LED Taillights, Diy

[Matt21lutz]

So I'm working on my LED tail lights and I need some help, I've made a schematic for the circuit including a voltage regulator, somebody help me out? I want to make sure that I have everything I need and it's all right.

[J Knight]

Are you building an LED light from scratch?

[Matt21lutz]

Originally Posted by J Knight

Are you building an LED light from scratch?

Yes I am. Using super bright LEDs

[ray_mcavoy]

Hi Matt,

A few issues I noticed:

(1) This is just a matter of terminology, but PWM stands for Pulse Width Modulation. The regulators you have listed are linear regulators ... they don't use pulse width modulation.

(2) Steady state DC current flow through a capacitor is zero. So the circuit will not work with the .33μF & .1μF capacitors connected as shown. Instead, they should be connected between ground and the regulator's input (& output) terminals respectively.

(3) Linear voltage regulators need a bit of "overhead" voltage to work properly. The NTE1970 you have listed requires a minimum input voltage of 14.8V ... see specs at: http://www.nteinc.com/Web_pgs/Positive.php?a=12 Even their low dropout NTE1954 12V regulator requires a minimum input voltage of 13.6V (http://www.nteinc.com/Web_pgs/pos_lowdo.php?a=12) It's very common for the voltage in a vehicle to drop below that level.

Having the input voltage go below the regulator's minimum requirement won't cause any harm but it will cause the output voltage to sag and therefore your LEDs will be dimmer than normal. So you might want to consider using lower voltage regulators and re-calculating your current limiting resistors to work with the lower voltage. That might also require reducing the number of LEDs in each series string.

Another alternative would be to directly regulate the current through each string of LEDs to the required 20mA. Take a look at the "1A Current Regulator" example on page 16 of the LM317 datasheet (http://www.alldatasheet.com/datashee...NSC/LM317.html) for an idea of how to wire it as a current regulator. You'll obviously have to change the resistor from the 1.2Ω for their 1A example. You can calculate the value using Iout = Vref / R where Iout is the current, Vref is the LM317's 1.25V reference, and R is the resistance. So for Iout = 20mA you'd want 20mA = 1.25V / R. Or R = 1.25V / 20mA = 62.5Ω. The only catch with this method is that you'll need a regulator for each series string of LED's.

[Dead Parrot]

Don't forget that the normal diodes have a voltage drop themselves. Silicon is .7v and germanium is .3v IIRC.

[nekkidhillbilly]

also there cheaper options that work just as good as superbright. order mine from china for about half work great.

[Dead Parrot]

Here is a hasty circuit that should be somewhat simpler and cheaper. The 3 zeners in the top will limit the voltage to about 10v. Size the resistor to limit the current so the zeners are not dissipating too much power. This is important since they could be on for hours. The one on the bottom is set for 12v. Again, size that resistor so not much flows through the zener. The zeners cost about 5 cents each with a minimum order of 60.

[DREAMER1]

Originally Posted by ray_mcavoy

Hi Matt,

A few issues I noticed:

(1) This is just a matter of terminology, but PWM stands for Pulse Width Modulation. The regulators you have listed are linear regulators ... they don't use pulse width modulation.

(2) Steady state DC current flow through a capacitor is zero. So the circuit will not work with the .33μF & .1μF capacitors connected as shown. Instead, they should be connected between ground and the regulator's input (& output) terminals respectively.

(3) Linear voltage regulators need a bit of "overhead" voltage to work properly. The NTE1970 you have listed requires a minimum input voltage of 14.8V ... see specs at: http://www.nteinc.com/Web_pgs/Positive.php?a=12 Even their low dropout NTE1954 12V regulator requires a minimum input voltage of 13.6V (http://www.nteinc.com/Web_pgs/pos_lowdo.php?a=12) It's very common for the voltage in a vehicle to drop below that level.

Having the input voltage go below the regulator's minimum requirement won't cause any harm but it will cause the output voltage to sag and therefore your LEDs will be dimmer than normal. So you might want to consider using lower voltage regulators and re-calculating your current limiting resistors to work with the lower voltage. That might also require reducing the number of LEDs in each series string.

