# pwm for more brightness white led

Discussion in 'The Projects Forum' started by minh, May 3, 2014.

1. ### minh Thread Starter Member

Nov 24, 2010
24
1
Dear all.
PWM : voltage - frequency - duty cycle to used for make maximum but safe more brightness of the white led (please see photo thank you).
minh

File size:
183.3 KB
Views:
51
2. ### Little Ghostman Member

Jan 1, 2014
307
102
I am not really sure what your asking? The Led needs a minimum of 3.V-3.4V to light, so I would run it at 4V or above, the max current is 20mA, so take your input voltage (the Voltage your using) and minus the Forward Voltage, for example: say your using 5V supply take away the minimum voltage to light the Led (3.2V)
5-3.2= 1.8V, to find the resistor you divide 1.8 by the maximum current you want, so if you want max brightness 20mA, in A this is 0.020A

1.8V/0.020 = 90 Ohms so use the next highest resistor that is closest to the value you want, so if your resistor family dosnt have 90 Ohm as a standard use next highest value, normally 100 Ohm, or you can use 2 resistor's of 180 Ohm connected in parallel to give you the 90 Ohm.
As for frequency, thats upto you, @5V & 90 Ohm any duty cycle less than 100% will decrease brightness of Led, but remember that it isnt linear it logarithmic because of your eyes, so 50% duty dosnt automatically mean it will look half as bright as 100%, there are tables on the net to show you the relative brightness

if your using 12V supply then its 12V - 3.2V = 8.8V therefore 8.8V/0.020A = 440 Ohm isthe lowest resistor value you should use. you may also use variable resistor to get values you need if you need absolute maximum brightness, but DO NOT exceed the max If

Last edited: May 3, 2014
minh likes this.
3. ### shteii01 AAC Fanatic!

Feb 19, 2010
3,971
616
Forward Voltage turns the diode in the led On and Off. What it does not do is control brightness.

You use current to control brightness of led.

Normal people control led in the following way:
1. Turn led On by applying, in your case, 3.2-3.4 volts across the led.
2. Control led brightness by changing current through led: 20 mA very bright, 15 mA good bright, 10 mA somewhat bright or dim.

I don't really see how PWM is useful in this application. Yes, it will let you turn led On and Off. But I don't see how it will let you control the brightness.

minh likes this.
4. ### #12 Expert

Nov 30, 2010
17,895
9,315
I can give you part of the answer.

You can drive 75 ma through the LED in short bursts, as long as you don't exceed the limit of 70 milliwatts on the average.

70 milliwatts is derived from 3.5 volts times 20 ma, continuous.
75 ma/20 ma = 3.75
That's 3.75 times the 20 ma typical use, so the time at 75 ma must be 1/3.75 of the on time. Lets multiply by 4 to get integer numbers: 4/15

75 ma for 4 milliseconds followed by zero current for 15 milliseconds will keep the LED within its limits for power and current. 19 milliseconds is 52.6 Hz, so your eye will see the light as continuous. The part I don't know is whether your eye will see the light as brighter than it would look at 20 ma continuous.

minh likes this.
5. ### DrRich New Member

Apr 21, 2014
5
9
PWM can be used to adjust the apparent brightness of a LED, due to our eyes persistence of vision i.e. our eyes inability to see rapidly changing features.

PWM changes the average power delivered to the load. Hence, with our eyes "averaging" effect, the LED appears at a brightness proportional to the duty cycle (the ratio of the on time into the period), if the PWM frequency is sufficient. If you had a high speed camera though, you could see the LED turning on and off.

The forward drop is the minimum voltage needed before the LED starts to conduct in forward bias. The LED needs to have its current set in the PWM on time either via a voltage source and a resistor or from a constant current source/sink.

minh likes this.
6. ### williamj Active Member

Sep 3, 2009
180
32
Common sense tell me I'm wrong but I can't shake the idea.

A PWM will take a full voltage and give it the appearance of a lesser voltage by altering the duty cycle of the pulses.

Would or could a PWM do the same for resistance? Could a PWM take a full resistance value and vary the apparent resistance by altering the duty cycle of the pulsed voltage going through the resistor?

7. ### Alec_t AAC Fanatic!

Sep 17, 2013
7,106
1,483
In some circumstances, yes (though actually it's current that flows). For example, consider a cap being charged via a resistor. If the resistor is switched in and out of circuit with a 50% duty cycle it will take twice as long to charge the cap. Hence the resistance appears to be doubled.

8. ### minh Thread Starter Member

Nov 24, 2010
24
1
Dear all...
First I am very sorry for my question is not clear and my English is not well.
I want to use a circuit PWM control more brightness of led But on Google I seen some site has this circuit with difference frequency from 50 hezt to over khz and 1/100 to 10/100 duty cycle. I wound like to try it,but I have not luxmeter for comparison brightness . So could you point me what (how much) frequency and duty cycle is best
ie: (10/100 duty cycle at 1khz frequency) to use for my white led (please see data on photo thank you)
Thank you again for your time
minh

Last edited by a moderator: May 4, 2014
9. ### wayneh Expert

Sep 9, 2010
13,624
4,416
The frequency you choose is not very important if you are only concerned with how the LED looks to a human. At the low end, any frequency above about 60Hz cannot be easily detected as "flashing" by the human eye. At the high end, the frequency is limited only by the circuitry and the components you use. For instance 80kHz is common and is far beyond the ability of a human eye.

LED brightness is nearly linear with current over most of its operating range (2-15ma or so for your LED). The time-averaged current, whether it is pulsed or constant, gives a similar perceived overall brightness. So a series of pulses that give 20mA for 50% of the time, will look much like continuous 10mA. There are some subtle effects of current on the emitted color, and the linearity is not perfect, but PWM works very nicely for most simple applications.

Take a look at Bill's blog on this forum, for instance Fig.10.4 gives a nice circuit for controlling frequency and duty cycle to produce PWM, and then use that for controlling LEDs.

minh likes this.
10. ### Markd77 Senior Member

Sep 7, 2009
2,803
595
Type fig 10.4 into the search box at the top right of the page

minh likes this.
11. ### shortbus AAC Fanatic!

Sep 30, 2009
4,549
2,588
Maybe I'm missing some thing in this thread, but with out cutting the life of an led, the brightness is set by the maximum allowable current limit. Isn't it? Pwm would allow it to be dimmed not made brighter??

12. ### #12 Expert

Nov 30, 2010
17,895
9,315

It's a question about using high current pulses with some, "off" time to get better brightness. Wayneh in post #9 seems to be saying you can not achieve much difference by using high current pulses.

Pwm is being discussed as a way to get more brightness instead of less because high current pulses require some, "off" time to avoid melting the LED.

minh and shortbus like this.
13. ### THE_RB AAC Fanatic!

Feb 11, 2008
5,435
1,309
Because of properties of human vision, a PWM driven LED can appear brighter and give more usable light to a human than a DC driven LED at the same average current or same average power use.

absf, minh, shortbus and 2 others like this.
14. ### Little Ghostman Member

Jan 1, 2014
307
102

So we are assuming he wants a Led that is brighter than it would appear when driven at 20mA continuous?

I play around with Leds alot and I can honestly say a led driven at 99% duty cycle looks no different to one driven at 20mA continuous.

Am I correct in assuming that what is being said, is to drive the led harder (70mA) for short burst's, in order to make it appear brighter?
If thats the case then isnt he going to have to characterize the led? Ok 3.2V isnt far from 3.4V, but if he is taking it to limits then its going to make a difference, unless of course he decides to take the lowest possible figure of 3.2V and drive it at 'Almost' the maximum peak mA.

A couple of things that would worry me.
The spec sheet dosnt say how long the 75mA can be held for.

It has a max of 75mW, so wouldnt it burn it up if driving it at 75mA even in very short burst's over say several minuets?
I am not understanding how he is going to maintain a higher fI even with PWM and stay within 75W??
I know this reads like I am disagreeing, but I am not. I am sure if you say it's do able then its do able, but I am very interested in how it's done, I make alot of Led cubes and such, and if I can find a way of making them appear brighter than normal, then it's of interest to me.
Sorry for sounding doubtful because I am not, I just wanted to word everything carefully.

Last edited: May 5, 2014
minh likes this.
15. ### Little Ghostman Member

Jan 1, 2014
307
102
Scrap the above, it has just sunk in what RB was saying , man that penny took ages to drop!

THE_RB likes this.
16. ### BobTPH Active Member

Jun 5, 2013
905
160
I don't think this is true. If it were, would not LED light bulb manufacturers use this method to make more efficient lighting?

For one thing, the light output of and LED is not linear, it is more efficient at lower currents. Running higher current pulses through the LED at the same average power will actually put out less average light. And secondly, from what I understand about the human eye, it integrates over a short period, so light at twice the intensity and 50% duty cycle would not appear any brighter.

Bob

minh likes this.
17. ### BobTPH Active Member

Jun 5, 2013
905
160
If this is true, why do we not see LED light bulbs that make use of this to provide more effective brightness at lower power?

The light output of LEDs is actually not linear, but curves downward with increasing current. I.e. the same LED at 20 mA does not put out twice as much light as it does at 10mA. So pulsing the LED at twice that current and 50% duty cycle actually puts out less light energy than running it continuously at the design current.

Bob

minh likes this.
18. ### wayneh Expert

Sep 9, 2010
13,624
4,416
I don't believe there is any route to get more brightness (as judged by a human) by using PWM instead of steady current. I'm open to learning I'm wrong on this! I do think there could be advantages for instrumentation, such as driving an IR pulse farther through tissue to measure pulse, while letting the LED cool between pulses.

The curve showing light output versus LED current bends down a bit, so it's a tiny bit more efficient to run at the lowest peak current you can, and then PWM down from that.

My reading of the OP's question was how to achieve dimming, not necessarily maximize output. But there may be a language issue.

minh likes this.
19. ### THE_RB AAC Fanatic!

Feb 11, 2008
5,435
1,309
You are correct about the LED light: power efficiency (instantaneous).

But I believe the effect is caused by the retentivity (persisitence of vision) of the light receptors in the human eye. It's been years since I read the article, sorry I can't remember where.

Many (if not most) LED lights these days use PWM current limiting, so even at "full" brightness setting they are PWMing the LEDs. At least the adjustable LED lights are this type, not the ones that use a LED bulb and a separate DC->DC converter.

(edit) This was near the top of a goolge search;
http://www.ledsmagazine.com/articles/2008/05/pulse-driven-leds-have-higher-apparent-brightness.html

Last edited: May 6, 2014
minh likes this.
20. ### wayneh Expert

Sep 9, 2010
13,624
4,416
My understanding of this is that, at the frequencies used for PWM dimming, that pulsing effect is not relevant. To get the brightening effect, you need to pulse in the range of 10Hz, i.e. strobe the LED, and you can easily detect the fact that it is pulsing, not really "dimming".

If the effect were relevant in normal PWM dimming, you would have the odd effect of an LED appearing brighter as the duty cycle is reduced from 100%, passing thru some maximum brightness at some duty cycle less than 100%, and then falling off the maximum as the duty cycle continues to drop.

I'm not aware of that being observed with PWM dimmers.

minh likes this.