Discussion Power supply discussion and recommended models thread

Page 26 - Seeking answers? Join the Tom's Hardware community: where nearly two million members share solutions and discuss the latest tech.
Apr 4, 2019
59
18
35
0
If capacitor degradation wasn't a thing, why is the tolerance for capacitance greater after 2000 hours of operation greater compared to initial tolerance? Why is tan delta (dissipation factor) allowed to increase by up to double compared to initial value?
probably because thats how the manufacturer expect the unit to behave over time. more insight i dont personally have over how this increase in tolerance effect the performance of cap.
 
Apr 4, 2019
59
18
35
0
So, they expect it to behave worse over time.
i dont have the insight in capacitor performance to know if that is what actually entails with the greater tolerances.

You just basically agreed that capacitors do degrade.
agreed that that is how the manufacturer expects the unit to behave, nothing extra.

and from what i can gather from someone i just doublechecked with. its values increasing doesnt entail the degredation of the cap, but more the death of the cap.

edit: and upon more info i got. the increase in tolerance isnt a big issue due ripple&noise being set low
 
Last edited:

TJ Hooker

Glorious
Herald
agreed that that is how the manufacturer expects the unit to behave, nothing extra.

and from what i can gather from someone i just doublechecked with. its values increasing doesnt entail the degredation of the cap, but more the death of the cap.

edit: and upon more info i got. the increase in tolerance isnt a big issue due ripple&noise being set low
Please define exactly what constitutes "degradation". Because if decreasing capacitance and increasing ESR aren't degradation, what is?

How are ripple and noise "set low"?

By all means, let us know who your secret source is, or encourage them to chime in directly.
 
Apr 4, 2019
59
18
35
0
By all means, let us know who your secret source is, or encourage them to chime in directly.
it would be the PSU tierlist community over at the LTT forum.

Please define exactly what constitutes "degradation". Because if decreasing capacitance and increasing ESR aren't degradation, what is?
from what i understand is that the increase in tolerances and values of a cap signals the end of the cap.

How are ripple and noise "set low"?
to quote the question and answer

Me: "do the increased tolerances actually entail degredation?"


other: "If the values of the Cap increases, the Cap is dying and on its way out
so its not that relevant
and also one of the reasons why the Ripple&Noise is set so low as well"
 

TJ Hooker

Glorious
Herald
Ah, so it's just some random chuckleheads on the LTT forums. Got it.

Here's an article from Avnet, a large components distributor.

https://www.avnet.com/wps/portal/abacus/resources/engineers-insight/article/understanding-esr-in-electrolytic-capacitors/

ESR is frequency-dependent, temperature-dependent, and changes as components age.
[...]
The electrolyte’s temperature-dependent conductivity is a major contributor to ESR. Ohmic losses produced by the aluminium oxide layer are frequency-dependent, reducing as frequency increases. Fluid electrolyte is lost over time by vaporisation and diffusion, causing a gradual reduction in the amount of conducting material, reducing the contact area, increasing the ESR and reducing capacitance.
 
Apr 4, 2019
59
18
35
0
Ah, so it's just some random chuckleheads on the LTT forums. Got it.

no, its as i said earlier. people behind the PSU tierlist


neat, still inline with what my source gave me and i quoted over here. as that is a sign of a dying cap. not to mention the very long lifespan of caps.


and as i said, more of an issue due to wrong input on mobo caps. either from group reg or other output issues.
 

TJ Hooker

Glorious
Herald
no, its as i said earlier. people behind the PSU tierlist
Do these people have any real experience or credentials that make them trustworthy sources, other than compiling a questionable tier list?

neat, still inline with what my source gave me and i quoted over here. as that is a sign of a dying cap. not to mention the very long lifespan of caps.

and as i said, more of an issue due to wrong input on mobo caps. either from group reg or other output issues.
No, that's not inline with what you said. It's not a case of where the values suddenly change at the end of the caps lifespan, just before it dies. As the source says, it's a gradual process over time. And after a sufficiently long time, after the values have drifted beyond some threshold, it is considered to have reached the end of its usable lifespan (assuming there was no catastrophic failure first).

And no, it's not just a matter of input voltage being high. Here's another quote from the article:
This “drying out” process is temperature dependent and accelerates in components used at higher temperatures or subjected to higher ripple currents, which dissipate more heat as part of their circuit function.
Higher ripple currents are correlated with higher load.
 
Last edited:
Apr 4, 2019
59
18
35
0
Do these people have any real experience or credentials that make them trustworthy sources, other than compiling a questionable tier list?
well other than creating a tierlist that serves is function perfectly and is updated according to findings by reviewers. they are also the people who "live and breathe" PSUs when it comes to tech.

And no, it's not just a matter of input voltage being high. Here's another quote from the article:
which falls inline with what i said with output issues from the PSU. which includes ripple and noise. which is what usualy ends capacitors early on motherboards.

No, that's not inline with what you said. It's not a case of where the values suddenly change at the end of the caps lifespan, just before it dies. As the source says, it's a gradual process over time. And after a sufficiently long time, after the values have drifted beyond some threshold, it is considered to have reached its usable lifespan (assuming there was no catastrophic failure first).

im sorry, ive sadly not got much more say regarding caps. i do not have insight in how the death of cap goes. possibly get back to it when Jonny comes along.

probably fairly annoyed at me i assume.
 
