Another thing, how can i find out the power dissipation for the MOSFET?

I know that for BJT, you use P=VI and V="the voltage across the collector and emitter".
so do i just do the same for mosfet? or it can be neglect since the voltage drop across the drain and source is small?

Thanks !!

 

Another thing, how can i find out the power dissipation for the MOSFET?

I know that for BJT, you use P=VI and V="the voltage across the collector and emitter".
so do i just do the same for mosfet? or it can be neglect since the voltage drop across the drain and source is small?

The same way; voltage x current, where voltage is Vds, and current is the sum of all of your series LED strings. Since it's a power MOSFET and LED strings present such a light load, your MOSFETs power dissipation should be quite low - as long as you keep the ON/OFF switching time low.

 

Great! I looks like i am going put three LED in series!

for the LTspice, how do i simulate LEDs? I right click on part and it poped up a diode properties window, it shows the forward current but it wont allow me to edit anything.

You'll have to either select a model from the library, or create your own model.

You might get by using a 3.3v Zener diode instead of LEDs, if you just want a quick solution. LEDs have a similar response curve to Zener diodes, although much more limited.

 

oh cool. I am going to make my own model then, i am in a processing of learning

And to control the brightness of the LEDs, all i need to do is to change the duty cycle of the PWM, correct?

 

oh cool. I am going to make my own model then, i am in a processing of learning

Making your own model is a bit advanced. Better to copy a model that has similar characteristics, and make small changes.

And to control the brightness of the LEDs, all i need to do is to change the duty cycle of the PWM, correct?

Early on, I suggested calculating Rlimit for 100% duty cycle; the maximum brightness that your LEDs can produce with a decent lifespan.

Reducing the ratio of ON time to OFF time won't change the peak current, but will reduce the average current. As you reduce the ON/OFF ratio, the LEDs will appear to become more dim.

 

Early on, I suggested calculating Rlimit for 100% duty cycle; the maximum brightness that your LEDs can produce with a decent lifespan.

So for this kind of LEDs (see link), the forward current is 20mA, but it doesnt say what is the max current it can draw. so should I use 20mA for my calculation?

Link: http://cgi.ebay.com/Lot-of-100-X-5m...ryZ66954QQssPageNameZWDVWQQrdZ1QQcmdZViewItem

 

So for this kind of LEDs (see link), the forward current is 20mA, but it doesnt say what is the max current it can draw. so should I use 20mA for my calculation?

Link: (omitted)

Those are cheap LEDs. You can give them more current if you want, but their lifespan will be greatly reduced.

As I read them, the specs say 20mA @3.2v typical, with a max Vf of 3.6v.

You interpret the specs as you wish. If you significantly exceed them, you'll wind up with a pile of dead LEDs.

 

Those are cheap LEDs. You can give them more current if you want, but their lifespan will be greatly reduced.

As I read them, the specs say 20mA @3.2v typical, with a max Vf of 3.6v.

You interpret the specs as you wish. If you significantly exceed them, you'll wind up with a pile of dead LEDs.

Ic. I am sure something will blow up or smoke for my first prototype . so i guess it's reasonable to use cheap LEDs.

By the ways, how much does the decent LEDs cost? and why are they better?

 

In that schematic, you had the transistor basically shorting out the supply instead of gating the current through the LEDs!

If you had instead used the NPN transistor to control the LED's ground path, your circuit would have been much better.

What is the meaning of "control the LED's ground path"? I have tried it, LEDs are not able to light up at all. I will post my connection. Let's see what is wrong!

 

Here's SonomaDog's original schematic:

As you can see, the transistor was connected to short the LED's supply to ground. This wastes a lot of power.

Here's an edit on the same schematic:

Now the transistor controls the ground path for the LEDs.