Workd Torch LM334 circuit experiment for MJ LED apllication

moraino

Enlightened
Joined
Jan 20, 2002
Messages
447
Location
B.C. Canada
I don't know how to name this thread but here is my story.

I am experimenting the World Torch circuit tonight and found something interesting. Here is the circuit:
http://www.worldtorch.com/LDO-regulator-schematic.php

I first wire up the LM334 according to this data sheet from STMicroelectronics:
http://www.st.com/stonline/products/literature/ds/2157/lm134.htm

The result of led current was a big surprise to me, 52mA instead of 1.5mA using 47 ohm resistor according to World Torch website. I was puzzled, what did I do wrong? I check the wiring and all is fine.

Then I notice on the National datasheet and on the Workd Torch schematics diagram the LM 334 V+ is on the left while STMicro is on the right.

So I reverse the V+ and V- of the LM334 on the bread board and get 0.15mA just like the schematic.

Now the question is since I can reverse V+ and V- on the LM334 and both works, I am wondering which way is the right way? Does STMicro have the wrong data sheet? The reference voltage on the World Torch is 0.065. With the reverse V+ and V- it is about 2.7...I don't understand.

I didn't measure the effeciency of the two but I will do it later to find out the difference...hmmmmm..


Henry
 
Last edited:

moraino

Enlightened
Joined
Jan 20, 2002
Messages
447
Location
B.C. Canada
Re: Workd Torch LM334 circuit experiment.

I did some resistor swapping as follow:

With V+ on the right, wrong way, and using 68 ohm resistor I get 37mA
P=0.037 x 0,.037 x 68 = 0.093 watt

With V+ on the left, right way, and using 2ohm resistor I get 30mA
p=0.03 x 0.03 x 2= 0.0018 watt

I think I'll never forget the polarity of the LM334, since V+ on the right is the wrong way and on the left is the RIGHT way.

So there is a big defferent in the energy wasted even though all in mini mini wattage. Please let me know if I am doing these wrong as I am refreshing my Ohms Law I learnt 20 years ago....

Still can someone answer me what will be the ill effect to reverse the V+ and V- in the circuit besides poorer effeciency. It seems does regulate the current in the reverse polarity of the LM334 though. I may be wrong.

Henry
 

rdshores

Enlightened
Joined
May 3, 2004
Messages
400
Location
Vermont
Here is the correct diagram:

LM334.jpg
 

Doug Owen

Flashlight Enthusiast
Joined
Jan 30, 2003
Messages
1,992
Henry,

While I haven't actually tried it, I'm quite sure the 334 is not regulating with it's plus and minus reversed. The bias on all six internal transisors would be backwards.

What you have, I think, is a case of the BE junction on Q2 being in reverse breakdown (rather than the normal forward bias), something loosely defined if at all. Have you actually varied the input voltage and seen the current hold constant?

I'd stick with the WT circuit as published. We've built a pile of 'em that way, they all work as National designed them to....

Doug Owen
 

moraino

Enlightened
Joined
Jan 20, 2002
Messages
447
Location
B.C. Canada
Doug Owen said:
Henry,

While I haven't actually tried it, I'm quite sure the 334 is not regulating with it's plus and minus reversed. The bias on all six internal transisors would be backwards.

What you have, I think, is a case of the BE junction on Q2 being in reverse breakdown (rather than the normal forward bias), something loosely defined if at all. Have you actually varied the input voltage and seen the current hold constant?

I'd stick with the WT circuit as published. We've built a pile of 'em that way, they all work as National designed them to....

Doug Owen

I followed STMicroelectronics data sheet because the LM234 I have are made by them. Now I realize the error and you are right, I don't think the reverse regulates the current.

I am thinking of using the circuit to power the MJ LED in Dorcy Cool Blue with 4AA. In 2AA light, the MJLED looks good at the beginning but when bettery drop down to 1.3 or 1.2 volt it's just not lighting up not to mention dimming gradually.

I think LM334 is a very good match to the MJled with any 4AA torch with 100mA. It is good and cheap. I will also make a few for the 4aa battery holder with 3AA as a very cheap TORCH. The circuit board with circuit and LEDs can fit in the 4th slot with same area at the 4th slot.

Thanks for the circuit and correction. Now the MJled can be put to good use. It is dimmable as well. Got to buy more battery holders.
 

Doug Owen

Flashlight Enthusiast
Joined
Jan 30, 2003
Messages
1,992
moraino said:
I think LM334 is a very good match to the MJled with any 4AA torch with 100mA. It is good and cheap. I will also make a few for the 4aa battery holder with 3AA as a very cheap TORCH. The circuit board with circuit and LEDs can fit in the 4th slot with same area at the 4th slot.

QUOTE]

Yup, that it does. And you can put a switch there as well. In fact if you go to the WT page (www.WorldTorch.com) then go to the gallery section 'WT in India' you can see how our partners there did exactly that. Using local made PCBs and Bombay bought parts they have the first run of 50 in service in 'the bush'. They plan another run as soon as they get some other stuff in order.

Anway, I fully agree with you, this circuit is an excellent fit to these sorts of applications. FWIW, it does a good job driving Luxeons at 350 mA as well......

Doug Owen
 

evan9162

Flashlight Enthusiast
Joined
Apr 18, 2002
Messages
2,639
Location
Boise, ID
With the right MOSFET, and a slight change in the circuit, the '334 can easily drive a R-O Lux III at 1.4A with about 0.1V of dropout (I've built it, works really well)
 

moraino

Enlightened
Joined
Jan 20, 2002
Messages
447
Location
B.C. Canada
evan9162 said:
With the right MOSFET, and a slight change in the circuit, the '334 can easily drive a R-O Lux III at 1.4A with about 0.1V of dropout (I've built it, works really well)

I probably read about it before. Are you the one who made those little cute sandwitch for driving Lexeon?

Which MOSFET did you use? And what about the effeciency and circuirt change comparing using IC such as FMMT717 PNP transistor?

I am think of driving a 1W Lux at below 350mA in Dorcy led lantern. It may not require heat sinking with decent brighteness in a amsll package. Thanks for sharing,

Henry
 

moraino

Enlightened
Joined
Jan 20, 2002
Messages
447
Location
B.C. Canada
QUOTE]

Yup, that it does. And you can put a switch there as well. In fact if you go to the WT page (www.WorldTorch.com) then go to the gallery section 'WT in India' you can see how our partners there did exactly that. Using local made PCBs and Bombay bought parts they have the first run of 50 in service in 'the bush'. They plan another run as soon as they get some other stuff in order.

Anway, I fully agree with you, this circuit is an excellent fit to these sorts of applications. FWIW, it does a good job driving Luxeons at 350 mA as well......

Doug Owen[/QUOTE]

I have been that website many times before and I really enjoyed. It's only now I have some time to sit down to experiment a bit trying to modify those light that I rarely use because of the way they are designed.

I hope there will be more CPF members building this great circuit to improve their lanterns. I think 18650 Li-Ionis a very good fit. You probably don't even need a over discharge protection with MJled or any led...
 

rdshores

Enlightened
Joined
May 3, 2004
Messages
400
Location
Vermont
I've been using the LM334 WT circuit to convert those round tap lights to LED. I'm using a 8mm 150ma 13 lumen LED LINK I got off ebay driven at 20ma(low) and 100ma(high).
 

evan9162

Flashlight Enthusiast
Joined
Apr 18, 2002
Messages
2,639
Location
Boise, ID
Heh.

Just so you know, the two pass MOSFETs I mentioned can't be found at Digikey - I had to go through Mouser to get those (order splits are a pain)
 

TorchBoy

Flashlight Enthusiast
Joined
Jan 15, 2007
Messages
4,486
Location
New Zealand
Thank you so much for this thread. I was trying to figure out how I was getting so much current through a LM334 and by comparing the ST and National datasheets I found they gave different pinouts. I had used the incorrect ST pinout because my LM334 is from ST, and I figured they should know how their own components operate. I then had a close look at the project I had pulled it out of, which showed it had been the other way around, possibly because the company I bought it from had the National datasheet on their web site. A Google search (not just CPF, either) for 'LM334 backward' listed this thread first, which clinched the matter, and explained exactly the symptoms I was seeing. Reversing my LM334 fixed everything. Thank you again, Henry and rdshores in particular.

BTW, I also downloaded the latest datasheet from ST's web site and it's no newer than the one I had, May 2003.
 
Top