The best explanation I have ever seen! Your series about electronics and the way of explanation is very communicative and easy to understand. Keep on going and ... stay academic teacher :)
It's getting kind of creepy. Once again we are working on exactly the same subject, haha. I was just diving into the same subject to make a 6Mhz simple sine wave generator. Well explained video, thanks 👍🏻😎
Very nice videos! Keep going! What about make the same parameter identification with more common equipment like a signal generator and an oscilloscope? How would you do it? What are the caveats?
As long as the signal generator can provide the necessary frequency step, the measurement can be done that way also. The setup would be the same - inject the signal from the generator to the quartz trough a known resistor and measure the signal being injected and the signal on the quartz with the oscilloscope. You need to step the frequency in fine paces, and them measure the amplitude ratio and phase difference between the 2 channels. See this document from tektronix on the topic - download.tek.com/document/75W_28152_1.pdf They explain measuring inductors and capacitors, but its the exact same principle.
@@FesZElectronics great! thank you very much for the attention. I think maybe a video with basically the content of the tektronics document will be a good video for the channel, right? People will identify themselves more because you are using hardware that they have. I don't know if the logic is right but I'm just trying to help as you helped me with your videos.
Hello Daniel! I guess at some point I will do a video on how to make bode plots without a network analyzer, since as you correctly said, not a lot of people have one of those. Thank you for this suggestion! In the mean time, I guess you can also see www.ti.com/lit/an/snva364a/snva364a.pdf?ts=1625457754591 Its a document by Texas Instruments about loop measurements, but the same principle applies - injecting a signal and measuring the phase shift and amplitude difference.
@@FesZElectronics yeah. A while back I made a python script that automatizes it. It set parameters on my (very cheap) signal generator and reads the data on the oscilloscope, both via usb. Then it stores the raw data: the input and output of the system (channel 1 and 2 of the oscilloscope) for each frequency but it is quite slow. Anyway it do the job. After this I take the raw data and post process it, using parameter estimation techniques to extract the gain and phase and plot or export this data. The only problem is that it's too slow, taking little bit less than 1s for each data point. But thanks for the attention and, again, congratulations for the excellent job here on youtube! :)
In my experience the biggest issue with this method of extracting parameters is extracting the phase shift, especially if there is a lot of noise - for example when measuring the loop response of a switching converter. I'm not sure what method you are using for the data processing but a while back I found some formulas to process a set of points to extract the gain and phase difference using an FFT algorithm - www.analog.com/en/technical-articles/ltspice-extracting-switch-mode-power-supply-loop-gain-in-simulation-and-why-you-usually-don-t-need.html its somewhere in the middle of the page. I implemented this in an older LTspice tutorial and it worked. I guess that even if it takes ~1sec per point, its still faster than doing it by hand.
If possible, could you please refer to how you have derived the motion capacity formula from 2 parallel resonance frequencies (11:50)? Using a common quarts parallel resonance frequency formula, I am getting another expression...
@@FesZElectronics Yes, the reference is exactly there. They do not derive it however and I suspect that a small frequency shift approximation is used - I will try again. Thanks a lot!
I will be honest, I did not check how the formula was derived, I just trusted the source; In general all sorts of simplifications are used when deriving nice simple formulas. In the end its all about how accurate the result is - may I ask if starting from the same values, is there a significant result difference using your equations?
@@FesZElectronics I agree with your point and the approach. Ideally, when an author comes up with a result, mentioning assumptions might be useful if a reader needs to use it. I will plug in the values and let you know if discover a noticeable discrepancy. P. Horowitz's book is a trustable source.
I tried to simulate a colpitts oscillator with 100Mhz quartz oscillator, but I failed . I did a quartz model that function in simulate but it not funnction in colpitts oscilator . The colpitts oscilator is that presented in @devttys0 cannel for 12Mhz but I tried to adapted it for 100Mhz and I do not had success
Man, you keep on hitting the topics that overlap with areas I work in, but don't always have time to dig into as deeply as I would like.
Excellent video. This channel is an underrated gem.
I really learned some things about crystals here. Thank you!
I'm embedded programmer and I want understand electronics in deep level. Your channel is very informative. Thanks a lot!
I'm happy my content is helping you! Let me know what particular subjects you are most interested in.
ua-cam.com/video/AfQxyVuLeCs/v-deo.html
Amazing presentation, FesZ! You really are among the very best electronics educators on UA-cam.
The best explanation I have ever seen! Your series about electronics and the way of explanation is very communicative and easy to understand. Keep on going and ... stay academic teacher :)
Amazing and informative video once again. Thanks Fesz!
This tutorial is top quality! Thanks so much
You are the best teacher. Thanks!
Thank you! I'm happy you are enjoying the videos!
Hello and welcome back
Well done.
Thank you sir...
Very good!
