Explain direct drive please...

[Robocop]

Ok a search on direct drive just seemed to confuse me even more.I often hear people use the term direct drive and each time it is said to produce a very bright Luxeon however with many risks involved.Can someone explain exactly what direct drive is?
I thought it was simply using a board with no resistor in place but from searching old posts I think this thinking is not correct.
Lets take the little Dorcy AAA circuit for example...I have used this circuit for many mods and usually remove the 1 ohm resistor.Does this mean this circuit is now direct drive?I am trying to understand the term and am also wondering just what the order is to control the current or voltage in a direct drive light.I know it has something to do with the VF of the Luxeon and I assume it also has something to do with the voltage and current of the power source but I am still curious.So what exactly is direct drive?...Is there such a thing as a regulated direct drive circuit?
Thanks for any help on this.

 

[hawkeye]

Feel free to correct the Newbie, but I believe direct drive refers to the absence of a current regulator. The presence of a resistor alone would make a current limiter, but not a regulator. As for the Dorcy resistor, most FET based current regulators have a small resistor on the common-bias, but I'm not familiar with the Dorcy circuit, so I don't know for certain.

 

[Robocop]

Thanks for the imput and that is probably correct.I thought the absence of the resistor made it direct drive however found some old threads where people talked of adding more or less resistance to their direct drive lights to alter performance.This led me to believe that the resistor had little or nothing to do with direct drive.
If what you say is correct I guess that there is no such thing as a direct drive regulated light.This is usually where someone like MrAl steps in and clears it up for me..hehe...he is a genious when it comes to explaining things like this...Thanks again Hawkeye and welcome to CPF

 

[red_robby]

Robocop,
direct drive is simply driving the lux directly from the batteries with no circuit on the way, placing a resistor
is still direct drive since the resistor is only there to limit current flow.

 

[Robocop]

Thanks for the help...I thought it may be something simple but you never know until you ask.Is it safe to say the most important factor of direct drive is good heatsinking?
What are the best advantages of direct drive and can someone give me an example of a high end production light that uses direct drive?...I read some old posts where MrBulk stated he was one of the first to offer a custom unit with direct drive.It seems to me that the keep it simple rule would apply here as there are less electronic parts to worry with when using this type of light.
Thanks again and I learn something new each time I visit here.

 

[SilverFox]

Hello Robocop,

I believe the LionHeart is direct drive at the brightest setting.

Tom

 

[cy]

when using a boost circuit and you exceed vf of luxeon. then you are in direct drive with only burden of circuit to limit current.

certain combinations of nmh cells matched to vf of luxeon, direct drive may be more desirable.

3x C nmh cells has fairly flat discharge rate and voltage that almost perfectly matchs Kbin vf.

 

[red_robby]

good heat sinking is always first on the list...
like CY already said, you want to match(close to it anyway) your luxeon's Vf to your power supply(total voltage of cells).
do you remember the CMG reactor ?, that was a direct drive light, it was under driven to keep tings from melting.
then the reactor 3 came out, it had 3 cells instead of 2, now they put a resistor in the tail cap(- path) to limit current.
the advantage of direct drive is efficiency since there is no boost/buck circuit all the power goes right to you lux.
when using a resistor you will loose some power there but still much less then any circuit out there, but your light's output is not constant throughout battery life.

 

[Doug Owen]

[ QUOTE ]
hawkeye said:
Feel free to correct the Newbie, but I believe direct drive refers to the absence of a current regulator. The presence of a resistor alone would make a current limiter, but not a regulator.

[/ QUOTE ]

I sorta agree.

However, as *adding additional* resistance (there is already some in any real world circuit) does, IMO, constitue a reguator (or at least an attempt at current regulation). If you intentionally add any more resistance, it's no longer DD.

FWIW, 'one off', hand tuned DD lights are an interesting case. With the *exact* conditions specified met (temperature, battery, bin, heatsinking, etc) they may be safe enough. But, with things a bit outside the lines, problems can develop I strongly suggest we don't hear about as often?

Doug Owen

 

[Doug Owen]

[ QUOTE ]
red_robby said:
do you remember the CMG reactor ?, that was a direct drive light, it was under driven to keep tings from melting.
then the reactor 3 came out, it had 3 cells instead of 2, now they put a resistor in the tail cap(- path) to limit current.
the advantage of direct drive is efficiency since there is no boost/buck circuit all the power goes right to you lux.


[/ QUOTE ]

An excellent example of a (commercial) DD light that was a bitter disappointment and was quickly replaced by a resistored version. There's a clue in there for all the budding engineers out there......

And I think 'all the power goes to the lux' is far too simple a way to look at it. You (should) design a light to give a certain level of performance (light out). How many hours of *that level* you get is a useful measure. Ignoring 'lumens per Watt' when you discuss efficiency is (IMO) disingenuous.

