*Please note an error in the integrator function which should be (z^-1) / (1 - z^-1). Thank you to the people who spotted the error. This does not however change the behavior of the transfer function of the noise and signal being HP and LP.*
I am a silicon validation engineer and we work extensively with delta sigma ADCs. As validation people, we often tend to test ADCs as black box. Your video was truly worth it. Very lucid content, and a strong hold on fundamentals. Thanks a ton! Keep this up.
this video is a goldmine of information, most professor cant explain this content so well, in fact in my experience most people are afraid to go into this kind of detail because they dont understand it that well themselves
This is one of the best electronics tutorial videos on the internet. You covered a complex subject really well in only an hour. It's a subject which many people know of and make use of, but few bother to understand and those of us without access to higher education and expensive equipment couldn't have really covered ourselves. Great as always, I'd love to see more of the same, thanks so much!
My God. This was such a well done video. No glossing over of any salient detail. If I watch this about 5 more times over the next few weeks I have the distinct feeling that I will fully understand it.This video is everything a person trying to use RUclips as a learning environment could want:- Highly detailed without ratholing.- Information has a flow which builds on prior concepts, like building a pyramid. No jumping around to this or that BUT with the proper amount of references back to prior material pretty much exactly where it is needed in order to clarify the current point.- Both the abstract and the concrete were presented- CLEAR English spoken with a good mic used.I cannot say enough good things about this presentation.THANK YOU for taking the time and obvious thought to share your expertise!!!
Wow. Thiis is one of the best videos yet. You apologized for using your tablet, but I think it led to a useful back & forth with your material that wouldn't have been possible with the (fun) single small whiteboard (used in the past). Today, I always knew when you were moving between 'slides' and understood how you were correlating newly introduced ideas on a later slide to fundamental ideas that you jumped back to. Please, please keep up the great work.
My master's thesis is going to be about VCO-based ADCs which are a type of the Delta-Sigma ADCs. Your video really helped me understand the concepts. Your explanation is better than even most professors'! Thank you, Shahriar!
This is by far the best video on this topic I've found in the internet. Despite having a very little knowledge about the topic I actually understood the discussion quite well. You are an excellent teacher and I hope you keep up the good work.
Thank you for producing these videos, Shahriar. You have a gift for making even the most complex stuff seem simple. On a side note, you don't need to apologize for having a bunch of theory in your video. It's an essential part of the explanation and a video without it would be meaningless.
This is my first youtube comment ever. I am a huge fan of the signal path blog. Even though this video is from 2014 I felt that it was important to weigh in because our host requested it. I LOVE the mixture of math, theory and hands on on this site. I don't mind the video length at all. This time is appropriate for the level of detail and care put into explaining the topics. I learn so much on the signal path. Please don't change anything. You are making me a better engineer. You are the mentor who I did not have when I got my first couple EE jobs. Thank You.
The two tones at 56:30 are the upper and lower images of the input signal offset at +/- 2 KHz from the sample rate (and harmonics). The sample rate doesn't come through because of the nulls in the sinc function at the sample rate ( and harmonics).
Many thanks Shahriar, I appreciate all your videos. Well above my level of understanding end to end but I gain many insights which reinforce what I do know, what I thought was well beyond me and the breadboarding of circuits give a good understanding of the workings at the component level. Good to see you back making videos. I appreciate that a significant amount of time and planning goes into every video. My preferences, not so much rip downs, but the theory and circuits with the frequency and time domain images. I look forward to every one and try to get fellow radio amateurs to view and suck up the good theory presented so convincingly.
Really great tutorial! RUclips needs more of these. You are able to glide through a complex topic from theory to actual experiments in a very interesting to watch fashion. Thanks a lot for sharing your knowledge. The experiments are really well thought of. Awesome!
Thanks for the great tutorial Shah! In a few key statements, you made some concepts click in my mind that haven't clicked during my many years of study in EE. I hope you keep educating the world for many more years!
Amazing man. I am a graduate student in Electronics, learning lots of design. I must admit that this video has taught me more than what I would learn in an ordinary class room :) Good job :)
Awesome flow of illustration going on here,,Theory and hands-on experience here are priceless..Best way I figured out finally how delta-sigma works Keep up the good work
Great video. Really helpful. Showing the averaging and noise shaping on the scope is a great way of bringing the theory to life. Fantastic job - keep up the good work! A big thank you for taking the time to share your insight.
Great tutorial! I liked the fact that it concentrated more on the intuition of how delta-sigma ADCs worked and then linked the intuition with actual physical measurements. Bravo!
