you do not follow my suggestions, then your results will vary considerably, up to the point of having to buy a new alarm clock.
     
  5. tomoimrt

    Thread Starter New Member

    Dec 1, 2011
    5
    0
    Thanks for the explanation of how this works.
    Very much appreciated!
     
Loading...