Another alternative would be to directly regulate the current through each string of LEDs to the required 20mA. Take a look at the "1A Current Regulator" example on page 16 of the LM317 datasheet (http://www.alldatasheet.com/datashee...NSC/LM317.html) for an idea of how to wire it as a current regulator. You'll obviously have to change the resistor from the 1.2Ω for their 1A example. You can calculate the value using Iout = Vref / R where Iout is the current, Vref is the LM317's 1.25V reference, and R is the resistance. So for Iout = 20mA you'd want 20mA = 1.25V / R. Or R = 1.25V / 20mA = 62.5Ω. The only catch with this method is that you'll need a regulator for each series string of LED's.

Ever feel like you walked in on a conversation from Star Trek?....I'm glad there's guys that kno this stuff that'll help out cause this is straight outta Mr. Spock's dialogue to me. LOL

[mrolds88]

Live long and prosper!! I understand some of it, then I go research to figure out the rest!!

[1774btcrew]

Ray is AWESOME and really knows his stuff! If you have a wiring question he IS the man to ask! He has helped me out more than once.

[ray_mcavoy]

Originally Posted by 1774btcrew

Ray is AWESOME and really knows his stuff! If you have a wiring question he IS the man to ask! He has helped me out more than once.

Thanks! And you're welcome for the help!

Matt ... I'd just like to add that the circuit that Dead Parrot posted should work good for your application since your LED strings present a relatively small fixed load that will always be connected across the Zener(s).

The Zener diodes also reminded me of something else I forgot to mention about the voltage regulator ICs. Should you still decide to use them in this or some other application, be aware that they also have a maximum input voltage of around 30 to 35 volts. Automotive electrical systems are prone to transient voltage spikes that can exceed that limit. Adding a Zener (connected between the regulator's input & ground) will help protect the regulator by shunting those voltage spikes to ground. For this case, the exact Zener voltage isn't critical so long as it's under the regulator's max rating but above the normal system voltage ... something in the 24 to 30V range would be good.

Originally Posted by DREAMER1

Ever feel like you walked in on a conversation from Star Trek?....I'm glad there's guys that kno this stuff that'll help out cause this is straight outta Mr. Spock's dialogue to me

Originally Posted by mrolds88

Live long and prosper!!

LOL

[Matt21lutz]

Originally Posted by ray_mcavoy

Hi Matt,

A few issues I noticed:

(1) This is just a matter of terminology, but PWM stands for Pulse Width Modulation. The regulators you have listed are linear regulators ... they don't use pulse width modulation.

(2) Steady state DC current flow through a capacitor is zero. So the circuit will not work with the .33μF & .1μF capacitors connected as shown. Instead, they should be connected between ground and the regulator's input (& output) terminals respectively.

(3) Linear voltage regulators need a bit of "overhead" voltage to work properly. The NTE1970 you have listed requires a minimum input voltage of 14.8V ... see specs at: http://www.nteinc.com/Web_pgs/Positive.php?a=12 Even their low dropout NTE1954 12V regulator requires a minimum input voltage of 13.6V (http://www.nteinc.com/Web_pgs/pos_lowdo.php?a=12) It's very common for the voltage in a vehicle to drop below that level.

Having the input voltage go below the regulator's minimum requirement won't cause any harm but it will cause the output voltage to sag and therefore your LEDs will be dimmer than normal. So you might want to consider using lower voltage regulators and re-calculating your current limiting resistors to work with the lower voltage. That might also require reducing the number of LEDs in each series string.

