Author Topic: Logic Inverter "noise" (Read 13185 times)

PA0PBZ · « **Reply #25 on:** July 18, 2013, 07:25:53 am »

I still think the chip was bad in the first place, so all you could have done is make it a little worse... good luck being patient

staze · « **Reply #26 on:** July 18, 2013, 07:32:17 am »

Quote from: PA0PBZ on July 18, 2013, 07:25:53 am

I still think the chip was bad in the first place, so all you could have done is make it a little worse... good luck being patient

Totally agree. And, yeah... I just want to get it off the bench and onto the shelf. I got other projects to tackle! =)

smashedProton · « **Reply #27 on:** July 19, 2013, 04:35:59 am »

Please forgive me if I say something dumb, I am a beginner.
Wouldn't it be best to just use a 74hc14 schmitt inverter and call it good?

David_AVD · « **Reply #28 on:** July 19, 2013, 04:40:55 am »

Quote from: staze on July 18, 2013, 07:23:30 am

Also note from looking online, CMOS logic chips should have all their inputs and outputs connected, even if not in use... so maybe that's part of the issue too.

Unused CMOS inputs should always be tied high or low. Outputs don't give a rats.

staze · « **Reply #29 on:** July 19, 2013, 05:10:25 am »

Quote from: David_AVD on July 19, 2013, 04:40:55 am

Quote from: staze on July 18, 2013, 07:23:30 am
Also note from looking online, CMOS logic chips should have all their inputs and outputs connected, even if not in use... so maybe that's part of the issue too.

Unused CMOS inputs should always be tied high or low. Outputs don't give a rats.

Ah, okay, that makes sense. just for S's and G's I'll try that tomorrow and see if my tests change at all.

David_AVD · « **Reply #30 on:** July 19, 2013, 05:30:12 am »

Floating inputs (even on "unrelated" gates) can cause all sorts of hell to break loose.

staze · « **Reply #31 on:** July 19, 2013, 06:00:40 am »

Okay, that didn't seem to make much of a diff. I tied all the inputs to my signal (10Khz, 2.5V +pulse), and all the outputs look pretty much the same... about 800mV, and not even inverted. =/

Is there something wrong with doing it that way? Interestingly, the input from my FG was DC offset by 1V (FG to Tee, one half to my scope, other to the circuit, FG set to HiZ). So the 74HCT04 was pushing 1V BACK into my FG output and offsetting the pulse. =/ Guessing that shouldn't be the case...

David_AVD · « **Reply #32 on:** July 19, 2013, 06:17:57 am »

Sounds like a fundamental issue at play. Got a picture of the setup?

staze · « **Reply #33 on:** July 19, 2013, 06:39:34 am »

Sure.

Inputs are 1,3,5,9,11,13. All are tied to the bus on the side, which I'm feeding with the FG. Red clip is from the FG, Black clip is the "gnd" from the FG.

Red wire heads off to my PSU positive with +5V rail. Black wire (where the black clip is) also heads off to the PSU gnd (negative). My scope probe is clipped to that wire out of frame. Obviously pin 7 is gnd, pin 14 is +5V (Vcc).

Just testing the output pins with my scope. Once I got it all hooked up, the chip was pulling 20mA of current at 5V from my PSU.

I don't THINK there's any issue with this, but I'm just a beginner, so I could be completely off base. Mind you, this is the chip that appeared to be bad from all other testing, and in circuit.

David_AVD · « **Reply #34 on:** July 19, 2013, 06:47:12 am »

Two problems:

1) No decoupling capacitor from Vss to Vdd.

2) It looks like you've gotten the pinout of the chip wrong. Check the data sheet again. (Vcc is pin 14 and Vss in pin 7)

staze · « **Reply #35 on:** July 19, 2013, 07:01:17 am »

http://www.nxp.com/documents/data_sheet/74HC_HCT04.pdf

Inputs are 1,3,5,9,11,13. Outputs 2,4,6,8,10,12. Vss is pin 14, Vdd is pin 7. Oh, son of a... how the hell did I get the impression the pins went clockwise... thinking about it... gar....

Didn't know you needed a decoupling cap between those. I just looked at this: http://www.vagrearg.org/content/decoupling

Still not sure what size cap would be appropriate... that said, the issue doesn't seem to be stalled output (failing to rise fully) (as seems to be indicated on the page above), it's that it just doesn't really come up at all... but I'll add one and see what I get.

Alright, I'll retest once I know what size cap to use. It better still be bad... I KNOW I was testing it right in circuit (since I referenced everything off Vcc and Vdd, and the inputs were all clean, and outputs bad).

