Newbie driver question

[Codswallop]

Hello all from the (not so) sunny UK. New here 1st post - facinating forum(s). I am in the early planning stages of building an led fishing headlight. I suspect that the final design/specs will have to be dictated entirely by the availabilty of parts. What I would ideally like to be able to do is create a two mode (high/low) light using a 3 pole on-off-on toggle switch. I have built similar in the past, but using two separate halogen bulbs direct driven. This time I would like to use a single led or single led array (if that makes sense!).

To this end I wondered if it is possible to say - direct drive one 'channel' for high power, and route the second 'channel' through a low output single mode driver running a single common return to the battery through the driver?.

I have only basic electronic knowledge and no experience of using drivers, so am unfamiliar with connections etc. Is the method viable? and are there any drivers that could accomodate it?. I will upload a drawing as soon as I am allowed.

Led considerations at the moment are: Triple Cree XR-E R2 or single SSC P7.
Existing battery options available are ideally 3 or 4 cell AA nimh enabling headband mounting, or (a poor 2nd), 10 cell C or 1-10 cell D nimh waist mounted.

Any early thoughts or observations from you enlightened folks would be greatly appreciated. (Phew!).

 

[TorchBoy]

:welcome: from the really sunny antipodes.

Yes, it's easy to do. The hardest part is probably deciding what to do.

 

[Codswallop]

Thanks for the confirmation TorchBoy, all I have to do now is source a driver!:thinking::thinking::thinking: Will upload a drawing to doublecheck when the forum allows.

 

[TorchBoy]

I hope you're not waiting to be allowed to upload pictures to this site. It's not going to happen, even if your post count was high enough (which you don't really seem to be working on); you need to use a third party picture hosting service.

 

[Codswallop]

Thanks for the advice, drawing is here:http://img691.imageshack.us/img691/300/picturecrop001small.jpg. Will this work ok (depending on specific driver)?. I have lots more questions, but just trying to work through one thing at a time.:duh2:

 

[TorchBoy]

Nice drawing (you can post them inline, you know), and very Christmassy... except the arrowheads on your LED appear to be pointing inward, making it a photo detector (eg, solar panel). But seriously, that looks fine. I've done something very similar with some headlamps using 3 mode switches, with two resistor modes and one driver mode.

 

[Codswallop]

Nice drawing (you can post them inline, you know), and very Christmassy... except the arrowheads on your LED appear to be pointing inward, making it a photo detector (eg, solar panel). But seriously, that looks fine. I've done something very similar with some headlamps using 3 mode switches, with two resistor modes and one driver mode.

OOps!. Like I said, my electronics knowledge is sketchy at best - (just like my drawings!) - sketchy geddit? Oh never mind!. Thanks again for the confirmation.

 

[TorchBoy]

sketchy geddit?

Ha ha.

Sorry, but on second thoughts (now that I've had a good sleep), that arrangement might not work, depending on the particular driver you were using. If it was a switched negative driver like an AMC7135 (where the LED goes on the positive side of the driver) for the low mode you would need the resistor on the negative side of the LED. That would be what I did with my headlamps.

 

[Magic Matt]

That circuit seems to have a switch that gives you two circuits... one via a resistor to the LED and back via the driver negative line... second circuit via the driver positive to the LED back through driver negative.

I don't understand... surely the driver should be in the circuit all the time, and the switch should be a DPDT with one side connecting and disconnecting the battery, and the other side taking the driver + output through two different value resistors depending on switch position, to the LED than back through driver -

...or am I totally wrong?

 

[TorchBoy]

Sorry Magic Matt, that's wrong. Depending on the driver type and input voltage, putting a resistor in series with the LED on its output may not change the driver's output current at all. A constant current driver will try to increase its output voltage to compensate for the voltage drop across the added resistor.

The problem with the circuit diagram is that some drivers may have no common rails, while with AMC7135 drivers it's the positive side that's common.

 

[Codswallop]

Thanks TorchBoy, that's exactly the sort of info that I joined the forum to try and learn.:thumbsup:

 

[TorchBoy]

There was probably a problem with what I just wrote as well.

 

[Magic Matt]

Oh balls... I just can't get used to this idea of a fixed current driver... all the drivers I've ever used are basically voltage regulators... hence the resistor.

Sorry - feel free to prod me with the stick.

 

[Codswallop]

There was probably a problem with what I just wrote as well.

I'm all ears

 

[TorchBoy]

I'm all ears

I noticed (and fixed) a typo. Well, more of a typo - I got the whole wrong word.

Voltage regulators have their place, MM, and I'm still learning. Codswallop, have you decided on a particular driver?

 

[Codswallop]

I was hoping to use 3 or possibly 4 NiMh cells to drive a P7, so to keep things simple (cheap), I was considering this driver:http://www.dealextreme.com/details.dx/sku.6190, and leaving it in medium or low setting - or maybe modifying if practicable - to a single mode depending on output.

 

[TorchBoy]

Why a Seoul P7 if you're going to be driving it at 200-350 mA (the low and medium modes of that driver)? To keep things inexpensive you could use a single Seoul P4 and a single mode driver even cheaper than that one (just disconnect two of the three AMC7135 chips to get 350 mA).

http://www.videofoundry.co.nz/ianma...st.php?price_max=3.19&type=linear&order=price

 

[Codswallop]

My idea was the DD side of the circuit would supply the high output, and the driver side the lower - whilst wasting less power. At the moment I don't know without seeing what the output levels look like, so some experimentation is needed before settling on a 'level'.

 

[TorchBoy]

My idea was the DD side of the circuit would supply the high output, and the driver side the lower - whilst wasting less power.

That's funny - I went the opposite approach, with the high mode having the constant current driver. The more volts dropped across a resistor the more even its "output" will be as the battery goes flat so I figured it made more sense to have the low modes with resistors.

Also, the AMC7135 driver you're planning on using is a linear regulator, which means it's basically just a resistor that varies its resistance automatically to produce a constant current output. That means it won't be any more efficient than an ordinary resistor. :sigh: Mind you, running from three NiMH cells they can be very efficient.

 

[Codswallop]

My thinking was that it would be run 90% of the time on low power and 10% at the most - on high, so a steady low output/runtime was the most important factor.

Also, the AMC7135 driver you're planning on using is a linear regulator, which means it's basically just a resistor that varies its resistance automatically to produce a constant current output. That means it won't be any more efficient than an ordinary resistor.

Interesting, does this mean I need to consider a boost and or buck driver, or a least one with a coil in it, or am I missing something?:duh2: