0

I am trying to increase the ON time of a Relay. The Relay is turned ON from ULQ2003 by a 1 Sec ground pulse on its OUT1 Pin. I am trying to increase it to approximately 3 Sec. The accuracy of pulse width is not important, even if it varies within +/- 1 sec that is fine. But i don't want to use 555 timer or 74121 monostable for increasing the pulse time.

Is it possible to achieve this using some simple passive components or a transistor or darlingtons from the ULQ2003 driver?

Edit:

  1. I have improved the image and included the pulse-time in it.
  2. The Relay is 12V relay now. In my earlier question it was a 9V relay.

Relay circuit

Elliot Alderson
  • 31,192
  • 5
  • 29
  • 67
alt-rose
  • 1,441
  • 14
  • 29
  • 1
    Unfortunately, no. Surely you've thought of a resistor, but that only makes things unpredictable. Without using a timer, I think you can try PWM. Relays are designed for speed, and you're going against that mechanical system. – Aloha Aug 02 '19 at 04:53
  • 3
    why you don't alter a signal before enters the uln, possibly program in the mcu? What pulse width are you talking about? Delay and PWM are totally different things – Marko Buršič Aug 02 '19 at 05:40
  • 3
    What is the input signal to your ULQ2003 coming from ? That's where you might able to add an RC time delay. You could use the bjt's in other parts of the ULQ if needed. But the original source & shape of your input signal is important. Also - isn't this the same 9v relay in your other question ? – Kripacharys Aug 02 '19 at 06:51
  • 1
    ... and apparently it's a 9 V relay on a 12 V supply. – Transistor Aug 02 '19 at 07:12
  • 1
    If you don't want to use 555, please remove the tag in your question – Huisman Aug 02 '19 at 08:47
  • 2
    "Is it possible..." Yes. You can do this using just passive components and transistors. Just copy the schematic of a 555 timer using your own discrete components. – Elliot Alderson Aug 02 '19 at 11:18

1 Answers1

2

Possible? Definately. There are many solutions, although not all of them make sense in all cases.

Whatever you decide to do, I would heavily suggest to put the delay circuitry on the input side of the ULQ2003: It is by far easier to build up the circuit you want on lower-power side (I guess the input is logic-level/CMOS/TTL/etc.). That is the reason why one uses a driver or transistor in the first place. To deliver more power than your control circuit is designed for.

The way I would approach the problem is the following: Use a transistor or switch to quickly charge up a capacitor, then slowly discharge it via a large resistor. Connect a comparator to the circuit, and whenever the voltage at the capacitor is above a threshold set by a voltage divider, the output becomes active. When the capacitor is discharged to some level, the threshold is not reached anymore and the ouput turns off. This output can be directly connected to the input of the ULQ.

Well, that might sound a little complicated. Let's draw a schematic:

schematic

simulate this circuit – Schematic created using CircuitLab

Actually, the design is not great overall. Especially the NPN transistor as emitter follower is not the right choice for this application, as there will be a voltage drop of 0.7V and the capacitor will never be fully charged. I would recommend to change this part to whatever fits your needs.

You can tweak C1 and R3 to match your desired pulse length (currently 3.3s). Simply multiply them to get the time in seconds.

There is some more or less complex math behind this, look for RC timing circuit on google to find massive tutorials and resources for learning. This circuit is adapted from "The Art of Electronics" (p. 24, Figure 1.38). You might want to check that out for explanations as well.

K. Krull
  • 336
  • 2
  • 8