LED sequencer using only JFETs

wouldlovesomehelp · Aug 16, 2011

Hi,

I am after some help and came across this forum which i believe could give me the answers that i require.

I have managed, with my limited knowledge, to get a LED sequencer working using only J112 JFETs. I need to develop a LED sequencer which does not use PICs or other ICs as I only have JFETs, capacitors and resistors available to me.

I have attached a circuit which does work but i require a mods which i have no idea how to achieve. I need to only flash one LED on at a time, at the moment they are all on and then one goes off. If any one can help i would be most grateful.

Many thanks in advance.

James

KJ6EAD · Aug 16, 2011

Perhaps you can adapt a version of the Discrete Multistage Light Sequencer shown on this page:
http://www.bowdenshobbycircuits.info/page5.htm#seq_1.gif.

(*steve*) · Aug 17, 2011

That's pretty much what he already has.

If I were him, I would look at monostables built using jfets (which he says he has). I would look for a design that is triggered on a negative edge, and produces a positive pulse.

Hooking these in a ring, and providing an initial reset/pulse would get a light chasing itself around the loop.

Although it might tend to be unstable, you might be able to get more than one light chasing itself by giving multiple spaced pulses when switched on.

KJ6EAD · Aug 17, 2011

He has a sequenced OFF circuit. The one I linked is the more traditional sequenced ON.

poor mystic · Aug 17, 2011

i thought n-channel jfets need a -ve control voltage? I must be confused. I don't see what makes this multivibrator work

(*steve*) · Aug 17, 2011

A JFET is a depletion mode device, that is, the (negative) control voltage restricts the channel, thus turning the device off.

Some amount of +ve bias can be given (up to 0.7V) before the diode that forms the gate junction conducts, however this mode of operation is seldom used.

In this circuit the gates are all held off by 100K resistors.

However, the gate of the first jfet is forward biased by the switch (and will actually conduct through the gate and the LED to ground). This is a very poor design, but nevertheless, it turns the jfet on.

As the switch is released the jfet turns off. The voltage at the junction of R1 and C2 rises and this, in turn makes the gate of Q2 more positive, turning it on.

As the capacitor C2 charges, Q2 turns off, and the voltage at the junction of R2 and C3 rises, which starts the process on Q3, which triggers Q1, then Q2, then Q3...