Excellent video, as always! The progression from theory to practical application really ties things together. So many university courses are entirely focused on the theory and it can be difficult to fully grasp the implications without practical application.
Fantastic - covered everything I was looking for and more! I'm coming at this from an audio perspective - I knew about noise-shaping but didn't realise it was intrinsic to the delta-sigma modulator. Your video will be a great help for my retro-computing project of building an audio ADC for my old Amiga. :)
Man I really like your tutorial videos Thumbs up! Delta-Sigma was one part I skipped in the exams due to time constraints because it was a really big chapter. But I wanted to understand them sometime and you just helped me with that extremely. Very much appreciated keep the good work up!
Excellent video-tutorial Shahriar! Thank you very much for all the time you invested in the preparation of both the theory explanation and the lab part in order to demonstrate how the theory is applied in real circuits and everything matches perfectly. Kind regards,
A very nice explanation of a sigma delta adc. A primer for me, to fully grasp the concept in its technical details. Thank you very much for a naother great video.
Thank you so much for the effort in this video. I never thought much about the inner workings of those compoents. The amout of theory was excellent to give me an understanding, and as a practicioner your breadboard example was fantastic to show it in real life. I wish school was as enaging as your videos back 30 years ago :)
Very good video, as always. A great review of transfer functions, from my collage days. Great explanations. I would like to say thank you for the great instruction.
Wow. More videos like this in specific topics of RF plz!. Also, would be great if you can provide some resources for self-learners like books or something that was helpful for you . Thank you
Great video. Just one thing I want to point out: for a 2KHz input signal, the nyquist sampling rate is 4KHz. If the sampling rate is 256KHz, then the oversampling ratio should be 64 not 128.
TheSignalPathBlog Well, I still miss the REASON. Why is like that? I´ll try. The funny thing that bad ass genius of Shahriar gave already the answer at 30:00, since the RIgol and the Agilent are just DAC. The outputs of the Rigol are two digital signals converted with a sample and hold correlated with the same internal Rigol clock, and in frequency there are shaped in the same way (Sync of internal clock Rigol, since there is not an tunable with frequency out analog filter at each Rigol output). Same thing for the Agilent input signal generator. Maybe those 2 tones in the spectrum when you showed us the Sync function are the related with the clock input and signal input considered as digital ones with a sample and hold to convert them to analog... And you are right no books are telling you that!!! How did you get to this point? Anyway smart girl Rosanah Murugesu and smart boy Shahriar. Why don´t do a follow up video to respond to your mysterious questions?
Coaina Zuppero If your clock frequency is a multiple of your sine frequency, you will be sampling your sine at the same points in each cycle of your sine. So, if you plot your quantization error vs. time, it will be repeating with the same period as your sine, let's call this period T0. A signal (in this case quantization noise) which repeats in the time domain with period T0 will have a fourier transform which is zero except for frequency points where f = k*(1/T0) where k is any integer. The spectrum of the quantization noise will no longer look like white noise! It will have spurs at multiples of your input sine frequency.
Rosanah Murugesu So by definition a DAC which has on the input a periodic N words sequence and a clock sampling/rendering of Fs will produce an quantization noise in the f domain as a impulse train at f=k*Fs/N (since T0=N/Fs). That´s why the two DAC Agilent and Rigol are generating the two tones at 56:10 ? Maybe I was thinking starting from the f domanin analysis, a sample and hold DAC with a numeric sine signal input will generate a train of impulse in fraquency shaped by a Sync. Maybe I am confusing signal output and noise output. I am still digesting this stuff, sorry if I was/am not precise and clean like you in your explanations. Definitely a smart girl.
Great video. It makes it very easy to understand the most important points, even if the finer details are two complicated for me right now. A video on PLLs would be quite interesting.
Great video you have everything knowledge, sense of humour and fantastic instrumentation. You hope we still there? You bet we are... Thank you and waiting for more...
There isn't many channels that I genuinely get excited for when they upload videos. But this channel - every video is a banger. Loved it. Learned some new stuff. Shahriar - you UofT grad ?
*Please note an error in the integrator function which should be (z^-1) / (1 - z^-1). Thank you to the people who spotted the error. This does not however change the behavior of the transfer function of the noise and signal being HP and LP.*
Really nice tutorial, Shahriar - I love the progression from theory, to practical, to real-world effects.
Thank you!
TheSignalPathBlog I agree with w2aew's assessment. Very nice tutorial!
You are mindblowingly smart, an amazing teacher, and an amazing video producer.
Dude, you're one hell of a beast! This is the best presentation I've seen in a really long time! Thank you so much!
I agree! Brilliant!
