I have made a pendulum clock that has a 90Ms output pulse every second. I need to make it drive a slave dial that requires a pulse every second of alternating polarity. The pulse current required is about 200Ma. Series resistance can be applied to suit voltages above what the dial actually requires.
I have zero experience with H bridges but think that's what i need. I have been looking at eBay for a suitable module. One comes up multiple times and uses an L298N chip. I've attached the schematic from one ad.
From the description:
When ENA enable IN1 IN2 control OUT1 OUT2
When ENB enable IN3 IN4 control OUT3 OUT4

Initially I would only need one channel. The clock dial could be put where the motor is and ENA is presumably put High?
Then, If I feed in identically timed logic pulses of opposite phase into IN1,IN2 do I get the same from OUT1,OUT2 but with power from Vs to drive motors (or clocks in this case). Is that how they work ? Or is there something more suitable? Or any other advice appreciated..

 

I think there may be an easier solution that doesn't require an H-bridge. Why can't you just add a flip-flop to the output of your 90Ms clock, followed by a transistor circuit to provide the 200mA?

 

I think there may be an easier solution that doesn't require an H-bridge. Why can't you just add a flip-flop to the output of your 90Ms clock, followed by a transistor circuit to provide the 200mA?

Thanks for that but I would need some more info on that solution.
In case I haven't described the problem properly, assume the clock drive coil has two terminals A and B. Assume single 5v supply.Assume 30 seconds between drive pulses.
First, A must be +5v and B 0v then, 30 seconds later, A must be 0v and B +5volts....repeating, ad finitum.
EDIT" 30 seconds later" should read " second 1 second later". Apologies for error..

 

Where did the 30 seconds come from? I thought the 90mS pulse came every second.

 

sorry, when you get old you type stuff and make mistakes :-(
I started thinking of the existing dial which IS impulsed every 30 seconds. Now I want to drive a different dial that does need a pulse every second of reversed polarity....it has a second hand ! And there is 1pps 5v cmos 90ms pulse available ..
The clock is described at www.rogerj.co.uk/clock.htm
The 1pps comes from ic3d at the bottom of the pendulum drive circuit on that web page.

 

In the pendulum drive circuit, I think that you should be able to use the Q and Q-not outputs of IC2B to get the timing you need. If so, you would need to add a drive transistor to each output to supply the 200mA drive current you need. Bear in mind that I am not an engineer, but just a tinkerer. Maybe a more knowledgeable forum member will chime in. I would breadboard a circuit to try, but I have another project that I must attend to today. Good luck.

 

Depending on your coil resistance/inductance properties, something as simple as this may do the job :-

 

Many thanks Alec_t..I will study that suggestion tomorrow and get back...

 

Many thanks Alec_t..I will study that suggestion tomorrow and get back...

It took me a while to work out how that works. The "coil" is the clock motor ? It's the input that I don't quite get. It's a 50/50 .5 hz square wave with a transistion every 1 second. If that's right I understand the circuit. But the available pulse is 90Ms with a 1 second period. I think I would need a 555 or something to re-shape it to a 50/50 square wave. Have I understood correctly ?

 

Have I understood correctly ?

You have. You'd need a flip-flop (e.g. 74HC74) to halve your existing 1Hz pulse rate.

 

Thanks Alec_t I will try that out suggestion..It's ingenious. Surprised though that no one seemed to know about the little cheap h bridge module I asked about in #1. Initially replies were slow to come, so I actually ordered one. I'll find out how they work when I get it. (and if they are even suitable )

 

The H-bridge IC would certainly do the job, but needs additional logic circuits to drive its inputs in the correct sequence.