does pwm for control brightness in flashlight works until the battery dead?

[rgbbv]

Hi

when using a flashlight with linear driver,,the current at the beginning is constant until the point that the battery voltage is too low and the flashlight getting to direct drive by passing the linear drive.
so when using a different flashlight that is direct drive with pwm.does the pwm working all the time untill the battery is dead or at some point when the battery voltage decrease,the pwm chip will be passed(like the linear drive)and the led will e feed directly from the battery?(whenever it in the high mode or lowers modes)

 

[dr pepper]

Depends on the driver, the circuit will work down untill there isnt enough volatge to run the electronics to generate the pwm.

If you monitor the voltage you can adjust the pwm automatically to compensate for battery volts drop.

 

[rgbbv]

hi,dr pepper.

so do you mean,that on direct drive+pwm flashlight, when the pwm stop working the led is turned off?do you mean that the pwm is not bypassed,gives the flashlight to keep working?
because at linear drive,(for example)like the amc7135,when the voltage source drop to about the vf of the led the chip is bypassed and the led keep on working.
and how much voltage is considered not enough to run the elctronics to generate the pwm?

 

[dr pepper]

All very much down to the electronics.

You could bypass the pwm below a certain batt v.

Some commercial led drivers go down to 1v, its possible to make one start at 0.6v and operate down even lower.

 

[ronsimpson]

Old type flashlights (light bulb type) get dimmer as the battery dies. There is plenty of time to get new batteries.

The first LED flashlights we built had a really good PWM that kept the LED current constant within 1% until the batteries dies. There was no warning. Most batteries recovered in 30 seconds and you got another couple of minutes of use before the batteries died forever. Most people did not like this.

The next versions of PWMs we used were modified to allow the light to get dimmer as the battery reached end of life.

As to using the batteries to 1V or 0.6V; We found that most of the energy of a battery comes from 1.5 to 1V and there is not much energy from 1V to 0.6V. We found that taking too much energy from a battery is not good. Over the next week or month a totally drained battery will often leak. We now build LED flashlights to stop pulling power from the battery while there is still a little energy in the battery to protect the flashlight from battery leakage.

 

[rgbbv]

.

The first LED flashlights we built had a really good PWM that kept the LED current constant within 1% until the batteries dies. .

are you talking here about constant current drive?
because i am talking about only direct drive with pwm.my question came up after i wonder what will happens to the current on the lower modes while i was thinking about the following example: on a simple direct drive+pwm flashlight there are 3 modes that controlled by pwm,100% duty cycle on high mode,30%-on mid mode and 5%-on low mode.when the battery is full charge each mode will recieve the max current accordingly to the duty cycle percent,but when the battery voltage decrease the current on the high mode,obvious, will decrease,but because the duty cycle percent on the lower modes doesn't change,the current on these modes also decrease.and this is continue all the way down with the same proportion.is this example right?
it seemed to me not efficient.and that is the reason that i thought,maybe,the pwm is being bypassed automatic at some point to allow the lower modes get more current direct from the battery rather than lost power by getting the same percentage pulses through the pwm from lower battery voltage,what cause them get less current.
am i right?

 

[dr pepper]

Direct drive as you call it if I understand what you mean isnt effecient either, the resulting current would be excessive.

Rons comments are totally correct for production models.

 

[rgbbv]

Rons comments are totally correct for production models.

i didn't said that rons comments are not correct,i just mention that i am talking about a simple direct drive with pwm for controlling brightness by 3 modes.
and what i would like to know,as i ask before,if the example that i gave is true?does the current on the lower modes is stay,always,30% duty cycle and 5% duty cycle even when the battery voltage drop?
because if it does,it is mean that the current on the lower modes will be lower than when the battery fullcharge,while the battery,practical,can supply to the lower modes the same current as before by automatic turn off the pmw or changing the duty cycle.
so what exactly happens?
i hope you can answer directly to my question.
thanks in advance.

 

[ronsimpson]

LED drive is a term that means different things to different people.
LED....PWM also mean different things.

