> -----Original Message----- > From: piclist-bounces@mit.edu [mailto:piclist-bounces@mit.edu] On Behalf > Of RussellMc > Sent: 14 March 2012 14:28 > To: Microcontroller discussion list - Public. > Subject: Re: [EE]:: "Interesting" LED flasher circuit >=20 > So. Here is how it works with no cap leakage. > E&OE. > Normal time constants do not work here as will be seen. >=20 >=20 > OPERATION: >=20 > Assume no cap leakage. >=20 > All off. > C1! discharged,. C2 discharged. > Power on. > C1+, C2+ pulled high via LEDs by battery. > As C1 and C2 have no V across them the -ve sides will also be pulled hig= h. >=20 > Bothe bases pulled high - boty transistors try to turn on. > One will win. > Say Q1 turning on faster: > C2 will have some charge as -ve side discharged via base (or left side > chgd via R1 D1). > QA Q1-c is driven to ground, C2- will be driven to BELOW ground. > Q2 driven hard off > If C2- driven more than about 0.6V below ground D4 conducts to clamp -ve > excursion. Also provides a discharge path via Q1 - see below. > . > Now Q1 on Q2 off. > C1+ will charge via R2D2 (not a robot). > Usually R4 is to B+ and it charges C2 so Q2 turns on. > Q1 is on as above as C1 high but C1 is charging via R2 D2 C1 R3 so -ve > end > falls until Q1 is not held on as Vb too low. > But now C2 is discharged due to path Ground-D4-C2-Q1. So when Q1 turns > off > Collector of C1 rises and lifts C2+ and so C2- follows so Q2 turns on and > identical second half cycle happens >=20 > This is a much damaged version of the original as usually bases are drive= n > negative and do not conduct during timing part of cycle and resistor > charges capacitor via a normal time constant. Here the caps forward bias > the bases when the opposite collector rises and you get massive base > current pulse via eg R1 D1 C2 Q2_base. So time constant of RC is much > shortened so oscillator will run rather faster than usually. T =3D RC of > bases is notionally 10k x 100 uf =3D 1 second but because capacitors are > clamped and discharged directly into bases this will run much faster. >=20 > So it works after a fashion but is crippled. I tried to simulate this in LTSpice over lunch; nothing I have tried so far= will make the circuit oscillate reliably in its original form, including m= ake the capacitors significantly different values, adding some leakage, cha= nging the collector resistors etc. I get some chaotic behaviour at power o= n, but not really what you'd expect from an LED flasher. Moving the base r= esistors up the Vcc immediately creates a perfectly operating astable. That's obviously not to say it couldn't work in practice, but I very much d= oubt it will be reliable or repeatable between different builds. I wonder = how many people have thrown their soldering irons away in disgust after was= ting hours with this? Mike =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D This e-mail is intended for the person it is addressed to only. The information contained in it may be confidential and/or protected by law. If you are not the intended recipient of this message, you must not make any use of this information, or copy or show it to any person. Please contact us immediately to tell us that you have received this e-mail, and return the original to us. Any use, forwarding, printing or copying of this message is strictly prohibited. No part of this message can be considered a request for goods or services. =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D --=20 http://www.piclist.com PIC/SX FAQ & list archive View/change your membership options at http://mailman.mit.edu/mailman/listinfo/piclist .