Zetex ZXSC300 single-cell LED driver

Status
Not open for further replies.
D

**DONOTDELETE**

Guest
I'm wondering if anyone has implemented the max-brightness single-cell LED driver circuit described at http://www.zetex.com/3.0/pdf/ZXSC300.pdf

Although the circuit is designed for the white Nichia NSPW500BS, I wonder if it would drive an amber Luxeon Star with R1 set to 0.04 ohm, which should pulse the LED at 475 mA.
 
<BLOCKQUOTE><font size="1" face="Verdana, Arial">quote:</font><HR>Although the circuit is designed for the white Nichia NSPW500BS, I wonder if it would drive an amber Luxeon Star with R1 set to 0.04 ohm, which should pulse the LED at 475 mA. <HR></BLOCKQUOTE>

Not sure myself. But, I think one might also have to change that one transistor as well. I beleive that National also has something in the same line. I went to the Zetex site to take a look and perhaps request a couple of samples. Not an easy thing to do. At least not as easy as getting samples from National.
 
<BLOCKQUOTE><font size="1" face="Verdana, Arial">quote:</font><HR>Originally posted by Duggg:
I'm wondering if anyone has implemented the max-brightness single-cell LED driver circuit described at http://www.zetex.com/3.0/pdf/ZXSC300.pdf

<HR></BLOCKQUOTE>

That's the exact chip i was talking about in
other threads :)
The minimum quantity for ordering around
here is 3000 pieces, which runs upwards of
about $1500.00 .
I inquired about getting smaller quantities,
but didnt get a response yet. It wont
help me that much to get samples if i cant
also buy in smaller quantities, because i
wont want to shell out 1500 bucks for these
chips.

The National chip is a charge pump device,
thats why there is no inductor.
The drawbacks to the National chip are:
1. input min voltage is 3.0 volts.
2. drives a max of 2 LED's.

Maxim has charge pump ic's that can
drive 2 LED's at lower input voltages,
if thats all you want to do.

So far, the Zetex chip is the best i've seen
so far. It does exactly what you would want
in an LED driver. You can even expand it
to power any number of LED's just by using
a different NPN transistor with it.
This is certainly my choice, but getting
some is another problem :)
Anyone have any luck with this company?

--Al
 
Yeah, I should have checked www.digikey.com before posting---they have a 3000-piece reel for sale for $1620 ($0.54 each), but if you buy just one ($1.05 each) you'll have to wait until January.

But, that may be okay---that's just in time for the release of the new white Lambertian Luxeon Stars, which promise to be awesome devices.

And, who knows---maybe DigiKey will cave in and peel off a few hundred chips from the reel for individual sale, if there's enough demand.

I like DigiKey: if you order $25 worth of stuff there's no handling charge, and if you send them a check with your order, there's no domestic shipping charge.

For $25 you can get the raw parts for both the Zetex solution AND the LM2621 solution, plus a bunch of other parts.
 
<BLOCKQUOTE><font size="1" face="Verdana, Arial">quote:</font><HR>Originally posted by Duggg:
...... wait until January.

But, that may be okay---that's just in time for the release of the new white Lambertian Luxeon Stars, which promise to be awesome devices.
............
<HR></BLOCKQUOTE>

Duggg, what are these promising Lambertian Luxeon Star thingies ? Specs please !

lightlover

(And that's about the only thing I understood while reading through this topic ...... )
frown.gif
blush.gif
shocked.gif
wink.gif
 
<BLOCKQUOTE><font size="1" face="Verdana, Arial">quote:</font><HR>Originally posted by *something ridiculous:
Duggg, what are these promising Lambertian Luxeon Star thingies ? Specs please !
lightlover
<HR></BLOCKQUOTE>
Other people on this forum call 'em "high-dome or hi-top" -Stingmon has quite a bit on this 2nd gen. Luxeon module on his site.

Doug
 
Dugg & MrAl

I guess I should have read through the datasheet again on the National IC.

The Zetex does sound ideal. Since I'm in the silicon valley area, I try asking around here at some of the parts houses and see if I can somehow get some samples out of Zetex.

I've got some National LM1117's coming now since I'm interested in a current limiting circuit.

Mercator
 
<BLOCKQUOTE><font size="1" face="Verdana, Arial">quote:</font><HR>Originally posted by Duggg:

but if you buy just one ($1.05 each) you'll have to wait until January.

I like DigiKey: if you order $25 worth of stuff there's no handling charge, and if you send them a check with your order, there's no domestic shipping charge.

For $25 you can get the raw parts for both the Zetex solution AND the LM2621 solution, plus a bunch of other parts.
<HR></BLOCKQUOTE>

Soundsss gooddd Duggg :)
Thanks for that info.

But what is this about waiting until January?
Why January?

--Al
 
Digikey's web site says individual ZXSC300's (part #ZXSC300E5CT-ND) won't ship until January 3, 2002. The page tells you this if you enter "1" and press the Quantity Requested button.

I read in another thread that LumiLeds will start producing the white Lambertian LS's at the end of the year. The batwing (aka "low dome") versions of the white LS are available now, but the batwing radiator doesn't produce as pure a beam as the Lambertian.
 
Dugg & MrAl

I'm still interested in the Zetex IC, I looked through the Zetex site again. They also have an IC simular to the one in this thread. It's the ZXCS100. Digikey has them in stock now for $1.75 each. Perhaps one of you can take a look and see if this other IC specs out to something close enough to the ZXCS300 number. They (Zetex) has an Application Note that addresses using this IC for single cell LED flashlight apps. Search for An-33.
BTW - Digikey has the ZXCS100 both in SO8 and MSOP-8 packages. I think that the SO8 is bigger and therefore easier to work with. I'm not much into designing circuits, but certainly can build then from a schematic.

This series of Zetex IC's could very well be what is being used in the Infinty and possibly the ARC AAA. Not sure about the ARC's "sun/moon" mode though.

Mercator
 
Yes, I saw the ZXSC100 too, but I rejected it mainly because of the efficiency difference of 82% versus 94%, and the 1-watt circuit (needed to drive the LS at full brightness) was considerably more complex than its 300 counterpart. (Note the 100 has twice the pins of the 300.)

But, you certainly have good points in that the 100 is available right now, and the SO8 is easier to work with than the SOT23-5, although the extra pins take away some of the advantage.

Still, I'm an efficiency freak so I'm holding out for the 300
smile.gif
 
<BLOCKQUOTE><font size="1" face="Verdana, Arial">quote:</font><HR>Originally posted by Duggg:
Yes, I saw the ZXSC100 too, but I rejected it mainly because of the efficiency difference of 82% versus 94%, and the 1-watt circuit (needed to drive the LS at full brightness) was considerably more complex than its 300 counterpart. (Note the 100 has twice the pins of the 300.)

But, you certainly have good points in that the 100 is available right now, and the SO8 is easier to work with than the SOT23-5, although the extra pins take away some of the advantage.

Still, I'm an efficiency freak so I'm holding out for the 300
smile.gif
<HR></BLOCKQUOTE>

Yeah i saw that 100 chip too. I guess im
just not in that much of a hurry anyway
smile.gif

Im looking at a ton of different circuit
configs right now. I found some interesting
stuff out there, but the myriad of possibilities takes a while to get through.
I think it will be hard to top the Zetex
300 chip though, except possibly with
a combination chip plus on chip
heavy duty Mosfet
and still keep all of the specs of the 300.

--Al
 
I asked afew of my parts sources around here this morning and it's looks like they don't buy directly from Zetex. So, that avenue didn't pan out for me.

I wrote Zetex directly and asked about purchasing samples. I received a reply this afternoon. Thay wanted my address so that a sales rep could send me information regarding samples. I wrote back, so we'll see what comes of that angle. I've got a feeling though that the sales rep is going to want to pay me a visit. Too bad Zetex isn't as easy to work with like National.

Both of your are right regarding the advantages of the 300 over the 100 chip. I sure would like to find something to build my LS around. I think the way to go is with a PWM circuit driving the LS at near max. I've been following the thread on the modified Brinkamn circuit and would like to see that simple circuit tweaked to LS specs. I would think it's possible once the right combination of componets are discovered. Perhaps the simple Sature design might work, but the drawback is the hand wound inductor. It's still a great design within it's limits.

I'll keep everyone posted if I have any luck getting samples out of Zetex. If I could buy a handful, I'd sell or trade them at cost.