Gar.... thanks, can't believe I did that.

David_AVD · « **Reply #36 on:** July 19, 2013, 07:04:49 am »

The screwed up pinout meant that +5 and ground weren't even going to the correct pins.

Use a 100n decoupling capacitor between pins 7 and 14.

EDIT: To be clear, pin 14 is Vdd (+5) and pin 7 is Vss (ground).

staze · « **Reply #37 on:** July 19, 2013, 07:16:39 am »

okay, just fixed it. Used a 100nF (tested at 92.6nF) cap across pins 7 and 14.

Inverter A1, A2, and A5 just give solid 5V output. A3 and A5 give 0V output. A6 comes close with an inverted pulse, but rather than being 5V to 0V, it's 5V to about 1V.

So, still dead dead dead. Though I'm sure driving it the way I did in previous tests probably did something bad. Replacement doesn't arrive until Tuesday. Gar.

David_AVD · « **Reply #38 on:** July 19, 2013, 07:19:33 am »

Can you post a picture of the corrected layout please?

staze · « **Reply #39 on:** July 19, 2013, 07:23:05 am »

Here ya go.

PA0PBZ · « **Reply #40 on:** July 19, 2013, 07:25:29 am »

That looks ok now.

David_AVD · « **Reply #41 on:** July 19, 2013, 07:26:16 am »

Looks correct now, but you may have toasted the chip with that arse about wiring earlier!

staze · « **Reply #42 on:** July 19, 2013, 07:29:08 am »

eh, like I said, it was seemingly bad in circuit (producing all kinds of weird output), so I pulled it reading for a replacement, and tested just because, well, why not. =)

But yeah, no doubt I could/may have toasted it. =) It's how we learn... potentially bust crap while testing it. =) I guess the biggest downside is not knowing how broken it was, since there's no way to know whether it's current state is my doing. Ah well... bad troubleshooting, good learning experience.

Thanks for the help! Now I just sit back and wait... =)

staze · « **Reply #43 on:** July 21, 2013, 06:08:05 am »

Interestingly, I just fired my test back up this evening, and the inverter seems to work.... so... not quite sure about that. I do note that the rise time is pretty quick, but the fall time is pretty slow...

The schematics for my meter don't seem to show any decoupling cap across the Vcc and Vdd. But maybe the 7805 regulator and it's filter caps fill the need?

So, I'm at a bit of a loss. Replacement part arrives Tuesday... guess we'll have to see. =/

Rerouter · « **Reply #44 on:** July 21, 2013, 06:13:34 am »

your inverters edges move fast which means its a decent spike of current, so the closer the decoupling the better...

if your main supply capacitance is more than 5cm away, a local decoupling cap is really in order,

staze · « **Reply #45 on:** July 21, 2013, 06:46:17 am »

Makes sense.

And David_AVD, I'd really like to thank you, seriously. Once I tested that chip again, with the proper pin-out, it worked. I dropped it back in the meter, and the meter now works.

So... either the chip is intermittent, or there was a bad solder on the Vcc or Vdd pins which was fixed by me putting in a socket. Either way, thank you for the reality check. I'm still going to replace the IC since it could be intermittent, but thanks again!

SeanB · « **Reply #46 on:** July 21, 2013, 06:55:49 am »

Likely the IC has a cracked lead frame or bond wire, and the heat of desoldering and the mechanical handling has remade the connection for now. Best to replace it.

staze · « **Reply #47 on:** July 21, 2013, 06:59:25 am »

Quote from: SeanB on July 21, 2013, 06:55:49 am

Likely the IC has a cracked lead frame or bond wire, and the heat of desoldering and the mechanical handling has remade the connection for now. Best to replace it.

That makes sense. But the fact that I verified the inverter was the cause of the malfunction is extremely pleasing. =)

Thanks everyone for all your help!

David_AVD · « **Reply #48 on:** July 21, 2013, 07:04:43 am »

A sanity check is always a good thing when things don't seem to be adding up.

staze · « **Reply #49 on:** July 21, 2013, 07:05:56 am »

Quote from: David_AVD on July 21, 2013, 07:04:43 am

A sanity check is always a good thing when things don't seem to be adding up.

Indeed. Love the avatar btw... very appropriate. =)


EEVblog Main Site	EEVblog on Youtube	EEVblog on Twitter	EEVblog on Facebook	EEVblog on Odysee

Author Topic: Logic Inverter "noise" (Read 13185 times)

Share me