Synchronous sampling and anti aliasing filter: a demonstration by LTspice simulation

  Рет қаралды 2,277

Sam Ben-Yaakov

Sam Ben-Yaakov

Күн бұрын

Пікірлер: 33
@RaananL
@RaananL 5 ай бұрын
Very nice presented! Great idea to show it with simulation. Thank you Prof. for your efforts making these great educational videos!- Raanan.
@sambenyaakov
@sambenyaakov 5 ай бұрын
Thanks Raanan
@Cydget
@Cydget 5 ай бұрын
One of the issues of adding the LPF is that it adds a substantial delay for low frequencies. Because the delay is so long, I've found reducing the zerocross jitter becomes a challenge. Thank you for showing how important the sampling frequency is, and that just by sampling closer to the nyquist frequency can yield different results.
@sambenyaakov
@sambenyaakov 5 ай бұрын
The relationship between BW and rise time is fixed. I am not sure I follow your comment. Can you be more specific? The application, some numbers?
@petrusbosman4264
@petrusbosman4264 5 ай бұрын
Thank you so much Sir!
@sambenyaakov
@sambenyaakov 5 ай бұрын
🙏🙂
@shamkumar730
@shamkumar730 5 ай бұрын
Thank you very much sir..
@sambenyaakov
@sambenyaakov 5 ай бұрын
🙏🙂
@byronwatkins2565
@byronwatkins2565 5 ай бұрын
To reproduce frequency f, you must sample for a time T=2/(f_s-2f) at sampling frequency f_s. So very near the Nyquist frequency it takes a very long time to gather enough data. But at f
@sambenyaakov
@sambenyaakov 5 ай бұрын
This is ( a textbook theoretical situation) for a fixed signal that is then averaged. Who is interested in that? In any practical use the signal is changing.
@byronwatkins2565
@byronwatkins2565 5 ай бұрын
@@sambenyaakov That is rather my point. Practical signals require significantly higher sampling rates than the Nyquist criterion... 4-5 times faster gets reasonable response times.
@sambenyaakov
@sambenyaakov 5 ай бұрын
@@byronwatkins2565 Isn't that what I am saying in the video?
@tamaseduard5145
@tamaseduard5145 5 ай бұрын
👍🙏❤️
@sambenyaakov
@sambenyaakov 5 ай бұрын
👍🙏🙏
@ats89117
@ats89117 5 ай бұрын
Very nice, but you've given the ideal synchronous sampling case where you are able to adjust the delays between the control of the SMPS clock and the A/D conversion clock so that the sampling occurs when the switching noise is minimal. In the real world case, it would be difficult to control this delay, and you might always be sampling at the time where switching noise is at a maximum. It should be possible in this case to put a small delay in the A/D's convert clock to get back to the case which you have assumed, but most systems that I am familiar with don't have this capability...
@sambenyaakov
@sambenyaakov 5 ай бұрын
Thanks for input. Not only that this is possible it is done very successfully. In the products that we have designed, the signals are sampled synchronously. Not that difficult to accomplish.
@ats89117
@ats89117 5 ай бұрын
@@sambenyaakov I wasn't claiming it wasn't possible, but if you look at precision A/D convertors from TI and Analog Devices, the app notes mention this subject exactly 0 times and I can assure you that outside of your community, this isn't even discussed. So synchronizing the frequency of the A/D with the SMPS is rare and adjusting the phase relationship to coincide with a quiet portion of the SMPS is unheard of. See for example Analog Devices app note CN-0137 where they discuss using a SMPS solution to increase efficiency for a precision A/D converter. In their circuit example, they don't even bother hooking up the switching frequency control. Or TI's Application Report on High Speed A/D Convertor Basics, where half of the long document is allocated to input signal conditioning without a single word about SMPS synchronization. I suspect that somewhere on these sites you would find this discussion, but that the overwhelming percentage of designers has never seen it.
@tomaszswiechowicz1181
@tomaszswiechowicz1181 5 ай бұрын
@@ats89117 I agree with professor. At my university in Poland and in the industry that I worked with using synchronous sampling is common practice. The app note you refer to is about ADC power supply, not application in power converter and that's why this information is missing. What systems are you familiar with? Are they equipped with microcontroller?
