LUXEON STAR, MAX756 ONE INCH DIAMETER REGULATOR

[ElektroLumens]

Here are some results of a new little regulator I just put together. It is just one inch in diameter, and uses the MAX756 8 pin DIP IC. I started with a single not real fresh 'C'battery. It lasted 5 1/2 hours before I shut it down.


12/19/2001

TIME Vin Vout
9:25 1.306 3.085 Start (Battery not completely fresh)
1.53 .915 2.937
2:16 2.928
2:38 2.916
2:50 .803 2.898
2:57 .775 2.878
3.08 .641 2.814 Shutdown, but still shining.

5 hours, 43 minutes on one 'C' battery.


I hope to do this same test with 2 'AA's and with 2 'C' batteries, and also do amperage testings as well.

The output of this one inch disc is a little less than with the 1" X 2" board I have, probably because of the small surface mount capacitors being used.

 

[jeff1500]

That's very good work. Look how well the chip stepped up and regulated the voltage, and how far down it drained the battery. Batteries are considered dead at .9 volts. This circuit took it down to .641

 

[MrAl]

Sounds good wayne, what kind of inductor are
you using?

--Al

 

[**DONOTDELETE**]

I've been following all the posts on the zetec regulator, but have not seen much info on the chip your using, could you possibly post some info on it and any schematics you may have.
Thanks

 

[ElektroLumens]

<BLOCKQUOTE><font size="1" face="Verdana, Arial">quote:</font><HR>Originally posted by MrAl:
Sounds good wayne, what kind of inductor are
you using?

--Al<HR></BLOCKQUOTE>

Hello Al,

I am using a 100uH surface mount inductor I bought from DIGIKEY, part # M9712CT-ND, and the inductors are the Kemet SMD 150uF caps from Mouser.

The efficiency I computed on this 1" board is 73%. Amps in = .69, volts in = 2.288, Amps out = .350, volts out = 3.296.

I also tested a MAX757 board I have. It measures 1" X 2", and I measured 76% efficiency from it. I am using 220uF caps, and the same inductor. Amps in = .74, volts in = 2.290, Amps out = .390, volts out = 3.347.


The MAX757 seems to have more oomph!

I want to do a test with 2 AA's, and then with 2 'C' batteries, and follow the current instead of the voltage. Assuming my calculations are correct. You can verify my calculations for me if you want. I think I hope I finally figured it out. Thanks for your help.

I think 2 batteries will give more satisfactory output and brighter light. I just wanted to see what would happen with just a single battery.

 

[ElektroLumens]

<BLOCKQUOTE><font size="1" face="Verdana, Arial">quote:</font><HR>Originally posted by Bill Aquino:
I've been following all the posts on the zetec regulator, but have not seen much info on the chip your using, could you possibly post some info on it and any schematics you may have.
Thanks<HR></BLOCKQUOTE>

Hello Bill,

The MAX756 is an older chip that's been around for awhile. I've been using the 8 pin DIP size, but there is a SO, smaller chip. I initially chose it because it is a larger chip and is easier to work with. It is rated at a max of 300mA, but with some tweaking more current can be obtained from it, especially if a higher input is used. I will give a link here, for you to read more about it. The MAX756 requires few external components: a inductor, a schottsky diode, 3 capacitors. There are newer chips with better efficiency, if you can solder tiny parts (I haven't figured out how to solder the smaller parts yet, but I'm working on it.)

MAX756 Documentation

 

[**DONOTDELETE**]

Hi Wayne,

Congrats on getting the MAX756 to work. I agree that the 8-pin DIP package is a lot easier to work with, though I'm getting better with this SMD stuff.

Are you using the Kemet T520 polymer tantalum capacitor for your 756 circuit? If so, I would be curious to find out if replacing it with an ordinary 220uF electrolytic lowers the efficiency, and by how much.

 

[ElektroLumens]

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

Congrats on getting the MAX756 to work. I agree that the 8-pin DIP package is a lot easier to work with, though I'm getting better with this SMD stuff.

Are you using the Kemet T520 polymer tantalum capacitor for your 756 circuit? If so, I would be curious to find out if replacing it with an ordinary 220uF electrolytic lowers the efficiency, and by how much.<HR></BLOCKQUOTE>

Duggg,
I'm using the Kemet tantalum capacitor you guys recomended on your Zetex thread. I saw it and thought I'd give it a try, in order to reduce the size of the board. I thought because I was able to place the capacitors so close to the regulating chip, and that they were tantanlum, the ripple would be reduced and the efficiency improved. Threre may be some factors involved here as to why I only got 73% efficiency with the 1" board. Surprisingly, with the larger MAX757 board, I did use the 220uF aluminum electrolytic caps, and it had a higher efficiency of 76%.

Size of the board was the major consideration here. Now, I need to work on efficiency, and try to get it up. The larger capacitors will in no way fit on this little board, which by the way, is only 3/8 tall. (The pin components are on one side, the surface mount are on the other side.)

I am planning to make a batch of about 20 one inch disc MAX757 variable voltage regulator boards in the near future, and sell them pretty cheap, (Not sure how much?) if there is an interest in them, here on the CPF? The variable voltage will have to be preset with small resistors, as there will be no room for a trim pot, as I can see at this time.

With these small circuit boards, it will make it easier to make Luxeon Star mods, or even Nichia flashlight mods. If we want 3.6, or 4.0, or 3.2, 3.4, whatever, with the MAX757 we can do it.

 

[MrAl]

Hi there wayne,

The results you got with the Max756 sound
pretty good, i'll have to get back to that
ic one of these days and do some tests at
higher current outputs.
The Zetex circuit is coming along pretty
good too and eventually it will be super
small.

One word of caution when experimenting
with the Luxeon Star though: the manufacturer
specifies a maximum pulsed current of
500ma peak. If you go over that for any
reason you take the chance of destroying
the LED or limiting its useful life or
causeing premature reduced light output.
Its not hard to do because even a small
change in ripple voltage could lead to
a large enough change in peak current that
the LS will see more then 500ma on a repetitive basis. Since the dynamic
impedance of the LS is so low, even a 100mv
increase in voltage would lead to 120ma
increase in output current. Now if your
already operating at an average current
of 350ma, adding 120ma to that sums up to
470ma, which is close to the max spec.
If this is the case, your probably operating
at a safe enough level, but if you have
150mv on top of 350ma average current you
end up going over the max spec by 30ma, so
you might see an early light output
degradation due to high peak current in
the LS. It pays to check this with the
network described in the Zetex thread.
One simple measurement, and you can be
much more confident that your circuit will
always provide a safe, reliable operating
current for your LS.

Good luck with your LED circuits,
Al

 

[ElektroLumens]

Hello MrAl,

As always, I appreciate your comments and advice. Yes, I have read the comments on ripple in the Zetex thread. Very helpful. I will definately check for ripple, and will probably follow some of the ways you have mentioned, to try to be sure ripple doesn't cause a proble.

Several ideas come to mind, one is to use a .1uF cap in series with the Vout cap. Or using the inductor on the Vout. I will experiment with these ideas. As also I hope to make the ripple testing circuit you had outlined.

One modification I have read for the MAX757 is to use two 1000uF 16V caps in parallel. The two caps help reduce ripple, if I understood the mod correctly. With 4.5 volts in, it can supply 5 volts out, at 1.5 amps! I hope to test this at some point.

Again, thanks for your advice.
Wayne