jarhead custom circuit

[shiftd]

Asked Jarhead if he is willing to make me a custom circuit that will fit into the arc LS's sink, can be run from 1AA, and 2AA (and also 3AA or 3.6V for the jsburly's rechargable /ubbthreads/images/graemlins/grin.gif) and more efficient and powerful than the current circuit available. After going up and down, slapping him and kicking him in the *** /ubbthreads/images/graemlins/tongue.gif , he finally agreed to make me one, on one condition: i have to supply the components (but it turns out later that the only components that i have to supply are the lux, sink and chip. He supplies the rest). I said sure. So he went and make me one, according to my spec (aint that cool? /ubbthreads/images/graemlins/grin.gif)
The circuit will run the lux on 2 stage, depending on the input V. He set the middle limit to be 2 V (as common battery never get to that V anyway. 2 NimH only 2.4V. 1 is 1.2V and lithium is either 3V or 3.6V for rechargeable, 1.7V for AA version). So anything below that V is considered low input V and the converter goes to low power, delivering only 250 mA to the luxeon, at +80% efficiency wiht 1.2V input V. If the voltage goes higher than that, the converter goes to high power mode and supply 750 mA (about that range) to the lux and has high 80's% from 2.4V input. Any Vin below the Vf of the lux will be step up according to the above spec. And if the Vin goes above the Vf, the converter switches to linear converter mode and acts as a resistor. I believe this thing can take more than 5V, but need to confirm with jar again for that.
I want to post comparison shot with my other BB750, but my other Arc head were away at the moment /ubbthreads/images/graemlins/frown.gif so that has to wait. The circuit was epoxied together in the sink, so cannot take pic on that.

 

[Klaus]

Where does the line start for that beast ?

Klaus

 

[shiftd]

behind me /ubbthreads/images/graemlins/grin.gif
i am asking him to make one more for me /ubbthreads/images/graemlins/grin.gif
but he is like "na ah, it takes me soooo much time"
hopefully, after some kicking and slapping, he wants to make more /ubbthreads/images/graemlins/grin.gif

 

[B@rt]

Hmm... JCC... /ubbthreads/images/graemlins/thinking.gif /ubbthreads/images/graemlins/cool.gif /ubbthreads/images/graemlins/drool.gif /ubbthreads/images/graemlins/wink2.gif

 

[NewBie]

Humm, heheh, thanks shiftd.

Lets see, lets write this up a little clearer.

0.8V-1.8V 250mA at 1.2V it is 80% eff.
1.8V-2.0V, about 400mA (warning range for rechargeable 2 AA NiMH)
2.0V- LED Vf 800mA 2.4V +80% eff, 3V +90% eff
Vf - 6V, circuit works like a linear regulator.

You're certainly welcome shiftd.

I was surprised I managed cram 1 IC, 9 capacitors, 12 resistors, a huge honking inductor (takes nearly 1/2 the board space), and a transistor onto a BB footprint, and slightly taller height. BTW, can you say 3D circuitry?

You talked me into making another one, not that I wanted to make more, lol.

I did notice it does a pretty decent job at sucking a 123 dry.


BTW, I take it that you liked it?

Enjoy!

 

[hotbeam]

Sounds like a very versatile driver. Good efficiencies too.

 

[Doug S]

[ QUOTE ]
NewBie said:
Humm, heheh, thanks shiftd.

Lets see, lets write this up a little clearer.

0.8V-1.8V 250mA at 1.2V it is 80% eff.
1.8V-2.0V, about 400mA (warning range for rechargeable 2 AA NiMH)
2.0V- LED Vf 800mA 2.4V +80% eff, 3V +90% eff
Vf - 6V, circuit works like a linear regulator.

You're certainly welcome shiftd.

I was surprised I managed cram 1 IC, 9 capacitors, 12 resistors, a huge honking inductor (takes nearly 1/2 the board space), and a transistor onto a BB footprint, and slightly taller height. BTW, can you say 3D circuitry?

You talked me into making another one, not that I wanted to make more, lol.

I did notice it does a pretty decent job at sucking a 123 dry.


BTW, I take it that you liked it?

Enjoy!

[/ QUOTE ]

Jarhead/Newbie: Nice job!! /ubbthreads/images/graemlins/thumbsup.gif
From the specs above, I would presume that this circuit is based on a TPS61020 IC. BTW, I started to fool around with this IC myself. Laid out a board and had it fabbed a couple of months ago but haven't loaded components yet. I got sidetracked by other stuff and I haven't gotten back to it. One thing that baffles me about your specs is that I don't see how using the components listed you could have it configured for constant current regulation and still get that 90% at 3Vin. Are you using a constant voltage scheme instead?

 

[darkzero]

Hey, don't forget me, hehe. I'm in for 1 or 2 of these right behind Klaus.

