The AL8871Q is an LED driver, controller IC for driving external MOSFETs to drive high-current LEDs. It's a buck-boost topology

 

Now I have seen the arrangement it looks like all will be lighted at one time. The first question is what is the basic source voltage? is it 12 or 48, as many EVs tend toward higher battery pack voltages. Does the light as shown need to be intensity controlled? As for DRLs, possibly??

13.5V source nominal from the harness. PWM bat pulse for DRL

 

I can disable the PWM but would need 2 more A and K outputs

 

The AL8871Q is an LED driver, controller IC for driving external MOSFETs to drive high-current LEDs. It's a buck-boost topology

OK, so that could be a driver for series strings in parallel, using a small value series resistor for each string. So now the challenge of finding the most efficient string length and voltage That might be two strings of 36 LEDs or 4 strings of 18 in series. That will require examining the application notes, because they often describe the sweet spot for top efficiency.

 

OK, so that could be a driver for series strings in parallel, using a small value series resistor for each string. So now the challenge of finding the most efficient string length and voltage That might be two strings of 36 LEDs or 4 strings of 18 in series. That will require examining the application notes, because they often describe the sweet spot for top efficiency.

Wouldn't that arraignment be to much of voltage drop for the 48V rail? Is this what you are referring too? each led would drop 3V

 

I was considering that the IC may have a higher than 48 volts voltage option. But possibly not

 

I guess my final concern is that the Amber/white dual die LEDs typically take 250mA to drive at a bright enough bin. I need to arrange the strings so I do not draw over 1.2 amps. Was thinking LEDs in series current limit resistor but can the strings be parallel to each other?

 

What I mean is that unless I am over looking something here, all LEDs will be powered of a boost converter power rail and the LEDs will be in series for equal current of 250mA per LED. Since I have to take into account 3vf and a rail of 48V then I would 12 LEDs with a series resistor. Then the next string would be on the same rail with LEDs placed in series but the string would be in parallel of position of the first string.

 

72 / 12 = 6 string
2 x 0.25 = 1.5A total current
1.5. x (48 - 36) x 6 = 108W dissipated

Are you planning on a cooling fan for the resistors?

 

Multiple strings of LEDs plus a resistorconnected across a power source are independent of each other. So adding LEDs until the forward drop total does not leave room for another, and then adding a series resistor to each string to set the required current from the supply current is the simple way. A switching LED driver that can deliver a higher voltage at a controlled current so that no resistor is required is more efficient. 18 in series with a 66 volt supply could work, about 2 volts margin, so might not be max light out but rather close. So it is all a trade-off, electronics or a lot of resistors, both can work.

 

72 / 12 = 6 string
2 x 0.25 = 1.5A total current
1.5. x (48 - 36) x 6 = 108W dissipated

Are you planning on a cooling fan for the resistors?

So in a boost led driver configuration how should I string these LEDs for max efficiency?

 

Each LED string is connected in series, meaning the positive terminal of one LED string is connected to the negative terminal of the next LED string, and so on taking into account 48v power rails

All the series-connected LED strings are then connected in parallel to the power rail of the boost converter the LED strings are in series, the same current will flow through each string, as there is only one path for the current to flow.

Therefore, the current draw will be the same for each LED string in this configuration. The boost converter can handle over an amp at 1.4 amps over the 1.2 amp allowance

 

Multiple strings of LEDs plus a resistorconnected across a power source are independent of each other. So adding LEDs until the forward drop total does not leave room for another, and then adding a series resistor to each string to set the required current from the supply current is the simple way. A switching LED driver that can deliver a higher voltage at a controlled current so that no resistor is required is more efficient. 18 in series with a 66 volt supply could work, about 2 volts margin, so might not be max light out but rather close. So it is all a trade-off, electronics or a lot of resistors, both can work.

So I was planning to have a series resistor with ccr control for each series string placed in parallel with each other over the strip

 

Therefore, the current draw will be the same for each LED string in this configuration.

No, each string is independent of the others. Each LED in a single string will see the same current. By current in srting 1 is not the same as the current in string 2. Forward voltage varies from one LED to another of the same type. Each string will have a slightly different forward voltage. Also, some LEDs will get hotter than others, which changes their forward voltage.

For good control of current, you will need a constant current driver for each string. That can be as simple as an LM317 plus a resistor. But again, without a switching regulator, you will produce a lot of extra heat. The LEDs themselves are going to be thermal management problem.

 

You can also go to a pulse width variation to hold the power dissipation down, because within reason, HEAT IS THE ENEMY. So with a bit of excess voltage and using PWM control to keep the dissipation within limits there will not be any hot resistors to deal with or to reduce efficiency. OR even go with a PWM LED driver that can also adjust the voltage.
Post #35 mentions the LED heat and that certainly does need to be considered, because with the higher current and te greater forward voltage there is power dissipated in each LED that shows up as heat, leading to a temperature rise. That can be a problem with the string of surface mount LEDs generating heat and not much room for it to spread out into. So that needs to be solved as well. So one more area for somebody to gain valuable experience in.

 

Taking thermal out of the equation for a minute, using heatsink and thermal tape or other strip technology. I still have to balance current so I was thinking about something like this. 160 mA forward current LEDs

 

Your new schematic has 12 LEDs in series and each LED is about 3V. Then a string of 12 LEDs needs 36V.
Your 82 ohms resistor has 48V - 36V= 12V across it then the current in the resistor and in the string of 12 LEDs is 12V/82 ohms= 146.3mA.
Then the heating power in the resistor is 12V x 146.3mA= 1.76W. A 1/2W resistor might catch on fire and even a huge 2W resistor will be very hot.

If you want a little more more light, hotter LEDs and a hotter resistor then the resistor is calculated as 12V/160mA= 75 ohms and heats with 12V x 160mA= 1.92W. Use an enormous 5W resistor.

 

Did the TS MEASURE the forward drop of the LEDs at 160 mA? My guess is that it is greater than 3.0 volts.
And to handle the resistor power drop another option is to spread the heat around with ten 8.2 ohm resistors Half watt devices will then be good for 5 watts, splitting the heat ten ways.

 

Put 15 in series and and use an LM317 as a current regulator. It will drop 3V, which is enough for it to regulate and only dissipate 1/2W each, which does not require a heat sink.

 

Put 15 in series and and use an LM317 as a current regulator. It will drop 3V, which is enough for it to regulate and only dissipate 1/2W each, which does not require a heat sink.

15 in series with a 48volt supply and a regulator will only work if they are 3 volt LEDs. Have you ever seen a white 3.0 volt LED?