Nice mount for home made lights.
- Thread starter Treadly
- Start date
H
Humbug
Guest
On 04/05/06 at 21:44:55 Treadly somehow managed to type:
>
> I found a 'good design'
> (http://www.hyperactive.oz.nf/Lights4/Luxeon4.htm) for mounting
> Luxeons. It's a simple but neat design.
I've seen that one before - pretty neat. BUT. Last night while I was
looking around for Luxeon lighting ideas [1] I stumbled across this
http://www.pilom.com/BicycleElectronics/LEDhead.htm
[1] For the last month or so I've been messing around with PIC micros
for another purpose and spotted a bike light thing spread over two
issues of Silicon Chip mag - Apr and May 06 [3]. It's vastly
over-the-top for straight bike lights what with a charging circuit,
more "modes" than you can shake a stick at, need two of 'em (front and
rear) etc. so I figured that I'd rat their design and make actual bike
lights, front and rear drivers in one unit.
For tail lights I reckon a couple of red 1W Luxeons - one flashing, one
steady (mounted as far apart as possible) should do the job. For
headlights I figured a 3W white luxeon for "low" beam and a 5W jobbie
for "high" beam and both the 3W and 5W Luxeons on for a "super high"
beam. All managed and driven by a PIC micro to get the last bit of
usable charge from a power source of between about 5 and about 16 volts.
Thoughts anyone.
[2] http://www.microchip.com/
[3] http://www.siliconchip.com.au/cms/A_106372/article.html
The question for Suzyj - or anyone else come to that. Do you have any
recomendations for PCB layout software ? GPL for preference but cheap
will do if it has to...
Linux, doze or even OpenVMS is fine.
--
Humbug
>
> I found a 'good design'
> (http://www.hyperactive.oz.nf/Lights4/Luxeon4.htm) for mounting
> Luxeons. It's a simple but neat design.
I've seen that one before - pretty neat. BUT. Last night while I was
looking around for Luxeon lighting ideas [1] I stumbled across this
http://www.pilom.com/BicycleElectronics/LEDhead.htm
[1] For the last month or so I've been messing around with PIC micros
for another purpose and spotted a bike light thing spread over two
issues of Silicon Chip mag - Apr and May 06 [3]. It's vastly
over-the-top for straight bike lights what with a charging circuit,
more "modes" than you can shake a stick at, need two of 'em (front and
rear) etc. so I figured that I'd rat their design and make actual bike
lights, front and rear drivers in one unit.
For tail lights I reckon a couple of red 1W Luxeons - one flashing, one
steady (mounted as far apart as possible) should do the job. For
headlights I figured a 3W white luxeon for "low" beam and a 5W jobbie
for "high" beam and both the 3W and 5W Luxeons on for a "super high"
beam. All managed and driven by a PIC micro to get the last bit of
usable charge from a power source of between about 5 and about 16 volts.
Thoughts anyone.
[2] http://www.microchip.com/
[3] http://www.siliconchip.com.au/cms/A_106372/article.html
The question for Suzyj - or anyone else come to that. Do you have any
recomendations for PCB layout software ? GPL for preference but cheap
will do if it has to...
Linux, doze or even OpenVMS is fine.--
Humbug
The problem with the Silicon Chip one is that not only is the design over the top, but the reliability of the circuit board too. The Luxeon pucks are sealed and I've only ever had one fail was when I connected the power up to it reverse polarity. I NEED reliability.Humbug said:On 04/05/06 at 21:44:55 Treadly somehow managed to type:
>
> I found a 'good design'
> (http://www.hyperactive.oz.nf/Lights4/Luxeon4.htm) for mounting
> Luxeons. It's a simple but neat design.
I've seen that one before - pretty neat. BUT. Last night while I was
looking around for Luxeon lighting ideas [1] I stumbled across this
http://www.pilom.com/BicycleElectronics/LEDhead.htm
[1] For the last month or so I've been messing around with PIC micros
for another purpose and spotted a bike light thing spread over two
issues of Silicon Chip mag - Apr and May 06 [3]. It's vastly
over-the-top for straight bike lights what with a charging circuit,
more "modes" than you can shake a stick at, need two of 'em (front and
rear) etc. so I figured that I'd rat their design and make actual bike
lights, front and rear drivers in one unit.