Mercator
 
<BLOCKQUOTE><font size="1" face="Verdana, Arial">quote:</font><HR>Originally posted by Mercator:

Too bad Zetex isn't as easy to work with like National.

I've been following the thread on the modified Brinkamn circuit and would like to see that simple circuit tweaked to LS specs
Mercator
<HR></BLOCKQUOTE>

Yeah Zetex seems to be a pain, too many
sleeping employees :)

What are the LS specs?

--Al
 
<BLOCKQUOTE><font size="1" face="Verdana, Arial">quote:</font><HR>Originally posted by Duggg:
The spec sheet for the Luxeon Stars can be found at http://www.luxeon.com/pdfs/Luxeon_star_DS.PDF<HR></BLOCKQUOTE>

Thanks for the link Duggg.
They are spec'ing the white LED light output
at 180 candela, while a 'regular' led is
about 5.6 candela. A little math says this
is over 32 times as bright as a regular white
LED. Is this correct? How can they get that
much brightness out of a single LED?
Wouldnt this make the brightest flashlight?
(even if it draws more current).

--Al
 
Comparing LEDs on the basis of candela output alone is like comparing how big homes are based on the size of their front yards.

Candelas measure the peak brightness, whereas lumens measure the total brightness.

A better way to do the comparison is to multiply the electrical efficiency (in lumens per watt) by the power dissipated.

According to http://www.misty.com/people/don/led.html the Nichia puts out 16 lumens per watt and the Luxeon puts out 18.

The Nichia can dissipate 120mW and thus produce 1.92 lumens.

The typical Luxeon draws 350mA at 3.42 volts for a total power of 1197mW and thus produce 21.546 lumens.

So the Luxeon should be equivalent to 11.22 Nichias in total output---assuming the 18 lumens/watt figure is correct.

Of course, the Luxeon is physically much larger than the Nichia. The Star/O collimator is 21.5 mm in diameter, for a total frontal area of 363 mm^2.

The Nichia comes in a typical 5mm package for an area of 19.6 mm^2.

You could theoretically cram 18.5 Nichias into the Luxeon's collimator, although in practicality, 12-14 is probably the best you can do.

So, keeping flashlight size the same, a 12-14 Nichia flashlight should be a tad brighter, but note that neither of these comparisons take heat into account. Both LEDs will produce much less lumens per watt if the heat is not properly dissipated, and I think heat would be more of a problem for the Nichias in such close proximity.

So I think the true winner would be the Luxeon, and that's why I'm excited about it.
 
<BLOCKQUOTE><font size="1" face="Verdana, Arial">quote:</font><HR>Yeah Zetex seems to be a pain, too many
sleeping employees :)

What are the LS specs?<HR></BLOCKQUOTE>


I heard back from Zetex today. They provided me with my local rep's phone number. I'll try and call them in the morning. If I can get them to sell me several samples at a couple of bucks apiece. I'm going to go for it. If anyone wnats some for themselves, let me know and we'll work the details out later.

MrAl... for some reason I thought you already owns an Luxeon Star. Dugg's explaination is correct regarding the equivalent light output. There are several advantages in favor of the LS compared to the equivalent number of single LED's. Top on the list are overall cost and packaging. I've bench tested my two LS modules using different batteries, voltage and currents. The more I play with the LS the more impressed I am. The one thing I have yet to try is a circuit that will drive the LS with PWM at or near the spec'd max of 500mA. With non-oscillating voltage/current the LS does get quit warm at max input. I suspect that with a plused input the heat problem would not be as severe.

Although I/m interested in the Zetex IC, I'm not sure if it will be the right choice for the lithium 3.6v batteries I want to run the LS on. Without any regulation one of these batteries could drive the LS at over 800mA.
I've limited this high current with a resistor, but of course that has major drawbacks regarding efficiency throughout the battery discharge cycle. I figure that with a PWM circuit running at about 450mA would deliver a flatter level of light output for a longer time verses just opperating off the battery and a dropping resistor.

I've mentioned this before... I'm not all that knowledgable to design my own circuits, so I guess I'm pretty much waiting for something to come along like the Zetex chip.
So far from what I seen and read, it looks like it might fit the bill.

Mercator
 
Status
Not open for further replies.
Top