Static electricity in ICs

[gboethin]

Hi all, I'm a total newbie to circuits, and I'm having a very frustrating problem. I'm basically just testing a quad NOR gate right now. I've got it on a breadboard, with a battery power source. I'm using two of the gates, and the output of each is connected to an LED which shows me whether or not power is flowing to the output. Here's what happens:

Occasionally, both LEDS light up, when no power is supplied to either of the inputs. With a NOR gate power should be going to the output when both of the inputs are low. And when both LEDs are lit, I can kill one of them by supplying power to either of the inputs (which is how it should be). But when I do that, it will stay off. It won't light up again after I remove the power to the input.

Here's what's really strange. By touching the breadboard (with my finger), I can make the LEDS light up, like they're supposed to. What's even stranger... is sometimes when I sit in my chair in front of the breadboard, and take my feet off the ground, the LEDS die, but when I put my feet back on the ground, they light up again.

I can actually see this could be used for something useful, because right now this NOR gate is essentially detecting pressure on the floor. But I don't want it to do that right now; I want it to work like a normal NOR gate.

Like I said, I'm kind of a dummy when it comes to electronics, but I'm suspecting static electricty.

What is going on here, and how can I fix this problem?

Thanks in advance,
Gregg

 

[gboethin]

No ground?

Hi. I just pulled the ground off the NOR gate and noticed that it behaved the exact same way. So would that mean that something's wrong with the gate? I checked the ground on the electrode and it's good. I mean, the electrode going to the ground is grounded. Any thoughts?

Thanks,
Gregg

 

[dknguyen]

Nor

2nd paragraph: You are providing +V to produce a logic 1. But are you grounding the terminal to produce a logic 0? Or are you leaving it floating?

I am not sure about the LED's lighting up. If static was flowing directly through them I would think it would just blow the circuit outright. EM induction?

"By touching the breadboard...":Maybe your breadboard backing is not flat against the breadboard? Maybe you have a faulty breadboard? I dont know about the feet on and off the ground thing.

 

[gboethin]

Hi, thanks for the reply. I'm not exactly sure what I'm trying to do. I'm basically just testing the NOR gate to see if it behaves the way it's supposed to in a diagram I'm following. As near as I tell I've perfectly replicated the diagram. What's weird is that the circuit behaves the same way with or without the ground connected. It's as if the ground has no effect on the IC.

I've found a few things cause changes in the circuit. Waving my hand over the board causes the lights to dim/brighten. Taking my feet off the ground cause it to dim. Touching the metal part of the board causes it to brighten. Touching one of the wires in the circuit causes it to dim. Pressing down on the plastic part of the board, with heavy pressure, causes it to brighten.

BTW, there's an LED that set to always be on when the power is supplied to the circuit (it doesn't go through an IC... it's wired directly into the + and -, with a resistor). It never changes. Only the LEDs powered by the output of the IC changes.

I just noticed that the board is sitting directly above a power strip, on a table. It's also close to a laserjet printer, computer, and LCD, on oh yeah, speakers. It's within about 3 feet of the speakers. That shouldn't be causing the problem, should it?

 

[dknguyen]

Waving will cause the circuit to behave differently so the dimming is normal. MOSFET transistors like those (most likely) used in the op-amp are temperature sensitive and waving cools them down.

The IC power leads (like the ground) should definately be grounded. Maybe there is a faulty connection in your breadboard causing unwanted connections (or disconnections)

Are you grounding the input terminals whenever you want to provude a logic 0 to the NOR gate or are you just disconnecting the power from the NOR gate pin when you want a zero (you should be connecting it to ground)? I know you are connecting +V to the input terminals of the NOR gate when you want to get a logic 1. (You do understand what I mean by logic 1 and 0 right?)

Maybe you just have a bad breadboard with lots of finicky connection strips and bad springs to hold the wires in properly. Touching the breadboard possible puts pressure on the connections in the possibly faulty breadboard and makes a better electrical connection.

Or possibly you damaged your NOR gate from not following proper electrocstatic discharge procedures (ESD)procedures. Its possible you zapped the IC or it was packaged in a regular plastic bag or container and got zapped in there.

Being near the printer and speakers and stuff shouldn't do anything.

I dont know about the feet thing. Are you touching the board when you lift your feet? Maybe its making pressure differences or something on the breadboard? Have you touched a screw beside a light switch to make sure you have no static on you? Or a ESD wriststrap?

 

[gboethin]

Hey, you nailed it! I hadn't connected the inputs on the IC to the ground. I didn't realize that I had to do that. I really appreciate you helping me here! Like I said, I'm kind of a dummy when it comes to this stuff.

FYI ungrounded inputs will actually sense foot steps and proximity... just getting your finger close will trigger them.

Hey, is this something you do professionally? I ask because I'm going to need to have a circuit designed for a product I'm working on. I'm sure it'd be easy for someone with a background in circuits. We're willing to pay what it's worth. Let me know, because I'm going to be looking for a company over the next few days.

Thanks again,
Gregg

 

[jbeng]

Detecting pressure on the floor? Not likely. What is the part number of the NOR gate you're trying to use? Maybe a 4001 (or 14001)? Sounds to me like you're allowing the input of the gate to "float" when you don't have a logic level signal connected to it. Basically, the input pin is connected to nothing. 4000-series CMOS chips don't like that at all. (Nor do any modern day chips for that matter.) When you allow a high-impedance input to float like that, it is likely to sense all sorts of energy from stray electric fields (including static) and it will do very unpredictable things.

What you need to do is have a pull-up or pull-down resistor on the input. For a pull-up, connect a 10K resistor between the gate input and V+ (pin 14). That way, there will always be a logic "1" on the input unless you connect it to ground, at which time it will be a "0". A pull-down works the opposite way: a 10K resistor from the gate input to ground (pin 7) will always put a logic "0" on the input unless you supply the input with a logic "1" by connecting it to V+. If you have the input connected to the output of another gate, you can leave the pull-up or pull-down resistor out of the circuit.
JB

*edit* Also, all inputs from any unused gates should be connected to V+ or ground.

 

[dknguyen]

Student

Im just a 3rd year electrical eng student...not doing this professionally....yet...

 

[ZIGGY_DAN]

I have used cmos logic chips previously and leaving any input or ouput floating can have undesired affects, al unused gates should be grounded also. A floating cmos logic gate can detect very small amounts of static, hence waving your hand made the LED's dim or brighten.

 

[Rixeh]

*Edited for STUPIDITY*

Oops.