1 sec alternating polarity circuit

[wombweller]

Hello all.
Please see the attached photo of a circuit I’d like to copy. I don’t have the actual board a friend of mine sent me the photo.

It’s a circuit to produce 1 second alternating pulse using the following IC’s-
CD4017B x2, CD4060B x1 and a CD4024B.

Can anyone point me to a circuit diagram for this, maybe there is something online, if there is I’m unable to find it.

I think this circuit is used so it can run on 3v so low power. It’s for pulsing a low voltage Brillie slave clock and needs to be a small footprint so it can go in to the back of the clock and run on batteries.

I look forward to hearing from you.

Regards
Mark

 

[ZipZapOuch]

I did a pulse clock driver back in the 70's for a TV station, but I suspect that things have changed a bit in 50 years.

ak

I think it's an "antique" system - a set of slaves they are trying to synch without a master.

 

[wombweller]

Bit strange he didn't bother writing the actual value down?

 

[wombweller]

Yeah but he said that this is all the info on it that he can see.

 

[wombweller]

Yeah but he said that this is all the info on it that he can see.

 

[augustinetez]

The value will be written under it where it's sitting on the board.

He should be able to carefully rotate it just a little to see the value with a bit of luck.

 

[wombweller]

Just to wish everyone who has contributed to this thread an happy new year and thank you all for your kind support.

 

[ZipZapOuch]

There are some reports online of these slave Brielle clocks with a 1-minute pulse from the master to ratchet the minute hand and keep it aligned with the master. Delivering a 30-second pulse high and 30-second low is kind of a waste of power. A 1-second pulse (or less) is likely enough to ratchet the minute hand each minute.

I'm guessing that the two transistors are an AND gate to make a 1-second pulse from the two CD4017 chips as follows (a 1Hz pulse to the first CD4017 is supplied by the CD4060+CD4024 output).

The green 1k is the Brillie clock in this case.

 

[AnalogKid]

There are many ways to gate the 4017 outputs to produce shorter pulses. However, your AND gate will not produce a bipolar pulse sequence.

To get the shortest output pulses without any additional parts, the two 4017's can be re-wired to reduce the output pulse width from 10 seconds to 6 seconds. As it is now, the two 4017's are divide by 5 and divide by 6. Change this to divide by 3 (by changing the Reset connection) and divide by 10 (no Reset connection).

ak

 

[AnalogKid]

AND - If you don't mind hand-soldering an SMT component, the CD4060 and CD4024 can be replaced with one TSSOP part:

74AHC1G4215-Q100

ak

 

[AnalogKid]

I'm puttering around with ways to improve the pulse outputs. Is there a specific sequence that is required? For example, should the negative-going pulse always come shortly after the positive-going pulse? Where I'm headed is 1 second positive pulse, 1 second rest, 1 second negative pulse, 57 seconds rest.

Side question - can anyone recommend a low voltage, full H-bridge device with on-board power transistors?

ak

 

[ZipZapOuch]

There are many ways to gate the 4017 outputs to produce shorter pulses. However, your AND gate will not produce a bipolar pulse sequence.

To get the shortest output pulses without any additional parts, the two 4017's can be re-wired to reduce the output pulse width from 10 seconds to 6 seconds. As it is now, the two 4017's are divide by 5 and divide by 6. Change this to divide by 3 (by changing the Reset connection) and divide by 10 (no Reset connection).

ak

From what I've read on the Brielle clocks with 1-minute pulses, they don't need bipolar pulses. That's why I didn't add one.

 

[wombweller]

I'm puttering around with ways to improve the pulse outputs. Is there a specific sequence that is required? For example, should the negative-going pulse always come shortly after the positive-going pulse? Where I'm headed is 1 second positive pulse, 1 second rest, 1 second negative pulse, 57 seconds rest.

Side question - can anyone recommend a low voltage, full H-bridge device with on-board power transistors?

From what I've read on the Brielle clocks with 1-minute pulses, they don't need bipolar pulses. That's why I didn't add one.

Hi. There may be a few Brillie slave clocks out there but everyone I’ve had and seen are alternating polarity and in most cases 30seconds impulsing, although it seems that the impulse unit sold to my friend isn’t is no use to him being a 1min pulse unit.
Each slave will run on around 1.5v sometimes a bit more at around 65 Ma. and alternating polarity.
Hope that helps.

 

[Diver300]

The capacitor on the output of the circuit shown will mean that the pulse is only a fraction of a second long. There doesn't seem much point in reducing the time that the transistors are turned on for.

 

[ZipZapOuch]

The capacitor on the output of the circuit shown will mean that the pulse is only a fraction of a second long. There doesn't seem much point in reducing the time that the transistors are turned on for.

Except to save battery life.

 

[AnalogKid]

Another design option is to move the pulse-forming in front of the output transistors.

ak

 

[schmitt trigger]

AND - If you don't mind hand-soldering an SMT component, the CD4060 and CD4024 can be replaced with one TSSOP part:

74AHC1G4215-Q100

ak

This is such an awesome device.
Since tuning fork resonators are 32768 Hz, the ubiquitous CD4060 will provide a 2Hz output which needs an additional divider to produce the (almost always) required 1 pulse per second.
Thanks for sharing.

 

[wombweller]

Another design option is to move the pulse-forming in front of the output transistors.

ak

I’m not sure I understand this ?

 

[Diver300]

The big capacitor limits the duration of the pulse that is sent to the clock. The function of limiting the duration of the pulse is what is meant by "pulse-forming".

It would also be possible to arrange a circuit that only allowed the output transistor to turn on for a fraction of a second, and then the output transistor could be directly connected to the clock input. That is what is meant by moving the pulse forming in front of the output transistor.

On a related subject, the existing design puts out a positive and a negative pulse. Is that negative pulse is needed or is it just to discharge the capacitor that becomes charged during the positive pulse? If the negative pulse is needed, does it have to be negative, or could it be a second positive pulse? If there is a circuit that has the pulse forming before the output transistors, knowing the answers to those questions could allow the circuit to be a lot simpler.

 

[wombweller]

The big capacitor limits the duration of the pulse that is sent to the clock. The function of limiting the duration of the pulse is what is meant by "pulse-forming".

It would also be possible to arrange a circuit that only allowed the output transistor to turn on for a fraction of a second, and then the output transistor could be directly connected to the clock input. That is what is meant by moving the pulse forming in front of the output transistor.

On a related subject, the existing design puts out a positive and a negative pulse. Is that negative pulse is needed or is it just to discharge the capacitor that becomes charged during the positive pulse? If the negative pulse is needed, does it have to be negative, or could it be a second positive pulse? If there is a circuit that has the pulse forming before the output transistors, knowing the answers to those questions could allow the circuit to be a lot simpler.

Thanks for the information. The movement in those slave clocks is an electro magnetic movement and does require an alternating polarity pulse, I’ll take some photos later on today so you can see why.
The owner of the pulsar unit is sending me it so I will be able to investigate it in more detail.

 

[Diver300]

If you do need reverse polarity pulses, then the capacitor is doing two things. One is limiting the pulse duration and the other is producing the negative voltage for the negative pulse.

If you have pulse shaping before the output transistors, you would need a full bridge driver with four transistors to so that the pulse could be reversed.