Need help in designing LED automotive circuit board

Here is the scenario. It is an LED design for an automotive 12v DC application. I have used LED Dynamics Buck Pucks in the past to feed 1 Lux III. I am having too many problems with them, so we need to design something new. Here are the problems

-They are causing FM interference in some cars due to we think..the switching frequencey
-The LED is blinking in some cars when the ignition is turned on, this needs to be fixed
-Too expensive for our application

So, I want a driver that will
-Power 1 Lux III at 900-1000ma(TBD) from 12v DC
-We want to design the circuit board to fit into our device(we have diagrams, but it needs to be a 32mm circle)
-Have no FM interference
-Will keep the LED at the same brightness for a long period of time
-Have no "flickering" issues
-be able to meet our budget

Does anyone have any ideas who can help design/make these. We want around 500-1000 of them. I spoke with George @ TaskLED already...so we are looking for other people that can help!

Does anyone have any idea whats causing the flickering and FM issue?? Any help would be greatly appreciated.

Thanks!

dlou87 said:

Does anyone have any idea whats causing the flickering and FM issue?? Any help would be greatly appreciated.

Click to expand...

This thread should give you some ideas. In a nutshell, forget the switching regulator and go with a linear one. That will solve the FM interference issues. Specifically, what causes the FM interference is the high frequency AC going through the inductor of the switching regulator. Sure, you can buy shielded inductors, and use bypass capacitors on the power supply leads, and perhaps put the entire thing in a metal case. That would probably work most of the time but it would add greatly to the expense, and you still might get interference. The flickering is likely because the feedback in the switching regulator can't adjust quickly enough to the fluctuations in voltage (a car is a very dirty source of power)

Now as I'm sure you know, one of the big problems with a linear regulator is the heat dissipation. However, there's a really neat way around this-use more LEDs! With a linear regulator you'll still have to deal with approximately 13.5 watts of power (1 amp from a "12V" supply which is typically about 13.5V) whether you have one LED or three. With three in series you'll be using about 11.4 V of the 13.5V to power the LEDs instead of using only about 3.8V. The resulting efficiency would rise from only about 28% up to around 84%-nearly as good as most switching regulators. And you'll get three times the light! The cost of two more luxIIIs would more than be offset by the lower cost of the linear regulator. All you need are a few bypass capacitors, an LM324 opamp, a diode, and a MOSFET. Total parts cost is probably under $2. Total cost of the parts mounted on a circuit board would probably by in the $4 area.

An alternative idea is the use about two dozen superflux (i.e. pirahna) LEDs instead of a LuxIII. Luminous flux would be about the same. Driver cost would be about the same (you would be using 8 series strings of 3). The cost of 24 superflux would probably be around $7 or $8 (about the same as three LuxIIIs). Mounting costs are more simply because you're dealing with more parts but this is offset by avoiding the need to have a heatsink. As you can see any design involves tradeoffs, and there's more than one way to skin a cat.

I may be able to help you with this project. Let me know what your cost constraints are. I actually already have the circuit board for the superflux design laid out and several samples made so I could make a prototype for you fairly quickly if you're interested in going that route instead of using LuxIIIs. We didn't use the superflux design for the taxi project because our cost constraints got tighter, not because there was anything wrong with the design.