If you knew a bit more about _analogue_ signals, you'd know that the inverter is simply an inverting voltage amplifier (although that should be pretty obvious in the CMOS version). If you provide 100% negative feedback from the output to the input of an inverting amplifier, you should expect it to settle to a steady value somewhere midway between the extremes of its voltage swing. The TTL is a bit harder to analyse, but it is made up of a common-base input stage followed by a splitter stage that acts as an emitter follower to the lower output transistor (a common emitter amplifier) and a common emitter amplifier to the upper output transistor (an emitter follower). That is also an inverting voltage amplifier. There are no mechanisms for oscillation without a phase change to create positive feedback. The oscillations you are seeing are most likely spurious cross coupling from the other gates that you left floating or pickup noise from stray electric or magnetic fields from the long wires you used. If you want to create oscillations with logic gates, use a Schmitt trigger like the 7414 or CD40106. A resistor from output to input and a capacitor from input to ground will allow the capacitor to charge and discharge between the hysteresis levels to create a stable square wave at the output.
I am a noob at electronics and I immediately spotted sonething was off. Stacking inverting logic like this with short wires and tiny capacitances will produce up to several GHz oscillations that a shitty 100 mhz scope will not show properly
Normally you should get oscillation with TTL gates because there IS a small delay betwen input ant ouput. So if the input goes from 0 to 1, after the delay, the output will switch to 0 and the input too at that same time. But because of the delay, this input 0 will not be reflacting a 1 at the ouput until the delay has passed and will allow th 0 at the output to settle down and exists. Same for the other edge.
@@johnyang799 Not yet - but next week I'll wire up a comparison and use a thermal camera to see if there's a big difference (along with measure the current draws)
Sorry, I came back to apologize, Just plug an odd number of them together, but not 1, you are shorting the transistors. With 3 you will create what is called a ring oscillator, it will go fast. Understand your oscilloscope is unlikely to measure this properly, but should get the frequency. The tau value for this is only 2 x propagation delay of your inverters x N. Also, if you chain 3 inverters together, you create 3 outputs each very roughly 120 degrees out of phase. This "very roughness" is sometimes used for random number generation, although I have read papers that say it's a bad idea.
Satellites can program laptops, battery packs to become better. Maybe, satellite companies play a game with the engineers, and they don’t know what causes the behaviour of chips or even smaller transistors, because it’s secret. When a bad guy uses a device, satellites shut down the components, and overheating happens. When the same device gets to good working people, satellites fix the components fully offline, without paying net or mobile provider. During a fix, you have to get to a list, that’s similar to a Napster download. Asus gets easier to the toplist to get the fix, other brands may take months. Asus is tricky, they may attack the device by a drone, if you play Doom, and those benefit, who praise Jesus with the text input.
If you knew a bit more about _analogue_ signals, you'd know that the inverter is simply an inverting voltage amplifier (although that should be pretty obvious in the CMOS version). If you provide 100% negative feedback from the output to the input of an inverting amplifier, you should expect it to settle to a steady value somewhere midway between the extremes of its voltage swing.
The TTL is a bit harder to analyse, but it is made up of a common-base input stage followed by a splitter stage that acts as an emitter follower to the lower output transistor (a common emitter amplifier) and a common emitter amplifier to the upper output transistor (an emitter follower). That is also an inverting voltage amplifier.
There are no mechanisms for oscillation without a phase change to create positive feedback. The oscillations you are seeing are most likely spurious cross coupling from the other gates that you left floating or pickup noise from stray electric or magnetic fields from the long wires you used.
If you want to create oscillations with logic gates, use a Schmitt trigger like the 7414 or CD40106. A resistor from output to input and a capacitor from input to ground will allow the capacitor to charge and discharge between the hysteresis levels to create a stable square wave at the output.
I am a noob at electronics and I immediately spotted sonething was off. Stacking inverting logic like this with short wires and tiny capacitances will produce up to several GHz oscillations that a shitty 100 mhz scope will not show properly
It would be nice to see you trying to make a ring oscillator by adding more inverters in the loop.
Normally you should get oscillation with TTL gates because there IS a small delay betwen input ant ouput.
So if the input goes from 0 to 1, after the delay, the output will switch to 0 and the input too at that same time. But because of the delay, this input 0 will not be reflacting a 1 at the ouput until the delay has passed and will allow th 0 at the output to settle down and exists. Same for the other edge.
It would be interesting to see the behaviour of a Schmitt trigger inverter in the same circuit.
These cmos circuits can use a boatload of power when the inputs are indeterminate (both transistors are somewhat 'on').
That's a good point! I didn't even think to feel if the chip was warming up. Sounds like I've got another measurement to take.
@@DrShaneMTU Is there a follow-up to this?
@@johnyang799 Not yet - but next week I'll wire up a comparison and use a thermal camera to see if there's a big difference (along with measure the current draws)
ua-cam.com/video/WZJ1xLhIEXg/v-deo.html Here you go! Thanks @ByWire-yk8eh for the idea!
I made an oscillator on a CPLD using inverter.
You haven't tied the unused inputs to GND or VCC. It tells you to do this in the datasheet. Did you read it?
Sorry, I came back to apologize, Just plug an odd number of them together, but not 1, you are shorting the transistors. With 3 you will create what is called a ring oscillator, it will go fast.
Understand your oscilloscope is unlikely to measure this properly, but should get the frequency. The tau value for this is only 2 x propagation delay of your inverters x N. Also, if you chain 3 inverters together, you create 3 outputs each very roughly 120 degrees out of phase. This "very roughness" is sometimes used for random number generation, although I have read papers that say it's a bad idea.
Satellites can program laptops, battery packs to become better. Maybe, satellite companies play a game with the engineers, and they don’t know what causes the behaviour of chips or even smaller transistors, because it’s secret. When a bad guy uses a device, satellites shut down the components, and overheating happens. When the same device gets to good working people, satellites fix the components fully offline, without paying net or mobile provider. During a fix, you have to get to a list, that’s similar to a Napster download. Asus gets easier to the toplist to get the fix, other brands may take months. Asus is tricky, they may attack the device by a drone, if you play Doom, and those benefit, who praise Jesus with the text input.
Thank you for your kind words.
well thats a computer, with a clock cycle, only a clock cycle generator to be exact.
I made an oscillator on a CPLD using inverter.