jonnyGURU Forums

jonnyGURU Forums (http://www.jonnyguru.com/forums/index.php)
-   PC Power Supply Discussion (http://www.jonnyguru.com/forums/forumdisplay.php?f=3)
-   -   Seasonic Prime 850W Review (http://www.jonnyguru.com/forums/showthread.php?t=13643)

quest for silence 10-02-2016 04:47 AM

Quote:

Originally Posted by Per Hansson (Post 130833)
I'm curious, what kind of PSU are you referring to here?

I meant anything working close to this way: https://www.silabs.com/marcom%20docu...rol.pdf#page=3; so, if I didn't grossly misunderstood the whole thing, with the feedback loops controlled in the digital domain according by a stored program/firmware (and not by an analog network).

Please also note that, even if that's not quite correct in english, I often use quotation marks for emphasis, and more rarely for irony/sarcasm (given that's how we may use them in italian language/mindset): that's how I used them for the words "real" and "digital".


Quote:

Originally Posted by Per Hansson (Post 130833)
In my mind a real digital PSU is one where the secondary side lacks electrolytic capacitors.

I didn't get you here, sorry: even if OT, providing you have time and patience, can you explain me why there must be no ecaps to call such a power supply as "digital"?

Per Hansson 10-02-2016 05:22 AM

Quote:

Originally Posted by quest for silence (Post 130836)
I didn't get you here, sorry: even if OT, providing you have time and patience, can you explain me why there must be no ecaps to call such a power supply as "digital"?

Capacitors are only required to fill in the "deadtime" in the control loop of a power system.
In the example of converting AC > DC you have no choice; the voltage really is zero when the sine wave crosses the middle point and thus the PSU must run from the bulk (electrolytic) capacitors at this stage.

But when it comes to the secondary, be that on a mainboard or a SMPSU the only reason electrolytic caps are required is because the switching frequency is so low.
I.e. they perform the same job here as the bulk filtering caps do, they fill in the voids caused by the control loop clipping the output to achieve regulation.

Think of it another way: You could also increase the frequency of the AC sine wave to get rid of the capacitors here too.
But of course designing generators that run at high speeds is a tricky job, moreso a hundred years or so ago when the system we have today was built!
And of course everything that assumes a 50/60Hz system would fail quite miserably if we increased the speed to say 2MHz :D

ITelektro 10-02-2016 07:59 AM

Quote:

Originally Posted by Per Hansson (Post 130838)
But when it comes to the secondary, be that on a mainboard or a SMPSU the only reason electrolytic caps are required is because the switching frequency is so low.
I.e. they perform the same job here as the bulk filtering caps do, they fill in the voids caused by the control loop clipping the output to achieve regulation.

This is possible if the frequency is infinite. Such a concept would be an ideal power supply. But if the frequency is infinite, then there remains the problem feedback that must also be infinite. If it is not then at the case of dynamic loading power supply, will occur certain deviations depending on the frequency of this dynamic loading. Manifested the same as ripple.

Can you explain why this is applicable only in case of digital control rather than analogue,What is the benefit of digital control in this case?
Another problem that I see in this concept of ripple itself. From what I had seen at the ATX power supply, ripple most important is the one at frequency of the mains. Switching ripple is typically much lower. And if we had a perfect secondary, there is a problem with PSRR (power supply rejection ratio).
What is the benefit at the case of digital control feedback?

The_Assimilator 10-04-2016 06:45 AM

Quote:

Originally Posted by quest for silence (Post 130689)
And because no other Prime unit tested by any major reviewer (I mean the 750 tested by Wolf here, the 650 tested by Aris @ THG, the 750 tested by Spectre @ [H]) performed "that well" (extremely well but not as well as Wolf's 850 sample).

[H] just reviewed the Prime 850W and got voltage regulation and ripple results that are similar to the amazing numbers Wolfy noted.

So unless Seasonic is cherry-picking PSUs to send to reviewers, the statement "I think for every hundred or so Prime 850s, you might find one or two performing like this at most" may be incorrect. Which, quite frankly, would be incredible.

quest for silence 10-04-2016 07:21 AM

Quote:

Originally Posted by The_Assimilator (Post 130942)
results that are similar to the amazing numbers Wolfy noted.

:eek: :crazy: :wtf:


Quote:

Originally Posted by The_Assimilator (Post 130942)
So unless Seasonic is cherry-picking PSUs to send to reviewers

Unlikely, moreover often [H] buys from retail: Spectre, did you buy your reviewed specimen?


Quote:

Originally Posted by The_Assimilator (Post 130942)
Which, quite frankly, would be incredible.

:eek: :crazy: :wtf:

I'm eagerly waiting for Aris' review: three in a rows? :beer:

McSteel 10-04-2016 07:43 AM

The -12V rail could use some work... :p

quest for silence 10-04-2016 07:54 AM

Quote:

Originally Posted by The_Assimilator (Post 130942)
[H] just reviewed the Prime 850W and got voltage regulation and ripple results that are similar to the amazing numbers Wolfy noted.

BTW, I checked twice the figures (just for shit and giggles), and splitting hairs they are not as incredible as Wolfie's cold tests ones (0%, 0%, 0.08%), albeit close (0.3%, 0%, 0.16%), and also close to the AX860i reviewed by Tazz in 2012 (0.2%, 0.3%, 0.11%, if ever comparable).

jonnyGURU 10-04-2016 02:09 PM

Quote:

Originally Posted by quest for silence (Post 130944)
Unlikely, moreover often [H] buys from retail

Not as often as reviewing units sent from the manufacturer.

But I would still like to know if it was a retail sample or a press kit sample.

kimobc 11-19-2016 10:16 PM

I like the new PSU but the top con for this model is the high inruch current over 55A, so I wonder if that can cause problem specially when using with 230v.

Stefan Payne 11-20-2016 04:07 AM

Quote:

Originally Posted by jonnyGURU (Post 130967)
Not as often as reviewing units sent from the manufacturer.

But I would still like to know if it was a retail sample or a press kit sample.

http://www.tweakpc.de/hardware/tests...tanium/s01.php

They tested TWO retail samples and two Press Samples.

Spoiler:
One of the Retail units was BETTER than the press sample.


All times are GMT -4. The time now is 03:36 PM.

Powered by vBulletin® Version 3.8.7
Copyright ©2000 - 2017, vBulletin Solutions, Inc.
Copyright © JonnyGURU.com 2000 -