LED ballast resistor - sizing

superheat6k

superheat6k

Well-Known Member
Joined
10 Jan 2012
Messages
6,803
Location
South Coast
Visit site
I am re-making a VP Engine warning block, where the lamps are pretty rotten. So I intend to replace these with LEDs. Maplins have got some super bright white LEDs that look suitable, but are only rated at max 3.6v, with a current draw of 100ma.

Maplins expert offered 150 ohms 2W as the ideal choice, but is he correct ?

LED power = 3.6 x 0.1 = 0.36 W

LED resistance V/I = 3.6 / 0.1 = 36 ohms

Assuming peak voltage on the system is 14.6 V, I need to drop the voltage by11 V, passing 0.1 amps.

This is 1.1W,

Resistance V/I = 11 / 0.1 = 110 ohms

Resistor value 110 ohms 1W (at max voltage). Is 150 ohms 2W the wrong choice here or erring on the side of caution ?

These LEDs will mostly function without the alternator running so need to be bright enough to be seen on the display. I am concerned they could become too dim.

NB I will be fitting another resistor in series with the Alt LED to provide alternator excitation current.
 
Hallo.
By 150 ohm the current will drop to 70 milliamps. The Led will be only a little bit
darker. Perhaps you'll try 120 ohm. But there is only a bit more lightness.
In the alternator circuit I think you will do better using the 2 watt edison lamp.
By Led you have to use a serial resistor of 120 - 150 ohm and and parallel
another 120 - 150 ohm to get about 200 milliamp to the alternator.
Gruss
Hermann
 
I understand it is more about current draw than voltage, so hopefully someone more knowledgeable will be along soon.
 
How do the warning lights work? If it's the earth side which is switched, as it often is, why not supply the positive side of the LEDs with a stabilised 3.6V supply from a DC-DC converter?
 
superheat6k said:
I am re-making a VP Engine warning block, where the lamps are pretty rotten. So I intend to replace these with LEDs. Maplins have got some super bright white LEDs that look suitable, but are only rated at max 3.6v, with a current draw of 100ma.

Maplins expert offered 150 ohms 2W as the ideal choice, but is he correct ?

LED power = 3.6 x 0.1 = 0.36 W

LED resistance V/I = 3.6 / 0.1 = 36 ohms

Assuming peak voltage on the system is 14.6 V, I need to drop the voltage by11 V, passing 0.1 amps.

This is 1.1W,

Resistance V/I = 11 / 0.1 = 110 ohms

Resistor value 110 ohms 1W (at max voltage). Is 150 ohms 2W the wrong choice here or erring on the side of caution ?

These LEDs will mostly function without the alternator running so need to be bright enough to be seen on the display. I am concerned they could become too dim.

NB I will be fitting another resistor in series with the Alt LED to provide alternator excitation current.
Click to expand...
Correct but your arithmetic seems to take a very round about route.

You need to drop 14.6 - 3.6 = 11 volts

the resistance which will drop that at 0.1amp is 11/0.1 = 110 ohms

The power dissipated in the resistance is 11 x 0.1 = 1.1 watts

So 120 or 150 ohms at 2 watts should be fine.
 
JumbleDuck said:
How do the warning lights work? If it's the earth side which is switched, as it often is, why not supply the positive side of the LEDs with a stabilised 3.6V supply from a DC-DC converter?
Click to expand...

This is not really feasible as the already complicated multiple diode circuit required for the associated buzzer and test circuit would preclude this - see recent thread on rebuilding these modules.

I have re-designed the circuit to allow the test button to light the lamps as well as the buzzer, but found doing this and including the Alt light in the test was just too complex, so the Alt light is excluded from the test circuit.

I do have one usable bulb mount so may re-use this for the Alt function.

Thanks for those who have responded.

My re-designed Module circuit is here, with thanks to earlier posters who helped me understand it.

https://www.dropbox.com/s/u86u3aivn4mqce3/VP Alarm Panel - Revised 2014.pdf?dl=0
 
superheat6k said:
This is not really feasible as the already complicated multiple diode circuit required for the associated buzzer and test circuit would preclude this - see recent thread on rebuilding these modules.

I have re-designed the circuit to allow the test button to light the lamps as well as the buzzer, but found doing this and including the Alt light in the test was just too complex, so the Alt light is excluded from the test circuit.

I do have one usable bulb mount so may re-use this for the Alt function.

Thanks for those who have responded.

My re-designed Module circuit is here, with thanks to earlier posters who helped me understand it.

https://www.dropbox.com/s/u86u3aivn4mqce3/VP Alarm Panel - Revised 2014.pdf?dl=0
Click to expand...

What is the problem with the other bulb mounts? Corroded spring? If so I have a fix for you.
 
Comrade Red said:
What is the problem with the other bulb mounts? Corroded spring? If so I have a fix for you.
Click to expand...