Another alternative would be to directly regulate the current through each string of LEDs to the required 20mA. Take a look at the "1A Current Regulator" example on page 16 of the LM317 datasheet (http://www.alldatasheet.com/datashee...NSC/LM317.html) for an idea of how to wire it as a current regulator. You'll obviously have to change the resistor from the 1.2Ω for their 1A example. You can calculate the value using Iout = Vref / R where Iout is the current, Vref is the LM317's 1.25V reference, and R is the resistance. So for Iout = 20mA you'd want 20mA = 1.25V / R. Or R = 1.25V / 20mA = 62.5Ω. The only catch with this method is that you'll need a regulator for each series string of LED's.

uhhhhhhhhhhhhh.. i feel like an idiot now. no idea what almost any of that means... crap. but All I know is i want to limit the voltage to the led's to a max voltage of 12v. of course. each one of my led's run at 2.2v at 20ma each. Really don't want to drop down to a 10v circuit would be to much really, even if I might not even create my own voltage regulator, might just buy one cause this is overly complicated.

[ray_mcavoy]

Originally Posted by Matt21lutz

uhhhhhhhhhhhhh.. i feel like an idiot now. no idea what almost any of that means... crap. but All I know is i want to limit the voltage to the led's to a max voltage of 12v. of course. each one of my led's run at 2.2v at 20ma each. Really don't want to drop down to a 10v circuit would be to much really, even if I might not even create my own voltage regulator, might just buy one cause this is overly complicated.

Okay, a pre-made regulator circuit will probably be your best choice then. But you still have to check the specifications to make sure it'll be suitable for your application. For example, you weren't completely off track with your original circuit using the NTE1970 regulator IC's ... you simply had the capacitors oriented incorrectly and overlooked the regulator's minimum input voltage requirements.

As far as the voltage goes, I agree, you wouldn't want to go much below 12V for your existing arrangement using 5 of those 2.2V LEDs in each series string. To work with a lower voltage, you'd have to reduce the number of LED's in each string. And that would mean adding more strings to keep the total number of LEDs the same.

[Matt21lutz]

Originally Posted by ray_mcavoy

Okay, a pre-made regulator circuit will probably be your best choice then. But you still have to check the specifications to make sure it'll be suitable for your application. For example, you weren't completely off track with your original circuit using the NTE1970 regulator IC's ... you simply had the capacitors oriented incorrectly and overlooked the regulator's minimum input voltage requirements.

As far as the voltage goes, I agree, you wouldn't want to go much below 12V for your existing arrangement using 5 of those 2.2V LEDs in each series string. To work with a lower voltage, you'd have to reduce the number of LED's in each string. And that would mean adding more strings to keep the total number of LEDs the same.

Ohhh I thought i had the whole ciruit wrong besides the led wiring setup. But I found something that might work, but i really want to make my own. This is what i found IF i did buy one but not sure, http://www.reuk.co.uk/buy-12-VOLT-RE...-WITH-FUSE.htm. But I continued to look on their site and found this voltage regulator, http://www.ti.com.cn/cn/lit/ds/symlink/lm2940-n.pdf, now it has a .5v dropout and I believe it can handle up 26v. I would like your thoughts because people are saying you know what your talking about and i believe them!

Also if it is and if u have the time could you make a circuit diagram of what i would need to make and o it properly? I would greatly appreciate the help!

[ray_mcavoy]

Originally Posted by Matt21lutz

Ohhh I thought i had the whole ciruit wrong besides the led wiring setup. But I found something that might work, but i really want to make my own. This is what i found IF i did buy one but not sure, http://www.reuk.co.uk/buy-12-VOLT-RE...-WITH-FUSE.htm. But I continued to look on their site and found this voltage regulator, http://www.ti.com.cn/cn/lit/ds/symlink/lm2940-n.pdf, now it has a .5v dropout and I believe it can handle up 26v. I would like your thoughts because people are saying you know what your talking about and i believe them!

From the looks of it, I think the LM2940 regulator IC (integrated circuit) you found in the 2nd link is what they're using on that pre-built circuit board in the 1st link. Specifically, it appears to be the LM2940CT-12 or LM2940T-12 version in a TO220 package.

I took a quick look through the LM2940 datasheet from your 2nd link and it appears as though it would be a good choice for your application. In fact, the description on the first page says it's designed for vehicular applications and has built-in protection against reverse polarity, transient voltage spikes, etc. so that's a plus.

The low 0.5V dropout or "overhead" is also good for your application. That means the regulator only needs a 0.5V difference between the input & output to maintain regulation. In other words, you'll have the full 12V output you're expecting as long as the input stays above 12.5V ... that's much better than the NTE1970's 14.8V min requirement.

As long as your truck's charging system is working properly, you shouldn't have to worry about input voltages lower than that. But as you can see from the "low voltage behavior" (figure 21 on the LM2940 datasheet), it's a fairly linear drop-off once the input goes below about 12V. So a low battery situation would allow the LEDs to dim, but the regulator isn't going to suddenly "switch off" under those conditions.

Originally Posted by Matt21lutz

Also if it is and if u have the time could you make a circuit diagram of what i would need to make and o it properly? I would greatly appreciate the help!

The "typical application" circuit (shown near the bottom of the first page of the LM2940 datasheet) is all you really need. Notice how the capacitors are connected ... that's what I was trying to describe when I said they were connected incorrectly in your original diagram.

So basically, take that "typical application" circuit for the LM2940 and substitute it into your original diagram in place of the NTE1970 & the 2 capacitors and you should be all set. But if that doesn't make any sense, just let me know and I can draw up a diagram for you.

Finally, your original circuit shows 56Ω resistors in series with each LED string. That's a good value to use for a 12V input, but like Dead Parrot pointed out earlier, don't forget about the voltage drop from the diodes being used to isolate the tail from the brake/turn circuits. With silicon diodes (0.7V drop) you'll only have 11.3V available to the LEDs. So you'll need smaller current limiting resistors in order to maintain the full rated 20mA through the LEDs. I'm calculating 15Ω resistors for five 2.2V, 20mA LEDs in series with an 11.3V supply.

[Matt21lutz]

Originally Posted by ray_mcavoy

From the looks of it, I think the LM2940 regulator IC (integrated circuit) you found in the 2nd link is what they're using on that pre-built circuit board in the 1st link. Specifically, it appears to be the LM2940CT-12 or LM2940T-12 version in a TO220 package.

I took a quick look through the LM2940 datasheet from your 2nd link and it appears as though it would be a good choice for your application. In fact, the description on the first page says it's designed for vehicular applications and has built-in protection against reverse polarity, transient voltage spikes, etc. so that's a plus.

The low 0.5V dropout or "overhead" is also good for your application. That means the regulator only needs a 0.5V difference between the input & output to maintain regulation. In other words, you'll have the full 12V output you're expecting as long as the input stays above 12.5V ... that's much better than the NTE1970's 14.8V min requirement.

As long as your truck's charging system is working properly, you shouldn't have to worry about input voltages lower than that. But as you can see from the "low voltage behavior" (figure 21 on the LM2940 datasheet), it's a fairly linear drop-off once the input goes below about 12V. So a low battery situation would allow the LEDs to dim, but the regulator isn't going to suddenly "switch off" under those conditions.

The "typical application" circuit (shown near the bottom of the first page of the LM2940 datasheet) is all you really need. Notice how the capacitors are connected ... that's what I was trying to describe when I said they were connected incorrectly in your original diagram.

So basically, take that "typical application" circuit for the LM2940 and substitute it into your original diagram in place of the NTE1970 & the 2 capacitors and you should be all set. But if that doesn't make any sense, just let me know and I can draw up a diagram for you.

Finally, your original circuit shows 56Ω resistors in series with each LED string. That's a good value to use for a 12V input, but like Dead Parrot pointed out earlier, don't forget about the voltage drop from the diodes being used to isolate the tail from the brake/turn circuits. With silicon diodes (0.7V drop) you'll only have 11.3V available to the LEDs. So you'll need smaller current limiting resistors in order to maintain the full rated 20mA through the LEDs. I'm calculating 15Ω resistors for five 2.2V, 20mA LEDs in series with an 11.3V supply.

awesome, I did see that it said for vehicular use on that just wanted to make sure. and as far as the diodes actually taking away some voltage.. i would have never thought about that unfortunately when i ordered all my resistors i got 56 ohm as the smallest ones, i could always buy more but then im paying triple just for shipping a few dollars worth of resistors, but i might. The only thing would that would be different is the 56 ohm would make the led's dimmer , just not sure how much actually... we'll find out! I'll probably make a few tester first then make the final product.