For tail lights I reckon a couple of red 1W Luxeons - one flashing, one
steady (mounted as far apart as possible) should do the job. For
headlights I figured a 3W white luxeon for "low" beam and a 5W jobbie
for "high" beam and both the 3W and 5W Luxeons on for a "super high"
beam. All managed and driven by a PIC micro to get the last bit of
usable charge from a power source of between about 5 and about 16 volts.
Thoughts anyone.
[2] http://www.microchip.com/
[3] http://www.siliconchip.com.au/cms/A_106372/article.html
The question for Suzyj - or anyone else come to that. Do you have any
recomendations for PCB layout software ? GPL for preference but cheap
will do if it has to...
--
Humbug
I've been using reflector brackets for my homemade lights. They're pretty adaptable. Another sort of bracket I've used for rack mounting and map holder mounting is one that I got from a marine supply place (Elizabeth St, Mel). They're for mounting things to railings on boats so they can do a fair range of diameters. Strong nylon and two screws. About $9 a pair.Humbug said:
Not sure about mixing Luxeon wattages. Since they need different currents you might need separate drivers for each type.I've seen that one before - pretty neat. BUT. Last night while I was
looking around for Luxeon lighting ideas [1] I stumbled across this
http://www.pilom.com/BicycleElectronics/LEDhead.htm
[1] For the last month or so I've been messing around with PIC micros
for another purpose and spotted a bike light thing spread over two
issues of Silicon Chip mag - Apr and May 06 [3]. It's vastly
over-the-top for straight bike lights what with a charging circuit,
more "modes" than you can shake a stick at, need two of 'em (front and
rear) etc. so I figured that I'd rat their design and make actual bike
lights, front and rear drivers in one unit.
For tail lights I reckon a couple of red 1W Luxeons - one flashing, one
steady (mounted as far apart as possible) should do the job. For
headlights I figured a 3W white luxeon for "low" beam and a 5W jobbie
for "high" beam and both the 3W and 5W Luxeons on for a "super high"
beam. All managed and driven by a PIC micro to get the last bit of
usable charge from a power source of between about 5 and about 16 volts.
Thoughts anyone.
[2] http://www.microchip.com/
[3] http://www.siliconchip.com.au/cms/A_106372/article.html
The question for Suzyj - or anyone else come to that. Do you have any
recomendations for PCB layout software ? GPL for preference but cheap
will do if it has to...
--
Humbug
G
[email protected]
Guest
Humbug wrote:
> The question for Suzyj - or anyone else come to that. Do you have any
> recomendations for PCB layout software ? GPL for preference but cheap
> will do if it has to... Linux, doze or even OpenVMS is fine.
Suzy's praises are being sung on
http://www.hackaday.com/entry/1234000273073676/ There's the odd bit of
leching going on too
Graeme
> The question for Suzyj - or anyone else come to that. Do you have any
> recomendations for PCB layout software ? GPL for preference but cheap
> will do if it has to...
Linux, doze or even OpenVMS is fine.Suzy's praises are being sung on
http://www.hackaday.com/entry/1234000273073676/ There's the odd bit of
leching going on too
Graeme
T
Terry Collins
Guest
slaw wrote:
> I've been using reflector brackets for my homemade lights. They're
> pretty adaptable.
How reliable do you find these?
what sort of load does it take to twist them around the bars?
> Another sort of bracket I've used for rack mounting
> and map holder mounting is one that I got from a marine supply place
> (Elizabeth St, Mel). They're for mounting things to railings on boats
> so they can do a fair range of diameters. Strong nylon and two screws.
> About $9 a pair.
Good tip. Thanks.
> I've been using reflector brackets for my homemade lights. They're
> pretty adaptable.
How reliable do you find these?
what sort of load does it take to twist them around the bars?
> Another sort of bracket I've used for rack mounting
> and map holder mounting is one that I got from a marine supply place
> (Elizabeth St, Mel). They're for mounting things to railings on boats
> so they can do a fair range of diameters. Strong nylon and two screws.
> About $9 a pair.
Good tip. Thanks.
