DCAudioDIY.com

DC Area Audio DIYer's Community
It is currently August 11th, 2020, 2:39 am

All times are UTC - 5 hours [ DST ]




Post new topic Reply to topic  [ 8 posts ] 
Author Message
 Post subject: Power Supply Regulation
PostPosted: July 4th, 2020, 12:20 pm 
Offline
User avatar

Joined: July 24th, 2015, 4:17 pm
Posts: 1239
Location: Parkville, Maryland
I stumbled on this video currently playing on YouTube. It is slow moving and the presenter could have delivered the same information in a more efficient manner. :eh:

https://www.youtube.com/watch?v=p9aaPZwRAs0

That said -- it clearly explains why there is no benefit to hanging large storage capacitors (C) on the load (output) side of a 3-pin regulator I.C.

It cleared up for me as to why I got such good results keeping the load C small and the input C large (that is -- up to the limit of the power transformer's ability).

The late Walt Jung explained this to me many decades ago, but I was unable to pass on that knowledge to this group in a way that cleared the controversy that eventually ensued.

Anyhow -- everyone on this blog can benefit from watching this video in its entirety. If you are prone to nodding off during boring seminars, drink some coffee first. ;)

_________________
Walt


Top
 Profile  
 
PostPosted: July 4th, 2020, 1:21 pm 
Offline
User avatar

Joined: February 28th, 2013, 3:31 pm
Posts: 1413
Several things he got wrong or missed.

First, on the spec sheet he points to the min and max voltage and says the regulator will hold the voltage to within +- 200 millivolts. Actually, the spec he is pointing to is the factory tolerance on the output voltage of any single regulator from production, not the output regulation which is covered in the two sections on differences due to input fluctuations of load changes.

Second, He is talking mainly about the ability of the regulator to suppress line frequency changes. He briefly mentions high frequency transients but does not go into any detail. For line frequencies, the regulator does not need much output filtration because the regulator can respond fast enough to keep the voltage constant. What he does not mention is fast changing loads that you get in audio circuits where the regulator cannot keep up with the rapid current changes. Also, the same is true for high frequency noise on the input of the regulator which for most part will be passed when the frequency gets high.

He also does not address the characteristics of the circuit being fed by the regulator. Most differential circuits including op amps have high power supply rejection ratios at lower frequencies but that starts to drop off at a rate of 12 dB/octave starting somewhere around 100 Hz. So you have a decreasing ability of the regulator to reject those high frequency aberrations at the same time that the audio circuit is suffering the same problem.

The best answer is to have a passive 12dB/octave low pass filter between the regulator and the audio circuit. That way any high frequency interference gets attenuated at the same rate that the ability of the circuit to reject that interference occurs. I routinely come out of the regulator into a 1 mh choke and then a 10,000 uf cap that is further bypassed with small HF caps. By doing that the roughly 120dB/octave PSRR of the op amp essentially stays the same for any frequency. In addition, that large reservoir essentially makes the power line disappear to the audio circuit.


Top
 Profile  
 
PostPosted: July 4th, 2020, 2:51 pm 
Offline

Joined: January 15th, 2015, 7:19 am
Posts: 1227
Location: Baltimore MD
thanks, Walt the video was informative to me being just a dumb lump of poop on a log.
I did not understand Toms remarks. Did not dumb it down enough for me.

I do have a question about the load on the PS in the demo. He varied the load from 47 ohms to 5 ohms.
would this be the what amps do to their PS as you raise the volume level?


Top
 Profile  
 
PostPosted: July 4th, 2020, 3:56 pm 
Offline
User avatar

Joined: July 24th, 2015, 4:17 pm
Posts: 1239
Location: Parkville, Maryland
Pelliott321 wrote:
thanks, Walt the video was informative to me being just a dumb lump of poop on a log.
I did not understand Toms remarks. Did not dumb it down enough for me.

I do have a question about the load on the PS in the demo. He varied the load from 47 ohms to 5 ohms.
would this be the what amps do to their PS as you raise the volume level?


Not to that extreme, but yes. Line level amps. (pre-amps and such) are much more benevolent in terms of requiring current delivery then power amps.

_________________
Walt


Top
 Profile  
 
PostPosted: July 5th, 2020, 11:08 am 
Offline

Joined: March 2nd, 2013, 2:43 pm
Posts: 182
Location: Potomac, MD
I second Tom's response. Connecting your audio circuit directly to the output of a regulator is basically connecting a high-feedback circuit not particularly designed for transient signals to your amplifier circuit. You are best served by isolating the high-feedback regulator circuit from your amplifier circuit using the power (load circuit) by an impedance network which can be as simple as a resistor, and then following that with lots of energy-storage capacitors. Tom uses an inductor for the isolation, where as I might use either an inductor or a resistor. I might be more inclined to use the inductor if I was concerned about power loss, but issues can develop such as induced fields causing a modulated voltage drop between the output of the regulator and the load circuit. The resistor degrades regulation performance if the load current varies, but this is why you should have lots of energy-storage caps following the resistor. So, my top choice is a resistor, which can be quite small in value. For op-amp load circuits 10 ohms is fine. Higher-load circuits can probably use one ohm, but at this point I would be considering the inductor.


Top
 Profile  
 
PostPosted: July 5th, 2020, 11:26 am 
Offline
User avatar

Joined: July 24th, 2015, 4:17 pm
Posts: 1239
Location: Parkville, Maryland
Well there are some other issues to consider. I would expect that for the large storage on the load side of the 3-pin regulator to be effective you need a Dianix Sigma 1000 galvanic demodulator for that to be effective. Additionally, a Transverse Axial Polarizer (TAP) is required to minimize a form of micro-distortion identified as “electromagnetic polarization” that reduces “sonic glare” while increasing clarity and coherency. Lirpa Labs, located in New Mexico, has done a lot of research regarding these issues. :thumbup:

Also, if you are going Tom’s route you should include a single channel ULF (Ultra Low Frequency) RF pulse generator that acts as low-frequency dither to overpower the ambient RFI and EMI fields in your listening. Another feature that serves Tom’s method is a high-speed, active ground block that is powered separately, which provides a conductive path separate from the normal current-carrying path of your components carried by a pure 120-Hz. sine wave which can reduce noise by as much as 10-dB! :o

_________________
Walt


Top
 Profile  
 
PostPosted: July 5th, 2020, 11:34 am 
Offline

Joined: January 14th, 2015, 11:15 pm
Posts: 368
If you want to spend a long afternoon you could wade through the D-Noizator thread.

https://www.diyaudio.com/forums/power-supplies/331491-noizator-magic-active-noise-canceller-retrofit-upgrade-317-based-reg.html

ray


Top
 Profile  
 
PostPosted: July 5th, 2020, 11:34 am 
Offline

Joined: January 14th, 2015, 11:15 pm
Posts: 368
If you want to spend a long afternoon you could wade through the D-Noizator thread.

https://www.diyaudio.com/forums/power-supplies/331491-noizator-magic-active-noise-canceller-retrofit-upgrade-317-based-reg.html

ray


Top
 Profile  
 
Display posts from previous:  Sort by  
Post new topic Reply to topic  [ 8 posts ] 

All times are UTC - 5 hours [ DST ]


Who is online

Users browsing this forum: No registered users and 11 guests


You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot post attachments in this forum

Search for:
Jump to:  
cron
Powered by phpBB® Forum Software © phpBB Group