But i would like to get your opinion on something else that i thought about, what do you feel is the best way to dim the led's? with a 1/2 watt resistor or to use a voltage regulator with a lower output voltage? for instance a 10v output.

but thank you for all the help so far as well! i greatly appreciate it!

[kikkegek]

he guys,

I have already build a couple of these tail lights for myself.

see my sig for the end result.

here is a link to the build topic of it on HID PLANET
http://www.hidplanet.com/forums/show...or-my-Suburban

they came out really nice.

as far as LEDs....dont go for anything else then true "Philips Lumileds"...as you can read in the thread there is a lot of crap on the market and for two reasons:
- most of them can only be driven on 20mA in stead of 70 mA
- and color is wrong. Should be orange-rd for maximum output and not red. when using normal red leds a lot of the output will be lost behind the lens.

there are a coupler of guys on HIDPLANET selling them. Their price is OK and they ship fast. Even to me accross the ocean to the Netherlands.

[ray_mcavoy]

Originally Posted by Matt21lutz

awesome, I did see that it said for vehicular use on that just wanted to make sure. and as far as the diodes actually taking away some voltage.. i would have never thought about that unfortunately when i ordered all my resistors i got 56 ohm as the smallest ones, i could always buy more but then im paying triple just for shipping a few dollars worth of resistors, but i might. The only thing would that would be different is the 56 ohm would make the led's dimmer , just not sure how much actually... we'll find out! I'll probably make a few tester first then make the final product.

You're right ... they'll definitely be dimmer than normal with 56Ω resistors on a 12V supply after you add in the diode drop. And without some additional data on the LEDs you're using, it's hard to say just how dim they'll be.

You can put some of the 56Ω resistors you already have in parallel to obtain lower resistances. To calculate the total resistance, add the reciprocals and take the reciprocal of the sum. Rtotal = 1 / ((1/R1) + (1/R2) + ... + (1/Rn)). So 3 of those 56Ω resistors in parallel = 18.7Ω ... and that's pretty close to the 15Ω I calculated earlier. Using 3 parallel resistors for each LED string obviously isn't something you'd want to do for your final design (it'd be a waste of resistors & board space). But I thought I'd mention it because it's something you could do for preliminary testing on a string or two of LEDs without having to buy more resistors.

Originally Posted by Matt21lutz

But i would like to get your opinion on something else that i thought about, what do you feel is the best way to dim the led's? with a 1/2 watt resistor or to use a voltage regulator with a lower output voltage? for instance a 10v output.

Either one will work (and there are several other ways as well). The resistor is simple and will get the job done. Unless you've already done so, you might need to try a few different resistance values to get the brightness of your tail lights to the desired level.

If you use a lower output voltage regulator, you might want to consider an adjustable regulator instead of a fixed 10V output. That'd allow you to easily change the output voltage in case 10V ends up making the LEDs too bright or too dim. The LM2941T is an adjustable version of the LM2940T-12 fixed 12V regulator you're considering. The output voltage is adjustable using the ratio of a couple of resistors. And for testing purposes you could use a potentiometer.

Originally Posted by Matt21lutz

but thank you for all the help so far as well! i greatly appreciate it!

You're welcome!

Originally Posted by kikkegek

I have already build a couple of these tail lights for myself.

see my sig for the end result.

here is a link to the build topic of it on HID PLANET
http://www.hidplanet.com/forums/show...or-my-Suburban

Nice job! I didn't get a chance to read through all the details, but it looks like you used the LM2940T-12 regulators that Matt is considering, correct?

[kikkegek]

Correct Ray. They are lowdrop regulators en thus very efficiënt and dont create a lot of heat.

[Matt21lutz]

Originally Posted by ray_mcavoy

You're right ... they'll definitely be dimmer than normal with 56Ω resistors on a 12V supply after you add in the diode drop. And without some additional data on the LEDs you're using, it's hard to say just how dim they'll be.

The LED's that im using are from superbrightleds.com, http://www.superbrightleds.com/morei...-mcd/356/1303/ they are EXTREMELY bright when wired correctly, ive got one small board sitting on my wall hooked up and working, i use them as a night light when i'm working late at night lol well with a resistor and witch they're a night light and then flip the switch for full brightness.

You can put some of the 56Ω resistors you already have in parallel to obtain lower resistances. To calculate the total resistance, add the reciprocals and take the reciprocal of the sum. Rtotal = 1 / ((1/R1) + (1/R2) + ... + (1/Rn)). So 3 of those 56Ω resistors in parallel = 18.7Ω ... and that's pretty close to the 15Ω I calculated earlier. Using 3 parallel resistors for each LED string obviously isn't something you'd want to do for your final design (it'd be a waste of resistors & board space). But I thought I'd mention it because it's something you could do for preliminary testing on a string or two of LEDs without having to buy more resistors.

as far as wiring in parellel... yeah that'd be too much for me, i could do it but I'll just end up buying the right resistors online for 10 cents each or something like that lol

Either one will work (and there are several other ways as well). The resistor is simple and will get the job done. Unless you've already done so, you might need to try a few different resistance values to get the brightness of your tail lights to the desired level.

If you use a lower output voltage regulator, you might want to consider an adjustable regulator instead of a fixed 10V output. That'd allow you to easily change the output voltage in case 10V ends up making the LEDs too bright or too dim. The LM2941T is an adjustable version of the LM2940T-12 fixed 12V regulator you're considering. The output voltage is adjustable using the ratio of a couple of resistors. And for testing purposes you could use a potentiometer.

I think I might try the adjustable one, I could always set it to 12v and use a resistor if that is what i go with.

[kikkegek]

Matt

couple of thoughts here:

1. LEDS:
those leds you mention are the LEDs that are run on 20mA (DC foraward current) and thus NOT the brightest on the planet.

These leds http://www.philipslumileds.com/uploads/3/DS05-pdf

are Philips Superflux and have a forward current of 70mA. that makes a big difference.

2. COLOR:
your leds have a wavelength of 640nm...that is RED...behind your red lens of the taillight a lot of the output will be lost beause of that.

the leds I used (and that are used by allmost all on hidplanet.com) are of the orange-red kind that have a wavelength of 626nm.....difference seems little, but in output makes a hell of a difference:


I have bought 6 different LEDS to test these things and found out myself. And the guys from the hidplanet.com forum had told me in their first comment to go with the Lumileds.

resistor calculation:
keep in mind that you are not feeding the LEDS 12V, but 11.3V because there should be a diode inline after the voltage regulator.

as far as voltage regulator. why use an adjustable? to control brightness?
brightness isnt only partially controllable by voltage...under a certain voltage they just turn off...that is why some guys build the PWM circuits for dimming teh arrays. I just did some testing behind the lens with several resistors and found the correct size for my application.

keep in mind to allways check if the wattage of the resistor is big enough to dissipate the heat it will build up.

just some thoughts....but go for the Lumileds Orange-Red...you wont regret it...they are better quality, last longer and blind the **** out of the cars behind you when you hit the brakes...hahahaha

[Matt21lutz]

Originally Posted by kikkegek

Matt

couple of thoughts here:

1. LEDS:
those leds you mention are the LEDs that are run on 20mA (DC foraward current) and thus NOT the brightest on the planet.

These leds http://www.philipslumileds.com/uploads/3/DS05-pdf

are Philips Superflux and have a forward current of 70mA. that makes a big difference.