H
Humbug
Guest
On 05/05/06 at 07:07:31 Treadly somehow managed to type:
<snip>
> The problem with the Silicon Chip one is that not only is the
> design over the top, but the reliability of the circuit board too.
> The Luxeon pucks are sealed and I've only ever had one fail was when I
> connected the power up to it reverse polarity. I NEED reliability.
I can't see why a well put together controller would be any less
reliable than the Luxeon pucks. A well designed controller would've had
reverse polarity protection built in.
--
Humbug
<snip>
> The problem with the Silicon Chip one is that not only is the
> design over the top, but the reliability of the circuit board too.
> The Luxeon pucks are sealed and I've only ever had one fail was when I
> connected the power up to it reverse polarity. I NEED reliability.
I can't see why a well put together controller would be any less
reliable than the Luxeon pucks. A well designed controller would've had
reverse polarity protection built in.
--
Humbug
S
Stuart Lamble
Guest
On 2006-05-05, Humbug <[email protected]> wrote:
> On 05/05/06 at 07:07:31 Treadly somehow managed to type:
>> The problem with the Silicon Chip one is that not only is the
>> design over the top, but the reliability of the circuit board too.
>> The Luxeon pucks are sealed and I've only ever had one fail was when I
>> connected the power up to it reverse polarity. I NEED reliability.
>
> I can't see why a well put together controller would be any less
> reliable than the Luxeon pucks. A well designed controller would've had
> reverse polarity protection built in.
Indeed. Diodes are dirt cheap; sure, you lose a little power (0.7V drop,
IIRC), but if you're going to blow out the entire board if somebody does
the wrong thing, it's well worth the cost.
I'm reminded of the tale of a guy who was designing systems to be
installed by drongos out in the middle of nowhere. The first lot was
wired up with reverse polarity, so the next revision had the terminals
shaped: positive was a square; negative a circle. The second lot was
wired up with reverse polarity: they'd hammered the connectors to fit
the wrong terminals. I seem to remember that there were a couple of
other similar instances before they finally built in a bridge rectifier,
and accepted the 1.4V drop in available voltage.
--
My Usenet From: address now expires after two weeks. If you email me, and
the mail bounces, try changing the bit before the "@" to "usenet".
> On 05/05/06 at 07:07:31 Treadly somehow managed to type:
>> The problem with the Silicon Chip one is that not only is the
>> design over the top, but the reliability of the circuit board too.
>> The Luxeon pucks are sealed and I've only ever had one fail was when I
>> connected the power up to it reverse polarity. I NEED reliability.
>
> I can't see why a well put together controller would be any less
> reliable than the Luxeon pucks. A well designed controller would've had
> reverse polarity protection built in.
Indeed. Diodes are dirt cheap; sure, you lose a little power (0.7V drop,
IIRC), but if you're going to blow out the entire board if somebody does
the wrong thing, it's well worth the cost.
I'm reminded of the tale of a guy who was designing systems to be
installed by drongos out in the middle of nowhere. The first lot was
wired up with reverse polarity, so the next revision had the terminals
shaped: positive was a square; negative a circle. The second lot was
wired up with reverse polarity: they'd hammered the connectors to fit
the wrong terminals. I seem to remember that there were a couple of
other similar instances before they finally built in a bridge rectifier,
and accepted the 1.4V drop in available voltage.
--
My Usenet From: address now expires after two weeks. If you email me, and
the mail bounces, try changing the bit before the "@" to "usenet".
H
Humbug
Guest
On 05/05/06 at 14:20:23 Stuart Lamble somehow managed to type:
<snip>
>
> I'm reminded of the tale of a guy who was designing systems to be
> installed by drongos out in the middle of nowhere. The first lot was
> wired up with reverse polarity, so the next revision had the terminals
> shaped: positive was a square; negative a circle. The second lot was
> wired up with reverse polarity: they'd hammered the connectors to fit
> the wrong terminals.
As someone once said - "never underestimate the power of human
stupidity". The biggest bang I've seen was at the RA transmitter site
at Shep. Someone (NOT ME) tried (accidently) to see how low the
impedence of a 22KV line was by switching a BIG genset in out of phase.
