1. Home
  2. Electronics Design
  3. Watt ever happened to Watt Sun anyway?

Watt ever happened to Watt Sun anyway?

I'm going to build what I'm pretty sure is the biggest Joule Thief, if not the biggest blocking oscillator alone, on the internet. Which is pretty sad when you think about it. But in any case, what I had in mind was...

formatting link

In simulation (here at school, I happen to have access to a simulator but not a Bench...I'm so ashamed), this circuit runs around 5A peak collector current. I have the inductance set to 400nH and it runs around 500kHz. Input is a couple of watts. Efficiency is around 70%, which is quite good for silicon at this voltage.

I haven't decided if I want rectified and filtered output or if I'll drive the LEDs directly. They may be more efficient with direct current, justifying the 0.3V drop of the diode. The ten LEDs will probably share current well enough...we'll see.

I remember Watt Sun whining that there are (almost?) no good transistors that run low Vce(sat) and high Ice. The 2SC5001 is under a buck from Mouser, so for the scale this sucker is built at, that's not too bad. For medium stuff, you can't beat Zetex parts, but they're still expensive, even for one offs, and especially compared to something like a 2N4401, which I think works just fine in a regular (one or two LED) inverter.

Tim

Reply to
Tim Williams
Loading thread data ...

Tim, try putting in real d.c.r. values for the inductor and the battery ... /then/ check the efficiency.

Cheers, James Arthur

Reply to
James Arthur

Ah, not content to trust a mere simulator, I already thought of that! :-) A couple turns of #18 should have much less than 20mohms resistance, no? Maybe about right at operating frequency. I also added parasitic capacitors representing LED/collector output and base- emitter capacitance.

Tim

Reply to
Tim Williams

Go for it, dude, and be sure to show us what you produce.

I was wondering about WattSun myself a couple days ago when I ran across several of my own blocking-osc-suck-a-battery-dry circuits hiding in a box amidst several other small things. I pulled them out and put them on the bench to play with anew (as soon as I remember where I stashed my pile of low Vce-sat devices).

Reply to
Michael

An alkaline AA has 300 milliohm typical esr. It won't like your 5A peaks! The inductor? Depends on core size and skin effect. 50 milliohms DCR would be a feat in a small unit.

Figure about 7-10% loss in the inductor, and about

10% in the schottky, maybe 3-5% core loss, and the rest in saturation and switching.

Zetex's FMMT618 lo-sat NPN is in your ballpark and cheap; you could switch to ZXT11N-series if need be. Expect ~180mV Vce(sat) @ 5A if you drive the hell out of it, and about 210mV if you don't :-).

You can cut the schottky Vf-caused loss in half by putting the LEDs in 2x5 series-parallel, which helps current sharing too.

After you go through this whole exercise you'll see why we like to use continuous mode circuits: they cut those peak currents in half. i^2*R and all that.

Cheers, James Arthur

Reply to
James Arthur

Note: Vce(sat) figures were for the heftier of the ZXTN11N-series; the FMMT618 is too wimpy. But, the ZXTN25020, for example, is better still, and just 12 cents if you buy a bunch.

James

Reply to
James Arthur

Some 10uF ceramics should help with that. And if still more is needed (I don't think it is, 10uF is 32 mohm at 500kHz and less than a third of that for the triangle wave's harmonics), I probably have some tantalums laying around, or if nothing else, I could slap on some electrolytics. But after the ceramic I think the cell will keep up -- relatively speaking, of course. This thing will drain a AA pretty quickly, after all. What's left can go to the two-LED thief I made earlier. ;-)

I was thinking 4 x 24AWG for the primary, sound reasonable?

Come to think of it, I wonder what core loss is like. 500kHz is on the high side for generic black ferrite.

The 2SC5001 I've ordered is in the same ballpark (under 300mV Vce(sat) at up to nearly 10A Ic, at hFE = 80), so I think I'll be good on that.

True, and with red LEDs, the extra voltage (about 4-5V) isn't a big deal with respect to the duty cycle.

Yeah I know, flyback sucks. ;-)

Tim

--
Deep Friar: a very philosophical monk.
Website: http://webpages.charter.net/dawill/tmoranwms
Reply to
Tim Williams

OMG, Tim you're everywhere, aren't you? I just happened to land here cause I'm in the process of making a fairly big Joule thief, too. Although I have a completely different approach in mind. Will post the results later.

Peace

Reply to
rouslan

Besides, if you're only gonna be lighting the 10 LEDs, 2 x AA (even rechargeable ones) in series will give you enough voltage to supply the red LED's with enough current and will have way higher efficiency, with no current limiting resistors it'll be as close to 100% as you'll ever get.

But I where the fun in that, right? :P

Peace

Reply to
rouslan

"Watt ever happened to Watt Sun anyway?"

His LED blinked ?:-)

...Jim Thompson

-- | James E.Thompson, P.E. | mens | | Analog Innovations, Inc. | et | | Analog/Mixed-Signal ASIC's and Discrete Systems | manus | | Phoenix, Arizona 85048 Skype: Contacts Only | | | Voice:(480)460-2350 Fax: Available upon request | Brass Rat | | E-mail Icon at

formatting link
| 1962 | I love to cook with wine Sometimes I even put it in the food

Reply to
Jim Thompson

He had health problems the last time I heard from him.

--

formatting link

aioe.org, Goggle Groups, and Web TV users must request to be white listed, or I will not see your messages.

If you have broadband, your ISP may have a NNTP news server included in your account:

formatting link

There are two kinds of people on this earth: The crazy, and the insane. The first sign of insanity is denying that you're crazy.

Reply to
Michael A. Terrell

Too soon old :-(

CT scan for me tomorrow morning :-(

...Jim Thompson

-- | James E.Thompson, P.E. | mens | | Analog Innovations, Inc. | et | | Analog/Mixed-Signal ASIC's and Discrete Systems | manus | | Phoenix, Arizona 85048 Skype: Contacts Only | | | Voice:(480)460-2350 Fax: Available upon request | Brass Rat | | E-mail Icon at

formatting link
| 1962 | I love to cook with wine Sometimes I even put it in the food

Reply to
Jim Thompson

Yep -- I figure this thing will be just the right size to fit on top of a D cell, and an appropriate load that'll last a few hours. I've already lit an LED by two half-dead AA's in series, which works quite well. But again, where's the fun in that, right?

Tim

--
Deep Friar: a very philosophical monk.
Website: http://webpages.charter.net/dawill/tmoranwms
Reply to
Tim Williams

D

it an

,

Is it those LED's that LEDdummy sent us some time ago you're talking about? If yes, then I've already put 10 of then to use exactly this way. My g/ f needed a new light for the bicycle, so I thought these would be perfect because of their wide viewing angle. Ended up just soldering

10 of them in parallel on a piece of stripboard and hooked them up directly to 2 non rechargeable AA batteries. The array draws about 550mA from fresh batteries, which is well within their specs of 70mA per LED. From a few feet away the light is nearly blinding, but I guess that's good, hopefully people will back off when driving right behind her. Oh, and that took less then half an hour to put together, with the thiefy thief I'd be still at the drawing board.

Well, I'm off. Peace ;)

Reply to
rouslan

Don't you wish you were a Europeon or Canuckistani now? You could wait another year for that.

--
Keith
Reply to
krw

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.