Timer on button

[Samonite]

I have a project were I need to keep a food waste disposer activated in about 15 seconds after a button has been pressed.

A standard disposer just have a switch to turn it on and off:
**broken link removed**

I have created a circuit that does exactly what I want:
**broken link removed**

Here is my problem:
The circuit I created uses an external power source which I don't want. But how do I create a circuit without the need of any other power source than the power from the wire of the disposer?

 

[Hayato]

Ok.
Let's imagine a 6V relay, with a 75Ω coil, that starts to operate at 4.8V. So you need at least 64mA of current running through the coil.

Taking β=100, you need at least 640 uA at the base. => (Vc - 0.7)/5000 >= 640uA

Vc >= 3.9V

How long does it take to the cap discharge from 6V to 3.9V?
3.9 = 6*e^-t/5
-t/5 = ln(0.65)

t =~ 2.15 s, not 15 s.

You would need a RC constant of 35 to get about 15 seconds. For a 5k resistor, you would need a 7 mF (70,000 µF) capacitor.

The best thing is to use a MOSFET, with despresible gate current, you can choose the RC constant that would fit for you.

And yes, you can use the power from the waste disposer, but that would be a transformless power supply (not good).

 

[Samonite]

The 15 seconds was calculated by the software..
The gain is 320.
The relay starts to operate at 4.8V and disables again at 2.3V.
I don't know if thats better because I don't know very much about electronics.. (I'm not in an electronics class, I just need this timer for a bigger project)

It would be so good if I were able to use the power from the disposer. I would believe that the voltage of the disposer wire is about 6-12 V. I can't check it because I don't have one but on the other hand it doesn't matter because the project is only theoretically (I can just pick a value).

Would something like this work?
**broken link removed**

 

[Hayato]

How do you know the gain is 320?

Well, you have to confirm the voltage and current ratings from your disposer.

 

[Samonite]

How do you know the gain is 320?.

Because the program says so:
**broken link removed**


As I wrote, I don't know the voltage and current.. We just have to make an approximate estimate.


Edit:
Sry the gain was 325

 

[Hayato]

Ok.
The program is being optmist. For BC548B the gain varies from 200 to 450. So you always have to consider the worst case, in this case, 200, and even so, you can't use 200, to assure the saturation, you must use a forced factor.

With the relay disabling at 2.3V, you get 4.75 seconds. So you are going to need about 30,000 uF to get the 15 seconds.

 

[Samonite]

I think found the problem.. For some reason the relay disables at about 1V? I have no idea why..

But anyways, thats not my question.
How do you keep the current flowing in about 15 sec if you press a button once? I have thought about a 555 timer but I don't know if that a good idea and I have no idea how to use it.

 

[Hayato]

What determines the charge rate is the RC constant.

When you press the button, the equivalent resistence is too low. Let's say 500 mΩ.
With 500 mΩ and 1 mF, your RC = 500 µs.
To charge a capacitor you need about 5*RC so that 5*500 µs = 2.5 ms (you need about 2.5 ms to charge the capacitor).

I think it is better to use the first setup.