The gen stopped and the big ex-ship diesel flung itself off its
mountings. BIG BANG followed by SILENCE. The answer to the impedance
question is VERY bloody low indeed...
You can save the voltage drop by putting a fast blow fuse between the
power supply and the equipment. The first thing AFTER the fuse is a
hefty diode connected across the supply rails. Connect it up the wrong
way and the diode conducts and kills the fuse - hopefully before any
damage is done.
--
Humbug
<snip>
>
> I'm reminded of the tale of a guy who was designing systems to be
> installed by drongos out in the middle of nowhere. The first lot was
> wired up with reverse polarity, so the next revision had the terminals
> shaped: positive was a square; negative a circle. The second lot was
> wired up with reverse polarity: they'd hammered the connectors to fit
> the wrong terminals.
As someone once said - "never underestimate the power of human
stupidity". The biggest bang I've seen was at the RA transmitter site
at Shep. Someone (NOT ME) tried (accidently) to see how low the
impedence of a 22KV line was by switching a BIG genset in out of phase.
The gen stopped and the big ex-ship diesel flung itself off its
mountings. BIG BANG followed by SILENCE. The answer to the impedance
question is VERY bloody low indeed...
You can save the voltage drop by putting a fast blow fuse between the
power supply and the equipment. The first thing AFTER the fuse is a
hefty diode connected across the supply rails. Connect it up the wrong
way and the diode conducts and kills the fuse - hopefully before any
damage is done.
--
Humbug
W
Wilfred Kazoks
Guest
Jaycar have a LED driver Kit for Luxeons which is a lot simpler, it drives
up to a 5W from a 12v supply with low battery cutout so you don't damage
your batteries.
I think it was Silicon Chip Nov 2004.
If you look at the latest article you will see they use a ex CPU heatsink to
cool the Luxeon. so keep that in mind when developing your setup. The older
article goes a little way to explaining how the choice of heatsink was done.
It also used a CPU heatsink. The article basically said the CPU heatsink was
marginal. but a moving bicycle should provide some airflow in the absence of
a fan.
It also discusses the option of driving a 3w LUXEON at 700ma instead of
1000mA to extend the life and decrease the intensity falloff due to running
near max output. Still 20 000 hour is probably enough but 50% reduction in
intensity is probably not and would be the limiting factor.
I am going to try the simpler circuit out with a single 3W LUXEON.
Wilfred
"Humbug" <[email protected]> wrote in message
news:[email protected]...
> On 04/05/06 at 21:44:55 Treadly somehow managed to type:
>
>>
>> I found a 'good design'
>> (http://www.hyperactive.oz.nf/Lights4/Luxeon4.htm) for mounting
>> Luxeons. It's a simple but neat design.
>
> I've seen that one before - pretty neat. BUT. Last night while I was
> looking around for Luxeon lighting ideas [1] I stumbled across this
> http://www.pilom.com/BicycleElectronics/LEDhead.htm
>
> [1] For the last month or so I've been messing around with PIC micros
> for another purpose and spotted a bike light thing spread over two
> issues of Silicon Chip mag - Apr and May 06 [3]. It's vastly
> over-the-top for straight bike lights what with a charging circuit,
> more "modes" than you can shake a stick at, need two of 'em (front and
> rear) etc. so I figured that I'd rat their design and make actual bike
> lights, front and rear drivers in one unit.
>
> For tail lights I reckon a couple of red 1W Luxeons - one flashing, one
> steady (mounted as far apart as possible) should do the job. For
> headlights I figured a 3W white luxeon for "low" beam and a 5W jobbie
> for "high" beam and both the 3W and 5W Luxeons on for a "super high"
> beam. All managed and driven by a PIC micro to get the last bit of
> usable charge from a power source of between about 5 and about 16 volts.
>
> Thoughts anyone.
>
> [2] http://www.microchip.com/
>
> [3] http://www.siliconchip.com.au/cms/A_106372/article.html
>
> The question for Suzyj - or anyone else come to that. Do you have any
> recomendations for PCB layout software ? GPL for preference but cheap
> will do if it has to... Linux, doze or even OpenVMS is fine.
>
> --
> Humbug
up to a 5W from a 12v supply with low battery cutout so you don't damage
your batteries.
