A new light

"Alan Braggins" <[email hidden]> wrote in message
news:[email hidden]...

Quoted message said:

Quoted message said:

Quoted message said:

Why rectify? That's going to drop some voltage and the LED will
rectify for itself, though I note that Luxeon's data sheet doesn't
list a permissible reverse voltage.

For the 5W, it says > 5volts. It also says 'LEDs are not to be driven in
reverse bias', but I'm not entirely sure if that means you'll knacker

them

Quoted message said:

Quoted message said:

if you feed them AC.

Even if it doesn't knacker them, you won't get useful light out of them in
both directions, and "6V" dynamos can certainly produce more than 5V

across

Quoted message said:

something not letting much current through. Using a second Luxeon for the
other direction might be efficient, but you will probably need a second

mount

Quoted message said:

and reflector or lens.

I was thinking of using a second light, with associated extra lens/size. I
reckon a physically bigger light is better for being seen by, so this isn't
actually a disadvantage.

I've already got a rear light with LEDs working on half-wave. (half pointing
electrically one way, half the other). It works very well.

Quoted message said:

(There was some talk of using FETs for active rectification to reduce the
voltage drop.)

How hard is this to do?

cheers,
clive

Alan Braggins said:

In article said:

"Ian Smith" <[email hidden]> wrote in message
news:[email hidden]...

Quoted message said:

Why rectify? That's going to drop some voltage and the LED will
rectify for itself, though I note that Luxeon's data sheet doesn't
list a permissible reverse voltage.

For the 5W, it says > 5volts. It also says 'LEDs are not to be driven in
reverse bias', but I'm not entirely sure if that means you'll knacker them
if you feed them AC.

Even if it doesn't knacker them, you won't get useful light out of them in
both directions, and "6V" dynamos can certainly produce more than 5V across
something not letting much current through. Using a second Luxeon for the
other direction might be efficient, but you will probably need a second mount
and reflector or lens.

Yes, but I'm not expecting useful light out of the reverse - if you
need that light, put another pair the other way.

Plus, I'm proposing using the fact that the dynamo will produce
higher-than-nominal voltages - two luxeon in series will have a
nominal voltage of 7V and a nominal current of 300mA. I reckon that a
nominally 6V / 500mA dynamo, which acts more like a constant current
than a constant voltage source, will manage to drive that setup in a
useful manner. I'd expect (on the basis of very hazy guesswork) it
would end up actually supplying pretty near the 500mA, but the luxeon
can handle that non-continuous. In fact, the light that triggered
this discussion demonstrated that they'll take even higher continuous
without immediately frying, though what that's done to lifetime, I
wouldn't like to guess.

regards, Ian SMith
--
|\ /| no .sig
|o o|
|/ \|

Ian Smith [email hidden] opined the following...

Quoted message said:

Yes, but I'm not expecting useful light out of the reverse - if you
need that light, put another pair the other way.

Plus, I'm proposing using the fact that the dynamo will produce
higher-than-nominal voltages - two luxeon in series will have a
nominal voltage of 7V and a nominal current of 300mA. I reckon that a
nominally 6V / 500mA dynamo, which acts more like a constant current
than a constant voltage source, will manage to drive that setup in a
useful manner. I'd expect (on the basis of very hazy guesswork) it
would end up actually supplying pretty near the 500mA, but the luxeon
can handle that non-continuous. In fact, the light that triggered
this discussion demonstrated that they'll take even higher continuous
without immediately frying, though what that's done to lifetime, I
wouldn't like to guess.

While they can take the current it relates to cooling. Mine was attached
to a reasonably sized piece of aluminium which was directly exposted to
the air and subject to an average of 15mph wind-assisted cooling! ;-)

However, these are not LEDs in the traditional sense and are not
designed to handle reverse bias at all. Simply putting another diode in
series with each would be a better solution.

However, a bridge rectifier and a smoothing capacitor should provide a
decent enough DC source to use a 7805 as a constant current device to
drive it properly. The major effect on lifespan is caused by over
current (I suspect mine won't make the full 20,000 hours!) so regulating
the current is a good idea.

The 1W ones are available in the Star/O configuration which saves the
complexity of trying to align the lens. If you can be bother to do that
though, the 3W ones would be very cool.

Jon

"Jon Senior" <jon_AT_restlesslemon_DOTco_DOT_uk> wrote in message
news:[email hidden]...

Quoted message said:

However, these are not LEDs in the traditional sense and are not
designed to handle reverse bias at all. Simply putting another diode in
series with each would be a better solution.

I was wondering if having normal (ish) red LEDs at the back would do this.
So we have something like:

---|>|---------------------------------|>|-|>|----------
|>| 2 luxeons in series
|>|

12 red LEDS in parallel

Two of these circuits, one built the other way round, joined at the dynamo.

