I am trying to adopt led lightning into the car interior. In stock it had incadenscent bulbs with a delay dimmer. Now i am trying to use a Nichia 219 LED with this driver Fasttech 1W - everything is fine except for the delay dimmer - it’s not working. Light is powered off instantly. Is there a way to modify the driver so it would work with the dimming function of the car?
I’m not sure, but if you use the right resistor instead of the driver the led should dim(faster of an incandescent light) driving it with voltage instead of current…
but you will get the cons of the resistors
Imprecise regulation, the battery voltage is variable especially if the car’s alternator is on, the volt can go up to 15 so lower life of the led
If the voltage drop on the led is not close to the source you will get insane low efficiency and so heat, if the led must be driven at 3.2v using 15v car’s source you will lose almost 10w in heat thorough the resistor, so is convenient just using 3 or better 4 led in series, with 4 in series you will have a voltage drop on leds of 4x3.2=12.8v and you will waste only 2 watt trough the resistor to drive 4 leds
here to calculate the resistor
another solution could be a driver with capacitors that have an auto dimming function… no idea how/where to get it
or wait some1 more skilled of me in that stuffs 
Hi,
It seems like that’d be the wrong type of driver for you if you want to have the dimming work?
It’s a “12V” and “constant current” driver, so it’s designed to output a constant current to the emitter, until the point that the input voltage drops below some level, which is probably why you’re seeing the behavior you’re seeing.
ohaya, please advice on the driver which will be able to drop the current on the led if current is dropped on its input.
Hi,
I’m assuming that the car circuitry had already provided the dimming functionality with the original incan bulb? If that’s the case, maybe use several LEDs in series with a resistor, as was suggested earlier, or find an already-made module with the same connector style?
Like this:
I’ve tested that driver a bit but FT somehow didn’t put up the review as yet. Off the top of my head, output decreases from 650mA with an input voltage of 19V to an output current of 610mA with an input of 7.35V… It had an output of 40mA with a single LiIon. In other words, the output of this driver is WELL regulated and when the voltage falls to too low a value, I’d guess just under 6V, output is drastically reduced: which is exactly what you observed.
On the back of the board I received, there was one resistor and a blank spot. Shorting the resistor with a pair of pliars (I guess it has some resistance) resulted in more than twice the output but, the light started blinking… Another time, I shorted the blank pads next to the resistor and fried the IC thingy… I’m pretty sure that a pot of some sort across either the blank pads or across the resistor would function as a dimmer but, I’m not good at electronics to can’t tell you for sure.
The advice ohaya gave above sounds good.
Well as the output from this driver was at 450mA i have already replaced the resistor - it had 30ohms - replaced it with 47ohms and got 200mA output which is really fine for this application. So only the dimming is giving a headache. Using a resistor will as i guess result in a really heavy heat wattage. So driver modification of some sort is preferable. On another forum someone has suggested to me placiung an extra capacitor and resistor on the output of the resistor so it will dimm when switched on and off. What do you think of this?
If the driver IC is dimmable, Pin 3 of the IC would be the DIM input pin. Try to apply a positive voltage between 0.5V to 2.5V or a PWM signal to pin 3 to test whether the IC is dimmable. This kind of buck driver for MR16 usually operates from 5.5-6V up and depends on how much LEDs are connected in serial.
The simple fact is that LED's don't dim the same way as Incan's. It's very likely that you won't be able to make the dimming function work without a fair amount of engineering effort.
I know it's not the answer you were hoping for, but sadly, it's the truth.
PPtk
You would need a MASSIVE capacitor for any human perceivable 'dim' of the emitter to occur.
Well i do believe that there must be a solution ) And really hope this forum will help me out ) What about getting another driver with PWM control and supplying voltage from the mains through the resistor to it for control? )
As for the capacitor - suggestion was using a 10V capacitor around 2200microfarad and a resistor of 5W and 100ohms
Adding a capacitor to the input will not allow it to dim as you want.
This is a switch mode power supply, it operates within a set parameter, IE 8V - 16V DC in (or whatever didn’t check the specs) will always = a set current output. This is the whole reason to use a constant current power supply.
The dimming in your car quite possibly uses pulse width modulation where it is chopping the power on and off, which quite possibly will harm your driver as well.
I suggest you abandon that driver and use a linear current limited driver, this will allow for the function you want. One possibility is a simply LM317 circuit designed as cc driver, or perhaps look at IC’s from Supertex.
ok than. will study the option of using LM317 or analogues and order the components. While awaiting for them and digging around the google i’ve found out that most probably driver is equipped with the good old PowTech PT4115.
So there is a dimmer pin available. So i was wondering if the car doesn’t utilise PWM method if it’s possible to supply the mains to the dim pin through the voltage divider (described here Divider Calculator - dropping the voltage to 2,5 and therefore controlling brightness?
I thought pow tech discontinued that ic, but not sure.
Either way, it would be possible, what isn’t, but would require some engineering, probably more than its worth.
First you would probably want to examine the waveform from your cars dimmer with an oscilloscope to ensure it even is PWM.
Then you would still want to apply a constant voltage to VDD of the IC. From there as you say the PWM signal voltage could be attenuated with a resistor divider circuit or a simple zener diode.
For linear drivers have a look at CL6 or CL7 supertex, or AMC7150 for instance. There are a few other single component or very low parts count cc linear drivers out there, but just going from memory.
3 series white LEDs would raise your forward voltage and increase efficacy in your circuit.
Ok now some dumb questions from my side ) I have inserted multimeter leads into the bulb socket of the car. During the dimming i saw the numbers gradually decrease from 12 to 5 Volts. Does that mean it’s now PWM driven? Or i still need oscilloscope and look at the waveform?
2nd was a discovery that this driver utilises CL6807 & not the 4115. I still have 4115 ones as well and will experiment on them as well. But looking at the data sheet of this chinese 6807 chip gave me this idea
So making a dimmable on|off feature might be easier than i thought. What condensators would you recommend for me to start with? I want dimming to last for 3-5 seconds.
I would also like to experiment with the dim pin. Do i have to cut it from the board first?
It could be PWM or it could be voltage control, a simple DMM isn’t fast enough to tell you.
You will still need a steady voltage for Vin. Max input on the adj pin of that IC is 6V, you will need to regulate the voltage going to it as well.
I don’t know currently how the dim pin is connected, but I am assuming it is either floating, not connected to anything, or tied to Vin. Check voltage at that pin to see, and then use ohm meter to try and trace where it goes. Obviously if it is tied to Vin you will need to cut that trace.
I don’t know what you mean by condensator, are you asking about capacitors for the soft start feature? Do you now also want to enable soft start?
CL6807 and PT4115 have the same electrical structure. The only difference is the I max for CL6807 is 1A which PT4115 is 1.2A.