Structural resonance is a really important topic - as you say, it's at the root of a great many compliance issues. Also, I need to get that VNA, it looks really good!
Even more interesting! Especially since I spend some time working on the common mode choke system to at least partially suppress the impact of interference on the measurement of electrical energy, using the example of my own apartment. Thanks to your demonstration, I can try to see if I can achieve results close to real with my simple, cheap Nanovna plus Nanovna Saver program, which will help me trust my Nanovna more. Many thanks!
Thank you for the video. I think the loop inductance increases with frequency because it becomes a transmission line or loop antenna with the first resonance at lambda/4. You can try to measure it with VNA at a higher frequency of ~3 GHz.
Yes, you are right. When measuring above 1 GHz, one can see the resonance of the loop. But I didn't want to confuse people, so only limited to 1GHz, and ignored the capacitance of the loop wire.
Cool experiment, I will try to repeat it myself using NanoVNA, which can also plot inductance and capacitance. Where exactly is your VNA calibration plane in relation to the plug on which you soldered these DUTs?
Good question. When I calibrated it, I used the N type connector, and I made a short, open and 50 ohm on the connector, so my calibration plane is the end of the N type connector. Then I made the loop wire and parallel plate at the same plane location.
I just checked my copy of "Noise Reduction Techniques in Electronic Systems", second edition, by Henry Ott, and I don't see "structural resonance" listed in the index. An oversight, maybe?
I have the second edition as well, Steve. But I also have his "Electromagnetic Compatibility Engineering", it is true he didn't use the term "structural resoance", however, you should be able to find "cavity resonance", "dipole resonance", "ground strap resonance" "cable resonance" "antenna resonance', all meaning structural resonance.
@@MachOneDesignEMC I suppose that one's opinion of "the largest cause of problems" depends on the problems you've worked on. For small sensors, I've had emissions troubles just from the need to limit the bandwidth on pwm signals, while maintaining the required drive current and required RF immunity. Switching power supplies are in the top 10 too, as is poor pcb layout, not understanding how to decouple to a metal enclosure, and many more. Since my testing required the use of a 1.5m(?) cable, it was also critical to control common mode emissions and susceptibility. I don't think resonance really factored into those problem areas.
@@MachOneDesignEMC another thought... I did deal with a few problems of resonance at the switch node on switching power supplies. Texas Instruments, I think, had a nice app note on how to design a snubber network for them. It helped me fix a supplier's problem. After a few problems of this type, I started recommending adding pads for a snubber network when reviewing peoples' board layouts for EMC issues.
Structural resonance is a really important topic - as you say, it's at the root of a great many compliance issues. Also, I need to get that VNA, it looks really good!
Thanks Julia, it is a good one, plus, it has the bode diagram function.
Even more interesting! Especially since I spend some time working on the common mode choke system to at least partially suppress the impact of interference on the measurement of electrical energy, using the example of my own apartment. Thanks to your demonstration, I can try to see if I can achieve results close to real with my simple, cheap Nanovna plus Nanovna Saver program, which will help me trust my Nanovna more. Many thanks!
Thank you for the video. I think the loop inductance increases with frequency because it becomes a transmission line or loop antenna with the first resonance at lambda/4. You can try to measure it with VNA at a higher frequency of ~3 GHz.
Yes, you are right. When measuring above 1 GHz, one can see the resonance of the loop. But I didn't want to confuse people, so only limited to 1GHz, and ignored the capacitance of the loop wire.
Cool experiment, I will try to repeat it myself using NanoVNA, which can also plot inductance and capacitance. Where exactly is your VNA calibration plane in relation to the plug on which you soldered these DUTs?
Good question. When I calibrated it, I used the N type connector, and I made a short, open and 50 ohm on the connector, so my calibration plane is the end of the N type connector. Then I made the loop wire and parallel plate at the same plane location.
I just checked my copy of "Noise Reduction Techniques in Electronic Systems", second edition, by Henry Ott, and I don't see "structural resonance" listed in the index. An oversight, maybe?
I have the second edition as well, Steve. But I also have his "Electromagnetic Compatibility Engineering", it is true he didn't use the term "structural resoance", however, you should be able to find "cavity resonance", "dipole resonance", "ground strap resonance" "cable resonance" "antenna resonance', all meaning structural resonance.
@@MachOneDesignEMC I suppose that one's opinion of "the largest cause of problems" depends on the problems you've worked on. For small sensors, I've had emissions troubles just from the need to limit the bandwidth on pwm signals, while maintaining the required drive current and required RF immunity. Switching power supplies are in the top 10 too, as is poor pcb layout, not understanding how to decouple to a metal enclosure, and many more. Since my testing required the use of a 1.5m(?) cable, it was also critical to control common mode emissions and susceptibility. I don't think resonance really factored into those problem areas.
@@MachOneDesignEMC another thought... I did deal with a few problems of resonance at the switch node on switching power supplies. Texas Instruments, I think, had a nice app note on how to design a snubber network for them. It helped me fix a supplier's problem. After a few problems of this type, I started recommending adding pads for a snubber network when reviewing peoples' board layouts for EMC issues.