Maker Pro
Maker Pro

556 timer circuit with a problem

Toby

Jun 15, 2011
19
Joined
Jun 15, 2011
Messages
19
Hello all,

I am trying to build a circuit that will operate a motor for about 3 seconds, this is via relays. The 556 timer operates the relays to provide power to the motor in either direction, depending on which trigger it recieves. The circuit is currently on proto-board.

Disregarding the motor, as that is a simple case of wiring power to it through the relays, I have an issue with the 556 circuit that drives the relays. The schematic is below...

I've been receiving help but they have now decided they no longer wish to so some of the design is things that have been suggested to me which I'm still not 100% on understanding.

The main issue is the fact that both relays are energising when I trigger just one side of the 556. If I trigger one side, one relay should energise. Trigger the opposite side and the other relay should trigger. It is a BJT 556 as opposed to a CMOS.

Also, of the things suggested I add to the circuit, I understand the purpose of the 10k pull-up resistors but I'm not entirely sure of the purpose of the 1N4148 diodes. It was explained briefly as preventing the trigger pin going higher than Vcc though perhaps someone could explain how that may happen without the diode.

The only other thing is the rating of the fuse. Is it a case of measuring the normal current draw of the circuit and selecting the appropriate fuse or should there be a way to calculate it? Is it even needed or am I being over cautious?

Any help is appreciated! Thanks in advance. If there is any more info that is needed, let me know.

Proto.jpg
 

poor mystic

Apr 8, 2011
1,074
Joined
Apr 8, 2011
Messages
1,074
Hi Toby
Last things first.
Fuses should protect the equipment from the effects of an uncontrolled draw of current or short circuit. In this case I'd want my fuse a good deal smaller than 10A since that's enough to melt the protoboard. Would the circuit still run (once it's running that is) on a far smaller fuse? What about 1A or even 500mA?

All kinds of things can go wrong with circuits built on protoboard. Here are some things I think need checked:
Are you sure that the protoboard is in good condition? Perhaps you could rebuild the circuit on a new section of the board, or move it along the board a bit.
Is the power supply good? Is the capacitor effectively smoothing the supply?
Is there a short circuit between the trigger pins? I could go on but there's no point, it is not possible to cover all eventualities.
Honestly there are really so many possible ways a circuit like this could go wrong you can hardly wave a stick at them let alone count them out in an off-the-cuff essay ike this.
If you think the circuit should be OK in other respects solder it up and try it out in the real world.

The most important thing is not to give up, so that in the end you'll succeed.
 

duke37

Jan 9, 2011
5,364
Joined
Jan 9, 2011
Messages
5,364
The trigger input will sit at 12V due to the 10k pull-up resistor. The pulse coming in from the 4069 will have a positive going edge and a negative going edge. On the negative edge, the trigger input will go low and then will rise to 12v as the capacitor charges. On a positive going edge, the trigger starts at 12V and will be driven positive so the trigger input will go above 12V. This is restrained by the diode which will limit the voltage to only a little above 12V by passing sufficient charging current.

If both relays could come in together, you should consider some protection otherwise the power supply will be shorted. If you have spare contacts, each relay should be powered through the back contacts of the other relay so that the first one to go in will switch off the other.

Try better smoothing of the supply to the 4069, provide power through a 10k resistor and add a 10uF capacitor in parallel with th 0.1uF, this may stop spurious triggering
 

Toby

Jun 15, 2011
19
Joined
Jun 15, 2011
Messages
19
Thanks for the help so far.

10A is probably far too high for the 556 circuit. I wouldn't know how to go about calculating how much current it should draw, and therefore the appropriate rating, without actually measuring once the circuit is built and connected in its final resting place.

Protoboard is brand new, the power supply is an old car battery but it is still sufficiently charged.

Duke, many thanks for the explnation. I can now see why the diode needs to be there. I had considered that possibility and as you suggest, wiring the relay contacts such a short is impossible is one way. I also came across an example of a lock out circuit that will prevent one timer being triggered if the other's output is already high which seems a more elegant solution.

The supply to both the 556 and the 4069 are decoupled by a 0.1uF as shown in the schematic but I also have a 47uF in parallel on the protoboard after reading that one larger cap as well as the smaller capacitor is more effective. Should have mentioned that beforehand, apologies.

When you mention a 10k resistor, can I confirm you mean the supply to the Vcc pins on the ICs? Triggering itself is fine, it only works when one side is triggered. There are no false triggers. The only problem is that both outputs are going high from one trigger.
 
Last edited:

duke37

Jan 9, 2011
5,364
Joined
Jan 9, 2011
Messages
5,364
The relay coils will take quite a large current and even a 47uF cap is unlikely to bypass a spike. The 4049 will take very little current and a resistor 1k to 10k in its supply should be OK and will supress any spike in association with a bypass capacitor to that particular chip.

