I am looking for information on how to build a crystal controlled LED array flasher capable of a 12V DC input and a 14Hz output @12v 900mA. 50% on-off cycle. Does anyone have plans or can steer me in the right direction?

 

Hello,

Did you already look at the 4060 Cmos oscillator/divider ?
See datasheet.
Output Q14 divides the crystal by 16384.
You will need a 14 * 16384 Hz = 229376 Hz crystal.
Feed pin 3 via a 220 Ohm resistor to the gate of a power mosfet and your flasher is complete.

Greetings,
Bertus

 

Hello,

Did you already look at the 4060 Cmos oscillator/divider ?
See datasheet.
Output Q14 divides the crystal by 16384.
You will need a 14 * 16384 Hz = 229376 Hz crystal.
Feed pin 3 via a 220 Ohm resistor to the gate of a power mosfet and your flasher is complete.

Greetings,
Bertus

As your signature says, "You don't have to know everything, if you know where to find it." Do you have any idea where he would find a 229376 Hz crystal? (unless he needs 10,000 of them)
It might be easier to use a microcontroller, and/or start with something like a 28kHz crystal. Dividing by 2000 is pretty simple, although it will take more than one IC if a μC is not used.

 

Hello Ron H,

You are right a crystal of that frequency is very hard to find.
The wanted frequency of 14 Hz is also "strange".
When I look at the list of available crystals from my local RF parts supplier there is not much to get this 14 Hz.

Laagfrequente kristallen high-Q Statek, alle RM 7.5 mm, freqs. in kHz:
   3.010 kHz (osc. in DIL) 10.000 kHz                                       4.90
  12.800 kHz                                                                4.90
  15.364 kHz                                                                4.90
  16.000 kHz                                                                4.90
  16.384 kHz                                                                4.90
  18.641 kHz                                                                4.90
  19.200 kHz                                                                4.90
  25.000 kHz                                                                4.90
  30.720 kHz                                                                4.90
  31.500 kHz                                                                4.90
  40.000 kHz                                                                4.90
  40.960 kHz                                                                4.90
  60.000 kHz                                                                4.90
  76.800 kHz                                                                4.90
 100.000 kHz                                                                4.90
 153.600 kHz                                                                4.90
 240.000 kHz                                                                4.90
 307.200 kHz                                                                4.90

Greetings,
Bertus
1000.000 kHz 4.90

 

Thanks for the information. I have not been able to find a crystal to make the 4060 get to 14 Hz. By the way 14 Hz is the frequency used to activate traffic control devices. I am trying to flash an array of IR LEDs instead of using the traditional xenon flash tube.

My experience with digital circuits is limited and the 4060 looked like the best solution, but the availability of the right crystal makes me wonder if the microcontroller might be the way to go ? Any sugestions on going that route ?

 

Hello,

How much difference may there be on the 14 Hz ?
Must it be crystal accuracy ?

Greetings,
Bertus

 

Bertus,
Stop reading my mind, willya?

Yeah, I was thinking the same thing. Microchip has PIC microcontrollers that have internal clocks that are factory calibrated within 1% of their rated frequency. Something like a PIC12Fx series, which is available in an 8-pin DIP package (and in industrial temperature ranges), might be a good candidate for experimentation. That series has 4MHz and 8MHz internal clocks; they can also be clocked externally using crystals, etc.

No point in looking for a uC with a PWM function, as 14Hz is far too slow; the lower frequency range for intrinsic PWM functions is around 1kHz.

The nice thing about using uC's is that your parts count is low, and changing the functionality generally just involves a change in software.

 

I hope he's not developing this for his own use.

 

I hope he's not developing this for his own use.

Yeah, I thought about that. I suppose there should be some kind of "Don't try this at home, kids!" kind of disclaimer made. I simply made the assumption that "E V Tech" stood for Emergency Vehicle Technician, but was a lot easier to type.

Just so it's said, the use or sale of traffic signal preemption devices in an unauthorized context was made illegal in the United States in 2005.

To further discourage unauthorized users, most signals go to 4-way red, so there would be no advantage to using one anyway.

 

Yeah, I thought about that. I suppose there should be some kind of "Don't try this at home, kids!" kind of disclaimer made. I simply made the assumption that "E V Tech" stood for Emergency Vehicle Technician, but was a lot easier to type.

Just so it's said, the use or sale of traffic signal preemption devices in an unauthorized context was made illegal in the United States in 2005.

To further discourage unauthorized users, most signals go to 4-way red, so there would be no advantage to using one anyway.

Unless said unauthorized user also had a red truck with flashing lights on top and a siren.
That would, of course, be an even deeper hole he would eventually have to try to climb out of.

 

The output does have to be within +/- .1 Hz or it will not work. You are correct in assuming that I am an Emergency Vehicle Technician and that unauthorized use of this type of device is illegal. I have several clients who require a device such as this and have asked me to look into building it for them. I have everything else ready to go except the timing circuit. There is too much variation in a 555 R C circuit as the temperature range can vary by 100 degrees F here.

 

A 32768 Hz crystal oscillator, divided by 2340 or 2341, will get you close enough. You can use a CD4059A to do the division.

 

Hello,

I go with Ron H.
I would take the 32768 crystal, divide by 1170, and then divide by 2 for the 50 % duty cycle.

Greetings,
Bertus

 

With the PIC12Fx series being within 1% of 4MHz with their internal resonators, they would easily have sufficient precision - and that's a single 8-pin IC; a 5v regulator would be necessary, along with logic-level MOSFETs to control the LEDs.

The big "learning curve" here would be the programming aspect. However, once that's done with (and a simple timing loop toggling I/O pins is rather elementary) then the resulting assembly will be quite a bit more rugged without a fragile crystal being involved.