I have a project that needs a simple circuitry powering up a LED (preferably extra bright white), which requires around 3.5V. Due to the fact that it is a compact unit, I have to use a 1.5-3V coin battery (probably 1.5V). Does anyone have suggestions on how to step up the voltage to the needed value of 3.5-4 V?

 

Use this:
http://www.national.com/mpf/LM/LM3355.html

 

Why not stack two coin units on top of each other, it's a lot simplier. You probably wouldn't need anything extra in the way of resistors.

In this case voltage isn't critical, the batteries are self limiting (as long as we're talking coin cells).

 

Stacking two 1.5V cells will not provide 3.5V.

There is a simple project called The Joule Thief that steps up 1.0V to 1.5V up to enough voltage to light a white LED. Look in Google.

 

Actually coin cells come in two voltages, 3.0V is the other voltage. 2025's and 2032's both come to mind, the latter used for a lot of computer motherboards for CMOS, and both available at any Walmart. They have both been used for exactly that LED application too, although as memory serves a single 2032 worked just fine.

It is all too easy to overengineer this application, when off the shelf tech works just fine.

 

Thank you for all of your replies!

Bill, the reason I am leaning towards 1.5V is that alkaline coin cells come only in the 1.5. This device is disposable and for hospital use. From what i understand, Lithium and Silver Oxide batteries are not hospital friendly. I cant stack up batteries because of weight constraints - every gram counts. (one alkaline coin is ~2.2g)

Audioguru, I liked the gadget, but i cant use this because it would be too bulky for my device.

Theamber, thanx for the link but i have to use 1.5V for input.

Looks like I need to move towards "DC-DC converter + capacitors" or "op-amps + capacitors". Does anyone have an idea for such circuit?

 

Alek,
Take a look at the "Joule Thief" on this page:
http://www.bigclive.com/joule.htm
It's very compact, but I have built them smaller.

 

Dang! A picture is worth a thousand words indeed. Looking at it I figure there is less than a $1 worth of parts, excluding the LED. Does anyone make those comercially?

 

Thank you for all of your replies!

Bill, the reason I am leaning towards 1.5V is that alkaline coin cells come only in the 1.5. This device is disposable and for hospital use. From what i understand, Lithium and Silver Oxide batteries are not hospital friendly. I cant stack up batteries because of weight constraints - every gram counts. (one alkaline coin is ~2.2g)

Audioguru, I liked the gadget, but i cant use this because it would be too bulky for my device.

Theamber, thanx for the link but i have to use 1.5V for input.

Looks like I need to move towards "DC-DC converter + capacitors" or "op-amps + capacitors". Does anyone have an idea for such circuit?

Take a look at charge pumps from On-Semi (MAX1720, MAX828/9), TI and Maxim.
Also look at the TI TPS61xxx series of converters.

(* jcl *)

 

Dang! A picture is worth a thousand words indeed. Looking at it I figure there is less than a $1 worth of parts, excluding the LED.

Actually, less than that INCLUDING the LED.
I bought several miniature ferrites for $0.05/each from Skycraft Surplus, some SMT 2N3904's (MMTD3904) for around $0.04/each, SMT resistors for less than a nickel each, and super-bright LED's for less than $0.20/each. A friend gave me a large spool of 34 gauge magnet wire in the early 80's and haven't made much of a dent in it, so not quite sure what cost would be on that nowadays. There are roughly 8,000 ft per lb.

Yikes! Just checked prices and holy cow has it gone up. Around $60/lb at Allied Electronics if you just buy one 1lb spool at a time. I could only get 7 pairs of turns on the tiny ferrites I used; had to increase the base resistor to 2.7k to avoid saturating the ferrite cores, but the LED was still very bright even running from a single watch battery. So perhaps 2 ft of wire per assembly. At $60/8000ft, that's less than $0.08 per assembly.

So, about $0.42/each. YMMV.

Does anyone make those comercially?

I wouldn't be surprised to find something like this as part of a larger device, but I haven't seen them being marketed.

 

John, thanks for the charge pump idea - it is now one of my options on table.

SgtWookie, wow, that Joule Theif is now looking really attractive. I need to check it with my patent agent. Meanwhile, do you know of a stock item I can use to replace the wired ferrite?

 

I hadn't bothered to look actually - but Coilcraft and a number of other manufacturers make miniature coupled SMD inductors (leaded ones as well) that would work with the Joule Thief. They can be remarkably small; much smaller than I could make with my fat fingers. The SMD inductors could be pick and placed on a small PCB.

Of course, the smaller the transformer, the higher the frequency you'll need to run it. Shielding will be a must to avoid interference with other instruments in proximity.

 

I need to check it with my patent agent.

I don't believe anything discussed on these forums is protected by non-disclosure agreements. You may want to check that aspect with your patent agent.

As for silver oxide batteries not being hospital friendly, as stated in an earlier post, can you reference that? What is the issue, compared to NiCd or NiMH?

John

 

I don't believe anything discussed on these forums is protected by non-disclosure agreements. You may want to check that aspect with your patent agent.

As for silver oxide batteries not being hospital friendly, as stated in an earlier post, can you reference that? What is the issue, compared to NiCd or NiMH?

John

I am actually trying to validate the "hospital friendly" part in parallel with circuit design. I cannot quote on silver oxide battery not being hospital friendly. However, I've heard that silver oxide can be toxic if burned and many larger hospitals burn their solid waste. Also, the device I am working on is disposable (maybe few per day will be disposed of). So, I bench marked similar disposable products and they all had alkaline batteries. I could be wrong in my assumptions and would love to hear any suggestions.

 

You are aware LEDs contain Gallium Arsenide (actually they are Gallium Arsenide)? The only reason most semiconductors are not considered toxic waste is they are permanently encapsulated, and do not present a hazard because thier packaging is forever.

 

In my experience, most hospitals have mandatory collection of batteries. It is simply impossible to separate those that are safe to burn from those that might present a slightly higher hazard, if burned.

As for burning hospital waste, some places separate plastic from paper. The paper is fine to burn day or night. The plastic is burned only at night...guess why. Ever burn polystyrene? There is so much more potentially hazardous material being burned, I doubt that a few batteries that might make it past the battery police would create much of a problem.

Of course if it is a regulation in your locale, it must be followed. I was just not aware of any regulation restricting use of silver oxide batteries in hospitals.

John

 

You are aware LEDs contain Gallium Arsenide (actually they are Gallium Arsenide)? The only reason most semiconductors are not considered toxic waste is they are permanently encapsulated, and do not present a hazard because thier packaging is forever.

Ah Oh... OK, thanx, nice to know.

 

Yikes! Just checked prices and holy cow has it gone up. Around $60/lb at Allied Electronics if you just buy one 1lb spool at a time.

Yet one more reason to avoid Allied like the bandits they are! Many internet sites offer magnet wire for $7 to $30 per pound, depending on quantity, gage ,and temperature rating.

 

Wish I had grabbed some literature, but I was several companies selling magnet wire at the last HamCon, some stuff never goes out of style.

 

I don't believe anything discussed on these forums is protected by non-disclosure agreements. You may want to check that aspect with your patent agent.

Unless the source and conditions are specifically attributed, then everything is in the Public Domain IAW the CCPDL. Please note the following from this license:

Dedicator recognizes that, once placed in the public domain, the Work may be freely reproduced, distributed, transmitted, used, modified, built upon, or otherwise exploited by anyone for any purpose, commercial or non-commercial, and in any way, including by methods that have not yet been invented or conceived.

Dave