I think it was Silicon Chip Nov 2004.
If you look at the latest article you will see they use a ex CPU heatsink to
cool the Luxeon. so keep that in mind when developing your setup. The older
article goes a little way to explaining how the choice of heatsink was done.
It also used a CPU heatsink. The article basically said the CPU heatsink was
marginal. but a moving bicycle should provide some airflow in the absence of
a fan.
It also discusses the option of driving a 3w LUXEON at 700ma instead of
1000mA to extend the life and decrease the intensity falloff due to running
near max output. Still 20 000 hour is probably enough but 50% reduction in
intensity is probably not and would be the limiting factor.
I am going to try the simpler circuit out with a single 3W LUXEON.
Wilfred
"Humbug" <[email protected]> wrote in message
news:[email protected]...
> On 04/05/06 at 21:44:55 Treadly somehow managed to type:
>
>>
>> I found a 'good design'
>> (http://www.hyperactive.oz.nf/Lights4/Luxeon4.htm) for mounting
>> Luxeons. It's a simple but neat design.
>
> I've seen that one before - pretty neat. BUT. Last night while I was
> looking around for Luxeon lighting ideas [1] I stumbled across this
> http://www.pilom.com/BicycleElectronics/LEDhead.htm
>
> [1] For the last month or so I've been messing around with PIC micros
> for another purpose and spotted a bike light thing spread over two
> issues of Silicon Chip mag - Apr and May 06 [3]. It's vastly
> over-the-top for straight bike lights what with a charging circuit,
> more "modes" than you can shake a stick at, need two of 'em (front and
> rear) etc. so I figured that I'd rat their design and make actual bike
> lights, front and rear drivers in one unit.
>
> For tail lights I reckon a couple of red 1W Luxeons - one flashing, one
> steady (mounted as far apart as possible) should do the job. For
> headlights I figured a 3W white luxeon for "low" beam and a 5W jobbie
> for "high" beam and both the 3W and 5W Luxeons on for a "super high"
> beam. All managed and driven by a PIC micro to get the last bit of
> usable charge from a power source of between about 5 and about 16 volts.
>
> Thoughts anyone.
>
> [2] http://www.microchip.com/
>
> [3] http://www.siliconchip.com.au/cms/A_106372/article.html
>
> The question for Suzyj - or anyone else come to that. Do you have any
> recomendations for PCB layout software ? GPL for preference but cheap
> will do if it has to...
Linux, doze or even OpenVMS is fine.>
> --
> Humbug
F
Friday
Guest
Humbug wrote:
> On 05/05/06 at 07:07:31 Treadly somehow managed to type:
>
> <snip>
>
>>The problem with the Silicon Chip one is that not only is the
>>design over the top, but the reliability of the circuit board too.
>>The Luxeon pucks are sealed and I've only ever had one fail was when I
>>connected the power up to it reverse polarity. I NEED reliability.
>
>
> I can't see why a well put together controller would be any less
> reliable than the Luxeon pucks. A well designed controller would've had
> reverse polarity protection built in.
>
>
I think for a road bike it might be ok but for a mountain bike it would
be too fragile.
> On 05/05/06 at 07:07:31 Treadly somehow managed to type:
>
> <snip>
>
>>The problem with the Silicon Chip one is that not only is the
>>design over the top, but the reliability of the circuit board too.
>>The Luxeon pucks are sealed and I've only ever had one fail was when I
>>connected the power up to it reverse polarity. I NEED reliability.
>
>
> I can't see why a well put together controller would be any less
> reliable than the Luxeon pucks. A well designed controller would've had
> reverse polarity protection built in.
>
>
I think for a road bike it might be ok but for a mountain bike it would
be too fragile.
F
Friday
Guest
Humbug wrote:
> On 05/05/06 at 14:20:23 Stuart Lamble somehow managed to type:
>
> <snip>
>
>>I'm reminded of the tale of a guy who was designing systems to be
>>installed by drongos out in the middle of nowhere. The first lot was
>>wired up with reverse polarity, so the next revision had the terminals
>>shaped: positive was a square; negative a circle. The second lot was
>>wired up with reverse polarity: they'd hammered the connectors to fit
>>the wrong terminals.
