Thank you for that Schmitt Trigger explanation! I tried reading the Wikipedia article on the Schmitt Trigger, and it kept mentioning hysteresis (which I understand as a mechanism that behaves in such a way that depends on its own history). It'd say hysteresis this, and hysteresis that, and I wasn't any closer to understanding what a Schmitt trigger actually did. But this actually helped me tie in the concept of hysteresis with a Schmitt Trigger for me. When the noisy signal first touches the upper threshold, that threshold gets "de-activated" and the output becomes HIGH, while the lower threshold gets "activated". And so on.
Exactly. Hysteresis basically means that the threshold *depends on* the previous state of the comparator - i.e. the threshold moves when the comparator changes state such that tiny wiggles/variation/noise on the signal can't cause chatter.
This is one of my favorite vids that you made. I built the oscillator using a 74AC14 I found on ebay (I wish I can send you photos of my Manhattan Style build) and got about a 2-4 nS rise/fall which is much more than you got, or the other viewer in the update video who built a PCB around the TDR. Thank you for all those years of education, your the man!
"The Schmitt trigger was invented by American scientist Otto H. Schmitt in 1934 while he was a graduate student,[1] later described in his doctoral dissertation (1937) as a thermionic trigger.[2] It was a direct result of Schmitt's study of the neural impulse propagation in squid nerves.[2]" Surely a squid trigger ?
The statement “we want the output impedance to be equal so we can get a 50% duty cycle”. Is it as simple as, If they were unequal, the feedback would vary causing the duty cycle to change?
You can do this with TTL logic, but then are limited to fairly low resistor values. Another variation is to make a schmitt trigger by wrapping an op amp with a couple of resistors to provide some positive feedback, then couple that to an integrator (the whole thing can be easily done in a dual op amp package). The output of the integrator goes to the input of the schmitt trigger. The advantage of this one is _very_ linear triangle wave, rather than the ones you get here. Also higher amplitude, if you want, and then by doing a bit of diode steering you can get narrow pulses and sawtooth rather than triangle waves, or even variable duty cycle.
2/ another tip! If you put a transistor (pnp ie bc107) Emitter to ground , Collector to input across C , potential devider Base and the top resistor to the out put of the Shmit, there will be a saw tooth wave form across the C.
Excellent video. Would this type of oscillator be fast enough to function as a local oscillator in an RF radio circuit? What would be the practical upper freq limit? Could it cover the entire 30 Mz hf band? If so, this could simplify homebrew radio construction tremendously. Thanks for the video.
Speed will depend on the logic family used for this, and the resulting propagation delay of the gate. You'd want to follow it up with a divide-by-2 to cleanup any duty cycle distortion, so that pushes the frequency even higher. So, I'd say it likely would not be a good for an HF LO.
Great video, Great lesson...Thank You! Question: As an existing signal passes through a Schmitt Trigger Oscillator, would it's phase change in any way?
Your videos are great, Alan. Picked up a couple of 74ac14's and I am planning on building this. I had a thought that I might just add the parallel trigger array, in the output of your circuit, to the output of my inexpensive function generator, using it as a conditioner for the square wave output. That way I could vary the frequency and duty cycle. For TDR measurements, would a very narrow pulse make it easier to visualize the delay ?
Adding an external input to the driver is a good idea - it lets you play with pulse widths, frequencies, etc. Certainly a narrow pulse can be used to measure delay - potentially with less ambiguity, as long as the pulse width is shorter than the cable delay.
Did you know if you replace the feedbach resistor with a coil you get a sighnwave on the gate input ! (It dose have little glitches at the historisis points but can be easly filtered ).
1/ tip of many , did you know if you replace the resistor with a coil, the input will resemble a sine wave (with a little trigger glitch that can be filterd) .
Hi. Under specific conditions (low power, low frequency) an op amp-based gyrator can simulate a coil's inductance, so can this circuit replace the resistor as well? Thanks. en.wikipedia.org/wiki/Gyrator
Another great video! I lashed this together on the breadboard tonight, substituting a 74C14 and an 8k2 for R and run it in 5V. I get a great triangle oscillation in the input, but can't find the PWM or square on the output (or anywhere), I can only get the same triangle wave in both the input and the ouput. Any ideas?
I have an RA gray control system that uses manchester fm2 code. What would be the best way to make a new controller/programmer so that I can create new shows?!?
Terminology Question: By definition(?) an op amp is a differential amplifier yes/no? Is a Schmitt Trigger an OP amp, or just a Schmitt Trigger made from an Op amp. I believe I’ve seen Schmitt Triggers as part of logic groups including TTL AND, NAND, AND FLIP FLOP, etc.
Yes, an op amp itself is a differential amplifier at it's heart. It may not be configured as a differential amplifier when in use (such as an inverting amp, etc.). A Schmitt Trigger is not an op amp. It is essentially a comparator circuit whose decision threshold changes whether the output is HIGH or LOW. A Schmitt Trigger can be made from an op amp, but most are made from comparator circuits, or are embedded in logic circuits. The common feature is that the decision threshold changes with the state of the output.
Great explanation and extremely practical. Still, can u please explain why the Calculations are not the standard f=1/(2(Pi)RC) which would give, in your example, close to 500Hz not 5000HZ.Thanks
The 1/2piRC refers to the lowpass filter characteristic. In this case, the voltage is only moving up/down on the capacitor by a small amount, so the cycle time is faster.
Does not the Schmitt trigger when reshaping a wounded square wave change its peiod (since it triggers on Voltage at might be different points at the time domain)?
It won't change the period (since the same rising edge on the next pulse will have the same delay). However, it can change the duty cycle if the rising and falling edges are different.
Hi, Thanks for these great videos. I have built the circuit and used it as TDR to measure coax cables. Now I am wondering if this circuit with its fast rise and fall times (and subsequent harmonic content) would be an effective wideband noise source? I recall seeing circuits from a few decades ago for an antenna noise bridge that used a 555 timer and zener to generate the wideband noise. However, it seems this might be a better approach. What are your thoughts on this. Also, I hope that you are able to get back to you lab very soon.
The tone spacing will be equal to the odd harmonics of the oscillating frequency, which may provide fine enough granularity to work as a noise source, or at least a comb generator (see my video on that). Note that the amplitude of the harmonics will fall off as you go higher in harmonic count (frequency), but that too might be OK for many applications.
Hello sir and thank you for your very informative videos. I have a question for you: I would like to make a circuit that can give me a rectangular signal between 0V and 5 V. A kind of switch, but with a smaller rise time in the order of 50ns to 100ns. I tried with the Mosfets, but the output siganl to a lot of swings and noises. What you can advise me or have already made a circuit like that. I like to test the filters placed at the retraction of a microcontroller by placing this signal recregular and read the response of the filter. Thank you
50-100ns rise and fall times are pretty slow. Just about any low-cost CMOS logic gate can swing rail to rail, so a simple oscillator using these devices should work fine.
hey just a quick question, how far can that oscillator go? i was trying to make a variable clock for an atmega but the 555 doesnt want to go beyond 2MHz (kind of a noob here)
Excellent video as usual. What determines the threshold voltage, as in this case of 1 volt, is it the Schmitt IC itself? Can you change the threshold voltage using this Schmitt IC using external components or is it pretty much fixed at 1 volt? Thank you
There are many causes for the ringing: sub-optimal circuit construction/layout, lack of impedance matching, long ground lead length on a scope probe, insufficient power supply filtering. All of this is exacerbated by the fast edges of the logic signal, because faster edges contain higher frequency content - where all of these factors matter.
Thanks for posting. Question have you in the past done business with Newark Electronics? Just would like some input on online vendors in general? Thanks.
Yes, I have done business with Newark Electronics. Good company. The online vendors for electronic components that I have dealt with most often are Digikey, Mouser and Newark.
Thank you very much for the nice video! I made a bread board circuit following your drawing. However, the rise time is about 8ns. Although it is already quite fast, I would like to at least reach below 5ns. It seems to be independent of the capacitor I put. I guess maybe the rise time only depends on 74AC14 itself? Different brands matters?
What is the Bandwidth of the oscilloscope and probe that your using to measure this? Your equipment might be limiting the rise time. If the device is truly in the AC family, it should be faster. Also, achieving fast edge speed requires very careful component and circuit layout and construction.
Thanks for your prompt reply! I am using Tektronix TBS1102B 100MHz and a 10x probe. I have the feeling that my breadboard may be the culprit. I could pick up the signal at a free port. I will try soldering a circuit board.
Yes, you'll never get fast edges using a plug-in breadboard. Also, you'll need to use an extremely short ground lead on the probe (ground lead inductance will slow the measured edge speed). Check out my video on making your own high performance scope probe socket. That scope/probe combination has about a 3.5ns risetime. Thus, the measured risetime will be the SQRT(actual^2 + 3.5ns^2)
Thanks for the video. I'm wondering if there are other applications? Can you use a Schmitt trigger to convert a frequency to a pulse? I'm trying to display speed in mph from some Hz freq, I need to be very accurate.
Great video. I know this is old now but you mentioned small difference between the 74HC14 and 74AC14 (speed). I had an issue recently when I replaced the HC14 with a AC14. I was passing a signal from a simple function generator chip to an Arduino which ,in turn, provided the output to a 7 segment display. Everything worked perfectly with the HC14 (frequency verified with oscilloscope and function counter). When I inserted the AC14, very low frequency signals (19Hz - 60Hz) were incorrectly displayed on the display. 19 Hz was displayed as 38 Hz. Most other frequencies were fine except at the 1.9 kHz - 4kHz range where again, the displayed frequency was roughly twice the input frequency (1.9kHz displayed as 3.8 kHz). I tried several HC14s and AC14s and the result was the same. Any idea why the AC14 could cause this anomaly?
I would bet that the faster rising/falling edges of the AC part are causing the circuit to momentarily ring or oscillate - especially possible on a breadboard where effective grounding and power supply decoupling is difficult to do. If you have an oscilloscope, take a look at the circuit's output.
Many thanks for your reply Alan. I suspect, you are right and there is some oscillation happening . The output of the circuit looks and reads fine on both oscilloscope and frequency is displayed correctly on a 'Leader' Function counter I have. Indeed, the input to data pin 5 of the Arduino reads correctly on the scope but the Arduino output pins are sending double the frequency (only at the frequencies I referred to) to the 7-segment display. The issue seems to be the way that the correct frequency is interpreted by the Arduino hardware or software (I think software and will play with the 'gate time' in the code). Absolutely weird but if I can't tweak the software I guess I will just live with the HC14 :-) Thanks again for your time.
Hey sir, What would be the benefits/benefit of using a schmitt trigger oscillator over using a 555 astable oscillator circuit to provide pulses? I cannot find a comparison between the two anywhere. Thanks.
+SPENCER1997ful There are a LOT of ways to create pulses - a schmitt trigger and a 555 are just two ways. Each has advantages and disadvantages. A schmitt trigger oscillator can run a lot faster than a 555 timer based solution - as one advantage. But, it has a more restrictive power supply requirement, and doesn't provide all of the functionality of the 555 (one-shot, triggerable, etc.).
Awesome explanation of Schmitt trigger operation. I knew the concept, but I was unsure of how to use one in a circuit. I assume the upper and lower thresholds are based on the particular IC you use (along with the voltages)?
Gotcha, thanks. I built a similar oscillator circuit using a NAND gate with both inputs tied together, 0.1 uF cap, and a 10k resistor, but couldn't see any oscillation on the output. I even tried some circuits I had found using 2 and then 3 gates and various caps, but still no luck. Just wondering if I'm missing something here.
I understand that it wouldn't function as a Schmitt Trigger. I was just trying to make a simple oscillator in the same sort of way. I probably just need to review the inner workings of oscillators a bit before attempting to make one.
Hello sir. I have a question about Wein Bridge Oscillators, but I couldn't find a video on that specific topic, so I decided to post my question here. Hope it's ok... I'm trying to build a Wein Bridge Oscillator that I found in National Semiconductor's App Note 31, but my Proteus 8 simulation of the circuit does not work and I can't figure out why. Unfortunately I couldn't find any additional info about the design. So I was wondering if you could be so kind and maybe take a lot at the circuit. It's a project that I have to build and turn in, so grades are on the line :D The links of the pdf picture and the simulation is below. BTW, there is -15 written on the bottom right corner. I can't figure if that's -15 volts or something else, cause almost all voltages on this App Note have a "v" written next to them and this one doesn't. I have also supplied the Op-Amp + & - 15 volts. Though the documents does not specify any value. PDF Schematic: drive.google.com/open?id=1SmV6B4-9yGnFe-V_tkdoIO7R3o0KZb12 Simulation: drive.google.com/open?id=1WjgPlmipImQD9Kkb9V2PxBVilSLpFOUn
Oscillators are notoriously difficult to simulate. Simulators work hard to determine a stable operating point at startup, which often prevents the oscillator from starting. You usually have to "kick start" them using an initial condition or a current impulse injected into the loop. And yes, the -15 implies -15V.
74LS series does not have a symmetrical output stage, so it won't work very well (can't source a lot of current in the high state). Stick with a CMOS variety of logic.
Thank you . Found out later and you did mention use a CMOS. I thought 74ls series was a CMOS. I have trouble reading does data sheets TTL higher current usage etc etc. Thanks again
This might be a silly question, but did you make that little piece of PCB that you soldered your Schmitt Trigger to? If not, where did you pick it up? Thanks & 73! K1TCP
A question from a noob. What's the difference between Schmitt trigger and those comparator ICs which we see in Chinese i.e. arduino shields? Are they schmitt triggers also?
