|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
206.255.198.169
In Reply to: RE: Jameco 5V 1A wallwart linear PSU posted by AbeCollins on February 15, 2017 at 09:26:04
understands hyperlinks(!), I found that document yesterday before ordering. It even provides a circuit diagram and values for the capacitors.
We'll see how well it works out. I don't mind experimenting with cheap options. :)
Follow Ups:
We'll see how well it works out. I don't mind experimenting with cheap options. :)
Same here. If it doesn't work out it's not like you're suffering financially and need to put it up on Audiogon! ;-) Or they can always be repurposed.
Tell you what if you guys don't like it...I'll buy it from you.Just shoot me your paypal email and I will send you the cash.
It's not going to make the sound stage bigger, create more bass, make the vocals jump out etc. As you know, the best gear gets out of the way, so what you should hear is nothing, that is no bad effects associated with bad power.
Edits: 02/16/17
What are the "bad effects" you are speaking of and how do I identify them? Are they identifiable by listening?
You should be able to elaborate on that better than me, being a reviewer. You come across a lot more hear, including gear designed to mitigate bad effects.
Therefore I ask you to impart your knowledge based on your experience
Let's be straight with each other. You ridicule me if I make comments on how things sound. I'm just trying to figure out how you describe the sound you hear. I'm not trying to be mean to you, I just don't understand how you communicate with other people. Is it all based on particular measurements? John Swenson has commented that it is difficult to predict how an LPS will sound based on measurements.I am working on a review that will involve the Sonore Signature Power Supply, the Uptone Audio UltraCap LPS-1, the SOtM mBPS-d2s, and the Fidelizer Nikola LPS.
The review will revolve around the SOtM sMS-200.
Again, I'm not looking for an argument. I just don't understand what you are saying.
Edits: 02/17/17 02/17/17
"John Swenson has commented that it is difficult to predict how an LPS will sound based on measurements."
John is definitely a reputable source in my book. In addition, un-reputable sources ie average music listeners have also echoed this by their own testimony of perceptions. Usually those people are ridiculed and confronted with measurements as proof of their lunacy.
Fact is, yes John Swenson is right. In terms of listening perceptions, measurements alone can not be used as a reliable predictor.
Not to pick a fight. But you can see how fast measurement specs came into play as a predictor of how the Jameco LPS might sound.
Not to pick a fight. But you can see how fast measurement specs came into play as a predictor of how the Jameco LPS might sound.Just to set the record straight, I have no idea how the Jameco wall wart linear PSU will "sound" on an RPi.
I just wanted to point out that just the fact that it is a "linear PSU" means nothing in terms of it's specs. It's specs are just so-so at best. Nothing special. Some people see "linear power supply" and immediately conclude that they must be awesome. Not saying this is you. In fact, there are switching wall wart PSUs with better specs vs the Jameco linear PSU.
How they SOUND and compare in your application, I have no idea.
Edits: 02/17/17
If you don't use GPIO its going to sound like crap....because as you already pointed out, you will still be using the Internal Regulators.Didn't mean to throw you under the Bus....:)
Edits: 02/17/17
No bus! Please no bus! Nah, I didn't see it that way.What puzzles me though is even if you go thru the GPIO, various components on the RPi still require 1.8V, 2.5V, 3.3V, and 5V. So those voltages must come from somewhere and I would guess that those onboard regulators are still in play. That is, unless you provide all of those voltages separately yourself via GPIO. I haven't looked at the GPIO schematic so I really don't know.
Edits: 02/17/17 02/17/17 02/17/17
You are right, many different voltages at play.
From what I gather, the regulation occurs at the main miniUSB input and voltages are supplied without further regulation. It would not make sense to regulate the incoming and then regulate the regulated again for each step. This would not only add cost but create a redundant step.
Again from reading the manual and other literature about the RPI all regulation is bypass when the GPIO is used. Therefore, there is a Warning to Not Use GPIO unless the PS is regulated otherwise you might damage the unit.
Don't worry about the Bus, I'm sure you have been run over plenty of times and know how to get out of the way by now.
:-)
From what I saw in the schematic the RPi takes in regulated 5VDC to power some components but the others are powered by additional regulators that take the 5VDC and regulate it down to 1.8V, 2.5V, and 3.3V. So there are three onboard regulators.
This is a partial schematic from an older RPi but I suspect the newer ones also require multiple voltages.
Thanks for the schematic, now we can know "For Sure" what other regulators are being used. See below data sheets from the manufacture.
Spoiler Alert: Looks like they are Linear Regulated.~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
The NCP1117 series are low dropout (LDO) positive linear voltage regulators that are capable of providing an output current that is in excess of 1.0 A with a maximum dropout voltage of 1.2 V at 800 mA over temperature. This series contains eight fixed output voltages of 1.5 V, 1.8 V, 2.0 V, 2.5 V, 2.85 V, 3.3 V, 5.0 V, and 12 V that have no minimum load requirement to maintain regulation. Also included is an adjustable output version that can be programmed from 1.25 V to 18.8 V with two external resistors. On chip trimming adjusts the reference/output voltage to within +/- 1.0% accuracy. Internal protection features consist of output current limiting, safe operating area compensation, and thermal shutdown. The NCP1117 series can operate with up to 20 V input. Devices are available in SOT223 and DPAK packages.
