driving IR LEDs with mosfet drivers.

[Jonathan Hallameyer]

Hello all,
I'm putting together a sound activated trigger for a friends camera, it uses
modulated IR (38-40 khz, like a remote control) and since this was meant for
outdoors use, and possibly for use with rifle and having to deal with muzzle
flash, and having to beat that out, I decided it might be worth it to try a
3W IR LED from deal extreme. It takes 1.5A @2v. Now for driving that, could
I get away with using a 2A peak mosfet gate driver?  I was going to put a
28ish uF capacitor in series to keep from damaging anything if it got stuck
on, and a diode in anti-parallel with the LED to discharge the dc block cap
on the off cycle.

The IR code for the shutter release works out to a 4% ratio of modulated
time to off time, and I'd only send the code once or twice giving it a 100
or 200 mS duration respectively.  Am I just working against average thermal
dissipation here, and not overheating the driver in the long run? or will
localized heating on the be an issue due to the prolonged pulse duration?
Anyone else use and abuse these in a somewhat similar manner?

Its a shame the D50 didnt come with a wired shutter release, just an IR
reciever, it's plenty capable otherwise, but the 100mS delay with the remote
(could be less, hopefully, there is 37 mS of pulses and then a 63 mS dead
time before the next pusle, ) makes it harder to catch fast events.  We'll
still be able to catch people mid recoil however though, just not the action
cycling if its a 100mS delay.

Thanks all,
--
Jonathan Hallameyer
--
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist

[ivp]

> 3W IR LED from deal extreme. It takes 1.5A @2v

Jonathan, common 5mm IR LEDs will handle 1.5A pulse current for
short periods. Use them with ZTX transistors, eg

http://home.clear.net.nz/pages/joecolquitt/irtx2.gif

--
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist

[ivp]

> Its a shame the D50 didnt come with a wired shutter release, just an
> IR reciever, it's plenty capable otherwise, but the 100mS delay with
> the remote (could be less, hopefully, there is 37 mS of pulses and then
> a 63 mS dead time before the next pusle, ) makes it harder to catch
> fast events.  We'll still be able to catch people mid recoil however though,
> just not the action cycling if its a 100mS delay.

I think many cameras will have perhaps 30ms from trigger to exposure
and the only way to catch the actual event would be with the shutter pre-
opened (and a front curtain flash if necessary)


--
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist

[ivp]

I wrote

> I think many cameras will have perhaps 30ms from trigger to exposure
> and the only way to catch the actual event would be with the shutter pre-
> opened (and a front curtain flash if necessary)

Oooops. Front curtain flash goes off when the shutter opens. The shutter
is already open, the flash would be triggered by the event

--
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist

[Jonathan Hallameyer]

Wow, that much from a 5mm IR LED? I knew they could take higher pulses but
I thought only a factor of 2-3 or so. Guess I'll give that a shot before I
use the star emitter.

Oh and yeah Ive seen various other methods of high speed photography, I'm
not looking to freeze frame water droplets hitting other water droplets or
anything, just get some action shots at the range. If the camera takes 150mS
from when the trigger box triggers to the shutter release, so be it, it'll
make for a neat electronics project. I might add an external input at some
time for other triggering methods.


Also, in the linked schematic, the representation of the pulsetrain looked
like it was a narrow duty cycle, do remotes typically use < 50% duty cycle
for the modulation?
Thanks.
-Jon
On Thu, Oct 15, 2009 at 2:29 AM, Jinx <joecolquitt@...> wrote:

> > 3W IR LED from deal extreme. It takes 1.5A @2v
>
> Jonathan, common 5mm IR LEDs will handle 1.5A pulse current for
> short periods. Use them with ZTX transistors, eg
>
> http://home.clear.net.nz/pages/joecolquitt/irtx2.gif
>
> --
> http://www.piclist.com PIC/SX FAQ & list archive
> View/change your membership options at
> http://mailman.mit.edu/mailman/listinfo/piclist
>

--
Jonathan Hallameyer
--
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist

[ivp]

> Wow, that much from a 5mm IR LED?