Apr 4, 2019
59
18
35
0
I made a comment about lack of otp being bad and an expert in the field clarified most PSUs have otp via a thermister if otp is not supported by the protections IC itself.
most units today have OTP, though there are allways exceptions. and i believe earlier in this thread there was a mention of it not being all that important, hence me pointing it out.

some statements in this thread can be interesting to say the least with some context.
 
Apr 4, 2019
59
18
35
0
Also the single rail units being limited to a tier and below is stupid. The only real benefit from extra rails it reliability.
that and its also a protection, hence a requirement for the higher placings. mostly a concern with higher wattage units.

Focus Gold units ripple is not that bad, and the units are superior to some higher ranked units.
the Focus fx units simply just dont meet the requirements for tier C and higher that have been set. its been brought up before, but per current ratings, its not going to be moved.

I disagree with the placement of many units
ive got some issues that ive brought up with the team, but each have a reasoning for their placements. if info or errors are discovered it gets corrected as soon as possible.

Its a good guideline, but not the best.
for someone seeking quick recommendation, its hard to get something better than a tierlist. its not perfect, but its not expected to be. its better now than it was at least, with it regularly being updated and input being taken in.

best is allways to start up your own thread and get recommendations that way, or offcourse independent research. something the tierlist does help with now as it sites sources for most if not all units (that have sources)
 

WildCard999

Titan
Herald
for someone seeking quick recommendation, its hard to get something better than a tierlist. its not perfect, but its not expected to be. its better now than it was at least, with it regularly being updated and input being taken in.
Not really, we get questions all the time about what someone will need for a PSU recommendation and usually they get a answer pretty fast as long as they post the current specs/budget. And well, if they don't know the current specs or can't figure them out then I may be hesitant to have them install a PSU.
 
Apr 4, 2019
59
18
35
0
is ranked higher than the Seasonic M12 units known to be very very reliable.
requirement set by the list. i believe its due to group regulation issues that make it run out of spec during crossloads. something that qualifies it for Tier D.

every tier has its requirements. its set to leave personal opinion out of the list.

the changes doen from V4 list form the V3 have been discussed and the previous upkeep just didnt get around to implementing the changes.

if you got some sources i might be able to see if the cx green to be changed if its info that hasnt been noiticed. any input is allways welcome.
 
Sure the g3 is slightly worse than the g2, however the g3 is way better than most PSUs.

I think the people that have created these lists are too caught up in the numbers, rather than reliability.

I don't care if a unit isn't the best performance, but if it is reliable there is no reason to mark it as worse than a unit that is known to fail.
 
Reactions: WildCard999
Apr 4, 2019
59
18
35
0
Stuff like that and seeing the EVGA G3 labeled as a low quality unit makes me think some of these people aren't really doing the research.
if you read the requirements for Tier D. all it says is that it has potencial to be dangerous in certain load scenarios.

sources are posted aswell.

Not really, we get questions all the time about what someone will need for a PSU recommendation and usually they get a answer pretty fast as long as they post the current specs/budget. And well, if they don't know the current specs or can't figure them out then I may be hesitant to have them install a PSU.
so does the LTT forum, but it doesnt stop the list from being a helpful tool for anyone new to PSUs. especially for people who arent in the common countries as some units simply arent avavible. making it a whole lot easier check with the PSU tierlist.
 
Apr 4, 2019
59
18
35
0
I think the people that have created these lists are too caught up in the numbers, rather than reliability.

I don't care if a unit isn't the best performance, but if it is reliable there is no reason to mark it as worse than a unit that is known to fail.

its based on reviews which shows it has issues. which is what the list is mostly concerned about. if certain stuff is noticed, its noted down by the members of the list.

Sure the g3 is slightly worse than the g2, however the g3 is way better than most PSUs.
its having good electrical performance sadly doesnt stop it from having issues, hence it being dropped to tier D. this has also been adressed multiple time.
 
if you read the requirements for Tier D. all it says is that it has potencial to be dangerous in certain load scenarios.

sources are posted aswell.
A G5 is not very good, but way, way superior to a 2012 CX430w. The updated list says otherwise.

I don't get the G3 called "potentially dangerous." All of the sources linked below the list are very positive to the G3.

Your sources link a 9.8/10 rating and recommendation from jonnyguru. You cited a Toms Hardware review that called the G3 a bargain and high performance.

Are the people making the list even reading their own sources?
You would not think these "experts" would conclude the PSU is dangerous after reading multiple positive in depth reviews and linking them as a source.
 
Last edited:
Apr 4, 2019
59
18
35
0
A Corsair CX green, a series known to fail and damage hardware, is ranked higher than the Seasonic M12 units known to be very very reliable.
units below 750w are going to be doublechecked if they meet requirement of tier C. there may be a change untill the next revision. comes out.
 
Apr 4, 2019
59
18
35
0
I dont get the G3 called "potentially dangerous." All of the sources linked are very positive to the G3.
issue is the OPP from what i recall, it being set too high and the PSU rippling out of spec on overload. and OTP which then decided its placement.

as per the older PSU tier list upkeeper. not the current one. The previous upkeeper being LienusLateTips. nolonger active due to "reasons"

id link the forum post regarding it, but i dont think i can link content from other forums
 

ASK THE COMMUNITY

TRENDING THREADS