There are ICs that have a PWM running at 500khz that regulate current. AND They also have a low frequency "PWM" that control how much of the time the LED is on. Example; the frequency might be at 100hz and 25%.

If the "PWM" only turns the LED on/off for a certain duty cycle it does not regulate. This is done for white LEDs because they will turn to a different color if not driven at full current. So they are run at full current (part of the time).

 

[rgbbv]

There are ICs that have a PWM running at 500khz that regulate current. AND They also have a low frequency "PWM" that control how much of the time the LED is on. Example; the frequency might be at 100hz and 25%.

If the "PWM" only turns the LED on/off for a certain duty cycle it does not regulate. This is done for white LEDs because they will turn to a different color if not driven at full current. So they are run at full current (part of the time).

i think i meant to the one that you mention in the end.maybe the one in the following image.
here is an image of flashlight's driver that run xr-e q5 led 3 mode.it is direct driver with pwm isn't it?
is the CAB1Q work as the pwm?
if it does,what type of pwm is it, from the types that you mentioned above?
is it the one that only turn the led on/off for certain duty cycle(that is not regulated)?or it is the IC that has pwm with certain duty cycle,not regulated?
and why there are two resistor at the driver?should it be only one resistor in series with the led?

 

[ronsimpson]

I do not see a inductor. If it had a inductor then it switches fast and has the ability to regulate with out power loss.

I have hundreds (thousands) of a very similar module in the office. There are many different versions, some regulate and some do not.

I have some modules that "blink slow, blink fast, on, off" and do not regulate.

The ones that regulate with no inductor do not really monitor the LED current but blindly guess at what the LED current is.
Example:
The LED and switch combination has a 2 volt drop.
The battery is 3V. The current limit resistor has 1V at full charge.
When the battery drops to 2.5V the resistor has 0.5V. (1/2 current)
When the battery drops to 2.25V the resistor has 0.25V or 1/4 current.
I choose the resistor to give the desired current at 2.25 battery and 100% on switch.
Then at 3V battery I tell the IC to be at 25%. (4x current at 25% gives approximately the correct current)
We program the IC to give (in this case) 25% at 3V, 100% at 2.25V and below. The regulation is very crude. We do not drive the LEDs to full current, or ever near full current.

Is this what you are looking fore?

 

[rgbbv]

so do you mean that from the image we can't tell how does this driver work because the cab1q chip can functions in different ways?
because i thought that it is less complicate,i thought that the resistor determine the value of the current(like any circuit of led in series with resistor)and the CAB1Q is only decrease the duty cycle from 100% on the high mode to lower duty cycle on the lower modes(in order to dim the led),with always the same duty cycles,even when battery voltage drop.

 

[audioguru]

I made some LED chasers that are powered from 2 AA alkaline cells (red LEDs) or from 4 AA alkaline cells (blue, white or bright green LEDs).
The cells last for months and the LEDs still light but are dimmed when each cell measures only 0.7V.
Energizer cells are fine but Duracell cells leak like MAD! So I don't use Duracell anymore.

 

[ronsimpson]

so do you mean that from the image we can't tell how does this driver work because the cab1q chip can functions in different ways?

You and I do not know what the IC is. There are many different parts. Now that I can see there is just a resistor + IC + cap + LED. So probably the resistor limits the current. There are micro processors that come in that package. Look at PIC for one example. Most probably it is not a PIC but I have "LED drivers" that are built on AVR or PIC micros.

 

[rgbbv]

Example:
The LED and switch combination has a 2 volt drop.
The battery is 3V. The current limit resistor has 1V at full charge.
When the battery drops to 2.5V the resistor has 0.5V. (1/2 current)
When the battery drops to 2.25V the resistor has 0.25V or 1/4 current.
I choose the resistor to give the desired current at 2.25 battery and 100% on switch.
Then at 3V battery I tell the IC to be at 25%. (4x current at 25% gives approximately the correct current)
We program the IC to give (in this case) 25% at 3V, 100% at 2.25V and below. The regulation is very crude. We do not drive the LEDs to full current, or ever near full current.

