PWM drivers on a video light???

[herculino]

Hi all! I'm impressed about how much information you are giving me.

George Thanks a lot for showing here, I've found drivers like yours (yet) I know that quality has a price but...me poor men have a lot of free time (at least till September 28) but little money. So I'll keep looking around for something cheaper. By the moment your H6CC is the best option I've in my wallet.

Thinking about AMC7135 options... It's possible to change current "activating" more or less chips. I mean for 2.8A to use 8 "active" and "disconnect" 4 for 1.4A. That makes sense or is another stupid idea?

This way I'll not use any kind of PWM. May be use two 1.4A drivers and put them to work on parallel or just one at a time. I don't know how to do it but may be it would work. At least until I have money for the much more elegant H6CC solution.

Hanachan, what to say awesome

Tplas could you give me a bit more information (leds, battery pack...)

Thanks for the help to every one.

Best regards from Zaragoza
Again thank to every one for your help.

 

[Packhorse]

Thinking about AMC7135 options... It's possible to change current "activating" more or less chips. I mean for 2.8A to use 8 "active" and "disconnect" 4 for 1.4A. That makes sense or is another stupid idea?

Yes it is.
Simply power 4 of them and you get 1.4 amp, power another 4 and you get 2.8amp.
I know if you put a switch on the negative or the output it will work but the switch will have to cope with the 1.4amp (per 4 chip AMC7135).
But you could put the switch on the positive input instead that doesnt have so much current going to it. You could use any size reed or even a hall effect switch to drive it.
Of course AMC 7135s only work at voltages below 6 volt but if you use this clever set up that can be extended. You still need to keep the max battery voltage less than 2.5 volts more than the total Vf of the LED's.



Of course the other option is to just use resistors. While not very efficient its simple.

 

[herculino]

I've been thinking and reading a bit more and I'm comming back to the original plan.

I need to confirm that I can drive 6 leds XR-E or XP-E or XR-G with a single driver, like H6CC.

The battery pack will be 7 18650 cells set in series, something about 26V and 2500mA

This would give more than two hours of operation if I'm not wrong.

I don't know if there is another better way to it, as I'm quite new at this things I ask for yor help and oppinions.

Best regards form Zaragoza

 

[jspeybro]

perhaps this driver could also be interesting for driving 5 or 6 cree leds: http://www.dealextreme.com/p/5v-8-4...circuit-board-for-flashlights-26mm-13mm-26106
There's a thread in the forum somewhere that gives more info on the circuit and how you can modify it to get a higher output current, allow a higher input voltage and even change the PWM frequency to e.g. 200kHz. I think this would be the cheapest solution. It gives you a few modes, but you can disable that easily if you want.
The only downside is that you'll need to do some soldering yourself. People have used this driver to drive 6 LEDs in series.

Johan

 

[herculino]

Thanks Johan for the answer, but this is a boot driver and if I'm not wrong is much less efficient than a buck driver, in the other hand uses PWM to regulate the power and I'm not sure if this could give me any trouble with the video.

Thanks for you help.

Best regards from Zaragoza

 

[herculino]

Hi all!

I've spend some time "playing" with the National Semiconductor's Web based workbench.

And got this:











I don't know if I'll be able to build this circuits but may be I should try. The only point missing is the analog regulation of the board, I've found some information but I was unable to make the web program to implement it on the design.

Best regards from Zaragoza

 

[tplas]

To Georges80 : I really dont understand electronics, sorry. Wish i could.
To Herculino :
I made the lights with 6 x xml each light.
Wiring : shown in post 22 from Packhorse, instead of a single led , i used 2 led parallel .
Batt: 3 x 26650 and 3 x 18650 , both in series. so about 12,4v with fresh batt.
0,16 A on low
0,91 on mid
2,82 on high , runtime about one hour with redilast 3100 mah .

I mostly use low and sometimes mid.
Some results with these lights : youtube : tplas1
Only do night dives.

Best regards
Ton

 

[herculino]

Hi Ton,

Please, would you be so kind to explain a bit more how did you do it?

I don't catch how did you place the batteries. Did you made two packs of 3 cells in series and then connect them in parallel? Mixing different kind of them?

The driver was giving about 3.5 volts to each pair of XM-L set in parallel with near 6 A.

I can't figure out how did you do it...I really don't understand electronics too.

 

[DIWdiver]

About AMC7135 based drivers...it sound well till you think in dimming the leds to extend battery life. If I'm not wrong this drivers "spend" the battery at the same rate despite of dimming leds, I mean, the power that doesn't goes to the leds leaks like heat in the AMC's. SO if you dimm you get hotter but not more durable.

This is partly true, but mostly not. I know why you think so, and I'll explain where where that idea comes from.

