74LS04 - quick question pleease

orion3311

Well-known member

Donor 2013
Joined
Nov 25, 2008
Messages
5,217
Reaction score
30
Location
Blue Bell, Pennsylvania
7404 is a logic inverter...thus...if a signal has peaks on the input, it should have identical spikes on the output, correct?

meaning...


INPUT: _/\__/\__

OUTPT: -\/---\/--

Right?
 
if a signal has peaks on the input, it should have identical spikes on the output, correct?

Well... for each inverter in the IC the input will be inverted on the output.

So if the input is high the output will be low. If the input is low the output will be high. I tried to draw it with ascii but it sucked. Just remember that the output is the inverse of the input.
 
So explain to me how the output would be the same across 3 different 7404 ic's?

Going from pins 3-4 on the o-scope are identical, even if I clip lead another 7404 to the existing one just for a comparison check?

YET

Pins 1 to 2 show an inverted signal, and pins 12 to 13 show an inverted signal. I can't have 3 bad chips in a row?? (They ARE used)
 
So explain to me how the output would be the same across 3 different 7404 ic's?

Going from pins 3-4 on the o-scope are identical, even if I clip lead another 7404 to the existing one just for a comparison check?

YET

Pins 1 to 2 show an inverted signal, and pins 12 to 13 show an inverted signal. I can't have 3 bad chips in a row?? (They ARE used)

Shorted trace on the board? If pin 3 and 4 were shorted I could see it doing that. Or a short somewhere else causing the input signal to short to the output pin. I would take a look at the schematic and try to rationalize what I thought was happening. You could test the gates out of circuit to be sure they're good but I would suspect something other than a bad IC at this point.
 
Your first picture was correct: (try using CODE tags to format it).


Code:
                __              _
INPUT:    _____|  |____________| |_____
 
          _____    ____________   _____
OUTPUT:        |__|            |_|
 
Thats what I thought, and thats what I"m getting between pins 1-2 and 12-13 as it should, but pins 3-4 and 5-6 which is a clock signal off the CPU is NOT inverting, more like clipping if anything.

The kicker is, I powered up a 7404 out of circuit - send it the same clock signal, and it does the same thing OUT of circuit! WTF!
 
See the attached - clock signal comes off the CPU and goes into pin 5 of the 7404 - that should be inverted at pin 6. Then pin 6 goes into pin 3 which should be inverted again at pin 4, meaning, pin 4 and pin 5 should almost be identical, not pin 3 and pin 4.

I also checked for shorts - no continuity between pins 3 and 4. Game was working then quit in the middle of a run, so unless a chip or cap shorted, it shouldnt show a short.
 

Attachments

  • Untitled.jpg
    Untitled.jpg
    5 KB · Views: 38
a bit hard to read the schematic sample but if i read it correcly, check the AND gate connected between the two inverters - if it is faulty it may be dragging down the output of the first inverter. or holding down the input of the second inverter (or the inverse)
 
Apparently youtube links aren't working anymore - anyhow - here's what I'm seeing. This is the first inverter off the clock signal. I sandwhiched an extra socket under the 7404 and bent out pin 6 so its connected to nothing, which eliminates anything else causing this to act up.

Pin 5 is the clock signal from the CPU. Pin 6 should be the inverse of it, but its not.

http://www.youtube.com/watch?v=j-pnDCs74a0

I've seen this now with 3 seperate 7404s I pulled from junk boards. I have a possible line on some new ones locally so maybe I'll try a brand new one tomorrow.
 
Last edited:
PS - The reason I tackled this part is because I've exhausted almost everything else on this board, I just can't find anything wrong, but this definitely seems strange.

On earlier discussions it seems that the address or data busses could have a problem, well everything checked out but today I realized that the address bus is switched between the CPU and some preset sync signals. Its switched using an inverse of the cpu clock signal, however it seems both buffers (244s) seem to be getting the same signal, which means the address bus is getting conflicting info from 2 seperate chips.

