Constant Current vs. PWM dimming Revealed

[NewBie]

I was referred to an article for further details:

On the Bandstructure in GaInN/GaN Heterostructures - Strain, Band Gap and Piezoelectric Effect

....The redshift approximately corresponds to one halfperiod of the Franz-Keldysh oscillation near the band gap. We therefore attribute this redshift to the Franz-Keldysh effect [23]. Spatially indirect transitions across the band gap appear at lower energy due to the tilting of the bands. At the same time the transition probability is reduced exponentially as the area under the triangular tunneling barrier grows without involving impurity or defect states (see schematic in Figure 4):


(14)



Within a halfperiod of the first Franz-Keldysh oscillation the transition probability has decayed to 4 % of its value at the DOS band gap [23]. In this framework initial and final states are states of delocalized carriers. The exponential "tail" is not induced by any disorder or impurities but rather by the tunneling process in the tilted bandstructure....


And a whole lot more here:

http://nsr.mij.mrs.org/3/31/text.html

 

[4sevens]

I still can't believe the staggering losses with PWM especially past
50% HUGE amounts of energy is lost in the pulsing especially at
higher rates like the lionhearts and lioncubs. LED's are about
efficiency - turning energy into light and not heat. And to believe
the denzens of these LH and LC using LED's like "light bulbs" as
Bulk said in the past. What a waste of LED to use PWM.

Remember theres a HUGE efficiency penalty at higher currents and these
PWM units are always hitting the LED at 1.3A-1.6A !!

 

[Doug Owen]

Gents,

Interesting thread. A couple of thoughts come to mind, first off CC regulators are not 100% efficient, typically far from it. And this is not being factored in. As a guess, this is probably more than the typical 'spread' shown for reasonable current levels above. This means that for most applications, PWM is more efficient. FWIW there's a good reason this technique is so popular.....

Secondly, 4sevens is spot on (again). Part of the potential loss issue is the 'bigger is better' view many have. If you severly overdrive (so 100% is really bright), you automatically impact the efficiency at 'useful levels'. If, however, you keep 'full blast' to reasonable levels, it's gonna be real hard to beat a well designed PWM system for efficiency.

Doug Owen

 

[NewBie]

Reasonable drive is key.

Notice the 253% increase in efficiency of the LED(in current dimming) at 3% PWM vs. 3% Constant Current, when the PWM is hitting the led with 1.140 Amps.

Thats less than a 50% efficiency level there.

It gets worse at higher PWM currents. Go back and compare it to the 350mA PWM chart.

If you go higher for current, with the PWM, the LED efficiency continues to drop when dimmed.

Yes, a well designed PWM, that takes in consideration of the LED, converter/PWM, and the battery.

Lately, Constant Current regulators in flashlights have made leaps and bounds, with efficiencies exceeding 90%.

Another consideration is regulation. Except on high, PWM can also regulate the light by adjusting the duty cycle as the battery depletes, but that isn't done on the mainstream lights which use PWM, like the Photons, LionHeart, LionCub. As commonly implemented it is more like direct drive. Here is an example of the unregulated runtime output on one of the common PWM lights:

Exactly. The whole light needs to be considered for the best solution, not just the converter/PWM circuit alone. Heat generated, thermal paths, temperature, reflector, front lens, battery contacts, converter/PWM, LED, battery, load pulses on the battery, etc.

 

[PeterB]

[ QUOTE ]
Doug Owen said:
If, however, you keep 'full blast' to reasonable levels, it's gonna be real hard to beat a well designed PWM system for efficiency.



[/ QUOTE ]
I've a different opinion regarding this statement. Even a linear regulation or a poti in series with the LED is more efficient than PWM (lumen out vs. current draw).
And how do you reduce the full blast? By a series resistance?

 

[NewBie]

Doug,

Some of the stuff a number of us talked about awhile ago is applicable to some of the points you brought up. It is a long read, but worth it, imho.
http://www.candlepowerforums.com/ubbthreads/showflat.php?Cat=&Board=UBB6&Number=600212

 

[Doug Owen]

[ QUOTE ]
NewBie said:
Doug,

Some of the stuff a number of us talked about awhile ago is applicable to some of the points you brought up. It is a long read, but worth it, imho.
http://www.candlepowerforums.com/ubbthreads/showflat.php?Cat=&Board=UBB6&Number=600212

[/ QUOTE ]

Yup, I recall it. FWIW, I was the 'Doug Owen' taking part....

Thanks for 'the trip down memory lane'.

Doug Owen

 

[NewBie]

No problem Doug.

Thanks for the contribution.

 

[cy]

an excellent old thread!

