FreeTrack Forum
|
Class : Moderator Off line |
|
|
Class : Habitué Off line |
|
|
Class : Moderator Off line |
|
|
Class : Habitué Off line |
|
|
Class : Habitué Off line |
|
|
Class : Moderator Off line |
And...? We are talking about voltage not current.
As I said above, you can drive LEDs without a resistor using a constant current regulator. Another way is to drive them with pulse width modulation, requiring at least a micro-controller. But why going the expensive way, when a resistor does the same job? Why using a resistor? Well, let's look at that closer. LEDs are semiconductors that behave different than resistors. If you apply a specific voltage to a resistor, you can describe the resulting current with: I = V / R That does not work with LEDs because they don’t behave like a linear resistor. To explain look for example at the data-sheet of the SFH485. It gives you two important curves: One gives you the max. forward current for DC which is: I(f)= 100mA  The other gives you the voltage-current characteristic line.  You can rise the voltage from 0 Volt to ~1.1 Volt without resulting in noticeable current. Apply a bit more voltage and there is current and the LED lights up. We have reached the forward voltage which is needed to open the pn-gate. From now on small changes in the voltage produce large effects on the resulting forward current. As you can see with a step off 1.38V to 1,44V the current rise up from 70mA to 100mA, our max. forward current for the diode. Apply a little bit more, and you quickly shorten lifetime off your LED. The reason why the LED does not blow up with more than 1.5V of a fresh batterie is because of the internal resistances of batteries. So, yes, you can drive LEDs without resistors, but this require current limitation in an other way, as you can see the voltage regulation is not appropriated. |
|
Class : Habitué Off line |
I=V/R V=IR If R is the same in the two equations then V and I are strictly and solely dependent to each other. In fact:
That's quite logical. Given R the same, If I varies then V varies proportionally to I, and vice-versa. LEDs are non-ohmic circuit, right, yet the whole circuit, except the LEDs, is ohmic. Since we have to control the circuit in order not to blow our LEDs up (and since we can't change the way LEDs behave), ohmic law or linearity still applies. We didn't need to calculate or extrapolate anything: according to Osram datasheet, nominal voltage is 1.5 V and maximum voltage is 1.8 V for SFH485. Our circuit would be composed as follows: Source (battery) -> Resistor -> Electrical utilizer (LED) As said before, if the source voltage is equal to the LED voltage drop and the current - and, therefore, the voltage - is constant (which is the case, since a battery is a DC source), then a resistor or current regulator is unneeded. Sir, pardon me, but still you didn't explain me the reason why resistors are required in the above-mentioned conditions. |
|
Class : Moderator Off line |
You said it yourself:
Ohmic law yes, but not linearity. The current flowing in an LED is an exponential function of voltage across the LED. You can’t say that LEDs have “resistance.” Resistance is defined as the constant ratio of voltage to current in a resistive circuit element. LEDs have diode like non-linear I/V characteristics. This picture will show it better than the logarithmic diagram from Osram datasheet.  The curve may also varies among different colors, different sizes, temperature and even tolerance in manufacturing.
Depending on what your DC source can handle, either you load down the supply until its voltage drops to Vf or current becomes so high that using E=IR, the voltage drop across the low resistance of the wires drops the remaining voltage and you still have Vf across the LED. Now, batteries are not regulated power sources. I have three new batteries here where I can measure: Duracel Alkaline - 1.59V No name Alkaline - 1.48V Panasonic Oxyride - 1.6 |
|
Class : Habitué Off line |
|
|
Class : Moderator Off line |
Call me Simson, but I don't understand this argumentation.   Sorry, perhaps I missed something.  Regarding the current-voltage curve of the sfh485 1.5V let more than 100mA current flow. But anyway, this is theory. Put a sfh485 on a 1.5 Alkaline and you never will measure more than 80mA in circuit. The internal resistance of battery prevents current rising. So I agree with you, you can drive a LED on battery without a resistor. |
|
Gian92;
You need to understand that the IR LEDs may have a Vf from 1.50 to 1.79 at 100mA. This is what the specs indicate for the SFH485P. In other words these are tolerances for Vf at 100mA. The specs do not even show a minimum Vf as some manufacturers do. I can tell based on your references to wikipedia, which has a lot of mistakes and misconceptions that you have a very limited knowledge of Electronics. There is the theoretical world and the real word applications. When you design for real world applications you must take into account worst case scenarios which includes all the variable tolerances that can occur. If you are disagreeing just for the sake of argument, then these diagrams will not make any sense to you. In the attached diagram you can see the difference what happens with a regulated 1.8 VDC regulated supply capable of 10 Amps. This is basically no resistors as you seem to think will work. The second diagram shows the resistor values to get 100mA. I used 4 LEDs that had different Vf at 100mA which will occur. This should help you understand it better.  |
|
> Stats
1 user(s) connected during the last 10 minutes (0 member(s) and 1 guest(s)).