I am a silicon validation engineer and we work extensively with delta sigma ADCs. As validation people, we often tend to test ADCs as black box. Your video was truly worth it. Very lucid content, and a strong hold on fundamentals. Thanks a ton! Keep this up.
this video is a goldmine of information, most professor cant explain this content so well, in fact in my experience most people are afraid to go into this kind of detail because they dont understand it that well themselves
This is one of the best electronics tutorial videos on the internet. You covered a complex subject really well in only an hour. It's a subject which many people know of and make use of, but few bother to understand and those of us without access to higher education and expensive equipment couldn't have really covered ourselves. Great as always, I'd love to see more of the same, thanks so much!
My God. This was such a well done video. No glossing over of any salient detail. If I watch this about 5 more times over the next few weeks I have the distinct feeling that I will fully understand it.This video is everything a person trying to use RUclips as a learning environment could want:- Highly detailed without ratholing.- Information has a flow which builds on prior concepts, like building a pyramid. No jumping around to this or that BUT with the proper amount of references back to prior material pretty much exactly where it is needed in order to clarify the current point.- Both the abstract and the concrete were presented- CLEAR English spoken with a good mic used.I cannot say enough good things about this presentation.THANK YOU for taking the time and obvious thought to share your expertise!!!
The most-comprehensible treatment of D-S conversion is in this video.
Shahriar has made the tutorial succinct and FUN to boot.
Thx!
This tutorial is just wow. Hats off to Shahriar to talk about complex theory of S-D A/D in as simple as words it could be.
Best DeltaSigma Modulation explanation in a simplified way. Keep on your excellent work.
Wow. Thiis is one of the best videos yet. You apologized for using your tablet, but I think it led to a useful back & forth with your material that wouldn't have been possible with the (fun) single small whiteboard (used in the past). Today, I always knew when you were moving between 'slides' and understood how you were correlating newly introduced ideas on a later slide to fundamental ideas that you jumped back to. Please, please keep up the great work.
First Downloaded offline, then subscribed after watching , and now going to analyse in Simulation tool, awesome video..
My master's thesis is going to be about VCO-based ADCs which are a type of the Delta-Sigma ADCs.
Your video really helped me understand the concepts. Your explanation is better than even most professors'!
Thank you, Shahriar!
This is by far the best video on this topic I've found in the internet. Despite having a very little knowledge about the topic I actually understood the discussion quite well. You are an excellent teacher and I hope you keep up the good work.
This is most definetly one of the best videos on youtube. Wish I had seen this while taking my classes on measurement systems!
With people like you looking after IC design etc Shahriar, things are in good hands.
Thanks for the great vid.
Excellent tutorial. Showing real signals on the scope really brought it home.
Thank you for producing these videos, Shahriar. You have a gift for making even the most complex stuff seem simple.
On a side note, you don't need to apologize for having a bunch of theory in your video. It's an essential part of the explanation and a video without it would be meaningless.
This is my first youtube comment ever. I am a huge fan of the signal path blog. Even though this video is from 2014 I felt that it was important to weigh in because our host requested it. I LOVE the mixture of math, theory and hands on on this site. I don't mind the video length at all. This time is appropriate for the level of detail and care put into explaining the topics. I learn so much on the signal path. Please don't change anything. You are making me a better engineer. You are the mentor who I did not have when I got my first couple EE jobs. Thank You.
The two tones at 56:30 are the upper and lower images of the input signal offset at +/- 2 KHz from the sample rate (and harmonics). The sample rate doesn't come through because of the nulls in the sinc function at the sample rate ( and harmonics).
Yes! f_s - 2 kHz and f_s + 2 kHz. Aliased all the way up to infinity
Thanks
Many thanks Shahriar, I appreciate all your videos. Well above my level of understanding end to end but I gain many insights which reinforce what I do know, what I thought was well beyond me and the breadboarding of circuits give a good understanding of the workings at the component level.
Good to see you back making videos. I appreciate that a significant amount of time and planning goes into every video.
My preferences, not so much rip downs, but the theory and circuits with the frequency and time domain images.
I look forward to every one and try to get fellow radio amateurs to view and suck up the good theory presented so convincingly.
The best explanation for delta-sigma ADC on the internet! Thank you so much!
from an electronic engineer here, in the development business for 8 years....awesome video
Really great tutorial! RUclips needs more of these. You are able to glide through a complex topic from theory to actual experiments in a very interesting to watch fashion. Thanks a lot for sharing your knowledge. The experiments are really well thought of. Awesome!
Thanks for the great tutorial Shah! In a few key statements, you made some concepts click in my mind that haven't clicked during my many years of study in EE. I hope you keep educating the world for many more years!
