What is PID control?

An intuitive power-electronics explanation of PID control
Please note: In slide at 13:00 minutes it should be 1/ 2piKd instead of Kd/2pi

[@jbshaik8512]

Very Nice and Clear Presentation Sir

[@sambenyaakov]

Thanks

[@TheGreatHammys]

Great job! More videos about control in PE, please!!

[@sambenyaakov]

Will try

[@alirezagalib657]

Dr. Sam, thank you very much for this wonderful lecture. I would like to request you to make some videos on digital control of power converters.

[@sambenyaakov]

Will try.

[@haneeshas5985]

Dr. Sam I really appreciate your presentations. They are really worthy. I have one doubt. In 13:00 slide is it 1/ 2piKd instead of Kd/2pi?

[@sambenyaakov]

Yes, thank you. Just added a note in description section " Please note: In slide at 13:00 minutes it should be 1/ 2piKd instead of Kd/2pi"

[@georgedaadouch1971]

As good as always. I think the first pole for last circuit you've shown must be rather 1/2piR2C2..cause C3 impedance is still higher than C2 impedance..thanks

[@sambenyaakov]

HI George, Thanks for discovering an error. You are right, it should be 1/2piR2C2 but this is not the pole location but rather the gain in this section. The location of the pole will be determined by the open loop DC gain of the amplifier.

[@georgedaadouch1971]

Hi prof..thanks for correction Hope you explain this region of the plot in a video like how open loop DC gain determine the pole location Always get benefits watching your videos

[@willytzui2407]

Very clear illustrations! Thanks!

[@sambenyaakov]

Thanks

[@WalidIssa]

Thanks. A suggestion for a future video about TypeII and III compensators especially for an application like converter and they are in this tool Power Stage Designer Tool

[@sambenyaakov]

See kzbin.info/www/bejne/jXSremxpp6t2aJo

[@user-power_electronics]

Thank you for your detail explanation on PID control. Could you explain why the first pole is 1/2*pi*R2*C3 ? I normally start at the zero 1/2*pi*R3*C2 with dc gain of R3/R2 and draw backward +20dB slope until it cut off with the op-amp open loop dc gain. Wonder if that the right way to do it ?

[@sambenyaakov]

Hi Thanh, The expression 1/2*pi*R2*C3 is not meant to indicate the the location of the first pole but rather the gain in this section. Unfortunately, there is a mistake in the drawing. It should read 1/2piR2C2. I hope other viewers will notice this correction. I might prepare a short correcting video.

[@HaykDingchyan]

We've already suffered for this :D

[@thezodiace7399]

thank you sir for the presentation. I have a question. Do we need to care about PID when using controller IC to driver the DC-DC converter we want to stabilise? thank you.

[@sambenyaakov]

Normally you do. It will be part of the feedback network. Mostly PI .

[@thezodiace7399]

@@sambenyaakov can you please elaborate more?

[@marcus_antonius]

Wow, you are great. Thank you.

[@sambenyaakov]

Thanks

[@stanleybane5483]

Hi sir, what is the difference between compensator and controller? when to use a compensator? when to use a controller?

[@sambenyaakov]

To me control is a general term independent of the method. Compensator is related to a control method that shapes the phase to compensate the phase lag that causes instability. PID is a compensator. Hysteretic control is not.

[@odissey2]

My experience over 15 years tuning galvos, voice coils, current drivers, ovens and TEC controllers, that D term can be successfully ignored. Occasionally, an automatic tuning is available (e g. Ziglier-Nichols), suggesting small D contribution, but tuning improvement is typically marginal (few %). I know that some authors are adamant about D term, which is contrary to my experience. My question is simple: where the idea of using P + I + D in feedback loop for system control comes from? What is a background for that (apart from empirical approach)? Most textbooks omit this subject entirely, succumbing for: "Let's try to control our system using PID..."

[@sambenyaakov]

D is used in control of power electronics systems! It would seem that you have not watched carefully the video. The fundamental reasons for using PID is in it. See also kzbin.info/www/bejne/qXLCZ6F7oNpmo9E

[@petergriffin760]

Thank you for this lesson! Every video on you channel enlights new interesting aspect of PE. In addition to this content, could you suggest readings about synthesizing a digital PID?

[@sambenyaakov]

Don't recall off hand, there are many textbooks and papers that you can find by googling.

[@swathi9086]

Sir coming to controlling of llc how does the error Amplifier or compensator ouput gives frequency that we need ,the same compensators are also used for getting duty,how is this possible please do answer or make vedio,please sir.

[@sambenyaakov]

The output of error amp drives an VCO. Not a duty cycle modulator.

[@swathi9086]

That's fine sir but how the same compensatory circuit is used for getting different control variables like duty, switching frequency .

[@sambenyaakov]

@@swathi9086 Why same?You need to design the controller per topology based on the particular transfer function of the system.

[@robson6285]

I think that 6dB/oct is much more clear than 20dB/dec It ís the same slope, or am i wrong? I like your lessons always, but this time i think a more practical approach was more usefull. But maybe you find (and maybe you then are right) that there are already enough of that sort videolessons online that are incomplete because they are missing much theoretical complexity for them who needs that deep understanding. And indeed its interesting to know.

[@sambenyaakov]

Hi Bob, I am leaving the Octaves for the musicians. Rarely, if ever, do power electronics workers use db/Octave. And by the way, 6 db /Octave is an approximation. It should be 6.0205999......... (-:

[@sambenyaakov]

One more point, the objective of this video is to unveil the mystery (to many power electronics workers) as to what is PID.

[@AspartameBoy]

Do z transforms

[@sambenyaakov]

This does not explain anything which is the objective of the video.

[@iftikharabid659]

Hi, Thanks for your nice explanation. With reference to your LinkedIn Comment. www.linkedin.com/groups/13606756/13606756-6428348019577618435 I am pasting your expression here for viewers for the correction in the last slide. "The expression 1/2*pi*R2*C3 in the last slide is not meant to indicate the the location of the first pole but rather the gain in this section. Unfortunately, there is a mistake in the drawing. It should read 1/2piR2C2. (this is for the approximation C2>C3))"

[@sambenyaakov]

Thanks for posting this.