Downside is we're looking at 9v or so to drive it. But I know a hub will
generate that even if driven very slowly open circuit, so it might work.

What's the light output vs current like for a luxeon? Ie are they efficient
if they're being underdriven? (thinking of the bigger ones here).

Pity they're a tenner each - for 2 quid a throw I'd go and try it!

cheers,
clive

Clive George [email hidden] opined the following...

Quoted message said:

I was wondering if having normal (ish) red LEDs at the back would do this..
So we have something like:

---|>|---------------------------------|>|-|>|----------
|>| 2 luxeons in series
|>|

12 red LEDS in parallel

Should do it. You will have no control of the current though so you can
probably expect the LEDs to require periodic replacement. Given that the
normal red LEDs are usually under a pound each this would compare
favourably with filament lamps.

Quoted message said:

Two of these circuits, one built the other way round, joined at the dynamo.

Downside is we're looking at 9v or so to drive it. But I know a hub will
generate that even if driven very slowly open circuit, so it might work.

What's the light output vs current like for a luxeon? Ie are they efficient
if they're being underdriven? (thinking of the bigger ones here).

I haven't done anything with a light meter but to look at it seems like
a reasonably linear dimming curve although there is a minimum current /
voltage below which they will not emit at all.

Quoted message said:

Pity they're a tenner each - for 2 quid a throw I'd go and try it!

They are quite pricey, but the prices do seem to drop rapidly. The 1W
ones were about £15 when I first came across them. You can now get the
3W for less than a tenner (ex VAT).

It's a bit of a gamble, but if it all pays off and everything works you
can look forward to 20,000 hours of night time riding before you need to
replace them (100,000 if you are using coloured ones!). They pay for
themselves rather nicely compared to filament lamps.

Jon

"Clive George" <[email hidden]> wrote in message news:<[email hidden]>...

Quoted message said:

Quoted message said:

(There was some talk of using FETs for active rectification to reduce the
voltage drop.)

How hard is this to do?

For a modest fee: http://makeashorterlink.com/?Q56F21499
Drives up to 6A with 95% voltage conversion efficiency with ~2V
output. Higher output voltages get more efficient.

Tony

Jon Senior said:

Clive George [email hidden] opined the following...

Quoted message said:

I was wondering if having normal (ish) red LEDs at the back would do
this. So we have something like:

---|>|---------------------------------|>|-|>|----------
|>| 2 luxeons in series
|>|

12 red LEDS in parallel

Should do it. You will have no control of the current though so you
can probably expect the LEDs to require periodic replacement. Given
that the normal red LEDs are usually under a pound each this would
compare favourably with filament lamps.

There's a certain amount of current control from the dynamo, isn't there?

Quoted message said:

It's a bit of a gamble, but if it all pays off and everything works
you can look forward to 20,000 hours of night time riding before you
need to replace them (100,000 if you are using coloured ones!). They
pay for themselves rather nicely compared to filament lamps.

re paying for themselves - I think I'd disagree for the front lamp, unless
starting from scratch. Rear LEDs are definitely the way to go (don't blow,
much brighter), and with the rear LED in place, the 3W front bulb seems to
last forever. (well, a couple of years, anyway). 5 years for the price of
one of the luxeons?
(ie having x000 hours of life doesn't matter if you're not going to use it!)

How does the light output/power compare for a luxeon vs a halogen?

cheers,
clive

Tony Raven said:

"Clive George" <[email hidden]> wrote in message
news:<[email hidden]>...

Quoted message said:

Quoted message said:

(There was some talk of using FETs for active rectification to
reduce the voltage drop.)

How hard is this to do?

For a modest fee: http://makeashorterlink.com/?Q56F21499
Drives up to 6A with 95% voltage conversion efficiency with ~2V
output. Higher output voltages get more efficient.

We've already been there - Jon's switched mode chip, or the more efficient
one I pointed out would seem to be equivalent.

I was actually asking about rectification, not switched mode power supplies.
Unless the latter can be used to do the former?

cheers,
clive

In article said:

Alan Braggins said:

In article said:

"Ian Smith" <[email hidden]> wrote in message
news:[email hidden]...
>
> Why rectify? That's going to drop some voltage and the LED will
> rectify for itself, though I note that Luxeon's data sheet doesn't
> list a permissible reverse voltage.

For the 5W, it says > 5volts. It also says 'LEDs are not to be driven in
reverse bias', but I'm not entirely sure if that means you'll knacker them
if you feed them AC.

Even if it doesn't knacker them, you won't get useful light out of them in
both directions, and "6V" dynamos can certainly produce more than 5V across
something not letting much current through. Using a second Luxeon for the
other direction might be efficient, but you will probably need a second mount
and reflector or lens.

Yes, but I'm not expecting useful light out of the reverse - if you
need that light, put another pair the other way.

