I am using a breadboard power supply that hooks into both power rails and has selectable 5v or 12v power. I have it mounted near the top of my breadboard and am constructing circuits near the bottom and am getting all kinds of strange circuit behavior. By plugging the power supply closer these problems go away. I threw away two 555s last night thinking they were broken since my 555 test circuit wasn't working properly.

Why does 4 inches of distance make such a problem?

 

4 inches of wire has inductance that causes the supply voltage to the 555 to jump up and down. the datasheet for the 555 recommends a supply bypass capacitor of 0.1uF 9ceramic) for high frequencies and an electrolytic capacitor of at least 1uF for low frequencies. Both capacitors must be connected very close to the power supply pins of the 555.

 

I did put these capacitors right next to the chip. But wouldn't get consistent results unless the power supply was really close.

 

An ordinary 555 causes current spikes of 400mA in the power supply. If the supply and its wires are not extremely strong then the supply voltage drops very low each time its output switches. Maybe you should use a Cmos 555 called an LMC555, TLC555 and ICM7555.

 

An ordinary 555 causes current spikes of 400mA in the power supply. If the supply and its wires are not extremely strong then the supply voltage drops very low each time its output switches. Maybe you should use a Cmos 555 called an LMC555, TLC555 and ICM7555.

400ma. Is this documented anywhere?

In general, bypass caps are always a good idea. If you don't need them it won't hurt, but the reverse is not true.

 

Whenever you are using a breadboard it best to put a sizable capacitor across the power of every single chip, as close to the chip as possible. Even a single breadboard pin connection can have a capacitance that will cause voltage drops when current is drawn suddenly - such as any time you switch in an sort of oscillating circuit.

If you have a cro slow down the time and watch the rise and fall time on the power rails at the chips power leads.

 

400 mA? Yes this switching spike has been known since the outset, but I think the actual value depends upon circuit conditions. It has upset many an apparantly promising design. And yes the Cmos versions are a great improvement - but only at low speed, it is often forgotten that as Cmos speed increases so does the current draw.

 

400ma. Is this documented anywhere?

Yes.
Intersil explains it in the datasheet of their Cmos 555 the ICM7555.