The Op Amp Gyrator Demystified

[@MuzixMaker]

Floobydust. Love it, you must also be an old timer! I still have my original copy of the AR Handbook.

[@oldhackee3915]

Yes that would be an accurate moniker.

[@MuzixMaker]

@@oldhackee3915 it takes one to know one! Cheers.

[@yorkyoung4449]

I'm so touch by your detailed math derivation steps! But seems not many one shares the same feeling! Thank you for your dedications, and i wish you a very pleasant future!

[@oldhackee3915]

Thanks so much for the kind words.

[@vutroslavgandic]

Amazing videos, exactly the topics I'm interested in. Really happy I found this KZbin channel.

[@oldhackee3915]

Glad you like them! Thanks.

[@hardrocklobsterroll395]

I look forward to all your releases. Thank you for such great information

[@oldhackee3915]

Thanks so much.

[@rayc1557]

Update on the headphone amp and EQ board. Executive Summary: It sounds Awesome!. High-level Outline: All first-pass PCBs (power supplies + Class-A transistor amp + 5-band EQ board) were created, populated and individually tested over the last several weeks. Today (9/16/2023) at 20:38, all the stereo pots were connected and everything powered-on as a system. Audio patched-in from a laptop and observed on a scope. Traces looked great. Volume gradually turned-up -and the sound was stunning. Clear, detailed and full-sounding. Right out of the starting blocks, it sounds as good as (or better) than my commercially purchased headphone amps. Those gyrators do indeed, work as described. Now that proof-of-concept is accomplished, I would like to build a 7-band EQ and fiddle with the Q's.

[@oldhackee3915]

Congratulations on your success. Great work.

[@rayc1557]

FYI: Version 2 of the Equalizer is in the works. Hopefully each equalized band will have less influence on the others. The plan is to add (sum) an inverted input signal to the output of each EQ band. In theory, for each EQ band, this will leave behind only the portion of the original signal whose amplitude was modified. The resulting sum of all those should contain just the frequencies that were adjusted by the user. The final step is use a cut/boost ckt to allow the user to fade-in some of the original signal to "smooth-out" or restore the gaps that could exist depending on how hard the user cranks the EQ bands. Let me know if that makes sense and if you think it will work.

[@rayc1557]

Just want you to know that your 3 videos (Boss GE-7 Graphic Equalizer, Cut & Boost Level Control and this one; Op Amp Gyrator Demystified) have been tremendously helpful. I'm building a headphone amp from scratch, designed from the ground-up. The front and back end are pure transistor/class A designs. The amp will accept pro or commercial audio, balanced or unbalanced. The output will be selectable for different ranges of headphone impedance. The front and back ends are working very well in prototype fashion. I decided to insert an equalizer and am using gyrators/opamps. The equalizer ranges are being fine-tuned to my liking with varying Q on several bands. Just finished the first gyrator for a single band and it works remarkably well. During this project (a first for me) matching all the stages and overcoming all manner of glitches has been a tremendous learning experience. At age 63, I feel that I've gotten smart too late and old too fast. I marvel at the myriad people before me whose efforts now make this a form of education and entertainment for me. All of your videos are top-notch... Thank you.

[@oldhackee3915]

Your comment means a lot to me. I’m so glad it was helpful. All the best on completing your project.

[@rayc1557]

@@oldhackee3915 Have now added a second band of EQ. The resulting Bode graphs, with the dials set to strategic positions, is a beautiful sight. Lest I jinks myself, it almost looks like I know what I'm doing. The big decision now: 5 or 7 bands of EQ??? That is the question. This project will indeed be completed... As you know -the DIYer's creed: Never pass-up the opportunity to make something yourself, that can be purchased outright at 1/4th the cost 🤣

[@oldhackee3915]

Love it. Let us know how it turns out.

[@VEC7ORlt]

Why would one need so many bands? One of the best eqs out there would be 'tilt equalizer' and should use very modest gain of +/-3dB - it basically makes sound darker/brighter, if you need more than that - those headphones are garbage to begin with.

[@LydellAaron]

Very nice. I was happy to watch your video. Well done.

[@oldhackee3915]

Thanks very much.

[@DustinWatts]

I am currently building a audio channel strip with EQ section. 4 individual cut/boost pots with selectable frequency. I have all the math done for those, but it is very interesting to have a peek at the math behind a gyrator. Thanks for the video.

[@oldhackee3915]

Thanks for the comment.

[@Abihef]

I can't describe how much i appreciate this video enough and how useful it is to me Thank you very very much and I'm off to check the rest of your channel now

[@oldhackee3915]

You're very welcome! Thanks for commenting.