Amazing man. I am a graduate student in Electronics, learning lots of design. I must admit that this video has taught me more than what I would learn in an ordinary class room :) Good job :)
Awesome flow of illustration going on here,,Theory and hands-on experience here are priceless..Best way I figured out finally how delta-sigma works
Keep up the good work
I agree too.
I agree too.
Great video. Really helpful. Showing the averaging and noise shaping on the scope is a great way of bringing the theory to life. Fantastic job - keep up the good work! A big thank you for taking the time to share your insight.
Alot of this was a bit beyond my level, however, as with all your videos, I ALWAYS learn something.. again.. another A+ video. Well done & thank you.
Your shop Is Nothing short of INCREDIBLE!!!
Great tutorial! I liked the fact that it concentrated more on the intuition of how delta-sigma ADCs worked and then linked the intuition with actual physical measurements. Bravo!
This is just amazing! Your teaching style is by far the best I have ever experienced.
I really enjoyed this video, going from theory to practice in just an hour makes an excellent balance !
Best video you have done to date. The tablet works very well. I think its a much better way to present this type of technical info. Very impressive!
I like it that you are always referring to signal-to-quantization noise instead of just calling it SNR
It is more clearer to me now with output confirming the theory behind this.I am thankful to you for this informative video.
Excellent tutorial. I especially appreciated seeing actual operation on a scope.
Excellent video, as always! The progression from theory to practical application really ties things together. So many university courses are entirely focused on the theory and it can be difficult to fully grasp the implications without practical application.
Very nice tutorial Shahriar. I really appreciate how you take the time to cover the theory and math behind this class of data converter.
Great video. Nothing else matches this depth on RUclips.
Well done and thanks.
Fantastic - covered everything I was looking for and more! I'm coming at this from an audio perspective - I knew about noise-shaping but didn't realise it was intrinsic to the delta-sigma modulator. Your video will be a great help for my retro-computing project of building an audio ADC for my old Amiga. :)
Man, you're an awesome teacher and the amount of knowledge in your videos are amazing.
Man I really like your tutorial videos Thumbs up! Delta-Sigma was one part I skipped in the exams due to time constraints because it was a really big chapter. But I wanted to understand them sometime and you just helped me with that extremely. Very much appreciated keep the good work up!
45:31 if you want to see a magic trick.
Another hugely enjoyable, interesting and educational video Shahriar. Thanks.
Fantastic demonstration! I really learned a lot about delta-sigma modulation. Designing the digital filter is the next step I would like to take.
Excellent video-tutorial Shahriar! Thank you very much for all the time you invested in the preparation of both the theory explanation and the lab part in order to demonstrate how the theory is applied in real circuits and everything matches perfectly. Kind regards,
Outstanding! Accessible at many levels of understanding, presented extremely clearly. Thanks!
This was a really great tutorial, and what I have come to love about The Signal Path. More this please!
Very Nice video to understand and visualize the Sigma-Delta ADC....
Amazing video. Very clear explanation. Loved the part where you demonstrate it with practical equipment to match the theory. Hats off!
very very good. very nice to see the teaching and application, as well as the use of the test gear to show it all.... very very nicely done
jim
I am an Electronics Engineer in the time domain, but in the s-Domain I am a Hairdresser.
Hahahahahaha.... man that's hilarious!
zaheran doroste k..... Honar nazd iranian ast o bas..... damet garm, aali bud!
Fantastic video as usual! I little over my head with the technical stuff, but you described the broad concepts very well.
I love your videos Especially the reviews and repair videos. I learn something new every time.
A very nice explanation of a sigma delta adc. A primer for me, to fully grasp the concept in its technical details. Thank you very much for a naother great video.
watched it all. you are a god amongst men, and your actions are commendable
Thank you so much for the effort in this video. I never thought much about the inner workings of those compoents.
The amout of theory was excellent to give me an understanding, and as a practicioner your breadboard example was fantastic to show it in real life. I wish school was as enaging as your videos back 30 years ago :)
Great presentation for introduction of sigma delta ADC design.
love the way the theory is visualized practically
Very good video, as always. A great review of transfer functions, from my collage days. Great explanations. I would like to say thank you for the great instruction.
Fantastic video as always. Love the mix of theory and practical. Can't wait for the next one.
Keep up with such good work and even more people you became your fans!
Thanks a lot for the tutorial Shahriar
Excellent explanation and demonstration. Keep it up.
I second Alans ***** comments, excellent tutorial on the inner workings of the delta-sigma ADC - thanks for posting. Gerry
Your video are first class. Excellent work. Much appreciated.
