Opamp power supply starting out bad and then making it better

Building an operational amplifier power supply for audio use. I start with a poor one (on purpose) just to hear if it is really that bad and then improve it in steps.
Patreon www.patreon.co...
Mail: PO box 362, Vandalia, OH 45377
Email: [e-mail]
Thank You!

[@guilldea]

good demonstration, I couldnt hear a darn difference!

[@msana4420]

Yeah, me neither. But then again, the audio has been processed on upload and conversion. It would help if you (JAT) would maybe upload the raw audio to some host (Google drive is free till 15 GB which is plenty for the ~60MB 10-20 second audio files.

[@akkudakkupl]

Multiplier supply is fine if you have a single winding transformer but need split rail supply for lower currents. Plop capacitance multipliers at the outputs and you have a solid hi fi preamp supply there.

[@1pcfred]

I was waiting for the full bridge rectifier myself.

[@dell177]

Early in my career (1970) i worked for Analog Devices as a engineering technician so I got a good grounding in opamps. Even early opamps have a pretty good PSRR (power supply rejection ratio) and that makes them somewhat forgiving with regard to ripple. That said a good clean supply is important if you want the best performance

[@worstuserever]

Good demonstration, definitely increased my understanding, thanks! Also a nice project to copy and experiment with. Time for a parts rummage...

[@VorpalForceField]

if the 10 ohm resistors are 1W resistors, the output will max out those resistors at 80mA, which will create thermal issues. 50mA is a more likely rating for its output.

[@ao6640]

Hi, Jhon! I've been playing with op amps recently (Frecuency response, output voltage, current supply, negative feedback, ground, bandwidth, slew, etc). I saw that some don't sound too well, some high performance ones sound better but there is a "point" when they are not too different between them. I'd like to ask, how to choose a good op amp for audio aplications? Maybe if you find this subject interesting, make a video in the future where we can see the difference between a non suitable one, a good one and when is an overkill. Thank you!

[@cassvirgillo3395]

Hello John, Good to see you and a very informative, as always video. Glad you are doing well and I hope your Brother is too. Last week, starting on the 13th, I had 12 little heart attacks and a Stent put in. Doing good, still doing my chores, a bit slower. It was my 1st Diagonal apparently, that had a blockage. 48 hours from getting admitted, to being released, I don't stay down to long, I've got JAT videos to watch. All the best from Oregon, C.

[@JohnAudioTech]

Hi Cass, hope all goes well with the recovery. My brother is doing well. He just needs to get off of that walker.

[@MsFireboy2]

Thanks for posting.

[@laxr5rs]

That was really good, John. I'm pretty sure we would have heard a big difference if you used super expensive cables.... ! Heh.

[@johncoops6897]

Need to use the special "AudioFoolery" mains cable for maximum effectiveness.

[@aagevaksdal]

Another great video! Can you teach us to build a switching power supply? Maybe a dual rail thing?

[@johncoops6897]

There is almost no point "building" a switching supply. They require so many specialised components, just the output inductors can be a massive problem to source (at an economically viable price).

[@roboticdart6138]

Awesome video 👍 I'd love to see one experimenting on power amplifier supplies

[@ferrumignis]

For a low current circuit like this were absolute voltage stability is not really an issue you could use a capacitance multiplier to pretty much eliminate the ripple.

[@suadcokljat1045]

Point taken. Now, let's make video on high performance, audiophile grade op amp power supply. Cheers!

[@Gersberms]

Great video, but because of what opamps do, EEVblog also explained it well saying that opamps try to equalize negative and positive inputs. I wasn't too surprised by this. However, I recently realized that transistors themselves also don't care much about their power input, because they are a current device. They just amplify the current coming in through the base, right? Depending on temperature and operating range. So power amplifiers, mostly, should also inherently not care much about some level of noise on the power input. Unless mistakes were made. Do I understand that correctly?

[@thevicol]

The professional use line level circuits operate on signals that can significantly exceed 2V. There must be enough headroom. Even newer opamp designs can work better with a reasonably high supply voltage(better linearity for signals close to rails, sometimes better slew rate,...). Sometimes it is necessary to use chips with j-FET inputs and common-mode input voltage range turns out to be a limitation.

[@aksela6912]

Would have been interesting to see the regulators performance with less pre-filtering. How little can you get away with?

[@JohnAudioTech]

I'll have to try that.

[@alicayden9792]

instablaster...

[@abhinb1998]

you rock man great work

[@amitghosh6966]

I have questions - 1. What should be the output capacitor value for the regulator IC if the required load current is 500 mA - 1 Amp (higher current version regulator IC) for running low power audio amplifier chips like Lm386, Tea2025 etc ? 2. Is it sufficient to have only one electrolytic capacitor at regulator output or a combination of ceramic capacitor (0.1 uF or so) for suppressing high frequency noises ?

[@KissAnalog]

Always a great video! Thank you - I look forward to your next video;)

[@rthu-tt1wx]

It would be interesting to see the half wave supply compared with a full bridge rectified with a resistor divider virtual ground supply.

