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.

LED-lights in a Car

Status
Not open for further replies.

Corday

New Member
I have made LED-lights for my car (dome-light, instrument cluster, HVAC, etc) to replace the stock bulbs. Although I have not installed any of these “bigger” units yet I have had a few test-series of LEDs connected in the car for some time. These are all still in good working condition.

Before installing the bigger pieces in the car i wish to take advice from people who know this alot better than me =)

The LED-lights are comprised of many series of 3 LEDs. Each series has a separate resistor. The LEDs Vf range from 3 to 3,5v and most are 30mA (they are standard high-intensive and not luxeon-stars or the like).

The issue here is the cars DC output vs the LED arrays resistors etc...
I have a scantool/datalogger software hooked up to the car and have monitored the voltage with alternator both on and off.. With alternator off, i get 12,5-12,8v.. and when its on i get around 13,7-13,8v.

I have googled and read alot and the biggest problem with car voltage in general seems to be "voltage spikes". Some say these spikes can become up to 150v..!

Although i have logged the car for hours and hours i have never seen a spike, but maybe i need an oscillator or something to see that properly.. But how big are the chances of a spike in a new car? Is it like once a day or week? Or is it more like "it could happen maybe if you were driving 350mph and lightning struck your car"?

Fact remains though that the test-LEDs are still working good...

My initial plan was to use LM317 to regulate the voltage to 11v - which would provide 11v for the LEDs both during alternator off and on. However 11v gives LM317 1,7v headroom when alternator is off and 2,7v when when its on.

I do realize that 1,7v may not be good enough and thats why im asking for help, because i dont want to waste time building something that wont work... How does the LM317 act when Vin drops below the required Dropout voltage headroom? Does Vout become 0 or does it just decrease – making the LEDs dim?

Would it be better/simpler to just design the LED-arrays for 14-14,5v and accept that they become abit dimmer when alternator is off? But then its the possible "spikes"... I just cant see a perfect solution, but can accept a compromise..
 
If you are going to use a regulator like the LM317, you would be better to use it as a constant current regulator.

There is a circuit on page 18 of **broken link removed**
You need to increase the resistor to about 60 ohms to get a constant 20 mA.

There are also other constant current circuits that will work with smaller voltage drop.

Very fast voltage spikes are caused by inductive loads being switched. They are not too big a problem because you can absorb a lot of them with capacitors and 18 V zener diodes.

The big problem is load dump, where the alternator is running when the load is sudden reduced, and the battery isn't there to absorb the current. It cannot happen if the battery is connected and working, even if the battery is very weak.

A small resistor, say about 10 ohms, in series, before the 18 V zener, will stop a load dump form destroying your circuit, but load dumps are very rare.
 
How does the LM317 act when Vin drops below the required Dropout voltage headroom? Does Vout become 0 or does it just decrease – making the LEDs dim?

The output voltage will track the input voltage down.
The 317 will offer some protection against normal noise on the 12 volt line over just running the LED's and a resistor.
 
Last edited by a moderator:
I tested and its like you say ronv =) They only glow marginally less as the battery provides less voltage so this is not a big problem... I am very willing exchange a few mcd for safety and stability..

I have wired the LM317 exactly as the datasheet suggests for "adjustable voltage regulator". Capacitors have me abit confused though.. I put a 0,1uF on the Vin-side and 1uF on Vout... But I have bought a box with lots of different electrolytes which range from 1uF to 4700uF (50v) and also a few ceramic ones.. Would i gain anything from putting some more capacitors on Vin or Vout? Im trying to understand when use them but cant really get a clear picture..

Diver300 said:
If you are going to use a regulator like the LM317, you would be better to use it as a constant current regulator.

Hmm how would this work? My plan was to use a LM317 to pass through a max of 11v to as many LED series as possible within heat-limit, then use resistors to limit the current... If i instead use the LM317 for current-limiting, how will this affect the resistors?

Hmm.. If i had enough headroom to use two LM317 in series, and use one for voltage regulation and one for current regulation, would it then be possible to connect the LED series without any resistors at all? (Assuming the voltage-regulator provides exactly the voltage the series needs and current-regulator@20ma)

Sorry im new to this and probably sound real dumb x)
 
If you use an LM317 arranged as a current regulator, it will hold the current constant over a wide voltage range, so you don't need a voltage regulator. You don't need separate current limiting resistors either, only the one that is part of the constant current circuit.
 
I will use the LM317 as a voltage regulator and put a 1,5A fuse on the Vin to protect against load dumps.

If i use it as a current regulator, wouldnt that be the same as using one resistor for a bunch of series of LEDs in parallel? Since all LEDs Vf differs (and then also current draw) it would lead to a domino-effect death of them all. I've seen audioguru warn against it several times =/

PlaneCrazy - Ehm "Okay"? If you read my first post properly, you will see what LEDs i used, and how i connected them. Cheaper than stock? The car didnt come with LEDs stock - obviously or i wouldnt have replaced the bulbs with LEDs....
 
Last edited:
PlaneCrazy - Ehm "Okay"? If you read my first post properly, you will see what LEDs i used, and how i connected them. Cheaper than stock? The car didnt come with LEDs stock - obviously or i wouldnt have replaced the bulbs with LEDs....

By "stock" I mean this:

I'm using two of the third type from the right, ans do far I'm not having problems. I can't dim yet, but that's another issue, and hardly missed.
 

Attachments

  • T1.75 LED.jpg
    T1.75 LED.jpg
    19.2 KB · Views: 350
  • T1.75 LED 2.jpg
    T1.75 LED 2.jpg
    15.3 KB · Views: 344
  • T3.25 LED.jpg
    T3.25 LED.jpg
    20.2 KB · Views: 342
  • T3.25 LED 2.jpg
    T3.25 LED 2.jpg
    28 KB · Views: 341
Last edited:
Status
Not open for further replies.

Latest threads

New Articles From Microcontroller Tips

Back
Top