>
>
> As someone once said - "never underestimate the power of human
> stupidity". The biggest bang I've seen was at the RA transmitter site
> at Shep. Someone (NOT ME) tried (accidently) to see how low the
> impedence of a 22KV line was by switching a BIG genset in out of phase.
> The gen stopped and the big ex-ship diesel flung itself off its
> mountings. BIG BANG followed by SILENCE. The answer to the impedance
> question is VERY bloody low indeed...
>
> You can save the voltage drop by putting a fast blow fuse between the
> power supply and the equipment. The first thing AFTER the fuse is a
> hefty diode connected across the supply rails. Connect it up the wrong
> way and the diode conducts and kills the fuse - hopefully before any
> damage is done.
>
That sounds like the best option. The problem has since been solved by
polarising the plugs. At the time I was just lazy and assumed the inputs
were protected, as are the outputs.
We did have a situation where we synchronised a 200 megawatt
generator not remembering the grid was disconnected and the output
phases earthed. I wasn't there at the time but it left huge burn marks
on the wall which had to be repainted, an area about 10 by 10 metres
where the busbars went through.
> On 05/05/06 at 14:20:23 Stuart Lamble somehow managed to type:
>
> <snip>
>
>>I'm reminded of the tale of a guy who was designing systems to be
>>installed by drongos out in the middle of nowhere. The first lot was
>>wired up with reverse polarity, so the next revision had the terminals
>>shaped: positive was a square; negative a circle. The second lot was
>>wired up with reverse polarity: they'd hammered the connectors to fit
>>the wrong terminals.
>
>
> As someone once said - "never underestimate the power of human
> stupidity". The biggest bang I've seen was at the RA transmitter site
> at Shep. Someone (NOT ME) tried (accidently) to see how low the
> impedence of a 22KV line was by switching a BIG genset in out of phase.
> The gen stopped and the big ex-ship diesel flung itself off its
> mountings. BIG BANG followed by SILENCE. The answer to the impedance
> question is VERY bloody low indeed...
>
> You can save the voltage drop by putting a fast blow fuse between the
> power supply and the equipment. The first thing AFTER the fuse is a
> hefty diode connected across the supply rails. Connect it up the wrong
> way and the diode conducts and kills the fuse - hopefully before any
> damage is done.
>
That sounds like the best option. The problem has since been solved by
polarising the plugs. At the time I was just lazy and assumed the inputs
were protected, as are the outputs.
We did have a situation where we synchronised a 200 megawatt
generator not remembering the grid was disconnected and the output
phases earthed. I wasn't there at the time but it left huge burn marks
on the wall which had to be repainted, an area about 10 by 10 metres
where the busbars went through.
R
Random Data
Guest
On Fri, 05 May 2006 10:36:52 +0000, Friday wrote:
> I think for a road bike it might be ok but for a mountain bike it would
> be too fragile.
Bollocks. If anything has a significant mass and is likely to vibrate
loose it can be potted. One of my quick and nasty methods to secure stuff
is to just chuck a (small) board and components in heat shrink. Even a
well soldered component will normally be OK, but a bit of mechanical
relief is a good idea.
--
Dave Hughes | [email protected]
"If you're bored, find something and break it" Jamie Rapson - 1996
> I think for a road bike it might be ok but for a mountain bike it would
> be too fragile.
Bollocks. If anything has a significant mass and is likely to vibrate
loose it can be potted. One of my quick and nasty methods to secure stuff
is to just chuck a (small) board and components in heat shrink. Even a
well soldered component will normally be OK, but a bit of mechanical
relief is a good idea.
--
Dave Hughes | [email protected]
"If you're bored, find something and break it" Jamie Rapson - 1996
H
Humbug
Guest
On 05/05/06 at 20:36:52 Friday somehow managed to type:
<snip>
>
> I think for a road bike it might be ok but for a mountain bike it
> would be too fragile.
A good quality PCB (not hard to make at home), bit of care in assembly
and the PCB mounted in a solid box and it'll be OK. Still doubtfull -
pot it.
--
Humbug
Today is Setting Orange, the 52nd day of Discord in the YOLD 3172
<snip>
>
> I think for a road bike it might be ok but for a mountain bike it
> would be too fragile.