In the real world I find PWM VS Constant Current regulation only matters if you are limited to battery choices.

if you are limited to or just prefer primary cells. then constant current regulation would be prefered.

with availibity of cheap hi-cap li-ion cells. like many on cpf, I'm using li-ion as first choice in almost all my lights.

effiency with li-ion/PWM matters little when you are running basically for free.

with PWM you are esentially running at direct drive on high. VF matters very much as that's what determines current going across luxeon.

with PWM first choice is Kbin (vf closely matches li-ion's voltage under load) which limits current to aprox. 1.3 amps. factoring in further VF shift downwards to Jbin, which will further increase current at lux to aprox. 1.5amps. Hbin will produce higher yet current draws of aprox. 1.75 amps. which is undesirable.

with CC regulation inverse is true. lower VF the better to increase runtimes from primary cells.

two different sets of requirements are emerging in PWM VS CC reguation camps.

 

[NewBie]

Thats not entirely true.

With a buck-boost, you can still run CC (constant current) off a single Li-Ion.

With a buck, you can run off two or more Li-Ion.

Efficiency is still essential, as it determines your runtime, and many would prefer not to replace cells in the field. On top of higher efficiencies when dimmed with CC for the LED, you also get higher efficiency out of the battery with CC. With CC, since you get more light out of the LED for the same amount of power when dimmed, you need less power to make a given amount of light, which means the runtime is increased. Now add on the higher efficiency of the battery with CC when dimmed, lower power draw, less heat, and things really start adding up fast.

And with the lower Vf LEDs, since power = Voltage
* Current, you are consuming less power for a given amount of light, yet again.

PWM for LEDs came out before 1975, when watch and calculators makers started PWMing to reduce power consumption, as at the time, the only other technique was resistor/linear regulation, which was very wasteful. At that time, switching power supply technology had not came of age for small portable items. Thats all changed now, most especially with cell phones over the past few years, greatly extending runtimes, laptops, MP3 players, portable games, GPS systems, etc.

 

[cy]

back to the real world again,

most usage of lights, PMW included if given a choice is at low levels. with high only used selectively.

this is especially true with todays overachieving luxeons, where high beams is esentially too bright for most closeup activities.

Just got back from a two week backpacking trip, where Ti lioncub was my primary light. Used that light constantly everynight without worrying about conserving the R123 li-ion cell. first R123 cell lasted for 12 days under extreme use conditions. naturally I changed to my backup cell.

My backup light hanging on my neck was Li14430/CR2 also li-ion cell, which made it thru the trip with original cell.

tent light was ARC AAA 2003 cpf edition which made trip with original cell.

yes a buck/boost can use a single li-ion cell, but is not very good at low level modes. besides most of my regulated lights are CC not buck/boost. I've purposly not purchased any wizard equiped lights due to reports of low level issues. which is of course 98+% of my light usage.

Wayne's new two stage CC board looks real promising. I'm looking forward to get one of these. don't have the details yet, but if it's CC only. it too would be running at essentially DD if VIN exceeds VF with li-ion.

 

[cy]

newbie, forgot to add, excellent point about PWM being used in early LED aplications.

here's an example of early LED's controlled by PWM, introduced in 1974 by National Semiconductor. Mathbox 650 a simple RPN calculator pictured with Ti Cub and Li14430/CR2

 

[NewBie]

Yes, for those who have been around long enough, they will remember the simple mod for many of these, adding a speaker in series with the battery, which causes the speaker to emit lots of neat beeps and buzzes as you operate it.

 

[NewBie]

CY,

Did you ever try putting a speaker in series with the batteries on that one?

 

[shiftd]

wow, totally missed the thread development.

plenty of useful info, need to catch up on the reading .

Never liked PWM, but i guess it has its own application

 

[cy]

WOW... a blast from the past!

certainly one of Newbie's best contribution to cpf and always worth another read.

 

[NewBie]

Thanks for the comments.

I had a lot of fun doing it!

BTW, did you ever try the speaker thing I mentioned before?

 

[NewBie]

I was looking thru some old pictures and realized I never posted this one:

 

[jsr]

Sorry for my lack of understanding here, but I'm wondering how everyone's distinguishing PWM from Constant Current. Here's my understanding...

A CC power supply varies the output voltage via a sense resistor to keep the current (the voltage at the sense resistor) the same. The way to vary the output voltage is via a PWM (or PFM) signal that adjusts the on-time (and thus average voltage and average current) at the output. The output is then fed through a filter (cap typically, and the inductor in a buck or boost (or sepic) topology acts as a filter also) so the output is not on or off, but a rippled voltage. That's the only method I know of on how to make a CC PS (by using PWM or PFM). Any dimming would just be a change in the feedback signal, reference voltage, or pulse width (or a combo thereof).

Unless there's another way to control the reduction of voltage/current, I don't see how PWM and CC is being distinguished. And I also don't see how people could be seeing flicker from PWM/SMPSs unless the output's not being filtered at all (but that's just a sloppy design if there's no output filter, unless space is really so constrained, but even a small cap would reduce the ripple to better than just on and off).

TIA for the education.

 

[evan9162]

The PWM we're describing does not have anything on the output to smooth the signal - so no cap, inductor, etc. Each pulse is a full current pulse to the LED, followed by some off time where the LED emits no light at all.

Adding a cap/inductor and some feedback gives you a switching DC-DC converter which gives you constant current on the output.