Cook Timer

After going through this you will start wondering what you have been doing without this. Yes, a simple kitchen timer with 14 possible settings—the minimum being 2.5 minutes and the maximum 45 minutes. What is more, the repeat accuracy is in direct proportion to your patience in setting a preset. In any case, an accuracy of two to three per cent is easily achievable, and for domestic needs even this is luxury.

To set a delay of 45 minutes, a timer in the monostable mode would have been sufficient- But the repeat accuracy would suffer due to the use of bulky electrolytic capacitors. And by the time you satisfactorily finish setting the timing preset, you will start wondering why you need a timer at all (each trial will need 45 minutes!). Alternatively, if you want to take life easy (without any trials), you have to spend a fortune on crystal-based clock generators and introduce a gang of ICs to divide and bring the clock pulse to humanly recognisable timing durations.

Well, we basically being smart, will do neither of the above. We operate a 7555 (CMOS timer) in unstable mode at a frequency of 54.6 Hz (clock pulse duration 0.018 sec). We then divide this 16384 times(!) in a single CMOS binary counter 4060B and get a basic clock pulse duration (BCPD) of five minutes. This pulse is inverted and fed to a decade counter 4017B (CMOS once again) and starting from 5, we can get a maximum interval of 45 minutes in steps of 5 minutes. This will be the `high` range.

For the ‘low’ range, starting from 2.5 minutes to 42.5 minutes, all we do is to directly feed the clock pulses from 4060B to4017B without inverting. The Q14 output of 4060B, after switch on will go high after 2.5 minutes (remember the BCPD is 5 minutes) and accordingly the Q1 output of 4017B will go high. Thereafter each successive output of 4017B will go high in intervals of 5 minutes up to 42.5 minutes (Q9 output). Thus with a basic clock pulse of 5 minutes, we gel a resolution of 2.5 minutes, in two ranges—High and Low.

The circuit can be assembled on a general-purpose PCB as there is not much complication, or the suggested PCB layout can be used. If you are sure you won’t make mistakes, D1 can be omitted. Just to safeguard against reverse polarity, it drops 0.7 volts out of the available 6 volts.


It’s presumed you have already crossed the stages of wrong connections, dry solders, missed components etc.

(a) Insert all three ICs in their socket, (Hope you have used IC sockets. All ICs are CMOS and they need lot of respect).

(b) Set a value of 4.5k in preset ‘ P1’ before soldering in place. For this, the single turn presets are totally banned. Use a good multi turn helipot.

(c) Switch-on ‘S3′ to test position (No. 1). SI should be in ‘Low’ position

(d) Have a stop watch (or an ordinary clock with seconds indication) handy; if you believe in god, pray and then switch ‘S2’ on. Following should happen:

(1) LED should blink at a rate of 3.4 Hz.

(2) After about 38 seconds the buzzer should sound.

If the buzzer does sound, the calibration for all the fourteen settings is over and you are the proud owner of a useful timer. Invariably, a few trials and minor adjustments on VR1 (1/4 to 1/2 turn at a time) may be needed. Once you are satisfied, switch S3 to position ‘2″ and switch-on S2. You should now time 150 sees. If you get this, no more verification is required.

Calibration using an oscilloscope is very simple, assuming that setting up the scope is easy. After patiently selecting the right time base, scale etc on the scope, measure the frequency from pin 3 of 7555 and set it to 54.61347 Hz. Obviously, if you can lay your hands on a frequency counter life will be much easier and pleasant.


Select the High or Low range and the exact time you need on S3. Switch-on S2, At the end of set timing, the burner will sound. Switch-off S2. When the timer is ‘on’ you can at liberty increase the setting in S3 but never touch SI. To alter SI, first switch-off S2.


(a) The circuit draws about 2mA during *on’ condition. Depending on the piezo buzzer, this will increase when the buzzer sounds.