 

[shiftd]

Doug, i think jarhead is using constant power instead. The current listed is taken from measurement using specific load (in this case, the lux).
And no, the circuit is not based on the TPS. I think it is more like LTI(something).

 

[battousai]

I'm in for one!!

 

[Justintoxicated]

Don't forget mine :P /ubbthreads/images/graemlins/hahaha.gif /ubbthreads/images/graemlins/poke2.gif /ubbthreads/images/graemlins/sssh.gif

 

[NewBie]

[ QUOTE ]
Doug S said:
[ QUOTE ]
NewBie said:
Humm, heheh, thanks shiftd.

Lets see, lets write this up a little clearer.

0.8V-1.8V 250mA at 1.2V it is 80% eff.
1.8V-2.0V, about 400mA (warning range for rechargeable 2 AA NiMH)
2.0V- LED Vf 800mA 2.4V +80% eff, 3V +90% eff
Vf - 6V, circuit works like a linear regulator.

You're certainly welcome shiftd.

I was surprised I managed cram 1 IC, 9 capacitors, 12 resistors, a huge honking inductor (takes nearly 1/2 the board space), and a transistor onto a BB footprint, and slightly taller height. BTW, can you say 3D circuitry?

You talked me into making another one, not that I wanted to make more, lol.

I did notice it does a pretty decent job at sucking a 123 dry.


BTW, I take it that you liked it?

Enjoy!

[/ QUOTE ]

Jarhead/Newbie: Nice job!! /ubbthreads/images/graemlins/thumbsup.gif
From the specs above, I would presume that this circuit is based on a TPS61020 IC. BTW, I started to fool around with this IC myself. Laid out a board and had it fabbed a couple of months ago but haven't loaded components yet. I got sidetracked by other stuff and I haven't gotten back to it. One thing that baffles me about your specs is that I don't see how using the components listed you could have it configured for constant current regulation and still get that 90% at 3Vin. Are you using a constant voltage scheme instead?

[/ QUOTE ]

It most certainly isn't a TPS61020. Way too much resistance in the internal MOSFETs. The 030 won't run down this low without some help, but it isn't that one either. You are going to have to look on the other side of the fence somewhere. : )

The 020 as a max current on the switch of about 1.5A at 275 mOhms. You'll need to find a part with about half this switch resistance, something that is running alot faster to keep the inductance down so you can use a lower DCR inductor. Though I ran the part at the bottom end of it's frequency range (~1 MHz) to save power in the switches... It can take over 3A in the internal switch.

Again, I used a truely massive inductor for this size footprint converter, about half the board space of a BB is occupied by the inductor. 18mOhm, 8A rated (heat), saturation current is 14A, and at the 14A, it still has 75% of it's inductance left, unlike some inductors, where you only have 35% of their inductance left at their saturation rating.

I also used 4 oz. copper board, and used really beefy wires in the current loops.

Not sure if it okay with shiftd to tell others what part number/brand of chip I picked out for this one, we'll have to defer to his okay.

 

[shiftd]

Sure Jar, i don't really care about secrecy /ubbthreads/images/graemlins/tongue.gif
All i know is that it is an linear something. (LTI). forgot the part number.
beside, most of the circuit is done by you and you are the one that suggest me the circuit, so the right is in your hand /ubbthreads/images/graemlins/smile.gif

been using the 1AA config, really nice. as bright as my grey. /ubbthreads/images/graemlins/grin.gif

 

[NewBie]

http://www.linear.com/pdf/3421f.pdf

BTW, I'd never hit that 90% point without a killer inductor, some very low ESR caps, and really heavy wiring. For the TPS61030, I coulda done it sleeping, but not with this LT part.

Here is a shot of it, before I cleaned it up and straightened things out. It fit within a BB cavity, plus 0.030" or so in height.

 

[shiftd]

did a run time test using single AA cell, quite fresh out of charger about 1-2 days ago. 1800 mAH, dynacharge cell (from the group buy).

here is the result

please note that the y axis does not represent actual lux reading. It is a lux reading, but not taken from 1 meter, so please use it only as comparison.
Also, the x axis is the minute (doh /ubbthreads/images/graemlins/tongue.gif)

 

[NewBie]

Shiftd, is that with one of those lossy kroll switches?

/ubbthreads/images/graemlins/thinking.gif /ubbthreads/images/graemlins/twakfl.gif /ubbthreads/images/graemlins/whoopin.gif

 

[PEU]

maannnn, that circuit doubles as a regulator and a sculpture... nice job

Pablo

 

[shiftd]

lol, yea jar. Don't have any other clickie switches so i just picked one kroll near me /ubbthreads/images/graemlins/tongue.gif
Next time will try doing one using twistie.

 

[NewBie]

Lol, I should order a twistie and send it to you shiftd.

PEU, tq.

 

[JonSidneyB]

Hey, is Jarhead going to become a new CPF manufacturer out here.