Constant-current LEDs?

Hah, I was thinking about an led in Fred's circuit, driving home. Put in two... they can be the indicator led, that saves a part. (do they make dual leds?)

George H.

plenty, but usually with two different colors ;)

-LAsse

If you use an LED in place of a Zener, the corresponding resistance needs to be changed. To keep from wasting half the current, I suppose the other resistor can be reduced. No need to run the same current in both legs, eh?

If you only need one LED per output, the other just takes up current.

Yes, usually bi-color, but there's nothing wrong with that. They could be run at different currents, as needed.

An LED-bipolar current source can have a near-zero TC and pretty low voltage drop.

For James' problem, it could be done with one LED, two transistors, and some resistors, mostly resistor packs. Somebody makes a dual pnp-npn transistor in one package. Or spread duals across the channels.

How about this?

Voltage limitations there. If you look at the OnSemi automotive grade CCRs, they're protection against overvoltage spikes (60V load dump) as well as LED driver- dual function.

as

istors in small quantity. Precision should be adequate for driving an indic ator LED. Three parts on bom. Dunno about your headroom, a done deal at 5.5 V.

each "leg".

I'm

about 47k output impedance- which is not bad for a LED driver. Advantage is it handles a LOT of voltage, and power dissipation is divided equally betw een the two transistors- each conducting half the current at half the volta ge- so runs pretty cool.

Correct my statement to half the current but full voltage minus the zener.. .need something else to halve the voltage :-)

So it might not actually melt or catch fire, then. Very reassuring. ;)

I did the calculation based on the numbers in the BCV61 datasheet, and the theta value I backed out was very similar to that of isolated transistors. Plastic is a really putrid thermal conductor.

Cheers

Phil Hobbs

The best idea so far is probably the LM317L/SOT89 and a resistor.

You call using a 317 an "idea"??? You ARE getting old.

Unlike certain parties here, I don't have trouble starting up.

Here's a start-up fix, dunno why it never made into the main thread, it shows on google:

Please view in a fixed-width font such as Courier.

. . . CCR . | . | . ----+---- . | | . --- [680] . ZD / \ | . --- | . | |< ZD = BZX84C2V4 or equiv. . +-------| T1 . | |\ TEMPCO -1.6V/oK TYP. . | | . +--[47k]--+ . | | T1/2=BC847BPDW1T3G or equiv. . \| | . T2 |-------+ .

That's the Thompson Methodology: just keep adding parts.

I wanted to invent something truly goofy for April Fools Day, but this is the best I can manage,

which, unfortunately, is pretty reasonable.

......

Wait a sec, this one is better:

Well, that's enough circuit design for now. I've got to take a shower and head out to work and design circuits.

The Zetex (Ferranti) E-line transistors are molded with silicon epoxy. They usually have better thermal spec than run of the mill TO92 that are a fair bit bigger.

Sometimes the LED parts in LED replacement bulbs are multi-chip white.

The configuration is at the manufacturer's whim - I've seen LED parts with multiple chips that have regular Vf, some have multiple chips in series in a single LED device. One bulb I stripped to see what's in it had 9 LED devices and the measured Vf-tot divided by 9 gave about 18V per device.

A green LED comes in pretty close to 2.0V and is claimed to be pretty stable.

One of the UK TV manufacturers used one as a bias generator for the video output transistors.

Never used the LND250, and never will apparently since the suppliers say it is obsolete and unavailable...

Yeah- don't want to overdo it...

The LND150 and LND250 are the same part. One of them has been EOL for years. I don't know why they did that. To add to the confusion, the Supertex web site now shows both as "in production" status.

Depletion mosfets can be really useful. This one (or two?) has Idss consistently close to 1.6 mA, and is rated for 500 volts, is gate protected, and avalanches nicely around 620 without drama. So you can make HV stacks with them.

Since you don't design electronics, it figures that you will never use either.

Apparently the channel conduction can be enhanced significantly with a positive VGS, the output characteristic showing over 5mA at VGS 1.0V for the 1.6mA IDSS part. How come you don't make use of that feature in your April Fool's circuits...