I do not understand the reference to 10A. The 556 will only have to supply the current to the relay coils, one at a time. Current = 12V / coil resistance. The relay contacts will supply the output current.

The 4049 is cmos and has a very high impedance input, the input diodes will charge up the inputs positive, there is no provision to discharge the inputs. How about a 100k pull down resistor?

Instead of the 4049 I would prefer a 4093 which is a Schmitt trigger, a slow input can be used (capacitor) and the output will click over when the threshold is reached, this gives much better noise imunity. I wanted a 4049 and asked a trader at a radio rally how much he wanted, I handed over my money and he handed me a tube containing a couple of dozen so almost everything I make has one!
 

Toby

Jun 15, 2011
19
Joined
Jun 15, 2011
Messages
19
Relay coils are rated at 33.3mA so a fuse as close as I can get to that? 10A was just a random value, mainly to demonstrate it was a fuse.

Does it make a difference to your reply that I'm using a 4069 rather than a 4049?

When you say the diodes will charge up the inputs positive, and that there is no provision to discharge the inputs. What is meant by this and what are the consequences of it?

So suggestions are to take my circuit as is currently. Then perhaps a 10k on the 4069 supply and a 100k pull down on the inputs to the 4069. Lets give it a try!

Apologies for the questions, I'm trying to learn :D
 
Last edited:

duke37

Jan 9, 2011
5,364
Joined
Jan 9, 2011
Messages
5,364
A 10A fuse seems to be a bit irrelevant, nothing in the circuit could stand that.

The 4069 is an invertor which changes output when the input crosses about half supply voltage. If there is noise on the input signal then you can get multiple output pulses. The 4093 is a Schmitt trigger which switches when the input goes above 2/3 of supply and below 1/3 of supply. In your case you would need a noise signal of 4V before it messes up and switches. The 4093 gives a rapid output change even with a low rate of change on the input so you get a good pulse to trigger the 556.

Thinking about it. the 10k power supply resistor may be too high, 1k may be better. You can measure what voltage you get.

Your input circuitry is not defined, what do the input diodes connect to?

Noise immunity is a big subject!
 

Toby

Jun 15, 2011
19
Joined
Jun 15, 2011
Messages
19
As I mention, it doesn;t need to be (and won't be, now I know) 10A. Maybe something like 50mA or 100mA - I'm not sure how low fuses go.

I'll look into the 4093. I'll have to get some 1k resistors. I've got 10k, 100k and 1M :D I also have plenty of 4069 ICs but if the 4093 works better I'll get some.

The input diodes are connected to a central locking system. Effectively one input goes high on locking, the other input is high on unlocking. Both signals are momentary (probably about a second or so - if that) When at rest, both are tied to ground.

One thing that makes me wonder about noise, this odd behaviour (having both outputs high together) is regular. Happens every time I trigger one side, rather than randomly. Makes me think that it is the way its wired on the protoboard but as far as I can see (and I've checked numerous times) it is all wired up correctly.

EDIT: I've just tried a 10k on the 4069 supply and the problem is still occuring.
 
Last edited:

duke37

Jan 9, 2011
5,364
Joined
Jan 9, 2011
Messages
5,364
I attach a diagram of what might do your job, I show only half the circuit, the other half is identical. The input is by-passed with a capacitor to reduce the possibility of noise input. The untriggerd input is tied down to ground with a resistor. Have you tried this on your circuit?
The two halves of the circuit are cross coupled so that only one relay will come in at one time.
I hope you get your original circuit working
 

Attachments

  • Toby.jpg
    Toby.jpg
    67.2 KB · Views: 1,593

Toby

Jun 15, 2011
19
Joined
Jun 15, 2011
Messages
19
Thanks for the drawing.

As per the schematic, both inputs are tied to ground when they are not active. When I wish to trigger the circuit, one input goes high, the other stays tied to ground. As mention, only one relay is meant to be triggered, but both are. The trigger pins on the 556 are driving both outputs high, rather than just the one they are supposed to.
 

duke37

Jan 9, 2011
5,364
Joined
Jan 9, 2011
Messages
5,364
You could pull the 0.1 caps, one at a time and see if the problem is still there. You can trigger the 556 by hand with the caps out. Try to simplify the problem by getting one part working at a time.
 

Toby

Jun 15, 2011
19
Joined
Jun 15, 2011
Messages
19
Well I THINK it is sorted. Though in the past when ever I say that I find something else wrong.

I tried you're suggestion duke and was having no joy.

Now, to my embarrassment, I found a reverse EMF diode in the wrong pins. I'd misplaced it by one row and it wasn't protecting the coil. I think this may have killed the output on the IC. A fresh NE556 later and all appears to work well.
 
Top