[@Stelios.Posantzis]

That was a quick turn-around indeed! I have spotted a couple more items in some other videos of your op-amp series. Would you like me to point them out? The thing is though, I can spot some stuff in parts of a video and I may miss others. While the videos are short, they are dense with info/formulae so it is hard to be thorough. I just thought I should mention this caveat. I really enjoy this series btw. Really well done, great presentation and choice topics. Looking forward to more uploads.

[@oldhackee3915]

Sure. Any comments are appreciated!

[@oldhackee3915]

Thanks for the kind words. You have been a great help to me and I appreciate it.

[@johndododoe1411]

Fun watching s equations again after 30 years.

[@oldhackee3915]

I love working in the complex frequency domain!

[@Alan96555]

Very complete video! Do you know if there is a way to demonstrate or find the matrix of the rotator from the circuit with the operational? I tried but it didn't work out. I also didn't find any books that talk about the subject in detail or that give demonstrations, most of them just spit out the equations without showing where they come from.

[@oldhackee3915]

Thank you. No, I do not.

[@alexloktionoff6833]

There must be gyrators implemented on one discrete transistor, they must be high-frequency capable, can you make a video of such a gyrators?

[@oldhackee3915]

Yes, I've seen them used in graphic equalizers. You just replace the opamp follower with an emitter follower NPN transistor.

[@alexloktionoff6833]

@@oldhackee3915 Wow! but emitter follower has amplification

[@oldhackee3915]

It doesn't need to be a gain of exactly one. It serves the role as a buffer for the RC network.

[@alexloktionoff6833]

@@oldhackee3915 please, please make a video with concrete gyrator implementation and oscilloscope measurements on gelly-beans, lm358 and/or 2n2222!

[@rayc1557]

Everything was going so well until... until I found-out that not all Ti 5532 opamps are created equally. I had a mix of different 5532 opamps and 4 of them were new originally from a sealed package. Upon adding a 5th band of EQ using a Ti 5532 with a different suffix (but also a new part), things went south. Both parts share the same datasheet and there are no apparent differences described therein. Also had some Ti TL072CP and generic LM833n laying around. The 072CP failed miserably and the generic 833's kinda worked but have distortions at higher gain. Any suggestions for a good opamp to implement gyrators?

[@rayc1557]

Update: The problem is not the gyrator ckt but, various combinations when joined with a cut/boost ckt. This was determined by implementing a single band of graphic EQ across two IC packages. One packaged housed the cut/boost ckt, the other held the gyrator. [In both packages, the unused OpAmp was tied down with unity gain as per your video on that topic]. A full mesh of 16 tests were performed using 4 different dual OpAmp packages. Only 2 combinations produce perfect results. Only one model could be used for either cut/boost or gyration. Some OpAmp packages simply do not tolerate being used for cut/boost and some cannot tolerate use as a gyrator. BTW: The test was done twice using different components to rule-out defective ICs. (Can you say: School of hard knock?).

[@rayc1557]

@7:10, depicts the passive and active version of an RL filter... What I want to know, is who the heck made the observation that the active ckt on the right produced an equivalent RL configuration? Indeed, the proof you (graciously) provided is very convincing; however, it does not explain HOW or WHY (not shouting) in the world, someone realized that the two circuits had parallel 1:1 impedance functions. That is the stroke of genius I'm curious about. ... Sheesh, just about the time I think I'm getting the hang of analog analysis, somebody pulls a rabbit out of their hat and knocks my socks off. I need a volume control that goes up to 11!

[@oldhackee3915]

Thanks for the comment Ray. The gyrator was actually conceived in the passive realm initially in 1948. The opamp version was a follow on to it. Good reading here: en.m.wikipedia.org/wiki/Gyrator

[@mr1enrollment]

Fun topic, I suggest follow up lab work with real circuit demo. ain't algebra great? cheers dan

[@oldhackee3915]

I may do that. Cheers back!

[@jacobfaseler5311]

Excellent presentation - my understanding of the functionality was definitely improved by understanding the derivation. Fun fact: these don’t have to be connected to ground. If you feed two into each other in antiparallel and connect the ‘ground’ nodes, you can simulate a floating inductor (see m.kzbin.info/www/bejne/d3atpZSglredldU) - which opens up myriad opportunities for application. Virtual inductance in a feedback loop, bias injection, you name it.

[@oldhackee3915]

Thank you. Glad it helped. The grounded one was the only one I had experience with. I was happy to finally get through the math. I wasn't aware of a floating one. I'll look forward to watching what you sent. Thanks so much.

[@stevekoehn1675]

Bummer you wasted time because of typos. This is stuff I have wanted to know. Thanks

[@oldhackee3915]

Thanks for the comment. I'm glad you found it useful.