A good quality PCB (not hard to make at home), bit of care in assembly
and the PCB mounted in a solid box and it'll be OK. Still doubtfull -
pot it.
--
Humbug
Today is Setting Orange, the 52nd day of Discord in the YOLD 3172
H
Humbug
Guest
On 05/05/06 at 18:06:02 Wilfred Kazoks somehow managed to type:
> Jaycar have a LED driver Kit for Luxeons which is a lot simpler, it
> drives up to a 5W from a 12v supply with low battery cutout so you
> don't damage your batteries. I think it was Silicon Chip Nov 2004.
Yeah but at Jaycar prices...:-( Those kits are around $30 EACH so for a
four LED system thats $~120 + LEDs = EXPENSIVE. I reckon I could build
my proposed system for less than about $50 + LEDs. STREWTH !!! - I've
just had look at the price (rather than just availability) of Luxeons
at Jaycar. Red 1W = $14.95 each, white 3W = $29.95 and white 5W =
$69.95. That's a lot of dollars for not a lot of LEDs.
>
> If you look at the latest article you will see they use a ex CPU
> heatsink to cool the Luxeon. so keep that in mind when developing
> your setup. The older article goes a little way to explaining how the
Getting rid of the heat shouldn't be too much of a problem. The real
problem, for me anyway, is going to be the code for whatever micro I
use - I'm leaning a bit towards something from the ATmega series ATM. I
need to do a bit of data sheet reading methinks.
<snip>
--
Humbug
Today is Setting Orange, the 52nd day of Discord in the YOLD 3172
> Jaycar have a LED driver Kit for Luxeons which is a lot simpler, it
> drives up to a 5W from a 12v supply with low battery cutout so you
> don't damage your batteries. I think it was Silicon Chip Nov 2004.
Yeah but at Jaycar prices...:-( Those kits are around $30 EACH so for a
four LED system thats $~120 + LEDs = EXPENSIVE. I reckon I could build
my proposed system for less than about $50 + LEDs. STREWTH !!! - I've
just had look at the price (rather than just availability) of Luxeons
at Jaycar. Red 1W = $14.95 each, white 3W = $29.95 and white 5W =
$69.95. That's a lot of dollars for not a lot of LEDs.
>
> If you look at the latest article you will see they use a ex CPU
> heatsink to cool the Luxeon. so keep that in mind when developing
> your setup. The older article goes a little way to explaining how the
Getting rid of the heat shouldn't be too much of a problem. The real
problem, for me anyway, is going to be the code for whatever micro I
use - I'm leaning a bit towards something from the ATmega series ATM. I
need to do a bit of data sheet reading methinks.
<snip>
--
Humbug
Today is Setting Orange, the 52nd day of Discord in the YOLD 3172
A
andrew Gamlen
Guest
Friday wrote:
> Humbug wrote:
>
>> On 05/05/06 at 07:07:31 Treadly somehow managed to type:
>>
>> <snip>
>>
>>> The problem with the Silicon Chip one is that not only is the
>>> design over the top, but the reliability of the circuit board too.
>>> The Luxeon pucks are sealed and I've only ever had one fail was when I
>>> connected the power up to it reverse polarity. I NEED reliability.
>>
>>
>>
>> I can't see why a well put together controller would be any less
>> reliable than the Luxeon pucks. A well designed controller would've had
>> reverse polarity protection built in.
>>
>>
>
> I think for a road bike it might be ok but for a mountain bike it would
> be too fragile.
I've made simple dc-dc converters for use with Li-ion batteris and
halogen lights on vero board. I've never had any reliablity issue and I
do 4 or 5 24 hour mtb races each year plus riding weekly at night during
the winter months.
> Humbug wrote:
>
>> On 05/05/06 at 07:07:31 Treadly somehow managed to type:
>>
>> <snip>
>>
>>> The problem with the Silicon Chip one is that not only is the
>>> design over the top, but the reliability of the circuit board too.
>>> The Luxeon pucks are sealed and I've only ever had one fail was when I
>>> connected the power up to it reverse polarity. I NEED reliability.
>>
>>
>>
>> I can't see why a well put together controller would be any less
>> reliable than the Luxeon pucks. A well designed controller would've had
>> reverse polarity protection built in.