When I tried this circuit with 47nF and 6.8KOhm components, 5V supply (with 100nF decoupling cap), I am not getting stable oscillation. Sometimes oscillation doesn't even start, sometimes it starts, and after some time stops. I have tried different ICs, cd74ac14, cd74hc14, and sn74ac14n variants as well, no luck. Tried different power supplies (switching, linear regulated down to 5V from 12V, linear PSU as well) to with and without 100uF capacitor, no difference. Observation - when it doesn't oscillate, the IC is consuming a lot of current (80mA or so - can go for many seconds like this), when oscillation starts, very little current is used. Observation - when probing at Pin 1 (where RC is connected), the triangle wave is the one that doesn't get formed properly. When it does, I see a lot of high frequency signal riding on the main triangle wave. Tried different R and C values, but still not reliable. Any ideas what I am missing?
The most likely scenario is that the chip is going into very high frequency oscillation. Possibly too high to see on your scope. Most likely due to insufficient decoupling and poor circuit layout. This circuit will be difficult to make on a plug-in protoboard due to these factors. Connections need to be short. You need a good solid ground plane. Decoupling with both large bulk capacitance, and good high frequency ceramic caps.
@@w2aew Yes, this is just what happened to me. I tried to build it on a protoboard and I can only see around 270-300MHz sinewave oscillation at the output. Once when I moved the board/connections it started to produce the squarewaves at around 3kHz, but after a few minutes it went back to high frequency oscillation again. I will build it on PCB... Thanks!
i received your schematic (not sure why comment removed) digilander.libero.it/beamweb/pwm%20osc.jpg pwm, meaning changing the resistance changes the PW instead of freq? how to change both? please explain. thx!
4 years later, but still, I was having the same question: where is the INPUT? [ If you would place the (missing) input between R and C, then there is, potentially, another stable solution : at the resistor, Vr = 5V , or whatever is the logical high voltage. With Vsource as input to the inverter, Vout = Vsource - Vr = constant which could satisfy the output of the inverter too, 0V, as long as the IC output can handle, without burning out, a not 0 voltage applied to its output. ] The circuit seems to forget (well, it definitively does not show) that the IC itself has a direct access to Vcc and to Ground (again, not shown on the circuit), so that the IC can sink or supply some current, by itself. And that is this connection of the IC to Vcc which is providing the required "input". Otherwise, it would be a "free energy" device. :-) It is not necessary evident, at first glance though. At least, not to me, too.
+Nasser Samim You can use the circuit shown in this video, simply adjust the capacitor value (lower) to increase the frequency to 65kHz. The exact value will depend on the IC you use, so you'll have to experiment with it.
Here's a schematic robotroom . com/PWM/VariableDutyCycleSchematic.gif Note that this schematic uses a NAND gate, which is the same as the NOT gate when you short the 2 inputs together. Let me know if you need to know what is happening in that schematic.
I had to stop the video and comment. This is the best explanation of a Schmidt trigger on RUclips. Period. Great job. Thank you!
Thank you for that Schmitt Trigger explanation! I tried reading the Wikipedia article on the Schmitt Trigger, and it kept mentioning hysteresis (which I understand as a mechanism that behaves in such a way that depends on its own history). It'd say hysteresis this, and hysteresis that, and I wasn't any closer to understanding what a Schmitt trigger actually did. But this actually helped me tie in the concept of hysteresis with a Schmitt Trigger for me.
When the noisy signal first touches the upper threshold, that threshold gets "de-activated" and the output becomes HIGH, while the lower threshold gets "activated". And so on.
Exactly. Hysteresis basically means that the threshold *depends on* the previous state of the comparator - i.e. the threshold moves when the comparator changes state such that tiny wiggles/variation/noise on the signal can't cause chatter.
Schmitt has hysteresis, regular inverter without schmitt when does it switch?
This is one of my favorite vids that you made. I built the oscillator using a 74AC14 I found on ebay (I wish I can send you photos of my Manhattan Style build) and got about a 2-4 nS rise/fall which is much more than you got, or the other viewer in the update video who built a PCB around the TDR. Thank you for all those years of education, your the man!
My email address is good on qrz.com
Very very nice. Your explanation of the two waveforms (input and output) illuminated that whole thing for me. thx
Best explanation of Schmitt triggers I've seen... Thanks!
"The Schmitt trigger was invented by American scientist Otto H. Schmitt in 1934 while he was a graduate student,[1] later described in his doctoral dissertation (1937) as a thermionic trigger.[2] It was a direct result of Schmitt's study of the neural impulse propagation in squid nerves.[2]"
Surely a squid trigger ?
It would be nice to replace the resistor with a potentiometer and add two diodes and you have a nice and simple PWM generator.
Great video, thank you.
Yes, 720p is a big difference, it's much more clearly. Thank you.
Perfectly explained as always, thanks thanks! PLEASE keep them up!!!
Thank you for an interesting Tutorial. Well explained and much apprecated.😊👍
The statement “we want the output impedance to be equal so we can get a 50% duty cycle”. Is it as simple as, If they were unequal, the feedback would vary causing the duty cycle to change?