@ats89117
@ats89117 5 ай бұрын
@@tomaszswiechowicz1181 The question is not about synchronizing the sampling frequency of the SMPS with the A/D converter. Many SMPS have a clock frequency input for synchronization. The question is about phase shifting the convert time to ensure that it occurs while the SMPS is at it's least noisy period. My application area is phased arrays where there are 10 to 1000 A/D converters each bringing in 1-2 million samples per second with 20 bits per sample. These systems are always controlled by FPGAs and the SMPS is frequency locked but I've never seen it phase locked. If there was a phase adjustment, which could certainly be added to the firmware, the only way to do the adjustment would be by trial and error because there is no visibility on the delays for either the SMPS or the FPGA clock circuit that generates the convert signal. I am designing one of these systems right now and have added this as a requirement because it seems like it could be very important. As an aside, thanks for your comment. I have a friend who is a professor at the University of Warsaw, and I am interested if this is a more visible issue in other places...
@tomaszswiechowicz1181
@tomaszswiechowicz1181 5 ай бұрын
@@ats89117 I think that my misunderstanding came from the fact that I'm designer of power converters (based on microcontrollers) and adding phase adjustment functionality is trivial. Looking at the market, however, I can't recall any SMPS IC capable of that. Maybe ICs for converters in interleaved configuration would have it. I have a fact that might be interesting to you: I built reactive power compensators that are supposed to work in parallel through fiber optic communication. One of my goals was to enable phase adjustment through it, because it greatly reduces conducted EMI below 1MHz when phase is selected for destructive interference.
@RogueTuning
@RogueTuning 5 ай бұрын
Thanks Professor! What was the simulation command you used to produce the spectrum plots?
@getusel
@getusel 5 ай бұрын
You add AC analysis instead of transient.
@RogueTuning
@RogueTuning 5 ай бұрын
@@getusel Ok. And what source to set for 'small signal AC analysis'?
@getusel
@getusel 5 ай бұрын
@@RogueTuning Right click to any independent source select advanced from the resulting window. Choose the source from the function menu. On the right side of the window you will see Small signal Ac analysis amplitude box. Insert 1 there and you should be able to run ac analysis.
@electronichome1153
@electronichome1153 5 ай бұрын
Mr. Yaakov, can you continue the series of simulations with classic PSpice, especially the free version of Texas Instruments - PSpice for TI? LTspice continues to be the same tool since its creation and I find it neither attractive nor better than the powerful PSpice engine and interface. It is so yeasy to import spice models into PSpice simulation... Thanks!
@sambenyaakov
@sambenyaakov 5 ай бұрын
I do mot like the TI Pspice. It is missing parts and the library is meager. Have you tried Qspice?
BJT filters and capacitance multipliers
19:51
Sam Ben-Yaakov
Рет қаралды 3,3 М.
Understanding Aliasing in Digital Down Sampling
9:40
Iain Explains Signals, Systems, and Digital Comms
Рет қаралды 18 М.
I CAN’T BELIEVE I LOST 😱
00:46
Topper Guild
Рет қаралды 114 МЛН
КАК ДУМАЕТЕ КТО ВЫЙГРАЕТ😂
00:29
МЯТНАЯ ФАНТА
Рет қаралды 1 МЛН
I2t. What? Why? When?: The electric fuse case
16:00
Sam Ben-Yaakov
Рет қаралды 4 М.
Aliasing and Anti-Alias Filters
14:16
Precision Filters
Рет қаралды 32 М.
Totem Pole PFC Design for E-Mobility
46:04
Wurth Electronics
Рет қаралды 10 М.
The inner works of Peak Current Mode (PCM) control
24:50
Sam Ben-Yaakov
Рет қаралды 3,6 М.
US New Supersonic Passenger Jet Finally Makes Its First Flight!
10:16
Military Affairs
Рет қаралды 22 М.
Ripple control of PWM converters
22:00
Sam Ben-Yaakov
Рет қаралды 3 М.
TI Precision Labs - ADCs: Aliasing and Anti-aliasing Filters
8:13
Texas Instruments
Рет қаралды 21 М.
Why Higher Bit Depth and Sample Rates Matter in Music Production
7:06
Audio University
Рет қаралды 201 М.
Common mode chokes: There is more than meets the eye
25:00
Sam Ben-Yaakov
Рет қаралды 4,6 М.
Estimating by time domain simulation the LoopGain transfer function
10:41