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.

Need Help With POP-UP GHOST Circuit

Not open for further replies.


New Member
I just found this site and I am hoping someone could help with a small project I am working on.

I am designing a circuit to control a pop-up ghost for our halloween hay ride party we do every year. Basically, a tractor will run over a pressure switch which will activate a screaming ghost that pops up with and air cylinder for 6 seconds. I have built the circuit using a LM386, but that was not load enough. So I am now using a TDA2003 with a 8ohm 10w speaker. I have tested the circuit on a bread board and it seams to work great.

I have taken electronics at a Votec about 15 years ago. So I understand how circuits work and such. But I am not sure about how to figure the math on this circuit. And I was hoping someone with a lot for experience and knowledge would look at the circuit and give me some pointers or ways to make it better.

Not sure if I need to explain what I am doing with the circuit, so here goes.

When the trigger switch is activated it sends a low signal to the 555 timer circuit which sends a high signal to the relay and a LM7808 (to kick it down to 5v). The out put of the LM7808 goes to 2 LED (for eyes) and an OR circuit to reverse the signal to low which goes to an ISD1110 (audio recording chip upto 11seconds). the recording chip then sends the audio to the TDA2003.

Any help anyone could give me would be great. Thanks


  • popup-ghost---2008-R4.jpg
    286.2 KB · Views: 313
You mean LM7805, the LM7808 gives 8V, I presume it was a typo as you've got the LM7805 in your circuit.

You don't need two LM7805s, U2 isn't needed, remove it and increase R5 to 22k.

It's not very good practise to parrallel LEDs, they can have slightly different forward voltages so one will be brighter than the other and under extreme circuimstances they could burn out (this won't happen here as R6 sets the current is too low). Connect D1 and D2 in series and use 470R for R6 as without U2 it's being powered from 12V.

It's a good idea to add a 100nF ceramic capacitor from U4's output to 0V to improve transient response.

What to you need help with calculating?

The time delay for the 555 is equal to 1.1RC which in this case gives 68.2s. You could easily get near 6s by reducing C2 to 10:mu:F and R1 to 560k which would give 6.16s but there's no point in selecting accurate values as electrolytic capacitors have a typical tollerance of +/-20%.
Ops, yes it is 7805.

I like the idea of not using the second 7805 but I was not sure how to calculate the resistor need. I know I*R=V but I have no idea how to apply that to the circuit. Could you or someone explain a little. I have no idea how to calculate the current across this circuit. A simple one yes, this one no.

What is transient response?

I had change the value of R1 to 62k ohms but forgot to update the schematic.

Thank you for your help an suggestion. That is the help I am needing.
The resistor limits the current through the LEDs to stop them from overheating.

The formular to calculate the resistor value is:
[latex]R= \frac{V_{IN}-V_F}{I_F}\\
I_F = \text{Foward current}\\
V_F = \text{LED forward voltage}\\
V_{IN} = \text{Supply voltage}\\
R= \frac{12-4}{0.02}= 400
It's better to go up than down so I selected 470R which will limit the current to about 17mA.
Last edited:
additionally R6 has a value much too high. It leaves an LED-current of ~9.6mA, that is 4.8mA per LED if two are connected parallel to one resistor.

I want to think you for all your help. That formula helps alot. I am correct in saying you figured the amps at 20ma based on a 1/4watt resistor? I know that is a basic question, I am just wanting to get my head wrapped around this correctly.
The NPN transistor's package says it's hfe is 200 (well it says typical hfe). So based on the formula 12 x 200/.02 = 24000, that I need to use a resistor 24k or higher, or would the 22k be alright? Just trying to understand this a little better.
All Q2 does is earth pin 24 of U3.

To ensure the transistor saturates fully a general rule of thumb is to supply the base with about 1/10 of the collector current. Because the base voltage is so low relative to the supply, we can generally make the base resistor 10 times the value of the collector resistor to achieve this. With the LM7805 now gone the supply to the base resistor is now 12V which is roughly double 5V ,I approximately doubled the value of the base resistor to keep the base current the same.

Nothing is critical here, you could use a 27k, 24k, 20k or 18k resistor I just chose 22k because it's an E3 preferred value and is very common.

Incidentally, something you should be aware of is that there are standard resistor and capacitor values.
Although it isn't shown on Wikipedia, the E3 series exists as well, it consists of 10, 22 and 47 and is the most common, especially for electrolytic capacitors.

In general try to design your circuits to use resistor values mostly in the lowest E-series you can, for example if you can don't use a 30k resistor (an E24 value), if you can use a 33k resistor or even a 22k resistor.
Ok, I built the circuit but I have had a few problems.

1)By removing the second 7805 the protection diode for the relay (D3 & D4) causes the voltage to be wrong on the output of the NPN, so the ISD1110 just replays over and over until the trigger (SW2) is pressed, then it does nothing. If I remove those 2 diodes the system becomes stable and funtions.

2) R1 and C2 set the time the high is on once the trigger switch is pressed. With a 620K (R1) and a 10uf (C2) it should be around 6 seconds. But it runs for over 30 seconds. In the end I had to use an 100uf and a 10k, which should be 1.1 seconds, but it runs for 6 seconds. This confussed me.

3) When turned on the circuit, the trigger is automaticlly tripped but the test circuit I build with a peg board (with out the amp) does not do this. I have the pull up resistor on pin 2, so I am not sure why my new circuit does this.

I have attached the updated schematic and board layout. Any help or explanations would be great.


  • popup-ghost---2008-R4.jpg
    284.5 KB · Views: 268
  • 2008color.jpg
    491 KB · Views: 311
Last edited:
I solved problem 3. I left off C1 in my board and that solves the problem of the auto trigger when power i applied.
Well, I guess I didn't solve problem 3. Adding C1 back in helps but in still will auto trigger some of the time.
Well, I guess I didn't solve problem 3. Adding C1 back in helps but in still will auto trigger some of the time.

If its triggering at switch ON, you need to hold RESET, pin4 of the 555 low for a short time at power ON.

Connect a 1uf cap to pin4, then connect the junction of pin4 and the cap to the +V line via a 47K or 100K resistor.

Is the 555 a CMOS version or TTL.?
An RC circuit on the reset will stop it from trigering when the power is turned on.

Pin 3 should be connected directly to the trigger transistor, not via the relay which can introduce all sorts of noise.

I don't know why the delay is so different from the calculated time.


  • popup ghost.GIF
    popup ghost.GIF
    45.6 KB · Views: 249
Not open for further replies.

Latest threads

New Articles From Microcontroller Tips