[@jp040759]

Great topic

[@amitghosh6966]

Please make a variable dual rail power supply circuit for powering any circuit including audio amplifiers

[@PauEstalella]

Wouldn't a full bridge reduce the capacitor requirements?

[@poptartmcjelly7054]

two half bridges are often used to produce split rails

[@HillsWorkbench]

Yeah, don't know why one would use a half-wave in modern times, it is a waste of tranformer capacity and diodes are cheap. May have made more sense when rectifiers were expensive.

[@HazeAnderson]

@@HillsWorkbench Because I can still get +/- 15V from a transformer whose output is only rated at 12V6.

[@JohnAudioTech]

@@HillsWorkbench Half wave is fine for a low current supply. Allows for a split supply with a transformer that is not center tapped. It does use both halves of the waveform (my transformer was not center tapped). More common voltages available, such as 9 or 12 volt secondaries.

[@mastroaka]

You could do another test with a capacitance multiplier. Trying it before or after the regulator. I've always wondered if such a circuit works well and if work better before or after the 78xx/79xx

[@mastroaka]

@Paweł Hadam Yes I was think about low power circuits. In my mind I was thinking to use 78xx/79xx to level the voltage and the capacitance multiplier to filter high frequency ripple noise. If I use the multiplier after the 78xx/79xx i will lose a little bit of headroom due to the transistor. I was wondering if putting it first could filter the ripple anyway.

[@mastroaka]

@Paweł Hadam i use LM7824+LM7924. I don't have an oscilloscope in my garage (and I cannot bring my circuit to work...), so I cannot check circuits in an accurate way. I will try the LM317 in the future and I hope this will perform better! Thank you for the advice!

[@robertneill3057]

@@mastroaka Way over for most op-amps. Most can handle a maximum of 30v to 36v on a single polarity supply voltage. Use a maximum of 15v per supply rail for a dual polarity supply that does allow some headroom. The LM 317 does have a negative counterpart the LM 337 regulator. Velleman does a kit that uses these variable regulators.

[@DAVIDGREGORYKERR]

What about virtual battery operation like that used in the TECHNICS SUA-900 MKII

[@trevorvanbremen4718]

SVR? I thought it was called PSRR on the datasheets?

[@EngineeringEssentials]

I have seen power amplifier transfomer center tap connected with pre amp ground and connected to metal body of amplifier. Somebody said connection with body of amplifier is mandatory to reduce noise. Is it? If so how it will reduce noise?

[@JohnAudioTech]

Having the metal casing at the same potential as the circuit common can reduce potential noise ingress but can introduce ground loops.

[@EngineeringEssentials]

@@JohnAudioTechnow I know . thank you 🙏. Hope you're doing great 😊

[@robertneill3057]

The 0v rail can be lifted by a 4.7Ω to 10Ω resistor rated at 2w to the earthed chassis. Cambridge Audio used this method in their A5 amplifier.

[@joeyscott4299]

Nice 👌 great information 👍

[@patrick_test123]

Am I missing something, but wouldn't you want a LC-pi-filter rather than an RC-pi-filter? Of course you would then need a series RC-network to prevent the filter from ocillating, but you should get a better result with that. Annotation: Expenation why this is a bad idea are in the comments.

[@patrick_test123]

And also you could get a desingn with that that can output much more power in theory.

[@notatoyotacamry195]

It's not necessary if you only need to send high frequencies under the noise floor, it seems. -20dB per decade is enough when the cutoff is only 15Hz. And low frequencies can already be rejected by the regulator.

[@patrick_test123]

@@notatoyotacamry195 Okay I just punchen in some combinations for RC an LC pi- filtes. And you will get ether huge inductors or capacitors to get your cutoff frequency as low as 15Hz. So it's just easier for the low currents to go with RC.

[@listerine-pr5lt]

If you use simple transistor regulator say zener diode and high gain single power transistor in pair to compare them with voltage regulator IC in next video it would be a perfect comparison.

[@JohnAudioTech]

I might have to try that.

[@jeffm2787]

Surprised you didn't use a capacitance multiplier. Dual of course.

[@rodrigobelinchon2982]

I hear more dynamic range and betetr high frequency response in the first sample , let`s see which is which

[@JohnAudioTech]

The first sample has the better supply (batteries). However, I doubt there was any audible changes to the music. If you listened to the second sample first you might have considered it better.

[@rodrigobelinchon2982]

@@JohnAudioTech I listened to it using professional audio monitors , regular hi-fi speakers mask a lot of what is going on on high frequencies and might make the two samples sound very close.

[@thevicol]

The difference is very subtle but it is. The second sample sounds a bit lifeless. The ripple is not loud, but a single stage with a moderate gain does not have a significant influence on relatively strong signals. The game starts when the signal has to be amplified strongly and the gain is applied in multiple stages. That is why good power supply is important.

[@handryraya2853]

100uF vs battery 08:51 1st sample 09:37 2nd sample My comment : at my 90% headphone volume from mobile, I just cant hear the difference, not even the noise. Are you serious john? ;)

[@handryraya2853]

I build my power supply for ne5532 preamp, i can hear the noise if i put my ear on the tweeter side.