Line Regulation for 3.3 V = 0.8mV or 0.0008V
http://www.onsemi.com/PowerSolutions/product.do?id=NCP1117
http://www.onsemi.com/pub_link/Collateral/NCP1117-D.PDF
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
LP29808.1 Application Information
The LP2980-N is a linear voltage regulator operating from 2.1 V to 16 V on the input and regulates voltages
between 2.5 V to 5 V with 0.5% accuracy and 50-mA maximum output current. Efficiency is defined by the ratio
of output voltage to input voltage because the LP2980-N is a linear voltage regulator.Line Regulation for ≤ 16 V = 0.7mV or 0.0007V
http://www.ti.com/lit/ds/symlink/lp2980-n.pdf
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~`
Edits: 02/18/17 02/18/17
No spoiler, I realized they were all three terminal linear regulators per the schematic. But I thought your goal was to bypass all the onboard regulators with something better.
Edits: 02/18/17
how relevant some specs are.Usually if some is good then more is better...we don't consider how much is enough, just more is better.
Same thing for specs, sure I've seen 0.003mV regulators. Even iFi specs rate the voltage as +/-.5V....which is actually much worse than the RPI internal regulator variance. But still, even if you use a 0.0000001mv regulator how much does the ARM processor pollute the whole system?
Pretty much, you can't obsess over one factor and ignore the whole chain. The RPI's regulators are but one factor in a chain of many many variables.
IMO, better to start with the low hanging fruit. Then work you way systematically and logically into other variables that produce the greatest effect on the whole system.
Edits: 02/18/17
That was my goal.
I bypass the switching with something better.
The ones left are linear with 0.8mV and 0.7mv regulation. I can live with that...that is to say my ears are not complaining.
They do have better regulators but I'm not trying to solder something that small...
I'm sure good measurements are not delitourous to the sound.
It depends on what and how one measures. Conventional PC type noise figures mean nothing.
Generally speaking I agree.
You asked me what the "bad effects" were.If you are truly interested I would pose this question to someone of more authority on the matter when it comes to objective, measurable effects, circuity design and electrical properties.
Hopefully you will ask Sonore et.al this same question and post it in your review or here as to educate us on some facts. I understand that some people might try to pick fights, but I'm not, I'd really like to know what they would have to say.
Subjectively speaking, as I am sure you know. Not all people can or will hear the same thing, and some will hear no discernible difference at all...
For me and my ears, bad effects present themselves as harsh sounding high to mid-high frequencies, sound that is not coherent in timber across the frequency spectrum and generally bad sound that makes long listening sessions unbearable.
In summary: Objective answers should come from the highest most credible professional. Subjective answers can be all over the map, due to individual perceptions. And I explained my own personal subjective perception.
Hope that helps - :-)
Edits: 02/17/17
I experimented quite a bit with both SB Touch and the microRendu player. For me, a better PS reduces a sense of "false brightness" that obscures true resolution.
At first, quiet linears sound almost "dark", but I realize that I am now hearing more of the recording and less of the HF noise. I find that to be true with other power related topics such as dedicated lines, aftermarket power cords and conditioners as well.
Yes after cleaning up the power things can seem darker due to less noise and blacker backgrounds. But as you said, after awhile you realize you are actually hearing more.
This is why I tried to start at the outlet and address everything I could [within reason]. Better outlets, Better cords, Regenerated Power, Balanced Power, Isolation, etc. etc.
I've even gone so far as to put the RPI in a Faraday Cage of sorts by the use of double layered Copper Tape inside the metal cases and in the case of my Havana Dac on the inside of the Dac.
Years ago when I lived in Atlanta, an elder in our church worked for Equifax at a large data center there. He asked if I would enjoy a tour and as an IT guy, told him sure. While they used huge Amdahl mainframes at the time, all the incoming data lines were sourced in a small room that was completely lined with copper!
BTW, received the power supply today and it is running the Touch quite nicely. Unit gets warm, but not hot. While it doesn't have the massive amount of filtering (~27000 uF) that the Welborne Labs unit did, first impression is good. Especially given the huge price difference. I think I paid about $250 for the other one.
I guess we are all on the same page.
"For me and my ears, bad effects present themselves as harsh sounding high to mid-high frequencies, sound that is not coherent in timber across the frequency spectrum and generally bad sound that makes long listening sessions unbearable."
"In summary: Objective answers should come from the highest most credible professional. Subjective answers can be all over the map, due to individual perceptions. And I explained my own personal subjective perception."
Thanks for the response.
It looks like the RPi has several of it's own onboard regulators for 3.3V, 1.8V, 2.5V derived from the 5v "bulk power" supplied by the wall wart. But it also uses 5v directly in other areas including the USB interface.The Jameco 5V +/- 0.3V spec isn't that bad for what it is powering (with a preference toward the + side rather than lower). According to the USB spec bus power across VBUS and GND should be a nominal 5V +/- 5%.
The data sheet for the Jameco PSU shows 250mV ripple at 1500mA load which is a greater load than the 1000mA the PSU is rated for so the ripple figure should be lower at lower loads. But 250mV is very high.
RPi Power
Edits: 02/16/17 02/16/17 02/16/17
When I get home I will test it under load UI see what kind of real world ripple it has....
Yes I think I mentioned it in another post that people should bypass the miniUSB power input due to the switching regulator in the RPi.
Using the GPIO is the only way to go if you are going to use a linear regulated power supply.
One of my 'edits' that I attempted in a previous post didn't get posted so I re-edited and reposted.
Are you talking about using a linear power supply on the GPIO for other devices?
No only for the RPI
Hmm, I didn't know you could power it via GPIO.
Yep I only use the GPIO
These are different RPI that I use around the house.
Post a Followup:
FAQ |
Post a Message! |
Forgot Password? |
|
||||||||||||||
|
This post is made possible by the generous support of people like you and our sponsors: