Homemade LED light inserts.

Jim Hawkins

Chief Petty Officer
Joined
Mar 11, 2013
Messages
499
Nothing earth shattering new here, just my version. My nav lights on the sides of my center console would not work so keeping the fixture, I pulled the lamp base guts and inserted my homemade LED. 4 super bright white LED's from radio shack, about $5 stuck through about a 1" X 1-1/2" piece of circuit board. The LED's have a forward voltage of 3.3 or 3.6 max which means 4 in series is just right for a 12 volt system that runs 12.8v to 14.4v. I also added 1, 100 ohm resister because I kept reading that the circuit should have some resistance, I don't know why. It seemed to work fine without the resister. 1 Ohm would have been better but I didn't have 1 handy like.
 

Attachments

  • photo257093.jpg
    photo257093.jpg
    278.3 KB · Views: 0

sam am I

Commander
Joined
Jun 26, 2013
Messages
2,169
I kept reading that the circuit should have some resistance, I don't know why.

At 3.6V max Vf on ea. LED x 4 with a supply voltage (Vd) at 14.4VDC, you'd be operating on the "knee" (see right hand side of graph), If Vd raises above the knee, say by a volt for example to 15.4VDC, the LED's Vf's are still clamped at 3.6V max per LED or the series combo14.4V, then 15.4Vd -14.4Vf = 1V and 1V/0'ish ohms = POOF!

e.g., There's no other series component in the series circuit to limit the current and the current will drastically raise (see right hand side of graph). The LED now overheat and can burnout in short time. A series resistor (you have one) limits the series branch current.

7f998478-c666-4fde-baf8-0e45e3959ae0.jpg
 
Last edited:

Jim Hawkins

Chief Petty Officer
Joined
Mar 11, 2013
Messages
499
At 3.6V max Vf on ea. LED x 4 with a supply voltage (Vd) at 14.4VDC, you'd be operating on the "knee" (see right hand side of graph), If Vd raises above the knee, say by a volt for example to 15.4VDC, the LED's Vf's are still clamped at 3.6V max per LED or the series combo14.4V, then 15.4Vd -14.4Vf = 1V and 1V/0'ish ohms = POOF!

e.g., There's no other series component in the series circuit to limit the current and the current will drastically raise (see right hand side of graph). The LED now overheat and can burnout in short time. A series resistor (you have one) limits the series branch current.


Soooo, I'm good as is? Good thing I added the resister, right?

Oh and the "POOF" made me laugh, out loud actually.
 
Last edited:

sam am I

Commander
Joined
Jun 26, 2013
Messages
2,169
Soooo, I'm good as is?

.........Yeah and assuming (RS dropped a lot of their online spec's) these are spec'd @ 20mA max, 3.6Vf LEDs, you'll be fine (see green box in plot for operational limits mentioned below).

Ya might be a bit on the dimmer side at 12V (just on battery) with around 2mA but, while running at 14.4VDC, the series branch will see around 8mA (bout half power if they're 20mA devices) and if by chance your reg takes a dump/spikes, whatever and Vd jumps to say 16VDC for example, current is limited at 16mA...so no POOF!

They'll see their max current (If) of 20mA at a Vd of around 17VDC (outside of op box)

Draft6.jpg


Good thing I added the resister, right?
yes
 
Last edited:

sam am I

Commander
Joined
Jun 26, 2013
Messages
2,169
If ya want/need to brighten up the 12V'ish op area and run about 7mA, you'll have to lose the 4th LED in the series branch and use just 3 LED's in series with a 300 ohm .5 watt limit resistor....Runs then about 12mA nominally and tops around 18mA if ya end up at 16VDC.

Draft7.jpg
 

tpenfield

Moderator
Staff member
Joined
Jul 18, 2011
Messages
17,624
Nice project . . . I gotta do that with some of my lights.

You generally find a resistor in series with LED's. In my layman's terms about the resistor . . . the resistor creates a voltage drop in the circuit that may be needed when there is too much voltage for the LED's to handle.
 

gm280

Supreme Mariner
Joined
Jun 26, 2011
Messages
14,585
I wanted LEDs in every lamp I installed in the boat as well. But once you buy specific LED navigation lights they cost two and three times as much as standard incandescent bulbed lights. SOooo I bought the standard nav lights and then bought LED replacement bulbs and installed. That was I can use LEDs and if they burn out I can replace with either LEDs OR incandescent bulbs. Works for me. All other lights are bought LED versions.

JH I can understand what you are trying to do. They make SMD LEDs to install and are extremely bright compared to the typical 3mm, 4mm and 5mm (T1, T1 3/4, etc. ) standard and defused LED types. JMHO
 

ondarvr

Supreme Mariner
Joined
Apr 6, 2005
Messages
11,527
SOooo I bought the standard nav lights and then bought LED replacement bulbs and installed. That was I can use LEDs and if they burn out I can replace with either LEDs OR incandescent bulbs.

I replaced all the incandescents with LED bulbs about 3 or 4 years ago, so far they all still work fine. It was cheap and easy, probably cheaper and better options are available now.
 

gm280

Supreme Mariner
Joined
Jun 26, 2011
Messages
14,585
I replaced all the incandescents with LED bulbs about 3 or 4 years ago, so far they all still work fine. It was cheap and easy, probably cheaper and better options are available now.

ondarvr, there is no reason LED light fixtures should cost more then standard incandescent light fixtures, but they do. So to beat them at their costly game, buy incandescent fixtures and replace the bulbs with LEDs. A lot cheaper and then you have the option to use either type. JMHO
 

ondarvr

Supreme Mariner
Joined
Apr 6, 2005
Messages
11,527
Between the auto parts store and eBay I had no problem finding all the correct bulbs to change over.
 

joeanna

Petty Officer 2nd Class
Joined
Jun 5, 2016
Messages
108
a friend has been having me redo some of his old "items" to leds. they save alot of amps and not as prone to burn out from excessive vibrations, there are some good "led resistor calculator" on the web witch also include the proper schematics to wire them up correctly, very helpful for those who don't know about led circuits. and also a word of caution when playing with leds, some are very concentrated radiation patterns being emitted and can blind you in a millisecond, so be careful when experimenting with them,.
 

Jim Hawkins

Chief Petty Officer
Joined
Mar 11, 2013
Messages
499
The reason I made my own was twofold.
1. I wanted them now and I don't know where to buy an LED replacement bulb at a reasonable price.
2. I wanted to use the existing fixture but the socket inside was kaput so I would have had to buy a LED bulb AND replacement socket and wait for it.
Here's a link to the LED circuit design tool. http://led.linear1.org/led.wiz
 

ondarvr

Supreme Mariner
Joined
Apr 6, 2005
Messages
11,527
The reason I made my own was twofold.
1. I wanted them now and I don't know where to buy an LED bulb at a reasonable price.
2. I wanted to use the existing fixture but the socket inside was kaput so I would have had to buy a LED bulb AND replacement socket and wait for it.
Here's a link to the LED circuit design tool. http://led.linear1.org/led.wiz


I figured the guts were bad in yours and that's why you made it yourself, mine were all good, so it was just pop in the new LED.
 

Jim Hawkins

Chief Petty Officer
Joined
Mar 11, 2013
Messages
499
If ya want/need to brighten up the 12V'ish op area and run about 7mA, you'll have to lose the 4th LED in the series branch and use just 3 LED's in series with a 300 ohm .5 watt limit resistor....Runs then about 12mA nominally and tops around 18mA if ya end up at 16VDC.


Sam I Am, what if I keep the 4 LED's and change the resister to 1 Ohm like the design tool said? Wouldn't that get me near max brightness when running with 14.4 V?
 

sam am I

Commander
Joined
Jun 26, 2013
Messages
2,169
Yes but, be advised the LEDs tap out much faster (di/dv slope is increasing) ..........e.g., right out of the shoot if Vd = 14.6 up from just 14.4V, the LEDs are at their presumed max spec forward current of 20mA (See green op box for 12-16Vd current values), if at 16V (reg craps out), 40mA = POOF!

Personally, I'd not run'um that marginal......they just don't run too long driving them at or over the top is all.




draft3.jpg
 
Last edited:

sam am I

Commander
Joined
Jun 26, 2013
Messages
2,169
Changing it up a bit with simple suggestion (electrically).......Design in a bit of a trade off of having to dissipate a little more heat over the operating conditions and also designing within the devices (design requires/requests 4 LED to be used) limitations, suggest moving the LED's operation off the non-linear "knee" area and onto the/a more predictable/stable linear (not curved) response area of the operational curve for much higher performance and predictability outcomes....

In a nut shell, optimizing the LED's operation on the linear response area will provide a higher constant power/light output over the varying supply range in the example of 12-16 Vd and will not be so touchy (dimming or over heating with only small changes Vd, and also allowing to run up closer against the max spec's output over the given Vd range) to varying supply deviations.

2 choices (note that the LEDs operation are again off the knee area similarly to post #5)........

1) Parallel Config: Highest constant brightness over Vd range, has most heat dissipated in series limiting resistor. Requires 160 ohm 1 **watt resistor minimum...


Parallel.jpg Parallel Config.jpg




1) Series Parallel Config: Less constant brightness (little higher slope) over Vd range, has lower heat dissipated in series limiting resistor. Requires 270 ohm .5 **watt resistor minimum...

Series Parallel.jpg Series Parallel Config.jpg

**Red trace is power dissipation in series resistor at the given Vd levels.
 
Last edited:
Top