The only thing I can think of at this point - and you can see this in the video, is that the clock signal is not a perfect square wave, in fact far from it, and I"m almost wondering if that lack of squareness on the front edge is what's causing the 7404 not to invert.

Why its not a perfect square wave I'm not sure - its generated from a counter, unless that counter is bad.

Just to recap - this game was on one night, and died during some gaming. Originally the processor would watchdog as soon as its turned on with a constant beeping sound (it beeps when first turned on). Test mode seems to work with the exception of video. Now it seems the processor randomly watchdogs or doesn't - no rhyme or reason. Video on screen is a solid sync'd screen with random garbage, and flickering lines amongst the screen. I think the flickering is due to the address bus getting 2 signals at once.

THe only thing I managed to fix so far on this board was the fact that it had no video whatsoever, due to a missing blanking signal. Replaced a chip @ K5 to fix that.
 
Ummm.....did you try looking at both signals at the same time? I bet it actually *is* inverting. But if you only connect one signal the scope is always going to trigger (start drawing) on a rising edge (or falling edge, depending on the "slope" setting). So yeah...try looking at both signals, say channel 1 = pin 5 and channel 2 = pin 6.

As for the non-ideal square wave, there's a good chance the CPU is NMOS. NMOS chips have an asymmetrical output stage. It's also normal for TTL chips to output about 3.5V for a high signal and 0.4V for a low signal. The falling edge will often be sharper than the rising edge, though not as blatant as an NMOS part.
 
Apparently youtube links aren't working anymore - anyhow - here's what I'm seeing. This is the first inverter off the clock signal. I sandwhiched an extra socket under the 7404 and bent out pin 6 so its connected to nothing, which eliminates anything else causing this to act up.

Pin 5 is the clock signal from the CPU. Pin 6 should be the inverse of it, but its not.

http://www.youtube.com/watch?v=j-pnDCs74a0

I've seen this now with 3 seperate 7404s I pulled from junk boards. I have a possible line on some new ones locally so maybe I'll try a brand new one tomorrow.

Umm...you need a second scope probe to be able to say if that's inverting or not. The scope is going to trigger on rising edges (or falling, whatever you've set), and since it's a square wave in both cases, it's going to look the same on both. You can see the invert better on non-clock signals because they tend to spend most of their time either high or low, and the inverted signal will be the opposite.

To tell if it's inverting, you'll need a second scope probe, so that you can see BOTH signals at one time (you'll trigger off just one of the signals). When you do that, you'll see that the chip is in fact inverting it's output.

As to the signal quality - I don't think it looks bad at all. As long as the edges are sharp (they looked OK), and the highs stay above 3 or so volts and lows below a volt, you should be fine. That bit of ramp is an artifact of the chip design, but as long as when the signal goes high, it goes to at least 3.5 or so and stays above that, you'll be OK.
 
Thanks guys - I am still learning this stuff so I figured I'd ask just incase that slope did matter. The signal looked good to me otherwise.

I may be able to get access to a second probe to try that out - I should just get a second one.

PS - that scope I'm using is a junker I got in one of my Quarterarcade stashes but save for one dirty knob it seems to work good. I have a better 100mhz Hitachi (don't ask me why I"m not using that instead).
 
Haha. Any scope will do if it shows you the signals. :D But two channels are the minimum for a practical scope. So yes, get a second probe. If you can't find any locally, let me know. I have way too many of them and I can send you one for next to nothing.
 
Maybe I missed it if you did it, but did you calibrate the probes to the scope? most scopes have calibrate test point on it somewhere and an adjustment to compensate for probe capacitance changes.
 
Of course what you *really* need is a logic analyzer.
icon10.gif


attachment.php
 

Attachments

  • READ_01.jpg
    READ_01.jpg
    19.6 KB · Views: 45
Back
Top Bottom