How can someone not like this?
Great, thanks.
Analogue Devices will probably make you head engineer in future. Thanks dude.
Very interesting and insightful presentation. Thank you
As always very much enjoyed your tutorial. Thanks Shahriar.
I like how rappers stack bills in their video and EE stack oscilloscope, signal generators in the background.
Wonderful overall work, wonderful presentation, and wow, what a great lab.
Excellent video! I really appreciate all the work you do here! Keep up the good work!
I most certainly learned from this. Keep up the good work.
I like your approach. It makes things so simple to understand.
Wow. More videos like this in specific topics of RF plz!. Also, would be great if you can provide some resources for self-learners like books or something that was helpful for you . Thank you
It's very good... Thank you very much for sharing your knowledge.
Thank you for the excellent video. It helped me a lot in Digital Signal Processing lectures! Thank you very much!
Great tutorial but I plan on watching it again to get the other half of what I missed. Great information!
Thanks for the tutorial along with the lab measurement to help solidify the theory.
Very clear explanation and demonstration, thanks.
شكرا جزيلا على الشرح الجميل المبسط حفظك الله من كل مكروه
Great tutorial, from theory to practice
Very interesting topic and perfect presentation! Thank you one more time!
Awesome video!
" I hope you're still there" - priceless.
Great video. Just one thing I want to point out: for a 2KHz input signal, the nyquist sampling rate is 4KHz. If the sampling rate is 256KHz, then the oversampling ratio should be 64 not 128.
Nice tutorial, now I am planning to check the DAC video.
Crystal clear explanation. Very helpful!!!
Great video... especially the choice of +/-VDD for supply helped me in visualizing it better, the output as pulse density modulated (PDM)
You can't use both signals from the same source because your quantization error is no longer randomly distributed?
Yes! You can't use two phase-locked correlated signals for clock and input signal to an ADC for exactly the reason you described. Smart girl! ;)
TheSignalPathBlog
Well, I still miss the REASON. Why is like that? I´ll try. The funny thing that bad ass genius of Shahriar gave already the answer at 30:00, since the RIgol and the Agilent are just DAC. The outputs of the Rigol are two digital signals converted with a sample and hold correlated with the same internal Rigol clock, and in frequency there are shaped in the same way (Sync of internal clock Rigol, since there is not an tunable with frequency out analog filter at each Rigol output). Same thing for the Agilent input signal generator. Maybe those 2 tones in the spectrum when you showed us the Sync function are the related with the clock input and signal input considered as digital ones with a sample and hold to convert them to analog... And you are right no books are telling you that!!! How did you get to this point? Anyway smart girl Rosanah Murugesu and smart boy Shahriar. Why don´t do a follow up video to respond to your mysterious questions?
Coaina Zuppero If your clock frequency is a multiple of your sine frequency, you will be sampling your sine at the same points in each cycle of your sine. So, if you plot your quantization error vs. time, it will be repeating with the same period as your sine, let's call this period T0. A signal (in this case quantization noise) which repeats in the time domain with period T0 will have a fourier transform which is zero except for frequency points where f = k*(1/T0) where k is any integer. The spectrum of the quantization noise will no longer look like white noise! It will have spurs at multiples of your input sine frequency.
Rosanah Murugesu So by definition a DAC which has on the input a periodic N words sequence and a clock sampling/rendering of Fs will produce an quantization noise in the f domain as a impulse train at f=k*Fs/N (since T0=N/Fs). That´s why the two DAC Agilent and Rigol are generating the two tones at 56:10 ? Maybe I was thinking starting from the f domanin analysis, a sample and hold DAC with a numeric sine signal input will generate a train of impulse in fraquency shaped by a Sync. Maybe I am confusing signal output and noise output. I am still digesting this stuff, sorry if I was/am not precise and clean like you in your explanations. Definitely a smart girl.
+Rosanah Murugesu
But what will be the problem/impact if the quantization noise is not white? Its not clear to me... can you eloborate more?
Great video. It makes it very easy to understand the most important points, even if the finer details are two complicated for me right now. A video on PLLs would be quite interesting.
Superb video!! Thanks for sharing.
Great video. This will help during my MSEE.
Great video you have everything knowledge, sense of humour and fantastic instrumentation. You hope we still there? You bet we are... Thank you and waiting for more...
Pretty cool explanation. Thanks a lot Shahriar.
great video as always, nice to see theory and example
There isn't many channels that I genuinely get excited for when they upload videos. But this channel - every video is a banger. Loved it. Learned some new stuff.
Shahriar - you UofT grad ?
great video, very insightful,'
thanks for sharing
well done... excellent work