Very good 🙏🙏🙏🙏🙏
👍👍 thank you
Nice job! I’m wonder if series load capacitors set up in a pi configuration are not more common.
Best!
Hey Fesz, very nice and informative video but can you elaborate or provide a schematic of the test fixture you built?
It's getting kind of creepy. Once again we are working on exactly the same subject, haha.
I was just diving into the same subject to make a 6Mhz simple sine wave generator.
Well explained video, thanks 👍🏻😎
Right?! He seems to have this 6th sense for that kinda stuff
So my crystal ball is working :D I'm happy I can help.
@@FesZElectronics Haha, nice pun!
Can you demonstrate working LTSpice simulation of Colpitts XTAL oscilator?
Excellent explanation, Is ESR the main component of a crystal for its energy dissipation? thank you very much.
Yes, all other elements are reactive so they do not waste energy.
Very nice videos! Keep going!
What about make the same parameter identification with more common equipment like a signal generator and an oscilloscope? How would you do it? What are the caveats?
As long as the signal generator can provide the necessary frequency step, the measurement can be done that way also. The setup would be the same - inject the signal from the generator to the quartz trough a known resistor and measure the signal being injected and the signal on the quartz with the oscilloscope. You need to step the frequency in fine paces, and them measure the amplitude ratio and phase difference between the 2 channels. See this document from tektronix on the topic - download.tek.com/document/75W_28152_1.pdf They explain measuring inductors and capacitors, but its the exact same principle.
@@FesZElectronics great! thank you very much for the attention. I think maybe a video with basically the content of the tektronics document will be a good video for the channel, right? People will identify themselves more because you are using hardware that they have. I don't know if the logic is right but I'm just trying to help as you helped me with your videos.
Hello Daniel! I guess at some point I will do a video on how to make bode plots without a network analyzer, since as you correctly said, not a lot of people have one of those. Thank you for this suggestion!
In the mean time, I guess you can also see www.ti.com/lit/an/snva364a/snva364a.pdf?ts=1625457754591
Its a document by Texas Instruments about loop measurements, but the same principle applies - injecting a signal and measuring the phase shift and amplitude difference.
@@FesZElectronics yeah. A while back I made a python script that automatizes it. It set parameters on my (very cheap) signal generator and reads the data on the oscilloscope, both via usb. Then it stores the raw data: the input and output of the system (channel 1 and 2 of the oscilloscope) for each frequency but it is quite slow. Anyway it do the job. After this I take the raw data and post process it, using parameter estimation techniques to extract the gain and phase and plot or export this data. The only problem is that it's too slow, taking little bit less than 1s for each data point.
But thanks for the attention and, again, congratulations for the excellent job here on youtube! :)
In my experience the biggest issue with this method of extracting parameters is extracting the phase shift, especially if there is a lot of noise - for example when measuring the loop response of a switching converter. I'm not sure what method you are using for the data processing but a while back I found some formulas to process a set of points to extract the gain and phase difference using an FFT algorithm - www.analog.com/en/technical-articles/ltspice-extracting-switch-mode-power-supply-loop-gain-in-simulation-and-why-you-usually-don-t-need.html its somewhere in the middle of the page. I implemented this in an older LTspice tutorial and it worked. I guess that even if it takes ~1sec per point, its still faster than doing it by hand.
Da bomb
If possible, could you please refer to how you have derived the motion capacity formula from 2 parallel resonance frequencies (11:50)? Using a common quarts parallel resonance frequency formula, I am getting another expression...
I think the source for these formulas was "The Art of Electronics" 3rd Edition. Chap 7.1.6 - section B pg 444-445 - There is a footnote with them.
@@FesZElectronics Yes, the reference is exactly there. They do not derive it however and I suspect that a small frequency shift approximation is used - I will try again. Thanks a lot!
I will be honest, I did not check how the formula was derived, I just trusted the source; In general all sorts of simplifications are used when deriving nice simple formulas. In the end its all about how accurate the result is - may I ask if starting from the same values, is there a significant result difference using your equations?
@@FesZElectronics I agree with your point and the approach. Ideally, when an author comes up with a result, mentioning assumptions might be useful if a reader needs to use it. I will plug in the values and let you know if discover a noticeable discrepancy. P. Horowitz's book is a trustable source.
And I forgot to thank you again for the great work you do for the tutorials. :-)
you didn't put jitter model inside model ? did you ??
Nope; I just focused on the resonance frequencies.
A 400$ tool only has a 10MHz Signal generator?
It used to be cheaper :D
I tried to simulate a colpitts oscillator with 100Mhz quartz oscillator, but I failed . I did a quartz model that function in simulate but it not funnction in colpitts oscilator . The colpitts oscilator is that presented in @devttys0 cannel for 12Mhz but I tried to adapted it for 100Mhz and I do not had success