Doug Owen

 

[andrewwynn]

I had originally considered anything w/o a controlling circuitry direct drive, but i think it's even covered in the CPF 'terms' that DD means no current control at all even a resistor, so typically people will call it a 'resistored' solution. I have some DD lights i made... and they start very hot on current.. 1200 to 1300mA, but have excellent heat sinks.. however it's so easy to make the LDO circuit MrAl designed for me that i 'tweaked' to fit into my AAA lights that i'd never do a DD design.. it costs about $1.40 in parts and i can solder one up in under 10 minutes now.. and they are inky-dinky (the circuit board i'm using is 1/4 the area of a penny).. the main FET is rated for up to 20A so there is no problem driving high-power stuff.. would take some fine-tuning to run at levels higher than what i've been using (1 1/2W).. but i have dialed it up to 700-800mA before even from an AAA bat.

I like the LDO solution a lot.. regarding 'all the power goes to the light'... it's nice that more of the power goes to the light as the battery voltage drops.. eventually the circuit almost becomes a dead short... roughly 0.15ohm resistance to the circuit when it goes out of regulation... i get efficiency levels in the high 90% when the battery gets low.. and a rock steady light output.. check out http://rouse.com/nano and look at the bright[ness] charts.

-awr

 

[bjn70]

I vote for "direct drive" to mean no resistor, as well as not buck or step-up circuit.

 

[Doug Owen]

[ QUOTE ]
bjn70 said:
I vote for "direct drive" to mean no resistor, as well as not buck or step-up circuit.

[/ QUOTE ]

Works for me, ya got my 'second'.

I call the question.....

Doug Owen

 

[Doug S]

[ QUOTE ]
Doug Owen said:
[ QUOTE ]
bjn70 said:
I vote for "direct drive" to mean no resistor, as well as not buck or step-up circuit.

[/ QUOTE ]

Works for me, ya got my 'second'.

I call the question.....

Doug Owen

[/ QUOTE ]
Aye, you have my vote. I might add, no gearing either /ubbthreads/images/graemlins/tongue.gif

 

[McGizmo]

I vote for DD to be any condition when the Vin is the Vf of the LED whether by design or hapenstance. (Thinking of Vin in excess Vf on a boost circuit or when a converter has bit the big one and not doing its thing /ubbthreads/images/graemlins/blush.gif ) A buck circuit may go into DD when Vin is below target Vf and steps out of the picture in idle.

Not doing well in explaination here. /ubbthreads/images/graemlins/frown.gif How about when Vf is free to move with Vin with little if any restraint? /ubbthreads/images/graemlins/icon3.gif

 

[gadget_lover]

[ QUOTE ]
Doug Owen said:
[ QUOTE ]
bjn70 said:
I vote for "direct drive" to mean no resistor, as well as not buck or step-up circuit.

[/ QUOTE ]

Works for me, ya got my 'second'.

I call the question.....

Doug Owen

[/ QUOTE ]

I vote with Doug.

As for Mc Gizmos's point, a circuit may go into "DD mode" without being a DD light. It's emulating "no resistor, as well as not buck or step-up" when in DD mode.

Daniel

 

[andrewwynn]

just like the LDO example.. the logic behind the circuit is that once out of regulation it is 'shorted out' or in DD mode, so that makes perfect logic describing the circuit. I used to think that if no active devices it was 'direct drive' but i prefer to call it 'resistored' because it's more accurate.

 

[Robocop]

Ok if I got this right let me give an example and tell me if this is correct.Lets say I have a 5 watt emitter.I know this will not run off of a single 123 being only 3 volts so I get 2 cells for a total of 6 volts.I tape the cells together and run one wire direct from the positive end of the battery and one from the negative end.I then attatch these wires direct to the positive and negative terminals of the Emitter.This should power the emitter and be a direct connection between the power source and Luxeon so is this what you mean by direct drive with nothing between the power and the emitter?

 

[Darell]

Bingo! Give the man his prize!

Yes, indeed, that's what everybody is talking about, and what Don sort of said /ubbthreads/images/graemlins/tongue.gif is also what I consider DD.

 

[hawkeye]

Instead of 'resistored', I would propose keeping to 'standard' electro-geek jargon and refer to our drive options as:

Direct Drive (DD) - No components other than those necessary for the creation of light. Batteries, wires, boards, switches, and LEDs only

Passive Regulation (PR)- Resistors, capacitors, inductors only. No device which requires bias (forward) voltage to operate.

Active Regulation (AR) - Devices with a PN junction

BTW- I know LEDs have PN junctions, but they're not part of the regulation circuit.