>>
>>
>
> I think for a road bike it might be ok but for a mountain bike it would
> be too fragile.
I've made simple dc-dc converters for use with Li-ion batteris and
halogen lights on vero board. I've never had any reliablity issue and I
do 4 or 5 24 hour mtb races each year plus riding weekly at night during
the winter months.
W
Wilfred Kazoks
Guest
"Humbug" <[email protected]> wrote
> Yeah but at Jaycar prices...:-( Those kits are around $30 EACH so for a
> four LED system thats $~120 + LEDs = EXPENSIVE. I reckon I could build
> my proposed system for less than about $50 + LEDs. STREWTH !!! - I've
> just had look at the price (rather than just availability) of Luxeons
> at Jaycar. Red 1W = $14.95 each, white 3W = $29.95 and white 5W =
> $69.95. That's a lot of dollars for not a lot of LEDs.
Yes a 4 LED system would be a lot of $$'s.
The Jaycar kit does drive a pair of 1W or 3W LEDs so that isn't as bad as 4
kits.
Wilfred
> Yeah but at Jaycar prices...:-( Those kits are around $30 EACH so for a
> four LED system thats $~120 + LEDs = EXPENSIVE. I reckon I could build
> my proposed system for less than about $50 + LEDs. STREWTH !!! - I've
> just had look at the price (rather than just availability) of Luxeons
> at Jaycar. Red 1W = $14.95 each, white 3W = $29.95 and white 5W =
> $69.95. That's a lot of dollars for not a lot of LEDs.
Yes a 4 LED system would be a lot of $$'s.
The Jaycar kit does drive a pair of 1W or 3W LEDs so that isn't as bad as 4
kits.
Wilfred
T
Terry Collins
Guest
Humbug wrote:
> Yeah but at Jaycar prices...:-( Those kits are around $30 EACH so for a
> four LED system thats $~120 + LEDs = EXPENSIVE. I reckon I could build
> my proposed system for less than about $50 + LEDs. STREWTH !!! - I've
> just had look at the price (rather than just availability) of Luxeons
> at Jaycar. Red 1W = $14.95 each, white 3W = $29.95 and white 5W =
> $69.95. That's a lot of dollars for not a lot of LEDs.
err, bulk buys and wrong supplier.
If other melbournites are interested, you need to do a bulk buy from a
related supplier.
> Yeah but at Jaycar prices...:-( Those kits are around $30 EACH so for a
> four LED system thats $~120 + LEDs = EXPENSIVE. I reckon I could build
> my proposed system for less than about $50 + LEDs. STREWTH !!! - I've
> just had look at the price (rather than just availability) of Luxeons
> at Jaycar. Red 1W = $14.95 each, white 3W = $29.95 and white 5W =
> $69.95. That's a lot of dollars for not a lot of LEDs.
err, bulk buys and wrong supplier.
If other melbournites are interested, you need to do a bulk buy from a
related supplier.
Humbug wrote:
> The question for Suzyj - or anyone else come to that.
> Do you have any recomendations for PCB layout
> software ? GPL for preference but cheap will do if it has
> to... Linux, doze or even OpenVMS is fine.
I use Protel (called altium these days). I believe there's free versions of the earlier stuff kicking around the web. Maybe google Autotrax. It only runs on windows tho.
I generally use BEC (www.becman.com) to manufacture PCBs. Mucking about with ferric chloride etc is a PITA, and getting nice double sided boards with plate throughs and solder masks are much better than dealing with single sided boards anyway.
Best of luck,
Suzy
> The question for Suzyj - or anyone else come to that.
> Do you have any recomendations for PCB layout
> software ? GPL for preference but cheap will do if it has
> to...
Linux, doze or even OpenVMS is fine.I use Protel (called altium these days). I believe there's free versions of the earlier stuff kicking around the web. Maybe google Autotrax. It only runs on windows tho.
I generally use BEC (www.becman.com) to manufacture PCBs. Mucking about with ferric chloride etc is a PITA, and getting nice double sided boards with plate throughs and solder masks are much better than dealing with single sided boards anyway.
Best of luck,
Suzy
Similar threads
