Continue to Site

Welcome to our site!

Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

  • Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

time delayed shut off

Status
Not open for further replies.

spicydeath82

New Member
hi, i need help coming up with a delayed shut off circuit.
here's what i need it for:
a: i shut off my video game console
b: the cooling fan(s) stay(s) running for lets say 5 minuets
c: then once time has reached around 5 minuets fan(s) shut off.

the reason for wanting this behavior is that the current design for the xbox 360's gpu heat sink is inadequate. so currently i have an externally powered fan blowing through the heat sink. but i forget to shut it off sometimes. not a huge deal but it does add to the power bill and i'd like to eliminate the externally powered fan.

Details:
fan(s)
power: 12v

supply(from console):likely 12v or 5v

also since this will be inside the console(hopefully) it needs to be somewhat small but not too small.
i have no restrictions on how this is done. could be a 555 timer or something else.

thanks in advanced.
 
There was a massive discussion on this very topic not long ago here in these forums. The discussion even included if it was necessary since in theory the moment you shut down the system quits producing heat. Anyway...

With a 12 volt fan running off the system PSU the moment shutdown happens the PSU shuts down. Thus nothing that is to run after shutdown, fan, timer or otherwise can be powered by the system PSU. I think anyway.

Ron
 
Yes, the conclusion was that, as soon as the power is turned off, the unit no longer produces heat, so the temperature of the components will get no higher then they were when the unit was running. Thus there is reason or advantage to keep the fan on after the unit is powered off.
 
actually pin 5 always has power @ 5V while the psu is plugged in. also pin 7 can be tied to ground to turn on pins 4,5 and 6 which are 12V.
and why not humor me? if i do it and it doesn't work like i had hoped then i've learned a lesson. also i will say that my current set up has extended my 360's life considerably by having an external cooling fan blowing through the heat sinks. now i'm not here for a debate, just looking for some possible circuits to accomplish this.
 
actually pin 5 always has power @ 5V while the psu is plugged in. also pin 7 can be tied to ground to turn on pins 4,5 and 6 which are 12V.
and why not humor me? if i do it and it doesn't work like i had hoped then i've learned a lesson. also i will say that my current set up has extended my 360's life considerably by having an external cooling fan blowing through the heat sinks. now i'm not here for a debate, just looking for some possible circuits to accomplish this.

OK, this could be done with a 555 circuit but I have a real simple suggestion that may work. See the attached few circuits. Here is the description:

Description
The two circuits above illustrate opening a relay contact a short time after the ignition or light switch is turned off. The capacitor is charged and the relay is closed when the voltage at the diode anode rises to +12 volts. The circuit on the left is a common collector or emitter follower and has the advantage of one less part since a resistor is not needed in series with the transistor base. However the voltage across the relay coil will be two diode drops less than the supply voltage, or about 11 volts for a 12.5 volt input. The common emitter configuration on the right offers the advantage of the full supply voltage across the load for most of the delay time, which makes the relay pull-in and drop-out voltages less of a concern but requires an extra resistor in series with transistor base. The common emitter (circuit on the right) is the better circuit since the series base resistor can be selected to obtain the desired delay time whereas the capacitor must be selected for the common collector (or an additional resistor used in parallel with the capacitor). The time delay for the common emitter will be approximately 3 time constants or 3*R*C. The capacitor/resistor values can be worked out from the relay coil current and transistor gain. For example a 120 ohm relay coil will draw 100 mA at 12 volts and assumming a transistor gain of 30, the base current will be 100/30 = 3 mA. The voltage across the resistor will be the supply voltage minus two diode drops or 12-1.4 = 10.6. The resistor value will be the voltage/current = 10.6/0.003 = 3533 or about 3.6K. The capacitor value for a 15 second delay will be 15/3R = 1327 uF. We can use a standard 1000 uF capacitor and increase the resistor proportionally to get 15 seconds.

Obviously the part labeled "Ignition" is from 12 volt power that goes away and the power to the relay is always present. This could be 5 volts using a 5 volt relay or whatever you come up with. I won't swear this is a solution but for a few parts it may be worth trying and modifying.

Ron
 

Attachments

  • Relay Delay.gif
    Relay Delay.gif
    3.4 KB · Views: 1,150
you read my mind in a way. i had thought of a small relay acting as the switch, i just wasn't sure of how to keep it on for a time then allow it to shut off. so the capacitor acts as a temporary battery in the circuit to keep the relay on, then once it gets to low the relay shuts off?
 
you read my mind in a way. i had thought of a small relay acting as the switch, i just wasn't sure of how to keep it on for a time then allow it to shut off. so the capacitor acts as a temporary battery in the circuit to keep the relay on, then once it gets to low the relay shuts off?

Sort of, the cap slowly discharges through the transistor when power is removed keeping the transistor on which keeps the relay pulled in. Once the cap discharges the relay opens as the transistor turns off. It's a simple little circuit to mess with rather than moving into a 555 timer circuit.

Ron
 
koo. i could make one of these on some simple strip board.

That was the general idea. :)

Work with it a little and test it. Try both versions and see what it behaves like.