If using an AMC7135 or other linear (as opposed to switching) regulator, you dim the LED by making the regulator pass less current to the LED. In a linear regulator, the load current at the battery is always the same as the LED current. So when the LED current is reduced, the battery current is reduced too, and the battery lasts longer.

Since the LED voltage is less at lower current, a switching regulator can make the battery current reduce MORE than the LED current reduces, for example the LED current is reduced 50%, the battery current may be reduced 52%. This gives the switching regulator a SLIGHT advantage over the linear regulator when dimming.

Where the linear regulator does not improve battery life is when you add more batteries in series (or use higher voltage batteries). Say you had one XP-G and one 18650 2500mAH battery, with a 1A linear regulator. You would expect about 2.5 hours life. You don't even have to think about efficiency. Then say you change to two 18650s in series. You still have 1A draw on them, so you still expect 2.5 hours life. All the extra battery does is heat up the regulator.

Now say you change the regulator to a switcher (in this case, a buck regulator). If the regulator is 85% efficient (not bad, but not impressive). You will get approximately 3.7V * 2500mAH * 85% = 7.9W-H out of the regulator. If the XP-G is typical, it will be about 3.3V at 1A, or 3.3W. 7.9W-H/3.3W = 2.4 hours. Note that this is actually WORSE than the linear regulator, though only a little bit.

But here's the big difference. If you switch to two 18650s, now you get 7.4V * 2500mAh * 85% = 15.7 W-H. 15.7/3.3 = 4.8 hours. That's a huge improvement over the linear reg.

This illustrates the main advantage of the switching regulator - it's ability to efficiently handle a large difference between battery voltage and LED voltage. If the battery voltage is only a little more than the LED voltage, the switcher has much less advantage, or none at all.

In your case with 7 cells and 6 LEDs, the efficiency of a linear driver would be expected to be around 76%, while a good switcher like a TaskLED might be around 90%. Thus the switcher might give you nearly 20% longer burn time. You'd have to decide it that's enough to make it worth using the switcher instead of the linear regulator.

By the way, just a quick plug for the IS1006 linear driver which can (in high voltage version) handle up to 30V input and 0-10A output. It can be dimmed with a potentiometer or by switching external resistors. No flicker issues!

Sorry for my bad English, I don't know how to express better the idea.

I understand you with no difficulty! Some Americans are harder to understand.

 

[MikeAusC]

. . . . Thinking about AMC7135 options... It's possible to change current "activating" more or less chips. I mean for 2.8A to use 8 "active" and "disconnect" 4 for 1.4A. That makes sense or is another stupid idea? . . . . .

Yes, that's a very efficient way to have variable brightness using Linear Regulators - the efficiency remains the same, no matter what brightness setting you use.

 

[tplas]

Hi Herculino
I cannot upload pictures of wiring diagram and my videolight.
So I send you a pm .

 

[herculino]

Hi all!

After more studying and reading and getting crazy :duh2: I end up with that idea.

I'm going to use XPG leds as they are very efficient running at lower voltage and able to give two times the lumen I need so I can use less amps in a more efficient way. Not XML as they cost more and I don't think I'll need 6000 lumen, may be too much.

I read somewhere that for diving candles is useless to have more runtime than needed, you are not using it anyway, and you will recharge anyway.
I figure out than 70-80 minutes is a good runtime as it lets you made two following dives without recharging the batteries. If you made another one later on the same day you probably will have time to recharge while denitro your body, or you can have extra batteries.

So I made a hipper-fast calculation: 6 XPG leds running at a max 1.3A for 90 min will req (80% efficiency) about 15Ah to throwout 2500 lumen.

4X18650s are able to "store" about 11Ah so about 70min at max power. Will set them in series so 14.8v and 2500-2800 mAh.

2500 lumen is double the planned intensity, as I don't expect to be able to develop a torch as efficient as Solo does, I'm planning for a 70% of efficacy in my DIY reflectors so I think that pointing to 1700 lumen is a good approach. That require 700mA for the leds and making the run time calculations again results in more than 120min.

So...6XPG moved by 4x18650..... I need a boost driver.

Even more complicated to build for my own, less cheap options.

Thankfully there are George's drivers

I'm not sure which one to choose: maxFlex6 or Hiperboost.

It sounds nicer the way maxFlex6 dim the lights, the only point is again PWM, I'll ask George about the frec of the output to the led when dimming, especially when dimming a lot as I need to be able to get as low as 100mA Iout for the leds without blinking.

I love all the "extras" that this driver has: Batt level, temp control, easy multimode, tiny tiny...:huh:

Will ask George what he thinks.

The other option is Hiperboost, where I can use an external pot to dim steeples and without any kind of PWM matters. The cons is that don't came with any extra:shakehead and is more expensive (not too much but 5 bucks is an XPG)

Please tell me what do you think about it, did I made wrong calculations? there's something missing? Which driver will you choose?

Best regards from Zaragoza. Juan

 

[herculino]

...sent twice...