Turn-key adjustable current regulated DC-DC converters?

flash....

Enlightened
Joined
Sep 16, 2002
Messages
366
Location
Winter Springs FL
I have done a few searches and have not been able to find a Turn-key adjustable current regulated DC-DC converter solution for some of my planned LS light mods I wish to build.
I would like something that will take 2.5 to 12VDC (ball park) and have an adjustable regulated current output... (upto 1.5A would be nice
grin.gif
)
I like to run My LS 1 Watt projects usually between 400 - 600ma.

Where can I find such a beast that is small, efficient and cost effective? Am I asking to much?

Any help would be greatly appreciated.
yellowlaugh.gif
 

dat2zip

Flashlight Enthusiast
Joined
Jan 5, 2002
Messages
3,420
Location
Bay Area
flash....,

I am working on a converter that will fit your bill.

It's not available right now but here it is in a minimag driving a 1W Luxeon. I call it the Wizard.

Wizard in minimag

Wayne
 

flash....

Enlightened
Joined
Sep 16, 2002
Messages
366
Location
Winter Springs FL
That looks really cool, how do you adjust the current ?? Do you have specs? when do you think you will go into production?
Does anyone make an off the shelf part?

Sorry for the 20 questions... I have been building resistor limited LS's so far and I'm ready to go regulated.

What would be really cool would be to have taps that one could wire to a switch for High\Low beams.
Maybe 350ma-Low to 600ma High or something similar.

Thanks..
 

PsycoBob[Q2]

Enlightened
Joined
Apr 9, 2002
Messages
518
Well, I have a 2aa Light that needs a refit- it has a pushbutton switch, almost like a miniture 2d Maglight, or something. The switch assembly was poorly designed, and needs a modded one anyway.
 

Scoper

Newly Enlightened
Joined
Aug 9, 2002
Messages
46
Location
Canada
Check out the MAX1698. If your voltage may drop below 2.7 volts, you will have to find a way to pwer the chip, althought the input voltage can be as low as 0.8 volts.
 

flash....

Enlightened
Joined
Sep 16, 2002
Messages
366
Location
Winter Springs FL
Thanks Scoper...
yellowlaugh.gif

But I was hoping for a complete turnkey item.
You know, 4 leads\contacts (2 in and 2 out) from a little black box. Black potted epoxy possibly?
Basically Vin 2.5 to 12V and Vout that is current and voltage load regulated by way of two very small adjustable potentiometers. One for Voltage and one for current. (or do this with one pot if possible)
I want to buy this. No fuss. just hook it up, set voltage, use an ammeter in series with load and dial in desired current.
Small,
Efficient,
Pipe dream???
Someone's gotta make one (or similar) ... custom perhaps?
confused.gif
 

Steelwolf

Flashlight Enthusiast
Joined
Feb 6, 2001
Messages
1,208
Location
Perth, Western Australia
flash....

We've all been asking Peter Gransee to sell us the circuits he uses in the Arc-AAA and Arc-LS as a small, black, epoxy-potted, "pill" for our mods and home projects. No dice yet, but that's understandable.

If you have the capability, the Zetex circuit being discussed in another thread may be what you are looking for. Only uses 7 other components aside from the chip. Maybe you can make them?
 

Jonathan

Enlightened
Joined
Dec 14, 2001
Messages
565
Location
Portland, OR
I've just finished a PCB layout for a current regulating buck converter, using the LT1767 and the ZXCT1010. A pot on the feedback for the ZXCT1010 could be used to adjust the output current. Maximum input voltage is 25V, maximum output voltage is 20V, maximum output current is about 1.3A. Somewhat larger than the 'wizard', but still pretty darn small.

It won't work for boost operation, the input voltage needs to be greater than the output voltage.

-Jon
 

Illuminated

Enlightened
Joined
Oct 26, 2002
Messages
886
Location
Tipp City, Ohio
Jon,
That sounds great! Any idea on efficiency levels? I'd be interested in buying such a circuit to drive 1 or 2 1-watt Luxeons efficiently from (4) NiMH's, or even a 5-watt from 7 or 8 cells (once the "good" 5W become more available).

Thanks - John
 

Jonathan

Enlightened
Joined
Dec 14, 2001
Messages
565
Location
Portland, OR
My best guess at efficiency would be > 80%, but I have not done any sort of spice simulation, nor have I build one to try out
smile.gif


Oh, a couple of more features. The chip has a shutdown pin that has an accurate voltage reference, so you can use it as an undervoltage lockout to protect rechargeable cells from over discharge, and I made my design single sided using larger components, so I'm pretty certain that it is larger than the 'wizard'...the PCB works out to 0.65 x 0.8 inches.

-Jon
 

Scoper

Newly Enlightened
Joined
Aug 9, 2002
Messages
46
Location
Canada
Flash...

If you find what you are looking for *please* let me know. It would move me months ahead in my project. In fact, this is exactly what I came to CPF to find in the first place.

Although, I really want the one hanging right beside it in the display case. The one that takes an input voltage of 3.2V down to 1.8V or less for a 2 cell configuration.
smile.gif
 

flash....

Enlightened
Joined
Sep 16, 2002
Messages
366
Location
Winter Springs FL
Jon...
That's awesome!!
Will you go into production with these??
Can I buy a few from you when your ready??
The step down works well for me as I want to use 12v sources for 5W LS's and 6v-12v for 1W LS's.

Thanks..
 

moraino

Enlightened
Joined
Jan 20, 2002
Messages
447
Location
B.C. Canada
Hi everyone,

Just want to get your attention here. I wonder if anyone has taken a look at this chip. Hey, it claims 97% effeciency...

http://www.national.com/pf/LM/LM2653.html

Parametric Table
Master No
Multiple Output Capability No
On/Off Pin Yes
Error Flag Yes
Input Voltage Min (Volt) 4
Input Voltage Max (Volt) 14
Output Current Max (mA) 1500
Adjustable Output Voltage Yes
Switching Frequency (Hz) 300000
Adjustable Switching Frequency No
Sync Pin No
Efficiency (%) 97
Step-Up No
Step-Down Yes

The Perfectlight uses it. http://www.resurgentsoftware.com/perfect_led_light.html

If you don't like it, could you also explain why? I am planning getting a sample board for $24.99 just want to throw it in my lantern so I can use any combination of batteries. Thanks for any reply.

Henry
 

Doug S

Flashlight Enthusiast
Joined
Jun 20, 2002
Messages
2,712
Location
Chickamauga Georgia
Originally posted by moraino:
Hi everyone,

Just want to get your attention here. I wonder if anyone has taken a look at this chip. Hey, it claims 97% effeciency...

http://www.national.com/pf/LM/LM2653.html

Parametric Table
Master No
Multiple Output Capability No
On/Off Pin Yes
Error Flag Yes
Input Voltage Min (Volt) 4
Input Voltage Max (Volt) 14
Output Current Max (mA) 1500
Adjustable Output Voltage Yes
Switching Frequency (Hz) 300000
Adjustable Switching Frequency No
Sync Pin No
Efficiency (%) 97
Step-Up No
Step-Down Yes

The Perfectlight uses it. http://www.resurgentsoftware.com/perfect_led_light.html

If you don't like it, could you also explain why? I am planning getting a sample board for $24.99 just want to throw it in my lantern so I can use any combination of batteries. Thanks for any reply.

Henry
<font size="2" face="Verdana, Arial">Henry, a few months ago I had quite a lot of off list communication assisting another CPF member who wanted to use this eval board to power a caving headlamp he was building. He was very pleased with the performance. As I am sure you know, you will have to change the value of R1 to get the output voltage you want. This is a surface mount resistor. If you are not experienced or equipped to work with surface mount stuff, it can be a bit daunting the first time but its doable. I note that this thread started as a constant current regulator thread. The LM2653 is a constant voltage output device. If the output is fed directly into a Luxeon, the current will increase as it heats up, this can be a rather large shift specially if you are driving it hard and if the dynamic resistance of the particular Luxeon is low. This effect can be reduced somewhat by adding a small amount of resistance in series with the LED. In the case of the particular light that I helped with, the builder was driving 3 1W Luxeons in parallel. We added a separate 0.51 ohm resistor in series with each LED. This not only helps to stablize the current w.r.t., temperature but also helps the 3 LEDs more equally share the total current. Other details about this particular lamp: LM2653 voltage and added resistance chosen to get about 200 mA/LED. Power source is 8 AA cells in a 4X2 series/parallel combination. Maybe this CPFer will drop in and tell you more about his light. I have only seen emailed photos, I have yet to see it in person as it living in another State.
 

Jonathan

Enlightened
Joined
Dec 14, 2001
Messages
565
Location
Portland, OR
I took a look at the 2653.

The high efficiency comes from the synchronous rectification (much lower 'catch diode' losses) and has switches with low Rds_on (0.1 Ohm)

The downside is the lower switching frequency, which means larger inductors and capacitors. There is an implication of a maximum output voltage of 5V, but I don't see anywhere in the datasheet that confirms this.

As was mentioned, this is a voltage regulated device, and to run an LED you want a current regulated output...but a ZCXT1010 fixes this, or at least should
smile.gif


Working with surface mount components is not too difficult; I just use an ordinary Weller soldering pen (the low end temperature regulated station). After a bit of practise, you will find the SMT stuff easier to do than through hole components.

-Jon
 

Doug S

Flashlight Enthusiast
Joined
Jun 20, 2002
Messages
2,712
Location
Chickamauga Georgia
Originally posted by Jonathan:
I took a look at the 2653.
The downside is the lower switching frequency, which means larger inductors and capacitors. There is an implication of a maximum output voltage of 5V, but I don't see anywhere in the datasheet that confirms this.

-Jon
<font size="2" face="Verdana, Arial">That 5V appears on page 1 of the datasheet but not in the "real specifications" part of the datasheet. The datasheet leaves at bit to be desired in this area but I believe that higher outputs would cause no harm. The amount of Vcb boost is not a function of output voltage so adequate drive for the "top" Mosfet is not an issue. In the case of the LM2653 eval board, the supplied output cap is only rated at 6.3V so this would be a limit unless you changed out the cap.
 

dat2zip

Flashlight Enthusiast
Joined
Jan 5, 2002
Messages
3,420
Location
Bay Area
Just FYI,

I expect my first proto boards for my latest creation, Badboy Wizard. Yes, this board can be configured as a step up (Badboy) or a dual mode step up/down converter. It's still the same 0.55" diameter, backward compatible connectors for use with the emitter board (Sandwich) and with minor component changes support input voltage to 16 Volts.

I expect them friday or early next week. I'll be building a few for testing and for another project. If the board works out as expected I should be making them in fairly large quantities in the near future.

The smaller board size means it will fit in more mods. The downside it is much harder to work with and doesn't include mounting holes or other niceties.
 

flash....

Enlightened
Joined
Sep 16, 2002
Messages
366
Location
Winter Springs FL
dat2zip...
That also sounds great...
any specs?? I am so ready to buy some of these as I think I'm getting a couple of 5w LS's real soon.
Can I get kits from you??
Any other specs?

Thanks...
 
Top