Anyone have a handy transistor array (NPN RET or logic level NFET) that'll easily drive LEDs? I need cheap (Joerg? ;) and small. Normal RETs are in the $.08 range but only come in duals in a SOT-363 (six lead) or some such. I'd like something with four or eight transistors with a common emitter (source). 5V 30mA each is good enough. I looked at the usual suspect's sites today but haven't seen anything interesting.
ULN2803 type? Kinda beefy though.
Open collector/drain logic inverters, like HC05? Dirt cheap.
You can also use an octal totem-pole part, HC240 or some such.
John
4mA per output max, 50mA through device gnd pin max....
8mA per output.
Right. I was looking at the LV logic family for a similar application a while back. Same issue, though then I could guarantee that not all LEDs would be lit at once (and I didn't need to see it outside).
LV240/4 is 20mA per, IIRC, but still 50mA package max.
I passed by similar stuff earlier. I really don't want the Darlington Vce(Sat) sucking up the power.
I'm looking for the similar things, so I'll watch along. (No darlington allowed.) I had given up and just used discrete PN2222A, which are large but cheap (less than 1/2 cent each.) The SMT variety in SOT-23-3 are way too expensive (an order of magnitude or more) and hard for students to work with, besides. A SIP would be the perfect form factor for me.
Jon
Why not? You have series resistors on LED's sucking up the power. Just consider it part of the total ballast.
In the pulldown direction, the big current is the chip to ground paddle, so it's probably OK to push them some. I suppose chips can eventually die from electromigration, but I've never seen it happen.
TI's AC05 is rated 50 mA/pin, 100 for the package. The ancient TI SN7407 is "recommended" at up to 40 ma/pin, with no total package spec.
John
A voltage source doesn't make a very good ballast. All it does is suck power that is otherwise usable.
It's certainly not a place I'd like to go. Certainly driving 160mA is out.
The AC05 (or '04, I suppose) is a possibility. Switching current (shoot through) in AC parts tends to be high. Though maybe it doesn't matter, compared to the LED current.
I need SMT. ;-) I'd like something smaller than eight SOT-23-3s, but that's what we're using now. I'm doing a cost reduction and part of that is trying to squeeze everything possible onto one board.
Well, I'll tag along for the ride. Maybe someone knows of a SIP that isn't an arm and a leg to buy. I'm also interested in SMT, though.
Jon
RET?
Tim
-- Deep Friar: a very philosophical monk. Website:
krw a écrit :
Not an array but small enough, maybe...
-- Thanks, Fred.
On a sunny day (Wed, 02 Sep 2009 19:23:17 -0500) it happened krw wrote in :
Any 7 segment LED driver? You'd have to do some BCD coding in your FPGA or micro.
OK forget that remark :-) Would be very limiting as to what LED was on or off.
On a sunny day (Thu, 03 Sep 2009 11:46:04 GMT) it happened Jan Panteltje wrote in :
It is also worth considering that modern LEDs need very liitle current. .5 mA would be more relalistic. And in that case you can drive those from any thing, even directly from a FPGA.
30 mA would light up a room with a high efficiency red LED, and blind the user.Tiny logic drives up to 32mA(50mA) per pin, 100mA per package, up to 3 outputs:
For more than 4 outputs, some analog switches may be a solution. See:
or some interesting special switches for high side multiplexing:
MIKE
-- www.oho-elektronik.de OHO-Elektronik Michael Randelzhofer FPGA und CPLD Mini Module Klein aber oho ! Kontakt: Tel: 08131 339230 mr@oho-elektronik.de Usst.ID: DE130097310
Do you really need that much current? How about super-bright LEDs run at
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.