Power Supply Control Loop Response Measurements using an Oscilloscope
Вставка
- Опубліковано 30 вер 2024
- Learn how to perform control loop response measurements (Bode gain and phase plots) to
test the stability of closed-loop feedback networks of switch mode power supplies.
Learn more! ► www.keysight.co...
Click to subscribe! ► bit.ly/Scopes_Sub
The mixed signal oscilloscope we used: keysight.com/f...
Learn more about using oscilloscopes:
oscilloscopelea...
Check out our blog:
keysightoscillo...
Like our Facebook page:
/ keysightbench
Download free 30-day trials of Keysight's design & test software at www.keysight.co...
It definitely is not accurate for sure. It is better to use a network analyzer ( like Bode 100 or AP300).
Based on what fact? I would be happy if you could refer to a source of comparison.
Anton: @ 5:52 the phase margin is ~55 deg and @7:31 the PM is 68 Deg. That's a huge difference. In the mid frequency range 2kHz - 50 kHz, even if the injection signal is changed from 200mVpp to 2Vpp, the phase margin would not change that much. Usually I use 1.72Vrms ( 5Vp-p) upto 1kHz and later I decrease the amplitude to just 30mVp-p. Switch mode power supplies are very noisy systems. They need an accurate measurement. The AP300 network analyzer cost $15,000 and its function is just to do Phase margin and gain margin (transformer inductance/impedance measurement can be done but this is secondary thing). Why do you think the equipment is so costly to just perform this GM/PM test? Especially for low bandwidth systems such as power factor correction circuits ( LED lighting circuits) you need very good accuracy. The bandwidth of these PFC systems is less than 20 Hz. I doubt this oscilloscope can do that job. Oscilloscope manufacturers can incorporate this frequency response measurement into their scopes, but it gets very costly ( looking at A/D sampling bits, noise issues, resolution etc- not an expert in these). The companies need to meet the cost vs performance in the industry. It is the difference between Toyota vs Mercedes Benz. You get what you pay for. Oscilloscope guys can build accuracy into this measurement but the cost of the scope goes up.It is a trade off.
Is that part of a recent firmware update? It looks much different from the X6000. I have struggled to overcome the noise on my SMPS tests - this seems to have much better performance.
Yes, we recently updated the firmware in the 3000T and 4000A X-Series. We implemented a completely different measurement algorithm that improves the dynamic range of frequency response measurements significantly. We also implemented a much better user-interface, as well as overlaying the gain and phase traces on the same plot. We plan to update the firmware in the 6000 X- Series next year. I don't know exactly when, but when it comes out, it will be a free upgrade.
The new firmware for my MSOX6000 just arrived! The FRA updates make it MUCH more usable, pretty happy to see this for sure. Thank you Johnnie.
Ummm, the magnitude response is coming back up above 0 dB. I was always taught that makes interpretation using a bode plot was nonsensical....anyone?
It is good but I thought it was a video on how-to do it with common scope without a built-in tracking generator and built-in software. In other words my garage.
WOW! This is one fantastic feature (among others) , FRA are just to expensive (bulky pc interfacing), showing time-domain distortion signals simultaneously just kicks it up a notch with effective bandwidth
The PICOTEST VRTS3 Power Supply Training and Test Board can be found here --> www.picotest.com/products_VRTS03.html
Thank you for the very good video. Is the power package available for the 2000 series?
It is not, 3000 series and up.
Is it possible to obtain the frequency response for current control loop of high power (about 25 kW) and high voltage (600 V) DC-DC converters? If yes, what are the things that needs to be considered?
Johnnie's on vacation right now, so we'll ask him for a more in depth answer when he gets back. But, you can do frequency response with the Keysight InfiniiVision oscilloscopes as long as you are able to safely probe your systems and safely connect the built-in function generator to your system. We haven't tested the control loop response as a current source for this application, so it's probably best to try to get a demo unit.
I know Bode didn't know how to pronounce his own name, but there is really no reason for you to perpetuate the error. And the word "across" does NOT end with a T