I just had a look at a typical 60mA IR LED. Pulsed current is 1.3A
for t = 10us. There are 5mm LEDs which can take 1500mA statically
(OP293 for example). I believe temperature is the determining factor
wrt the safety of the die and I think to some extent you can estimate
what's safe by working out the average wattage. It's not unusual to find
that eg 20mA LEDs in strobed applications like moving signs run quite
warm. The LEDs will have limiting resistors of perhaps 47ohms with
a 12V supply. Without strobing the LED would soon burn out with
this x10 current

Some IR controls do use pulses as short as 10us (which you could
use if the receiver can discriminate pulses that short), although I think
the common RC5 protocol is something like 1ms

--
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist

[Olin Lathrop]

Jonathan Hallameyer wrote:
> It
> takes 1.5A @2v. Now for driving that, could I get away with using a
> 2A peak mosfet gate driver?

The current the MOSFET is driving and the current requirement of the gate
driver are not much related.  2A is on the high end for gate drivers, and
should be able to overcome whatever gate capacticance there is quickly.  You
have rather low voltage and current requirements for the FET, so you can use
one with relatively low gate charge.  A much lower current driver should be
fine.

The IRLML2502 N channel FET can easily do this job, and can often be run
straight from a PIC pin.

> I was going to put a 28ish uF capacitor
> in series to keep from damaging anything if it got stuck on, and a
> diode in anti-parallel with the LED to discharge the dc block cap on
> the off cycle.

That's the wrong place for the diode to discharge the cap.


********************************************************************
Embed Inc, Littleton Massachusetts, http://www.embedinc.com/products
[phone].  Gold level PIC consultants since 2000.
--
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist

[Marechiare]

>> It takes 1.5A @2v. Now for driving that, could I get away
>> with using a 2A peak mosfet gate driver?
>
> The current the MOSFET is driving and the current requirement
> of the gate driver are not much related.  2A is on the high end for
> gate drivers, and should be able to overcome whatever gate
> capacticance there is quickly.  You have rather low voltage and
> current requirements for the FET, so you can use one with relatively
> low gate charge.  A much lower current driver should be fine.

Yes, 2A for the gate driver would mean that that's the current limited
by the driver's output resistance. That is, driver's output resistance
would be up to 10 Ohm, not of much use to drive 1.5A @2v.

--
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist

[Dwayne Reid]

At 07:11 AM 10/16/2009, Olin Lathrop wrote:

>Jonathan Hallameyer wrote:
> > It
> > takes 1.5A @2v. Now for driving that, could I get away with using a
> > 2A peak mosfet gate driver?
>
>The current the MOSFET is driving and the current requirement of the gate
>driver are not much related.  2A is on the high end for gate drivers, and
>should be able to overcome whatever gate capacticance there is quickly.  You
>have rather low voltage and current requirements for the FET, so you can use
>one with relatively low gate charge.  A much lower current driver should be
>fine.
>
>The IRLML2502 N channel FET can easily do this job, and can often be run
>straight from a PIC pin.

Hi there, Olin.

I may be wrong, but I think that what the original poster was asking
was: Can he use a Mosfet gate driver chip to drive his LED
directly?  In other words, the gate driver chip is being used to
drive the load directly without an intervening Mosfet.

dwayne

--
Dwayne Reid   <dwayner@...>
Trinity Electronics Systems Ltd    Edmonton, AB, CANADA
[phone] voice          [phone] fax
www.trinity-electronics.com
Custom Electronics Design and Manufacturing

--
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist

[Wouter van Ooijen]

Dwayne Reid wrote:
> At 07:11 AM 10/16/2009, Olin Lathrop wrote:
>> Jonathan Hallameyer wrote:
>>> It
>>> takes 1.5A @2v. Now for driving that, could I get away with using a
>>> 2A peak mosfet gate driver?