i read one more time the example and something is weired.
if the desired current at 2.25v battery is 1400mA when 100% switch.
it is mean that the current,when the battery 3v,is pulses of 5600mA current at 25%.
and it is too much current.am i miss here something?
and what will happens when we change to the lower modes?how does this integrated in your example?and when you use the percenrage values are you meant the dutycycle that turn on/off the led?
i read before that there is a constant current driver (like amc7135)that gave constant current,and if you want to use lower modes you use a pwm microcontroller that his only function is to turn on/off the led in specific dutycycle for controling the brightness,or you have boost driver to increase low voltage to high voltage or buck driver that does the opposite and they both use an inductor.
but your example make me very confused.

 

[ronsimpson]

The AMC7135 is a linear regulator. It outputs 350mA. It wastes any extra power in heat. It is not the part in you picture. If you can turn off/on the part at 10% of the time then the peak current will be 350mA the average will be 35mA. This is certainly not the type of regulation I was talking about. The AMC7135 does not use a resistor to limit current.
_____________________________________________
In my example of how to regulate using a resistor and PWM:
3V=25% 2.25V=100% (bright mode)
3V=12.5% 2.25V=50% (1/2 power mode)
3V=6% 2.25V=25% (1/4 power mode)

 

[rgbbv]

_____________________________________________
In my example of how to regulate using a resistor and PWM:
3V=25% 2.25V=100% (bright mode)
3V=12.5% 2.25V=50% (1/2 power mode)
3V=6% 2.25V=25% (1/4 power mode)

Hi ronsimpson.
this is,exactly,what i need to see to get out from my confusion.
now it is more clear.i see that the duty cycle on each one of the modes is not constant,but it being changed when the battery voltage decrease,while i thought before,that it stays the same,and only being changed when i click to other mode.
so,only to be sure that i understood it 100%.in case of drivers like in the image(ic+resistor+cap+led)does the duty cycle always function like you said(i mean the situation that the duty cycle value being changed when the battery voltage decrease)?
or it might be a case that the dutycycle won't be changed when the battery voltage decrease,but only when i click to lower modes?

 

[ronsimpson]

Yes... To your questions.
I can't know what you are looking at.
I have used ICs that change duty cycle when you change modes and do not change duty cycle when the voltage changes.
I have used ICs that change duty cycle when you change modes and change duty cycle when the voltage changes.
AND
I have used ICs that CAN NOT CHANGE duty cycle under any condition.
I have used ICs that change duty cycle when you change modes and change duty cycle to keep the current constant.

Most ICs will, at low voltage, stop functioning. They might stop with the LED on, or with the LED off depending on how the part is designed.

 

[rgbbv]

thank you very much for your replies.

this new technology of led flashlight is much more complicated from what i thought.
but i think that now i understand the subject better than before.

thanks again
and best regards

 

[rgbbv]

The LED and switch combination has a 2 volt drop.
The battery is 3V. The current limit resistor has 1V at full charge.
When the battery drops to 2.5V the resistor has 0.5V. (1/2 current)
When the battery drops to 2.25V the resistor has 0.25V or 1/4 current.
I choose the resistor to give the desired current at 2.25 battery and 100% on switch.
Then at 3V battery I tell the IC to be at 25%. (4x current at 25% gives approximately the correct current)
We program the IC to give (in this case) 25% at 3V, 100% at 2.25V and below. The regulation is very crude. We do not drive the LEDs to full current, or ever near full current. QUOTE]

Hi ronsimpson.i am sorry,but there one more thing that i try to understand.
you said

I choose the resistor to give the desired current at 2.25 battery and 100% on switch

if you choose the resistor to give the desired current,let say 1400Ma,at 2.25v battery and 100% on switch.it seems that when the battery is 3v the current seems to be 5600mA pulses(4*1400mA).

Then at 3V battery I tell the IC to be at 25%. (4x current at 25% gives approximately the correct current)

because at 3v the current will increase x4 to 5600mA ,and then 25% of this will give me the 1400mA current.and pulses of 5600mA is too much for the led.
i begin to get headache