Thank you for switching to HD! Looks a lot better now. :)
You can do this with TTL logic, but then are limited to fairly low resistor values. Another variation is to make a schmitt trigger by wrapping an op amp with a couple of resistors to provide some positive feedback, then couple that to an integrator (the whole thing can be easily done in a dual op amp package). The output of the integrator goes to the input of the schmitt trigger. The advantage of this one is _very_ linear triangle wave, rather than the ones you get here. Also higher amplitude, if you want, and then by doing a bit of diode steering you can get narrow pulses and sawtooth rather than triangle waves, or even variable duty cycle.
One of the problems with TTL is that is has asymmetric output drive capability, and can't drive a low impedance load.
could you perhaps do a video on parallel Schmitt trigger configuration and benefit of that optimization
Great Video as usual, thank you so much for teaching all your knowledge in such a didactic way!
2/ another tip! If you put a transistor (pnp ie bc107) Emitter to ground , Collector to input across C , potential devider Base and the top resistor to the out put of the Shmit, there will be a saw tooth wave form across the C.
Excellent video.
Would this type of oscillator be fast enough to function as a local oscillator in an RF radio circuit?
What would be the practical upper freq limit?
Could it cover the entire 30 Mz hf band?
If so, this could simplify homebrew radio construction tremendously.
Thanks for the video.
Speed will depend on the logic family used for this, and the resulting propagation delay of the gate. You'd want to follow it up with a divide-by-2 to cleanup any duty cycle distortion, so that pushes the frequency even higher. So, I'd say it likely would not be a good for an HF LO.
Very well thought out and produced. Thanks for the extra effort.
Great video, Great lesson...Thank You!
Question: As an existing signal passes through a Schmitt Trigger Oscillator, would it's phase change in any way?
Your videos are great, Alan. Picked up a couple of 74ac14's and I am planning on building this. I had a thought that I might just add the parallel trigger array, in the output of your circuit, to the output of my inexpensive function generator, using it as a conditioner for the square wave output. That way I could vary the frequency and duty cycle. For TDR measurements, would a very narrow pulse make it easier to visualize the delay ?
Adding an external input to the driver is a good idea - it lets you play with pulse widths, frequencies, etc. Certainly a narrow pulse can be used to measure delay - potentially with less ambiguity, as long as the pulse width is shorter than the cable delay.
Did you know if you replace the feedbach resistor with a coil you get a sighnwave on the gate input ! (It dose have little glitches at the historisis points but can be easly filtered ).
1/ tip of many , did you know if you replace the resistor with a coil, the input will resemble a sine wave (with a little trigger glitch that can be filterd) .
Hi. Under specific conditions (low power, low frequency) an op amp-based gyrator can simulate a coil's inductance, so can this circuit replace the resistor as well?
Thanks.
en.wikipedia.org/wiki/Gyrator
Another great video! I lashed this together on the breadboard tonight, substituting a 74C14 and an 8k2 for R and run it in 5V. I get a great triangle oscillation in the input, but can't find the PWM or square on the output (or anywhere), I can only get the same triangle wave in both the input and the ouput. Any ideas?
Nice and clear - excellent.
I have an RA gray control system that uses manchester fm2 code. What would be the best way to make a new controller/programmer so that I can create new shows?!?
Terminology Question: By definition(?) an op amp is a differential amplifier yes/no? Is a Schmitt Trigger an OP amp, or just a Schmitt Trigger made from an Op amp. I believe I’ve seen Schmitt Triggers as part of logic groups including TTL AND, NAND, AND FLIP FLOP, etc.
Yes, an op amp itself is a differential amplifier at it's heart. It may not be configured as a differential amplifier when in use (such as an inverting amp, etc.). A Schmitt Trigger is not an op amp. It is essentially a comparator circuit whose decision threshold changes whether the output is HIGH or LOW. A Schmitt Trigger can be made from an op amp, but most are made from comparator circuits, or are embedded in logic circuits. The common feature is that the decision threshold changes with the state of the output.
Thanks so much for your videos. I've learned a lot! Please continue making more videos like this!
Great explanation and extremely practical. Still, can u please explain why the Calculations are not the standard f=1/(2(Pi)RC) which would give, in your example, close to 500Hz not 5000HZ.Thanks
The 1/2piRC refers to the lowpass filter characteristic. In this case, the voltage is only moving up/down on the capacitor by a small amount, so the cycle time is faster.
Thanks for the prompt answer
👍Thank you sir.
Nice quality resistors
HEY ..I WANT TO MEASURE CABLE RJ45 LENGTH BUT MICRO CONTROLLEER ALONE CANT DO THAT ..WITH 74AC14 I CAN MEASURE ??