> I may be wrong, but I think that what the original poster was asking
> was: Can he use a Mosfet gate driver chip to drive his LED
> directly?  In other words, the gate driver chip is being used to
> drive the load directly without an intervening Mosfet.

If that is the question the OP should check the continuous allowed
current. My gut feeling is that it is maybe a factor 10 lower than the
peak current.

--

Wouter van Ooijen

-- -------------------------------------------
Van Ooijen Technische Informatica: www.voti.nl
consultancy, development, PICmicro products
docent Hogeschool van Utrecht: www.voti.nl/hvu

--
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist

[Marechiare]

>> I may be wrong, but I think that what the original poster
>> was asking was: Can he use a Mosfet gate driver chip
>> to drive his LED directly?  In other words, the gate driver
>> dchip is being used to rive the load directly without an
>> intervening Mosfet.
>
> If that is the question the OP should check the continuous
> allowed current. My gut feeling is that it is maybe a factor
> 10 lower than the peak current.

The driver's output resistance should be checked first, as for me.
Continuous allowed current may not be specified for such kind of
devices for their mission is to quickly and efficiently charge input
capacitance not to supply steady current.

--
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist

[Olin Lathrop]

Wouter van Ooijen wrote:
>> I may be wrong, but I think that what the original poster was asking
>> was: Can he use a Mosfet gate driver chip to drive his LED
>> directly?  In other words, the gate driver chip is being used to
>> drive the load directly without an intervening Mosfet.
>
> If that is the question the OP should check the continuous allowed
> current. My gut feeling is that it is maybe a factor 10 lower than the
> peak current.

And a gate driver will be bigger and cost more than a simple FET driven
directly from a digital output.


********************************************************************
Embed Inc, Littleton Massachusetts, http://www.embedinc.com/products
[phone].  Gold level PIC consultants since 2000.
--
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist

[Wouter van Ooijen]

> And a gate driver will be bigger and cost more than a simple FET driven
> directly from a digital output.

OTOH a gate driver is a push-pull, so a double gate driver (8-pin SOIC
or DIP) can make a H-bridge for a small motor. I am not sure I approve
of this use, but a Dutch PIC Basic site shows such a design, and I sell
a suspicious amount of TC4427A chips without any FET in the same order...

--

Wouter van Ooijen

-- -------------------------------------------
Van Ooijen Technische Informatica: www.voti.nl
consultancy, development, PICmicro products
docent Hogeschool van Utrecht: www.voti.nl/hvu

--
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist

[Marechiare]

>> And a gate driver will be bigger and cost more than a
>> simple FET driven directly from a digital output.
>
> OTOH a gate driver is a push-pull, so a double gate
> driver (8-pin SOIC or DIP) can make a H-bridge for a small
> motor. I am not sure I approve of this use, but a Dutch
> PIC Basic site shows such a design, and I sell a suspicious
> amount of TC4427A chips without any FET in the same order...

According to the datasheet TC4427A output resistance is typ 7 Ohm. At
1.5A it would consume 16W power and would require voltage 11V across
it.
A FET would consume only a small fraction of 1W. A push-pull
functionality is not needed for this thread's subject in my opinion.
--
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist

[ivp]

> A FET would consume only a small fraction of 1W. A push-pull
> functionality is not needed for this thread's subject in my opinion

For a simple remote control like the OP wants a low-side FET or
transistor is all that's needed. I suggest the ZTX transistors because
they are superior in this application IME to comparable transistors
(eg BC337, BC639). They have low saturation voltage and can
switch their rated current more quickly. For instance with battery-
powered transmitters with a reservoir cap, the leading edge of the
pulse is sharper. ie the battery's and/or capacitor's energy is initially
more efficiently dumped into the LED, giving a better range

--
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist