2 Cell Boost Circuit.... Edit !... IT WORKED!!!!!!

[ViReN]

Please View the following image

I have successfully constructed it using 2N3904 & 2N3906 Pair ...

The current is also very nicely regulated using the variable resistance.

My main quesiton is, can the output voltage of this circuit be increased ? (if i need to overdrive the LED's)...
? Also, can this circuit be modified for a single cell application ?

I had tried my hands on http://www.elecdesign.com/Articles/ArticleID/5886/5886.html circuit, as some one pointed it out. But.. it simply didnt work... i wonder what went wrong !!!....

I used Eagle 4.11r2 (PCB Design tool) and did not do any errors while constructing it.

Thanks & Regards,
ViReN

 

[MrAl]

Re: 2 Cell Boost Circuit....

Hi there,

This circuit looks a lot like the Brinkmann.

To begin with, there are a few problems. The main
one being that in order to obtain the 'boost'
this circuit puts out high current pulses (more then
1 amp possible) which averaged out equals the
normal current of an LED.
For a Nichia LED operating at 20ma, this isnt as much
of a problem because it is rated up to 100ma peak,
which is 5 times it's average rating.
For a Luxeon this poses a problem, because the
Luxeon peak is only twice that of it's average rating.

What this means is using a Luxeon with this circuit will
mean running it out of spec, and the possibility of
early failure looms.

Change #1: increasing current output
Increasing the current, which is probably easy enough to do,
will lead to even higher peak currents for the Luxeon, which
is not a good idea.

Change #2: decreasing input voltage
While it is probably possible to do this, doing so will
cause even higher peak currents in the Luxeon (4 amps or higher)
which is really not good at all and most likely will cause
early failure, if not very early.

So what to do about all this?

The most reasonable solution is to add a Schottky diode on the
output along with a nice size cap 100uf or better.
What this does is absorb the high current pulses so that the
Luxeon can run with a lower peak while still getting the
average current flow it needs to provide full brightness.

Of course there is another problem that comes in when you
try to lower the input voltage, and that is that the
sat voltage of the pnp added to the BE diode drop of the npn
comes just short of the 1.5 volt supply, which is the intended
Vin for one cell. Adding to this problem is the fact that
the inductor only gets Vin minus the Vsat of the NPN, so the
natural solution is to use two low sat transistors. Of course
two low sat transistors in this app runs the risk of overdriving
the npn so a 22 ohm base resistor on the npn seems like a good
idea, for protection.

Add to that a somewhat lower inductor value to increase frequency
and the circuit could put out 500ma if you want it too :-)

Let me throw some parts at ya...

FMMT617 Zetex
FMMT717 Zetex
22 ohm base resistor for the NPN
200uH inductor for L1
1N5822 Schottky (or Zetex's 2 amp Schottky)
200uf capacitor

The Schottky diode connects in series with the LED,
the 200uf cap across the LED, observing polarity.

Take care,
Al

 

[Doug Owen]

Re: 2 Cell Boost Circuit....

Al (he says, standing CPF hat in hand...),
awesome!!!

Doug Owen

 

[tonyb]

Re: 2 Cell Boost Circuit....

I had originally designed the circuit keeping it simple and using only readily availible components for ease of assembly. As I originally said when I first posted it a year ago "it is what it is and that is a simple dimmable 2-cell boost converter for a 1-watt Luxeon". The addition other components ie- schottky diode and cap is recommended for more output and less stress on LED (high voltage spikes into LED),and selection of a high Q inductor will also help eff. You may also parallel the output transistor with another matched beta transistor to increase eff.

I am glad to see people are experimenting and progressing the art. Thanks to Al for his input into component selection for increased eff.

Thanks!

Anthony Bakagios aka-tonyb

 

[ViReN]

Re: 2 Cell Boost Circuit....

I am sorry, there is some sort of confusion. I would like to clear it first.

1) I would be using this circuit to drive a couple of(3-6-or even 10-12) 5mm, 10,000 mcd, White LED's in parallel.

2) (please note... since i am a newbie) once this circuit is successful... i would then go on in modifying a couple of flashlights that i have at my disposal....

3) third step for me is to use these circuits.. for Lx's

tonyb: Thanks for designing the circuit.... are there any other circuits that you have designed ? (especially the ones that drive LED's.... from a 1.5 V cell).... I am currently playing around with the brinkmann circuit... (if you see in my other thread)....... i did try to search on google about your circuits with "Anthony Bakagios" in the search string... /ubbthreads/images/graemlins/jpshakehead.gif but could not get any results... /ubbthreads/images/graemlins/thumbsup.gif its amazing to get a responce from the author himself here...

MrAl: Awesome..... I would be playing around with the components and use various tricks that you had suggested... will report about the results on it by tomorow...

Thanks & Regards,
ViReN

 

[MrAl]

Re: 2 Cell Boost Circuit....

Hello again,

Doug Owen:
Well, thank you very much Doug! I was just trying to give
him some ideas so he could venture down the path to a better
circuit with approximately the same ideas in mind.

tonyb:
You're welcome, and yes, if you want to stick to the more common
parts then it will be harder to get to 1.5 volts with full
output current, but you could always try another circuit topology,
which might get you there faster.

ViReN:
Oh ok, good luck with the experimentation. If you find that
you cant get enough output perhaps another circuit would
be better (for higher outputs that is).


Take care,
Al

 

[ViReN]

Re: 2 Cell Boost Circuit....

Anthony Bakagios aka-tonyb: ... Hats OFF to you Mannnnnnnnnnn........ designing such a wonderful circuit.... ... (atleast from my standards)....

I did play around with some of the components and changing the values..... ... and who was at fault ???

THE TRANSISTORS !!!... they were FAULTY... i changed em.. and it was all that was needed....

the effectivity of the circuit... well this picture will say it all...

(sorry for the poor quality... my Mobile cant get better pic's than this)

YES... it can run 2 White LED's in PARALLEL.... and on a 1.5 V LR41(fresh one) .... it has been atleast 45 Minuites.... and LED's are still shining... so i guess.. its efficient too.... i wonder what would be the runtime.. if i run it on a AA Alkaline.

THANKS TO EVERY ONE WHO HAS REPLIED..... if it wasent of you guys... i would not have attempted this hard.... and finally was the result. /ubbthreads/images/graemlins/smile.gif /ubbthreads/images/graemlins/thumbsup.gif

Best Regards for every one,
ViReN

 

[ViReN]

Re: 2 Cell Boost Circuit....

Here is the Schematic .. and PCB... if any one needs....

Regards,
ViReN

 

[tonyb]

Re: 2 Cell Boost Circuit....

I am glad to hear you now have it working. I had put some other circuits on here in the past I will try to dig them up tonight and put them on again.

Thanks!

tony.

 

[ViReN]

Re: 2 Cell Boost Circuit....

tonyb: .... I am trying to search through your post... even went on to your site (probably http://www.na****i.n**/abakagios *ed for security reasons... hope you understand).. to look for some circuits designed by you.... but /ubbthreads/images/graemlins/frown.gif Listing denied....

I have been through some old post's.... dimmalux 300 some thing.. but no image /ubbthreads/images/graemlins/frown.gif ....

I have been searching... and will be searching... hope... you would help me in getting some circuits...

I must congragulate you for making such a circuit... i was just surprised.... when i connected the LR41 cell !!!.... i was just playing around... and never expected that this could work for that miniscule thing.... most importantly... those 2 LED's gave me a run time of 1hr 10min .... thats whooping !... after about 1hr 11min.. the light started to dim... its about 2 hrs now... and the battery is almost about to give up...

From the basic knowledge.... i know this circuit is not regulating one (please correct me if i am wrong)... how can we modify this to a regulating circuit ?

Also... I have a few queries....

1) I am not sure what value of inductance i have used in this circuit... i asked my local electronics dealer.... "Give me a GOOD inductor... which is small... and has 2 leads.... with some core..." he was uncertain... and was puzzeled... he gave me a pack of various sizes.. and dots.. and pins... and what not... ... and i closed my eyes.. and picked up one !!!!!..... it had 2 red dots (red for confidence !!!.. ) on top.... and 1 big brown (brown down to earth !!!) dot on side ... what is it's value ?
If i think of using the color code of resistors.... 2 - 2 - 1 ? ... umm 220 uH ???

2) while experimenting, when i connected another inductor in parallel... the light was slightly brighter... can you please give me any clues and ideas ? (inductors in parallel... will reduce the resultant inductance ???....)

3) what value of resistor should i put (100K ??? if i am not wrong)...... if i dont want to make it dimmable ... and have a max o/p at 1.5 v input (if you see my other post....(http://www.candlepowerforums.com/ub...&view=collapsed&sb=5&o=365&fpart=2#Post619572) ?...
(dimmable circuit is very useful for me for another mod that i am planning to make....)


Thanks guys... u all are awesome....
very special thanks to "Anthony Bakagios" ... tonyb /ubbthreads/images/graemlins/smile.gif

Best Regards,
ViReN

 

[MrAl]

Re: 2 Cell Boost Circuit....

Hello again,

Hey, that's great. Im happy to hear you're getting
some good results. Is this your first switching
circuit?

Take care,
Al

 

[ViReN]

Re: 2 Cell Boost Circuit....

Yes Mr Al.... it is indeed my first circuit.... and i plan to look for more brightness and more current ... along with the regulation... (no SMD's !!!.... Reason... i cant do it (i guess it requires special equipment))

Thanks & Regards,
ViReN

 

[jtr1962]

Re: 2 Cell Boost Circuit....

[ QUOTE ]
ViReN said:
(no SMD's !!!.... Reason... i cant do it (i guess it requires special equipment))


[/ QUOTE ]
SMDs don't really present any special problems except that you can't breadboard them on perfboards. The only key with using SMD is to not apply too much heat when soldering them to the circuit board. Also, I find the best way to install small SMD components like resistors is to first put a small solder blob on one of the pads on the circuit board, then grasp the component with a tweezer, melt the solder blob while you put the component into position, let the joint cool for a few seconds, and then solder the older lead(s) of the component. I can work with components as small as 0402 this way although I prefer 1206 or larger. I just hope they don't get any smaller than that! ICs with close lead spacing require a soldering iron with a fin tip. For example, on the CAT32 LED driver IC I'm using the leads are spaced on 0.95mm centers. Compared to this, a regular DIP IC or even an SOIC is positvely huge. The great thing about SMD is that you can make LED driver circuits with a very small footprint, in some cases smaller that the LED it is driving.

 

[ViReN]

Re: 2 Cell Boost Circuit....

jtr1962:

Thanks for sharing your views. I would be studying more about using SMD Components at home.
I am very uncomfortable with the breadboard.... so its good ... thing /ubbthreads/images/graemlins/smile.gif that SMD's cant be breadboard'd ....

Where can i get tools (specalised hardware like soldering iorn... fin tip ??? etc) to work on I do have a regular soldering iorn (25 Watt) ... but will that suffice ?... i guess... it will burn the SMD Component's :P !

Yep... world is shrinking... no doubt... LED Driving circuits are.. with the advent of SMD's.... and i wonder... when there would be a day.... when Lux's will come with a built in drivers (or are they already available) ....

Thanks once again,
Regards,
ViReN

 

[jtr1962]

Re: 2 Cell Boost Circuit....

Mouser has a wide selection of soldering irons and tips. A 25W or 40W iron is fine. As long as the tip has temperature regulation you won't burn any components since the heater supplies only enough heat to keep the tip at the set temperature. Cheaper soldering irons have no regulation and can burn up even through-hole components as well as circuit boards. The trick with SMD components is to only apply heat just long enough so the solder flows and creates a good joint. This usually takes about two or three seconds. I've been hand soldering SMD components for over ten years (mostly repairs) and haven't had any reliability problems using my methods.

 

[ViReN]

Re: 2 Cell Boost Circuit....

jtr1962 .. thanks for the input ... I will soon gear up for the SMD's /ubbthreads/images/graemlins/smile.gif

Thanks & Regards,
ViReN

 

[ViReN]

Re: 2 Cell Boost Circuit....

WARNING : ! .... /ubbthreads/images/graemlins/ooo.gif

This has happened with me TODAY..... /ubbthreads/images/graemlins/frown.gif

I Modified the 2 Transistor Circuit.... adding a Shottky Diode (1Amp) and 2.2 uF Capacator (63V, i could not get voltage rating below this) thinking... that i would get a better o/p .... /ubbthreads/images/graemlins/smile.gif .... and this is what happened....

1) I connected the supply....1.5 V,
2) Added Schottky diode at the o/p end... (+ve)
3) Connected a Capacator.....(With Proper polarity)
4) Connected about 3 parallel LED's.....

And it was expected.... that this will work fine...
but what actually happened was.... just a small flash ... and nothing else /ubbthreads/images/graemlins/ooo.gif ... when i tried to reconnect the led's (remember supply was on all the time ... even though just 1.5 V ... AAA) hmm... again same thing.... /ubbthreads/images/graemlins/thinking.gif

i was thinking.. /ubbthreads/images/graemlins/thinking.gif /ubbthreads/images/graemlins/thinking.gif that circuit was not working as intended... so i removed the added Schottky and Capacator...
connected LED's again... and this time... nothing happened... not even a flash !....
checked with DMM.. what's the o/p .... and it was fine... what could have possibly gone wrong ???

Boy oh Boy.... when i connected another led's .... they were lit (ofcourse without the cap and diode)....
again connected cap... diode.... SAME... Flash... later on... nothing !!!.... /ubbthreads/images/graemlins/rolleyes.gif

I wondered.... /ubbthreads/images/graemlins/poke2.gif and thought.... and a wild thought came in my mind.... are the LED's (6 by now... ) blown OUT ????!!!!!!................ /ubbthreads/images/graemlins/ohgeez.gif /ubbthreads/images/graemlins/ohgeez.gif /ubbthreads/images/graemlins/ohgeez.gif

YES THEY WERE !~!~!~!~!~!~!~!~!~!~!~! /ubbthreads/images/graemlins/banghead.gif /ubbthreads/images/graemlins/banghead.gif /ubbthreads/images/graemlins/banghead.gif /ubbthreads/images/graemlins/banghead.gif

Guys... reading this DONT LAUGH... its an Achievement... blowing up to 3 LED's using a AAA cell... and a fine piece of Circuitry... /ubbthreads/images/graemlins/cool.gif

Later on... i figured out that while the supply was on .... the capacator charges... (to the voltage peak's) .... and when we connect LED's ... that Flash... tells you about BLOWING LED's !.... /ubbthreads/images/graemlins/clap.gif

I Learnt the hard way.... but... for you guys (who might already know this... may be .. may not be)... ....

NEVER EVER CONNECT SUPPLY BEFORE CONNECTING LED'S TO THE INDUCTOR BOOST CIRCUIT ESPECIALLY WHEN SCHOTTKY DIODE AND CAPACATOR ARE USED... AT THE OUTPUT END

Later on when i connected the things... with the LED's ... the brightness was increased... and ofcourse.. that pinkish.. blue tinge ... on the beam was gone...

thankfully... the LED's that i blew up... were cheap ones ($0.1 .. a piece) .... and not the Nichia's ($2 a piece... which i procured recently) ....

Hope this will help fellow CPF Members & those who build their circuits for the first time.... ... (i think.... the transistor circuit is a good one to start with /ubbthreads/images/graemlins/smile.gif )....

Mr Al .... I have a request for you .. will post in the proper thread heading...

Thanks & Regards,
ViReN

 

[MrAl]

Re: 2 Cell Boost Circuit....

Hello Viren,

Yes, that's good to mention. As with most of these
switching circuits the load has to be connected before
turning the circuit on. Even without the LED connected
the output transistor can easily become damaged due to
the increased output voltage with no load.
I guess we need to mention that from time to time
when talking about these circuits so that we all keep
this in mind and new circuit builders wont have any
problem either.

Im just glad the LED's didnt cost you too much money.
Some people blew out Luxeons when experimenting with the
Zetex circuit.


Take care,
Al

 

[korpx]

Re: 2 Cell Boost Circuit....

nice job! I hope I can get through MrAl's ee course (and have it stay in my thick head) so I have the guts to start playing with my own circuits too. /ubbthreads/images/graemlins/ooo.gif

It's been some years since I studied all this in school, and alot of other important things have replaced it in my brain /ubbthreads/images/graemlins/grin.gif

 

[ViReN]

Re: 2 Cell Boost Circuit....

Thanks Korpx ....

MrAl's lessons are really good... you learn a lot...

-ViReN