I'm using a 47nf cap, and 10x on the probes. My scope says it's oscillating right around 15 kHz. I'm running mine at 5V, should I try 12?
Does not the Schmitt trigger when reshaping a wounded square wave change its peiod (since it triggers on Voltage at might be different points at the time domain)?
It won't change the period (since the same rising edge on the next pulse will have the same delay). However, it can change the duty cycle if the rising and falling edges are different.
Hi, Thanks for these great videos. I have built the circuit and used it as TDR to measure coax cables. Now I am wondering if this circuit with its fast rise and fall times (and subsequent harmonic content) would be an effective wideband noise source? I recall seeing circuits from a few decades ago for an antenna noise bridge that used a 555 timer and zener to generate the wideband noise. However, it seems this might be a better approach. What are your thoughts on this. Also, I hope that you are able to get back to you lab very soon.
The tone spacing will be equal to the odd harmonics of the oscillating frequency, which may provide fine enough granularity to work as a noise source, or at least a comb generator (see my video on that). Note that the amplitude of the harmonics will fall off as you go higher in harmonic count (frequency), but that too might be OK for many applications.
I've used it for excacly the same reason .
Hello sir and thank you for your very informative videos.
I have a question for you: I would like to make a circuit that can give me a rectangular signal between 0V and 5 V. A kind of switch, but with a smaller rise time in the order of 50ns to 100ns.
I tried with the Mosfets, but the output siganl to a lot of swings and noises.
What you can advise me or have already made a circuit like that.
I like to test the filters placed at the retraction of a microcontroller by placing this signal recregular and read the response of the filter.
Thank you
50-100ns rise and fall times are pretty slow. Just about any low-cost CMOS logic gate can swing rail to rail, so a simple oscillator using these devices should work fine.
How? Schematic plz. Thx!
hey just a quick question, how far can that oscillator go? i was trying to make a variable clock for an atmega but the 555 doesnt want to go beyond 2MHz (kind of a noob here)
High Def!
Another excellent video!
Excellent video as usual. What determines the threshold voltage, as in this case of 1 volt, is it the Schmitt IC itself? Can you change the threshold voltage using this Schmitt IC using external components or is it pretty much fixed at 1 volt?
Thank you
It is fixed in these ICs. You can setup something custom using a comparator with some positive feedback.
I built this pulse generator and used in my school, but the students don't like the ringing on the signal. Is there any method to decrease it?
There are many causes for the ringing: sub-optimal circuit construction/layout, lack of impedance matching, long ground lead length on a scope probe, insufficient power supply filtering. All of this is exacerbated by the fast edges of the logic signal, because faster edges contain higher frequency content - where all of these factors matter.
hey whiskey two alpha echo whiskey, I love your videos, you are very thorough and very concise, great explanation, thanks kj6onb mk
snaprollinpitts Thanks, much appreciated! Be sure to let me know if you have any topics that you'd like to see in a future video or two!
***** could you please do a video on constant current sources, what there good for and how to use them. that would be great!!!
Already did. See #190:
m.ruclips.net/video/xR0RfmmRhDw/видео.html
Thanks for posting. Question have you in the past done business with Newark Electronics? Just would like some input on online vendors in general? Thanks.
Yes, I have done business with Newark Electronics. Good company. The online vendors for electronic components that I have dealt with most often are Digikey, Mouser and Newark.
Thank you very much for the nice video! I made a bread board circuit following your drawing. However, the rise time is about 8ns. Although it is already quite fast, I would like to at least reach below 5ns. It seems to be independent of the capacitor I put. I guess maybe the rise time only depends on 74AC14 itself? Different brands matters?
What is the Bandwidth of the oscilloscope and probe that your using to measure this? Your equipment might be limiting the rise time. If the device is truly in the AC family, it should be faster. Also, achieving fast edge speed requires very careful component and circuit layout and construction.
Thanks for your prompt reply! I am using Tektronix TBS1102B 100MHz and a 10x probe. I have the feeling that my breadboard may be the culprit. I could pick up the signal at a free port. I will try soldering a circuit board.
Yes, you'll never get fast edges using a plug-in breadboard. Also, you'll need to use an extremely short ground lead on the probe (ground lead inductance will slow the measured edge speed). Check out my video on making your own high performance scope probe socket. That scope/probe combination has about a 3.5ns risetime. Thus, the measured risetime will be the SQRT(actual^2 + 3.5ns^2)
***** Hi, I just soldered the new cricuit. Rise time 3ns :-)
Once again, great posts!
Thanks for the video. I'm wondering if there are other applications? Can you use a Schmitt trigger to convert a frequency to a pulse? I'm trying to display speed in mph from some Hz freq, I need to be very accurate.
M. Baffa 1:1 freq*60=ROtations and then multiply ROtations by the tire’s outer diameter, divide by time per rotation and that’s speed.
Thanks. Helped me along in a project.
Great video. I know this is old now but you mentioned small difference between the 74HC14 and 74AC14 (speed). I had an issue recently when I replaced the HC14 with a AC14. I was passing a signal from a simple function generator chip to an Arduino which ,in turn, provided the output to a 7 segment display. Everything worked perfectly with the HC14 (frequency verified with oscilloscope and function counter). When I inserted the AC14, very low frequency signals (19Hz - 60Hz) were incorrectly displayed on the display. 19 Hz was displayed as 38 Hz. Most other frequencies were fine except at the 1.9 kHz - 4kHz range where again, the displayed frequency was roughly twice the input frequency (1.9kHz displayed as 3.8 kHz). I tried several HC14s and AC14s and the result was the same. Any idea why the AC14 could cause this anomaly?
I would bet that the faster rising/falling edges of the AC part are causing the circuit to momentarily ring or oscillate - especially possible on a breadboard where effective grounding and power supply decoupling is difficult to do. If you have an oscilloscope, take a look at the circuit's output.
Many thanks for your reply Alan. I suspect, you are right and there is some oscillation happening . The output of the circuit looks and reads fine on both oscilloscope and frequency is displayed correctly on a 'Leader' Function counter I have. Indeed, the input to data pin 5 of the Arduino reads correctly on the scope but the Arduino output pins are sending double the frequency (only at the frequencies I referred to) to the 7-segment display. The issue seems to be the way that the correct frequency is interpreted by the Arduino hardware or software (I think software and will play with the 'gate time' in the code). Absolutely weird but if I can't tweak the software I guess I will just live with the HC14 :-) Thanks again for your time.
Hey sir,
What would be the benefits/benefit of using a schmitt trigger oscillator over using a 555 astable oscillator circuit to provide pulses? I cannot find a comparison between the two anywhere. Thanks.
+SPENCER1997ful There are a LOT of ways to create pulses - a schmitt trigger and a 555 are just two ways. Each has advantages and disadvantages. A schmitt trigger oscillator can run a lot faster than a 555 timer based solution - as one advantage. But, it has a more restrictive power supply requirement, and doesn't provide all of the functionality of the 555 (one-shot, triggerable, etc.).
+w2aew great, thanks for that :)
The HD looks great!
Thanks
Awesome explanation of Schmitt trigger operation. I knew the concept, but I was unsure of how to use one in a circuit. I assume the upper and lower thresholds are based on the particular IC you use (along with the voltages)?
Yes, there is no "standard" for the schmitt trigger thresholds, so it will vary a little among manufacturers and device types.
Gotcha, thanks. I built a similar oscillator circuit using a NAND gate with both inputs tied together, 0.1 uF cap, and a 10k resistor, but couldn't see any oscillation on the output. I even tried some circuits I had found using 2 and then 3 gates and various caps, but still no luck. Just wondering if I'm missing something here.
A NAND gate has a single threshold, so it won't operate like the Schmitt Trigger circuit where the threshold changes based on the output state.
I understand that it wouldn't function as a Schmitt Trigger. I was just trying to make a simple oscillator in the same sort of way. I probably just need to review the inner workings of oscillators a bit before attempting to make one.
Can i get 10 MHz square wave with 74hc14
you should be able to
@@w2aew with 74hc14?
@@w2aew i will use clock oscillator with hc14?
Great tutorial. Many thanks.
Hello sir. I have a question about Wein Bridge Oscillators, but I couldn't find a video on that specific topic, so I decided to post my question here. Hope it's ok...
I'm trying to build a Wein Bridge Oscillator that I found in National Semiconductor's App Note 31, but my Proteus 8 simulation of the circuit does not work and I can't figure out why. Unfortunately I couldn't find any additional info about the design. So I was wondering if you could be so kind and maybe take a lot at the circuit. It's a project that I have to build and turn in, so grades are on the line :D
The links of the pdf picture and the simulation is below.
BTW, there is -15 written on the bottom right corner. I can't figure if that's -15 volts or something else, cause almost all voltages on this App Note have a "v" written next to them and this one doesn't.
I have also supplied the Op-Amp + & - 15 volts. Though the documents does not specify any value.
PDF Schematic: drive.google.com/open?id=1SmV6B4-9yGnFe-V_tkdoIO7R3o0KZb12
Simulation: drive.google.com/open?id=1WjgPlmipImQD9Kkb9V2PxBVilSLpFOUn
Oscillators are notoriously difficult to simulate. Simulators work hard to determine a stable operating point at startup, which often prevents the oscillator from starting. You usually have to "kick start" them using an initial condition or a current impulse injected into the loop. And yes, the -15 implies -15V.
w2aew I see! Thank you so much.
So I guess I'll just go ahead and build it up and see what happens.
Hello Can I use a 74LS14 instead of the 74AC14 or 74HC14... Will the circuit work??? Thank you
74LS series does not have a symmetrical output stage, so it won't work very well (can't source a lot of current in the high state). Stick with a CMOS variety of logic.
Thank you . Found out later and you did mention use a CMOS. I thought 74ls series was a CMOS. I have trouble reading does data sheets TTL higher current usage etc etc. Thanks again
This might be a silly question, but did you make that little piece of PCB that you soldered your Schmitt Trigger to? If not, where did you pick it up? Thanks & 73! K1TCP
A question from a noob. What's the difference between Schmitt trigger and those comparator ICs which we see in Chinese i.e. arduino shields? Are they schmitt triggers also?
A Schmidt trigger is simply a comparator with a little positive feedback to create a little hysteresis as explained in the video.
Sir..just one question...where are the thresholds set? Are those values predefined inside the IC or something else??? Thank you..
On the schmitt trigger ICs such as these, the thresholds are internally set.
Thanks a lot Sir!
When I tried this circuit with 47nF and 6.8KOhm components, 5V supply (with 100nF decoupling cap), I am not getting stable oscillation. Sometimes oscillation doesn't even start, sometimes it starts, and after some time stops.
I have tried different ICs, cd74ac14, cd74hc14, and sn74ac14n variants as well, no luck. Tried different power supplies (switching, linear regulated down to 5V from 12V, linear PSU as well) to with and without 100uF capacitor, no difference.
Observation - when it doesn't oscillate, the IC is consuming a lot of current (80mA or so - can go for many seconds like this), when oscillation starts, very little current is used.
Observation - when probing at Pin 1 (where RC is connected), the triangle wave is the one that doesn't get formed properly. When it does, I see a lot of high frequency signal riding on the main triangle wave. Tried different R and C values, but still not reliable.
Any ideas what I am missing?
The most likely scenario is that the chip is going into very high frequency oscillation. Possibly too high to see on your scope. Most likely due to insufficient decoupling and poor circuit layout. This circuit will be difficult to make on a plug-in protoboard due to these factors. Connections need to be short. You need a good solid ground plane. Decoupling with both large bulk capacitance, and good high frequency ceramic caps.
@@w2aew Yes, this is just what happened to me. I tried to build it on a protoboard and I can only see around 270-300MHz sinewave oscillation at the output. Once when I moved the board/connections it started to produce the squarewaves at around 3kHz, but after a few minutes it went back to high frequency oscillation again. I will build it on PCB... Thanks!
i received your schematic (not sure why comment removed)
digilander.libero.it/beamweb/pwm%20osc.jpg
pwm, meaning changing the resistance changes the PW instead of freq? how to change both?
please explain.
thx!
Sorry for the dumb question, but what exactly is your input for this oscillator?
Just the power supply.
Like, just a battery? (Vcc in video #88 diagram)
bboyHarrypotter Yes, a battery or other power supply.
4 years later, but still, I was having the same question: where is the INPUT?
[ If you would place the (missing) input between R and C, then there is, potentially, another stable solution : at the resistor, Vr = 5V , or whatever is the logical high voltage. With Vsource as input to the inverter, Vout = Vsource - Vr = constant which could satisfy the output of the inverter too, 0V, as long as the IC output can handle, without burning out, a not 0 voltage applied to its output. ]
The circuit seems to forget (well, it definitively does not show) that the IC itself has a direct access to Vcc and to Ground (again, not shown on the circuit), so that the IC can sink or supply some current, by itself. And that is this connection of the IC to Vcc which is providing the required "input". Otherwise, it would be a "free energy" device. :-) It is not necessary evident, at first glance though. At least, not to me, too.
Good job on the video,
I usually asume X= 0.6
Definatly true .
hello Sir ,
How i can generate 65kHz square wave would you please help me
+Nasser Samim You can use the circuit shown in this video, simply adjust the capacitor value (lower) to increase the frequency to 65kHz. The exact value will depend on the IC you use, so you'll have to experiment with it.
thanks Sir
Thank you for the reply!
VERY GOOD VIDEO, THANKS!!
Here's a schematic robotroom . com/PWM/VariableDutyCycleSchematic.gif
Note that this schematic uses a NAND gate, which is the same as the NOT gate when you short the 2 inputs together.
Let me know if you need to know what is happening in that schematic.
🌷
Advisable Alan each video to be named at start with : w2aew #87 Schmitt Trigger Oscillator / 74...…..
I am trying do this with 2222's