(b) Do not replace 7555 with ordinary NE555 as the current consumption will shoot up. Also do not lower the value of R3. This also will affect the current drawn.

(c) Cl should be tantallum and VR1 should be multi turn pot. Save on these and you will end up wasting a lot of time in calibration.

(d) With a current of 2mA, the four pen torch cells should at least last a year and a half. Hence, resist the temptation to use the timer through a battery eliminator. If you cannot, go ahead, but remember to retain the battery back-up or else a supply failure in the middle of timing can lead to unpleasant consequences.

(e) For BCPD other than 5 minutes use the following formula:


F1 —— F1 1N Hz


BCPD in seconds

Once FI is known, setting up 7555 to operate at that frequency is all that is to be done by you. For calibration, put S3 in position 1 and set VR1 for a time equal to BCPD/4 secs.

(f) Finally, during calibration don’t waste time in trying to get exactly 38 secs or 150 secs. Error of ±2 secs, in 150 secs is quite acceptable. An accuracy of *1.3 per cent is more than enough for the domestic front.

Readers’ Comments:

The article ‘Cook Timer’ in July’92 issue of EFY is very good. Please convey my thanks to the author.

But, in fact, the circuit has been made more complicated using IC1 7555 as the pulse generator.

I wonder why the author did not use the superb facility of on-chip oscillator available in IC2 4060B?



The most interesting aspect of the project is the binary counter 4060B which drives pulse 16,384 times. The author is requested to give more useful information about the counter, in-cluding the role of pins 01, 12 and 13 of decade counter CD4017.


The author, Mr R. Raghunathan, replies:

The idea to use on-chip oscillator of IC4060B is good. It will definitely bring down the component count. But, in the bargain, the circuit would lose its frequency stability when the supply varies. That was the single reason for using 7555—the industry workhorse. The oscillating frequency of 7555 is completely independent of the supply voltage variations.

Mr Kalpesh’s idea can give excellent results with a regulated supply but when battery supply is used it is better to include 7555.

4060B is a versatile IC having an on-chip oscillator and 10 binary outputs. When an external clock signal is available it can be fed to pin 11 and binary divisions of 2^4 to 2^14 are available as outputs, as shown. Alternatively, a capacitor and resistor can be connected, as shown, to pins 9 and 10 and the on-chip oscillator will oscillate at a frequency of 2.2xRxC (R in ohms and C in farads). The frequency will then be divided and available as outputs. But in this case, the frequency will slightly vary with supply variations and hence a regulated supply is recommended. Pin 12 is a reset pin which when held high resets all outputs to low, and blocks the oscillator.

IC 4017

Pin 1 to pin 7: Outputs Q5, Ql, Q0, Q2, Q6, Q7 and Q3 in that order.

Pin 8: Ground

Pin 9 to pin 11: Outputs Q8, Q4, Q9 in that order.

Pin 12: Carry out. Goes high after 9 clock pulses for a duration of 1 clock pulse. Used for cascading.

Pin 13: Clock enable. Counting is possible only when this pin is low. Pin 14: Clock input.

Pin 15: Reset pin. When this is taken low all outputs are reset to zero. During counting this is to be high. Pin 16: +Vcc (3 to 15 volts)

Pin connection of 4060B

I would like to make a few corrections in the circuit of ‘Cook Timer’ published in EFY Jul.*92 issue. In the circuit LED1 and LED2 connections should be as shown below.


The author, Mr R. Raghunathan, replies:

Mr Kothadiya’s suggestion is right. LEDs can be connected as shown by him in which case the IC will be sourcing the LEDs. Otherwise connections can be made as shown in the circuit diagram in the article with just the direction of the LEDs reversed.

There are further corrections also which should be made.

1. In the circuit diagram, position of resistor R4 is to be taken as shown in the figure ( below .). However, the PCB layout is properly drawn.

2.The value ofVRl should be read as 10k in the components layout where it is marked 1k.

SQL - 17 | 0,410 сек. | 7.4 МБ