High-Power LED Flasher

Wendy

Joined Mar 24, 2008
23,415
It would work, but it would be better to drop the voltage some. There are several ways to do this. Are you planning on using 3 CR123s? If so a simple regulator set for 9V would be good.

A LM317 is common as dirt (Radio Shack carries them, as well as everyone else).



This part will get hot, so attach the metal tab to a largish chunk of metal for a heat sink.

Figure R2 as two 1.5KΩ resistors in series with each other. This will create 9.06V on the output. Not shown are two 0.1µF capacitors, and maybe a 220µF on the input if the power supply is more than 6" away. This little chip makes a dandy variable power supply if you ever want to expand your test equipment, just use a variable pot for R2.

I'm fighting the urge to buy one of these power pucks and LEDs. Have I mentioned my parts box overflowith?
 

Thread Starter

turnertj

Joined Nov 4, 2009
78
Cool, I'll try that. Going to the shack at lunch.
Is the current okay at 1.2-1.7Amps?

I do have some extra 700mA buck pucks that I mistakenly ordered, without the control line to them...want one or two?
 

SgtWookie

Joined Jul 17, 2007
22,230
If you're going to the Shack, you'll probably wind up with an IRF510 MOSFET

Better get a heat sink for it, and some heat sink compound. It may get kind of warm with the slow switching speed.
 

Thread Starter

turnertj

Joined Nov 4, 2009
78
Just so I have this right, Bill, the voltage regulator is for the battery charger source I found at my house, right? And not for the actual final prototype circuit with the Lithium 123s?

Tj
 

Wendy

Joined Mar 24, 2008
23,415
Just so I have this right, Bill, the voltage regulator is for the battery charger source I found at my house, right? And not for the actual final prototype circuit with the Lithium 123s?

Tj
Right, I just want to give you a source to test this sucker without sucking a bunch of batteries dry. Once you have a prototype built, then suck a bunch of batteries dry.
 

Thread Starter

turnertj

Joined Nov 4, 2009
78
Hey guys...

I got a power source at the electronics store today. It should be good. It's a 9V 2A source.

Anyway, I got to playing with the circuit more today. And what's going on is this. When I measure the current draw from the source, I get approximately 530mA. When I go to BuckPuck 1 (the top buck puck) and measure the current from the LED, it's seeing 650mA. The lower buck puck is giving the second LED 450mA. The drivers (buck pucks) are both 1000mA rated, and when you take wither of them out of the circuit and hook them up to an LED and the source, they deliver 1000mA at the LED.

Any ideas? Did I do something wrong?

Thanks.

Tj
 

SgtWookie

Joined Jul 17, 2007
22,230
Wait a minute - aren't those LEDs rated for 700mA?

Didn't you get the 700mA BuckPucks?

The Cntl pins are supposed to have a 1k input impedance. 9v/1000 = 9mA. That's pretty dicey at best with CMOS 555's for source current, particularly if they aren't working at their rated voltage.
 

Wendy

Joined Mar 24, 2008
23,415
What is the problem with the rated voltage? The 7555 should be golden at 9V (or more, given the fluxuation from the batteries).

This unit does have a control to tweak the current, as I recall. You may need to tweak that current down, and be sure the LEDs have good heat sinks.
 

Thread Starter

turnertj

Joined Nov 4, 2009
78
I was going to drive them at 700mA, but when I called to order the buck pucks they told me I should have no problem driving them at 700mA. So I got the 1000mA driver...think I'm driving them too much?
 

Wendy

Joined Mar 24, 2008
23,415
It will shorten the lifespan. Depending on your use this may not be an issue, but one of the reasons LEDs are so popular is their long life.
 

Thread Starter

turnertj

Joined Nov 4, 2009
78
Ok. Right now though neither LED is pulling more than 650mA. Is there a reason one BuckPuck is delivering 200mA more than the other? Shouldn't they both be delivering 1000mA?
 

Wendy

Joined Mar 24, 2008
23,415
What is the diode drop on each? Power = Voltage X Current, so if one diode is dropping more voltage it is using more power.

If you could give us the numbers your seeing along with the color of the LED it would be helpful. That curiosity bump still itches, but we need the numbers to understand what you're seeing. It might make the difference.
 

SgtWookie

Joined Jul 17, 2007
22,230
What is the problem with the rated voltage? The 7555 should be golden at 9V (or more, given the fluxuation from the batteries).
Bill,
What I was trying to say (I guess I didn't word it very well) is that the control input has a 1k input impedance.

"The Cntl pins are supposed to have a 1k input impedance. 9v/1000 = 9mA. That's pretty dicey at best with CMOS 555's for source current, particularly if they aren't working at their rated voltage."

That means that the 7555 timer will have to source 9mA at 9v, which I don't know if it will be able to do. I'm looking at a Dallas/Maxim ICM7555 datasheet, and I'm having to extrapolate from the current source plot; it looks like the Dallas/Maxim ICM7555 might source nearly that much, but would lose a volt from Vcc.

I don't know offhand exactly which 7555 that Tj would up with.

If the control line isn't being pulled all the way up to the BuckPuck's supply voltage, it will probably interpret that as an order to output less than maximum current.

This possible problem could be solved pretty easily using an NPN-PNP pair of transistors with some base current limiting resistors.
 
Last edited:

Wendy

Joined Mar 24, 2008
23,415
Without actually having a device in front of us your guess is as good as mine. My feeling is that it is a digital device (but I will say this is a guess). I believe it will work as is.

As a backup I'm going to try taking a PuckBuck without a control input and use some digital MOSFETs. I have some IRFZ34N I'm going to try with it.
 

SgtWookie

Joined Jul 17, 2007
22,230
Without actually having a device in front of us your guess is as good as mine. My feeling is that it is a digital device (but I will say this is a guess). I believe it will work as is.
What led me to look at the possibility that the control pin wasn't going high enough was Tj mentioning that it worked fine without the circuit attached, but with the circuit attached, the LED current was far lower.

My only explanation for that, is the control pin isn't being pulled high enough, because the TLC555 (or whichever version he has) can't source enough current.

If you have other ideas about why he's experiencing the current drop, I'm all ears.

As a backup I'm going to try taking a PuckBuck without a control input and use some digital MOSFETs. I have some IRFZ34N I'm going to try with it.
psst... Bill ... they're logic level MOSFETs. ;)

IRFZ34's are N-ch standard power MOSFETs; you will need Vgs=10v to turn them on fully.
IRLZ34's would be logic level power MOSFETs.
 
Top