2. COLOR:
your leds have a wavelength of 640nm...that is RED...behind your red lens of the taillight a lot of the output will be lost beause of that.

the leds I used (and that are used by allmost all on hidplanet.com) are of the orange-red kind that have a wavelength of 626nm.....difference seems little, but in output makes a hell of a difference:


I have bought 6 different LEDS to test these things and found out myself. And the guys from the hidplanet.com forum had told me in their first comment to go with the Lumileds.

just some thoughts....but go for the Lumileds Orange-Red...you wont regret it...they are better quality, last longer and blind the **** out of the cars behind you when you hit the brakes...hahahaha

Just looked at the pdf in the link, the continuous forward current isn't listed, that 70ma rating is the peak forward current, i say this becuase yes it does say DC forward current but above that section it says "absiolute maximum ratings", so really that is the peak forward current. Which is actually a lower peak forward current then the ones that i have, mine run on a MINIMUM of 20, but you can run them at 20-75 or even greater, the peak for mine are actually 100ma. They run at 10 lumens at 20ma so if i beefed it up to 50 or 70ma they'll be closer to 15 or 20 lumens, the intensity for the led's that i have is about 5500mcd at 20ma. I don't want or need to blind people when i hit the brake i just want it to be clear that i am stopping, i don't need them rear ending me.

resistor calculation:
keep in mind that you are not feeding the LEDS 12V, but 11.3V because there should be a diode inline after the voltage regulator.

as far as voltage regulator. why use an adjustable? to control brightness?
brightness isn't only partially controllable by voltage...under a certain voltage they just turn off...that is why some guys build the PWM circuits for dimming the arrays. I just did some testing behind the lens with several resistors and found the correct size for my application.

keep in mind to allways check if the wattage of the resistor is big enough to dissipate the heat it will build up.

I have tried using resistors to regulate the brightness, 1/2 watt resistors, tried a few different ones, the problem with it is that get incredibly hot even when everything is wired correctly and you can not put a heat sink onto a resistor so I like the idea of a voltage regulator. More because if i don't want it as dim i can simple adjust the voltage a lot easier than i would be able to with a resistor, otherwise i have the desolder and put in a new resistor every time.

[ray_mcavoy]

Originally Posted by ray_mcavoy

And without some additional data on the LEDs you're using, it's hard to say just how dim they'll be.

Originally Posted by Matt21lutz

The LED's that im using are from superbrightleds.com, http://www.superbrightleds.com/morei...-mcd/356/1303/ they are EXTREMELY bright when wired correctly, ive got one small board sitting on my wall hooked up and working, i use them as a night light when i'm working late at night lol well with a resistor and witch they're a night light and then flip the switch for full brightness.

Okay, based on the Fig1. Forward Current vs. Forward Voltage curve shown in that link, it looks like you'll be operating them at roughly 2.1V, 15mA with 56Ω resistors once you add in the diode drop bringing the voltage down to 11.3V.

Assuming your test board is using the same configuration (strings of 5 LEDs in series w/56Ω resistors), it should be easy enough to power it through a diode to see how much it affects the brightness.

Originally Posted by kikkegek

as far as voltage regulator. why use an adjustable? to control brightness?
brightness isnt only partially controllable by voltage...under a certain voltage they just turn off...that is why some guys build the PWM circuits for dimming teh arrays. I just did some testing behind the lens with several resistors and found the correct size for my application.

True, lowering the voltage too much will turn off the LEDs. And PWM is a more efficient way to dim the arrays for the tail lights. But I get the impression that Matt isn't looking to build a circuit quite that involved yet.

Adding a series resistor is also dimming the LEDs by reducing their input voltage. So in that respect, it's the same as lowering the voltage via the regulator.

Matt was considering a fixed 10V reg and I suggested the adjustable one because a fixed 10V might be too high or too low. The adjustable reg can be easily set to whatever voltage ends up being ideal for his application. Since he needs a regulator anyway, might as well make it adjustable and eliminate the need for a resistor to dim the tail lights. The adjustable regulators are a little more expensive though.