The original lamp mounts were simply disintegrating. I managed to recover the plastic bases, and some of the surround with the metal shroud. I removed the spring inner terminal, drilled out the base 3.5mm, filled inside the shroud with hot melt glue and shoved the LED wires through the glue to sit prominent the same height as the normal bulb.

I then remounted the individual lamps, connected the +VE feed to all four, with four separate wires for the -VE to the individual alarms within the new Vero CB. The lamps once mounted and wired was then completelly immersed in hot melt glue

Nigel thanks for the selection guide. I used the 150 ohm 2W and the light was really bright at 13.8 volts from my bench power supply I mounted these within the separate module on the Vero CB.

Having now re-made 5 of these modules the latest was the neatest, although the lamp cluster from the original which was in poor condition, has been rebuilt for function rather than aesthetics, but it is out of site.

If anyone wants one re-making please PM me. Not free but I will try to keep the cost sensible.
 
nigelmercier said:
Presumably you keep the complexity of the Zener and transistor as you still use the alternator lamp? Otherwise the whole thing could be reduced to just the diodes and four LEDs with resistors.
Click to expand...

The Alternator circuit is effectively separate. The transistor / pot divider resistors and Zener / Diode pair simply cause a voltage drop to the buzzer. It would have been simpler if VP had used a buzzer suited to 12 - 14 volts.

Final version with LEDs here ...

https://www.dropbox.com/s/u86u3aivn4mqce3/VP Alarm Panel - Revised 2014.pdf?dl=0

This has the otherwise redundant Glow lamp used for exhaust alarm, and the test button sounds the buzzer plus lights Temp, Oil and Glow lamps. Oil & Alt come on with engine pre-start, then go off as normal.
 
superheat6k said:
It would have been simpler if VP had used a buzzer suited to 12 - 14 volts...
Click to expand...
It is certainly OK for 12V, I'll have a look for the thread where I found the specification...

Found it, it is fine for 12-24V:

nigelmercier said:
I've just checked, the horn is a SC 235 BF made by Sonitron: http://view.ceros.com/sonitron/catalogue/p/12

It works up to 35V, so I see no reason for the zener diode, except to reduce the volume. Probably H&S reasons :(
Click to expand...
 
Last edited:
JumbleDuck said:
How do the warning lights work? If it's the earth side which is switched, as it often is, why not supply the positive side of the LEDs with a stabilised 3.6V supply from a DC-DC converter?
Click to expand...

No not good. The nature of an LED is that it draws no current until the correct voltage is reached then high current. If this voltage is supplied from a source capable of supplying high current the LED will take as much current as it can (call it infinitely high current but not really) This results in destruction of the LED. A series resistor reduces the voltage to the LED as it draws more current so that it settles to a correct current on the knee of its current voltage curve. So it would be OK to run it off a 5v stabilised supply but would still need a resistor for each LED. It may be confusing because many up market LEDs run on a "stabilised power supply' but perhaps better called a stabilised current control. This is not usually like a DC buck regulated power supply like a 7805 but provides current limiting to often high current pulses to the LED. (which averaged over time equal normal steady current. olewill
 
Last edited:
1) you don't need hi power LEDs for warning lights, something drawing about 30 to 50 mA will probably be fine even in sunlight.
2) the alternator light is better done by a bulb. You need a very cunning and complex circuit to do the job half as well!
 
I'm always amazed at the level of misunderstanding about LEDs. Seems many try to apply the limited knowledge they possess about DC electrical theory and electrical conductors, to semiconductors.

Diodes and other semiconductors exhibit non-linear resistive characteristics, so they don't obey Ohm's Law. In other words, V=IR doesn't apply.
 
+ 1 to misunderstanding of LEDs and diodes.
However I wold disagree about ohms law I think he stands correct no matter what.
With a resistor if you double the voltage fed to it you double the current so 4 times power. This is a liear progression of current to voltage.
Not so in a diode. With connection in the forward direction a silicon diode will not allow any current until it reaches the breakdown voltage about .7 volt wherupon it will conduct showing very little resistance. In the opposite direction it will not conduct at all until breakdown voltage (usually hundreds of volts) when it conducts then destructs.
A zener diode is same as siliocon diode in forward direction but in reverse voltage when the designed breakdown voltage is reached it starts to pass current in an avalanche but provided the current is not too much it will not be destroyed but will stop conducting if voltage is reduced . This voltage can be designed into the diode from 3.3volts to 75 volts typically and is used as a voltage reference.
Light emiting diodes are more like a silicon diode in that the voltage required for them to start carrying current is in the order of 3.6 volts. At and above that voltage there is no limit to the current it will conduct up until destruction. It will aslo destruct in the reverse direction with excessive voltage often as low as 12v.
So at any given current voltage situation ohms l;aw does apply. However the apparent resistance varies with the voltage applied. Very non linear. olewill
 
You must log in or register to reply here.

Share this page

Top