Hex inverter buffer mystery

[CameronHanson20]

Hello everybody. I have designed a circuit that is supposed to turn on an led after a certain amount of time. The circuit as I have drawn it and excuse the poor handwriting, works exactly like I thought it would. I was trying to think of improvements on it and to reduce part count. So I took the output of my flip flop and bypassed it straight to the nand gate input or point B in the drawing, bypassing the hex inverters. Curiously enough, the circuit starts misbehaving and the output will immediately turn on as soon as power is applied. I noticed that when power is applied the 555 monostable will actually have a "hump" and this hump lasts for around 6ms and reaches a peak voltage of 1.3V but if you look at the circuit it shouldn't matter because the flip flop output is a 0, so the nand gate returns a logic 1, which then is inverted back into a 0 and the LED is off. Does anybody have any ideas on why bypassing those inverters causes it to misbehave?

[moffy]

Make sure that all the inputs to the hex inverters are pulled high or low else you will get high frequency oscillations!

[CameronHanson20]

Thank you moffy for your quick reply and yes you were absolutely correct, it did solve my problem. I guess when I started moving wires around I got a little careless and left some inputs floating. Now if you don't mind me asking, the hex buffer shares power and ground, correct? Do oscillations may their way back up and then interfere with other outputs?

[moffy]

Yes through both the power supply and by parasitic capacitive coupling (the air). The edges are very fast. Ran across the same problem just recently. One of my links to an inverter input wasn't making contact.