Ron
 
i may just make the circuit for the heck of it even though i don't need it now. i modded my Xbox so it's running 2 of the CPU heat sinks and it's running super cool now, the old GPU heat sink was tiny and in adequate and would get so hot it'd burn your fingers if you touched it. now the thing is very comfortable to the touch even after hours of playing.
so thanks Reloadron
 
No problem, just file it away for future refrence. :)

Ron
 
Hi guys. Interesting thread.
I just happened to be looking in need of this exact circuit!
I am trying to delay the switching off of a relay in an automotive circuit.

Reloadron
I have made your circuit (the second one, on the right) using a 4700uf capacitor, 50K trimpot and a 1N914 diode in parallel with the relay coil contacts.
I cannot seem to get the circuit to work. I can get the relay to trigger if I connect both the input of the 1n4002 and the collector to 12v but there is no time delay.
I have checked all connections but cannot find anything wrong. :confused:

Any chance you could suggest some items to check?
 
...
I am trying to delay the switching off of a relay in an automotive circuit.
...
I have made your circuit (the second one, on the right) using a 4700uf capacitor, 50K trimpot and a 1N914 diode in parallel with the relay coil contacts.
I cannot seem to get the circuit to work. I can get the relay to trigger if I connect both the input of the 1n4002 and the collector to 12v but there is no time delay.
...

The "+12V" input must remain connected to live 12V power AFTER the other input is disconnected. Bowden's circuit is naive, to say the least.

The delay will be very short. How much delay are you looking for?
 
Last edited:
Sorry. I figured out soon after posting that I had been given the wrong data sheet with the transistor, showing the E,B,C the opposite way.
I have replaced it and now have it placed correctly.
The circuit works, though as predicted with only a short delay.
I can adjust it with the pot until about half way after which it won't trigger anymore. The only way it will trigger again is if I turn the pot down below about half....
Ultimately I am looking to get around 10-15 mins delay if at all possible. Am also interested in the "safety" of the circuit. Is it possible or likely rather for the capacitor or any component to fail and start a fire in the vehicle? I am going to place the board in a jiffy box. Any thoughts?
 
At 10min (600sec), you will need a different technique.

What is the tolerance of the delay you need, e.g. would 15min +- 3min be ok?
 
Thanks for the quick reply Mike.
Yeah, tolerance isn't too much of an issue. Not wanting to go too long though. Less than 15 is fine but 18 mins and above is getting a little too long.
10 mins is good but I guess whatever is the easiest option. I already have this circuit so if it can be modified it would be great.
 
This will give you a long delay after the ignition switch is turned off. The NFet is not critical. Anything >50V Vds in T0220 package will likely work. I use a similar circuit to automatically turn off the Ham rig in my car. I also have a momentary push switch from +12V to the cathode of the diode so I can start the timer without putting the key in the switch.
 

Attachments

  • DF52.png
    DF52.png
    43.1 KB · Views: 380
Last edited:
Tried the circuit as suggested. Used a P16NF06, 2x 1N4004's and an electrolytic cap. I believe pin 2 of the MOSFET to be the drain – Is pin 2 also connected to the tag? Thanks for the circuit. I am getting around a 6min delay..
Now a few questions....:D

What is the best method of extending the circuit (pot?) to around 10 mins (or shortened also)?.
What is the overall potential timing range?
Is there any potential for fire etc.. if the circuit is insulated (Will be used in a new vehicle!) ?
You mentioned a momentary switch to the cathode side, obviously it was D2 you were referring to?
Does repeated switching (in quick succession) affect the circuit?
What would be the best way to reset the unit (drain capacitor to release relay)?

Ds.
 
...Used a P16NF06, 2x 1N4004's and an electrolytic cap. I believe pin 2 of the MOSFET to be the drain – Is pin 2 also connected to the tag?
Middle pin on TO220 Mosfets is the same as the tab.

Thanks for the circuit. I am getting around a 6min delay..
Now a few questions....:D

What is the best method of extending the circuit (pot?) to around 10 mins (or shortened also)?
The delay is proportional to the product R1*C1

What is the overall potential timing range?
Allowed range of R1 is 10K to ~5megΩ. C1 can be uF to mF.

Is there any potential for fire etc.. if the circuit is insulated (Will be used in a new vehicle!) ?
Run it through a ~1/4A fuse on the "always live" connection to the battery (not counting load being switched, fuse that separately). No fuse is necessary on the "ignition switch" connection.

You mentioned a momentary switch to the cathode side, obviously it was D2 you were referring to?
Yes, a momentary push switch from "always live" to the junction of D2 and R2 will pump-up the RC network for a new cycle.

Does repeated switching (in quick succession) affect the circuit?
No, each contact closure to R2 just recharges the RC network all the way...

What would be the best way to reset the unit (drain capacitor to release relay)?
Use a momentary push switch to ground the junction of D2 and R2.
 
How about using a thermistor in place of the RC timing circuit mentioned earlier?

Use a trimpot to set the temperature set point in a voltage divider circuit to the base of the transistor. Feed a resistor back from the output of transistor for hysteresis to stop relay chatter.

or even simpiler still use a bi metal strip

This way the fan will stop when it is cool enough.
 
Status
Not open for further replies.

Latest threads

New Articles From Microcontroller Tips

Back
Top