Flashing Police Lights

[Boncuk]

That won't work.

Both MOSFETs will stay on because the LEDs (which won't light) prevent the gat capacitance from being discharged when the IC's outputs go low.

With other words you are telling me that Lab Center Electronics (UK) is selling a pile of junk with "PROTEUS - ISIS" for 4,500.00 lbs.

Further there is only one LED contained in the schematic, which is the power LED. The lamps connected to the power MosFets are light bulbs with 12Ω resistance, which can easily be identified by the filament drawn in the symbol, one of them lit.

To make sure the gate discharges a small value resistor (1K) could be connected from ground to the gate.

PWM circuits made using an NE555 timer prove that the Fet gates discharge quickly.

Boncuk

 

[Boncuk]

What did Chuck Norris' flashing headlights look like? I don't remember them from the show. '

I am pretty sure U1 in Boncuk's schematic is a 555 timer. Its very common and easy to get. U2 looks like a 4017 Decade Counter. Also very common and easy to find.

The BUZ10's are power MOSFETs. The resistors and capacitors have values next to them.

The 555 is used in astable mode. If you google "555 astable" you will get lots of pages to learn about how this works and how to adjust the speed.

The 4017 just clicks along at the speed the 555 is going, setting one pin at a time high. Pretty much

Pulse 1 from the 555 sets Q0 high and lights lamp A
pulse 2 will make Q1 go high and Q0 go back low. Nothing is hooked to Q1 so lamp A goes out
pulse 3 makes Q2 high and lamp A goes back on
then we have 2 unused pins that give a delay and then we move over to lamp B and repeat.

Q8 is hooked to the reset pin on the 4017 so we can do it all over again. This just loops over and over and over.

Make sense? The diodes block the current from going back to the low pin on the 4017

The speed of 20Hz seems way to fast for a incadesent bulb to me. I don't think it will even flash at that speed, just look steady and a bit dim.

I am sure Boncuk will stop back in and clarify this in time. He's a regular around here.

Hi Andy,

some years ago there was a movie series on German TV with Chuck Norris as ranger driving a (obviously tuned) pickup with a second set of headlights (strobes) which flashed at a pretty high rate alternating left and right with a double flash each.

I imagine being followed by a policecar with that outfit has a strong psychological effect on the person being chased.

You probably didn't click on the image to magnify it. The part names are contained in the schematic.

A clock frequency of 20Hz is just right for the necessary effect. I used 1A filament type lamps to simulate and the really do flash. (500mA bulbs glow steadily and dimly) If 55W lamps are used they will also flash. Considering the frequency division of the decimal counter, the on and off times of the lamps is pretty long.

The OP might use the circuit and make the astable frequency adjustable within a range of 10 to 20Hz.

Regards

Boncuk

 

[Boncuk]

Revised schematic

Hi parkaroo13,

I have revised the schematic and added two pull-down resistors of 10KΩ to properly discharge the MosFet gates. I have further added a trim pot (100KΩ)in the timer circuit to make the frequency adjustable from 13 to 21Hz.

As is can clearly be seen from the simulation the circuit works fine (even without gate pull-down resistors).

If you want I'll make a single layer PCB layout for you.

Boncuk

 

[Duvig123]

If you want to buy a ready made flasher, Built for Emergency vehicles, look up "WIG WAG" on ebay. The dont cost too much and last practically forever.....

 

[Boncuk]

Schematic and PCB-layout

Hi pakaroo13,

in the meanwhile I made a schematic and PCB-layout of the circuit with some small modifications to have equal times on and off for both lamps.

The PCB layout is scaled 2:1. The actual size is 3.4X2.1inches.

The board is purely single sided with no traces between IC-pins.

Boncuk

 

[Hero999]

With other words you are telling me that Lab Center Electronics (UK) is selling a pile of junk with "PROTEUS - ISIS" for 4,500.00 lbs.

Just because it works on a simulator it doesn't mean it will work in real life.

Further there is only one LED contained in the schematic, which is the power LED. The lamps connected to the power MosFets are light bulbs with 12Ω resistance, which can easily be identified by the filament drawn in the symbol, one of them lit.

Yes you're right they're normal diodes not LEDs but it doesn't make any difference, the effect is the same. Picture the MOSFET gates as capacitors which will charge via the diodes but nothing will discharge them.

PWM circuits made using an NE555 timer prove that the Fet gates discharge quickly.

That's because a 555 has a totem pole output which can both push and pull current.
You couldn't simply add a pull down resistor on the gates of the MOSFETs could you?

To make sure the gate discharges a small value resistor (1K) could be connected from ground to the gate.

That will work and is fine for low frequency applications such as this but, just for future reference, it's no good for higher frequencies - 100kHz i.e. switching regulators.

 

[Boncuk]

You couldn't simply add a pull down resistor on the gates of the MOSFETs could you?

Yes, I can and I also did. Compare the latest simulation screenshots. Each gate has a 10K pull down resistor.

Those pull down resistors are also contained in the Eagle schematic and PCB layout.

 

[Ubergeek63]

i did this for someone who wanted to pack that into a matchbox car... I don't know why you guys are calling it a "chuck norris flasher" all unmarked vehicles here in the US use that sequence. In retrospect, he could have used a single IC in a 74HC4060 by driving the LEDs with Q6 and referencing them to Q4. (or any 2 outputs that skip one number)

Now if you want a real drastic effect you charge an inductor to an LED absolute max current and then dump it to the LED-normal headlights are slow so the instant on of an LED would heighten the effect.

Dan

 

[Hero999]

Yes, I can and I also did. Compare the latest simulation screenshots. Each gate has a 10K pull down resistor.

Those pull down resistors are also contained in the Eagle schematic and PCB layout.

That'll do the job.

Here's how I came to the conclusion that it won't work:

What's the reverse leakage current of the diodes?

The graph on page 5 of the datasheet says 0.011µA with a reverse voltage of 20V and a temperature of 20°C. At lower voltages it will be even lower.

https://www.electro-tech-online.com/custompdfs/2009/02/1N4148_1N4448_5.pdf

What's the capacitance of the MOSFET gate?

Page 3 of the datasheet says 820pF.

http://www.datasheetcatalog.org/datasheet/siemens/BUZ10.pdf

[latex] \frac{dV}{dt} = \frac{C}{I} = \frac{820 \times 10^{-12}}{0.011 \times 10^{-6}} = 0.0745s[/latex]

Note that this is how long it will take for the voltage at the gate to drop by 1V. If the initial gate voltage is 12V and the bulb will turn off when the gate voltage drops below 4V then it'll take 0.6s for this to happen.

If you're running this at a frequency of under 0.4Hz then you probably won't notice this but at 10Hz it probably will never turn off.

The resistors you've added will solve the problem by providing a discharge path for the MOSFET gate capacitance - the RC time constant will be 8.2µs which is more than fast enough.

 

[Boncuk]

Thank you for the thorough explanation, hero999.

Boncuk

 

[shokjok]

I agree with Andy1845c about the flasher idea. A wrap-around flasher loop will activate each light alternately. If you cascade three auto flashers together, with diodes at the outputs having the anodes tied together feeding the bulbs, you can experiment to get near-stroboscopic flash sequence seen on wide multi-bulb roof bars. Circuit to follow later, this idea has not been tried.