[kikkegek]

Matt...lumen outputs are not that simple. look into this factsheet of Philips

http://www.philipslumileds.com/uploads/207/AB20-1-PDF

the output of an LED is determined by its application and one LED is not gonna output 10 lumens

also standard specifications for LEDS use the "DC Forward current" for comparison. Your LEDs will run fine on 15mA...but will get screamingly hot if run for a couple of minutes on 50mA. believe me I have tried. where these Lumileds (datasheet calls them Superflux) run only hand-warm at 50mA.

the peak value is only there to tell you what it can very briefly during "peak" its able to handle and it of no use for running lights or turn signal.

I looked at the color again and I now see that yours are 630nm..the ones I used are 624. so pretty close. that would make these leds more orange-red then red. But that is what you want for tail lights with a red lens.

I dont understand why your resistors would get that hot. I even used SMD resistors for my arrays to control the needed current and I believe I used one 1/4Watt resistors for dimming the LEDs and they do not even get hand warm. using the right wattage will make sure they are able to dissipate the load they need to handle.

All my thoughts and all the answers are in my build thread, but feel free to ask if you need.

for turn signals I just calculated the needed resistor with a online ledcalc and for the regular running lights I just did a test behind a lens and just swapped in resistors without soldering, holding them in series without soldering them in.

[Matt21lutz]

Originally Posted by kikkegek

Matt...lumen outputs are not that simple. look into this factsheet of Philips

http://www.philipslumileds.com/uploads/207/AB20-1-PDF

the output of an LED is determined by its application and one LED is not gonna output 10 lumens

also standard specifications for LEDS use the "DC Forward current" for comparison. Your LEDs will run fine on 15mA...but will get screamingly hot if run for a couple of minutes on 50mA. believe me I have tried. where these Lumileds (datasheet calls them Superflux) run only hand-warm at 50mA.

the peak value is only there to tell you what it can very briefly during "peak" its able to handle and it of no use for running lights or turn signal.

I looked at the color again and I now see that yours are 630nm..the ones I used are 624. so pretty close. that would make these leds more orange-red then red. But that is what you want for tail lights with a red lens.

I dont understand why your resistors would get that hot. I even used SMD resistors for my arrays to control the needed current and I believe I used one 1/4Watt resistors for dimming the LEDs and they do not even get hand warm. using the right wattage will make sure they are able to dissipate the load they need to handle.

All my thoughts and all the answers are in my build thread, but feel free to ask if you need.

for turn signals I just calculated the needed resistor with a online ledcalc and for the regular running lights I just did a test behind a lens and just swapped in resistors without soldering, holding them in series without soldering them in.

There is not set way to determine the difference between the two LED's because they used different way to measure the intensity. though a single led CAN produce 10 lumens IF the viewing angle is narrow, the way the led's I have now were tested at a smaller viewing angle. I actually have ran the led's at 50ma at 2.3v before, for 8 hours straight to test them, they actually didn't heat up at all still cool to the touch, mine never even get warm at all, not sure why yours are.

One problem with the datasheets that the Lumileds have is that they are only showing the PEAK and MAX rating not the average, granted the charts show you but there is not defined average or what they should be ran at, they do not say it within the text or tables. I've been all through all the datasheets before, researched and tested before, they were simply not as bright as the ones that I have now, i tested them both behind the lenses already, the Lumileds are not as bright. even if I am mistaken, i don't care, i do not want or need them to be any brighter than they are. Once again I want people to see me stopping NOT TO blind them.

Also it's fine to run them at higher current for turn signals/brakes, not for running lights though because it is only momentary and not staying at that current for more then a few seconds at a time and when u brake, none the less I'm running them at about 30ma right now. So that's fine.

As far as the way I want to reduce the power to the leds for running lights, I like the idea of using a voltage regulator because it's one less thing to worry about. otherwise i would still need to run a voltage regulate plus the resistor, resistor tend to burn out after excessive use anyways. I need to test it still, but I'm only dropping the voltage low enough that they will dim not low enough that they turn off.