Plus, I'm proposing using the fact that the dynamo will produce
higher-than-nominal voltages - two luxeon in series will have a
nominal voltage of 7V and a nominal current of 300mA.

The original question was "would it be possible to run a 3W Lumeon".
Two one watt LEDs in series probably is a better match for the dynamo,
but doubling that then you are up to four heatsinks, lenses, etc., even
if the heatsinks will be smaller than if the LEDs were driven at 500mA
full time. (Or you are trying to squeeze two LEDs under the same lens
reasonably aligned for both.)
At which point you might be better off buying a 3W with full regulation
from Solidlights or someone (or building something similar yourself).
This - http://www.speleogroup.org/attinya.html - might be a good
starting point.
(Hmm - though the Solidlights web site only shows battery lights with one
and three LEDs, two is possible. I wonder how much extra it would cost to
have a second LED lighting automatically at high speed?
(In an ideal world, linked to a sensor so it the second LED only came on
at high speed when it was really dark, not just dull or under streetlights.))

Alan Braggins said:

In article said:

Yes, but I'm not expecting useful light out of the reverse - if you
need that light, put another pair the other way.

Plus, I'm proposing using the fact that the dynamo will produce
higher-than-nominal voltages - two luxeon in series will have a
nominal voltage of 7V and a nominal current of 300mA.

The original question was "would it be possible to run a 3W Lumeon".

I think you've missed some posts. Specifically Message-ID:
<[email hidden]>, in which I say "could you
not put two 1W luxeon in series across the dynamo". The conversation
has moved on. If it's not allowed to move on, teh original question
was not, in fact, would it be possible to run a 3W, but rather "any
more details on your switch-mode chip", being the oldest question in
ancestors to this post.

Quoted message said:

At which point you might be better off buying a 3W with full regulation

But that _has_ to lose some power, because it has voltage drops across
components that aren't producing light. If the point is to maximise
light output for a given power, methods that eliminate points of power
dissipation without light generation have got to have a head start.
The economics are not particularly relevant - if I was embarked on a
cost-driven exercise the solution would be to carry on using teh
lights I've got at the moment.

Regarding active rectification, does that actually work if you don't
have access to a power supply other than the unrectified level you've
got? I can think up an active rectification at small power levels
using an op-amp circuit, but that depends on a power supply to teh
op-amps at higher voltage, so obviously doesn't work here, unless
having rectified you step up to something to drive teh rectifier, and
then I think I've invented a perpetual motion thingy. I'm not
familiar with FETs, so I may be missing something fundamental - I need
to do some reading.

regards, Ian SMith
--
|\ /| no .sig
|o o|
|/ \|

Clive George [email hidden] opined the following...

Quoted message said:

There's a certain amount of current control from the dynamo, isn't there?

There is, but it may not control the right amount of current.

Quoted message said:

re paying for themselves - I think I'd disagree for the front lamp, unless
starting from scratch. Rear LEDs are definitely the way to go (don't blow,
much brighter), and with the rear LED in place, the 3W front bulb seems to
last forever. (well, a couple of years, anyway). 5 years for the price of
one of the luxeons?

The Luxeon could burn for 24 hours a day, for 2.5 years before it
reached its rated life. The problem with this as an investment for life
is that the technology will probably be superceded by then. :-)

The major advantage as far as I'm concerned is the reliability. Filament
lamps can suffer from knocks which cause them to blow early or cause the
glass to crack. And temperature changes can have a dramatic effect as
well. Solid state stuff will just keep going in a uniform and
(reasonably predictable) manner for ever'n'ever!

Quoted message said:

(ie having x000 hours of life doesn't matter if you're not going to use it!)

True, but there are beneficial side effects as well. You don't need to
carry a spare lamp and never have down time while you fumble to change a
small bulb in the dark.

Quoted message said:

How does the light output/power compare for a luxeon vs a halogen?

My 1W (When burning a little warm) was a little dimmer than my old 10W
halogen. I haven't taken it somewhere dark enough in the new
configuration to test it properly but will do so soon enough. I'll
report back with more information.

Jon

"Clive George" <[email hidden]> wrote in message news:<[email hidden]>...

Quoted message said:

We've already been there - Jon's switched mode chip, or the more efficient
one I pointed out would seem to be equivalent.

I was actually asking about rectification, not switched mode power supplies.
Unless the latter can be used to do the former?

The point with the switched mode power supply is it has the same
problem as you have which is the voltage drop across the rectifier
diode is a significant proportion of the output voltage and leads to
serious inefficiency. The way round that is to replace the diode
rectifier which will have, even for a Schottky diode, a voltage drop
of 0.3 - 0.5V, with an FET syncronously switched with the switching
cycles and can have series on resistances of 0.01 Ohm.

Tony