Beware of Exploding Gigabyte PSU's being Dumped by Newegg in Forced Bundles

Joined
May 16, 2007
Messages
628
Ask yourself these if your would:
  1. Is any of their OPP testing by some guidelines, a consistent testing procedure which can be peered reviewed? No
  2. Where is the control standard or samples? Did you see them doing the exact same test and samples for a number 750w Seasonics in similar price range and maybe one other like a number of 750w EVGA power supplies? Would they with their testing methodology have the same result? We don't know
  3. They did show that OPP worked on all samples, where does OPP protection state that the PS has to work afterwards, some PS have protective fuses on the inside which are not repairable by the user, meaning you have to have it professionally repaired and tested to make sure it was not the PS that failed vice a downstream load. Now granted I would like to know if the power supply after an OPP protection event is still good and not destroy my machine on a restart. Still there was no promise after an OPP event the PS is still good, now Master Cooler states their power supplies will protect themselves from an OPP event, if true or not I am not sure.
  4. They said this process was over 5 months in the making, they didn't have time to test their methodology against other power supplies to validate the results? Is that not somewhat disingenuous?
  5. Did any of those power supplies failed during normal operating specified conditions before the tests? Indicating the failure rate they plastered all over the video Newegg feedback chart? Note: people are more likely to leave feedback with a failure, skewing any meaning of those graphs.
All safety features to prevent fires, shocks etc. OCP/OVP/OPP/Short Circuit Protection are based on a single event failure, not multiple events. Meaning these are minimally design to handle one abnormal condition and not two. Gamer Nexus did two abnormal conditions: Running the power supply above normal operating power for a time period followed up to an OPP condition -> The power supply has a rating, max rating 750w. If you start operating it above 750w before testing any of the safety features you are causing 2 failed events together (operation above design and then testing OPP). If they took the power supply from 60% to OPP protection, a single failure event either internal or external of the power supply then that would indicate better if the Power Supply was usable after the protective feature was activated. I do think these are crappy Power Supplies but not from their testing.


I'm not sure any psu manual I've read says the protection only works once, maybe something the average user is unaware of. probably better if it failed gracefully and blew a fuse on a protection event rather than powering on after some time.
 

Ebernanut

[H]ard|Gawd
Joined
Dec 15, 2010
Messages
1,557
I'm not sure any psu manual I've read says the protection only works once, maybe something the average user is unaware of. probably better if it failed gracefully and blew a fuse on a protection event rather than powering on after some time.
I've never seen that, my current PSU manual does mention how to reset it after a safety feature trips if it doesn't start working again automatically(I believe you unplug it and switch it on and off a certain number of times).
 

[Spectre]

[H] Admin
Staff member
Joined
Aug 29, 2004
Messages
18,082
I've never seen that, my current PSU manual does mention how to reset it after a safety feature trips if it doesn't start working again automatically(I believe you unplug it and switch it on and off a certain number of times).
I'm not sure any psu manual I've read says the protection only works once, maybe something the average user is unaware of. probably better if it failed gracefully and blew a fuse on a protection event rather than powering on after some time.

Most manuals don't say anything because the company selling it is not the OEM and they 1) don't know or 2) one guy knows and he is not on the marketing team making the manual. You would be really surprised at basic questions I ask a Brand about their unit that they can not answer.
 

Ebernanut

[H]ard|Gawd
Joined
Dec 15, 2010
Messages
1,557
Most manuals don't say anything because the company selling it is not the OEM and they 1) don't know or 2) one guy knows and he is not on the marketing team making the manual. You would be really surprised at basic questions I ask a Brand about their unit that they can not answer.
It's a Seasonic, I think they know and like I said it gave instructions on how to force the safety features to reset after tripping if they didn't do so automatically.

My point was that I had never seen a warning that a PSU SHOULD die from a safety feature tripping and in fact some have specific instructions to reset them after a trip if they don't automatically work after a safety trip. Now power supplies can die and pushing them too hard is a good way to get one to go if it's close but most of the various safety features specifically state they're there at least partially to protect the unit itself from failure so if it's not doing that most of the time there's an issue. That also matches my observations involving power supplies tripping various protections, they typically just restart though a few times I had to fully drain them before they would work again.
 

Kingrat

n00b
Joined
Sep 28, 2004
Messages
19
I got tired of trying to buy a 30X0 GPU for any kind of sane price, and my PC began to show signs of instability anyway, so I ended up buying an ABS Gladiator prebuilt PC. I was happy with the components used and the case was OK, and it was a reasonable deal, so I went for it. When it came I checked it over to make sure everything was still plugged in, and it booted up without any issues. I saw when I was checking it over that it had a Gigabyte PSU, and didn't think twice about it. I didn't know about the problems that some Gigabyte branded PSU's had. After I watched a couple of YT videos and read some stull about them, I checked my 3 day old PC and there it was, P750-GM in it. I wasn't happy. I didn't want it in my PC, I didn't want it in my apartment. I ended up calling ABS support, and got nowhere, "You have a 1 year warranty" on what seemed like an endless loop, so I decided to just buy another PSU and swap it out, just for peace of mind. I noticed the Superflower/EVGA 750 GA that I was putting in was slightly heavier than the original, and the cables were not nearly as flexible, but other than that, nothing to really notice. So, now I have a PSU sitting and waiting for the (IMHO) eventual recall to happen. As far as Newegg itself goes, I have spent many thousands of dollars there over the last 20 or so years. My last 3 build parts were bought from them, and my last couple of laptops were too. I've only had one problem where something was packed badly, and they gave me a refund and a credit on top of it for my troubles. Every HDD has been in a seperate box inside the main one, and I never had, with the one exception above, an item damaged due to bad packing, etc. They pack stuff a lot better than Amazon does most of the time, a recent purchase from Amazon had an item about 4" X 2" inside a box a small child could hide in, with the only packing material being a small amount of kraft paper balled up in one corner. That's not counting the stuff Amazon tosses into a padded envelope and ships it, hoping it comes intact. Most of the time, it does, but not always. In the future, I will look elsewhere when I buy PC parts and if the price is close, NE will not be getting my business.
 

Johnx64

[H]F Junkie
Joined
Apr 22, 2002
Messages
8,320
I got tired of trying to buy a 30X0 GPU for any kind of sane price, and my PC began to show signs of instability anyway, so I ended up buying an ABS Gladiator prebuilt PC. I was happy with the components used and the case was OK, and it was a reasonable deal, so I went for it. When it came I checked it over to make sure everything was still plugged in, and it booted up without any issues. I saw when I was checking it over that it had a Gigabyte PSU, and didn't think twice about it. I didn't know about the problems that some Gigabyte branded PSU's had. After I watched a couple of YT videos and read some stull about them, I checked my 3 day old PC and there it was, P750-GM in it. I wasn't happy. I didn't want it in my PC, I didn't want it in my apartment. I ended up calling ABS support, and got nowhere, "You have a 1 year warranty" on what seemed like an endless loop, so I decided to just buy another PSU and swap it out, just for peace of mind. I noticed the Superflower/EVGA 750 GA that I was putting in was slightly heavier than the original, and the cables were not nearly as flexible, but other than that, nothing to really notice. So, now I have a PSU sitting and waiting for the (IMHO) eventual recall to happen. As far as Newegg itself goes, I have spent many thousands of dollars there over the last 20 or so years. My last 3 build parts were bought from them, and my last couple of laptops were too. I've only had one problem where something was packed badly, and they gave me a refund and a credit on top of it for my troubles. Every HDD has been in a seperate box inside the main one, and I never had, with the one exception above, an item damaged due to bad packing, etc. They pack stuff a lot better than Amazon does most of the time, a recent purchase from Amazon had an item about 4" X 2" inside a box a small child could hide in, with the only packing material being a small amount of kraft paper balled up in one corner. That's not counting the stuff Amazon tosses into a padded envelope and ships it, hoping it comes intact. Most of the time, it does, but not always. In the future, I will look elsewhere when I buy PC parts and if the price is close, NE will not be getting my business.


It's usually not the business who's model is to save time and money on packing but the low paid workers trying to cut corners to be productive(yes they want to save time and money but not if it means having an item damaged and it being sent back). Get fired from amazon go work at newegg. Get fired from newegg go work at amazon. Rinse and repeat with all retailers it's the same people with the same worth ethics. Sit in the bathroom on your phone then go cut corners to look productive at the consumers expense.
 

[Spectre]

[H] Admin
Staff member
Joined
Aug 29, 2004
Messages
18,082
It's a Seasonic, I think they know and like I said it gave instructions on how to force the safety features to reset after tripping if they didn't do so automatically.

Cool. Post a picture. I have seen a lot of Seasonic manuals and the one I remember has it described incorrectly.

My point was that I had never seen a warning that a PSU SHOULD die from a safety feature tripping and in fact some have specific instructions to reset them after a trip if they don't automatically work after a safety trip. Now power supplies can die and pushing them too hard is a good way to get one to go if it's close but most of the various safety features specifically state they're there at least partially to protect the unit itself from failure so if it's not doing that most of the time there's an issue. That also matches my observations involving power supplies tripping various protections, they typically just restart though a few times I had to fully drain them before they would work again.
Permanent latch off is an acceptable response to certain faults.
 

noko

Supreme [H]ardness
Joined
Apr 14, 2010
Messages
6,644
I'm not sure any psu manual I've read says the protection only works once, maybe something the average user is unaware of. probably better if it failed gracefully and blew a fuse on a protection event rather than powering on after some time.
That is correct, most also don't even mention them or say the PS will survive an abnormal condition because abnormal conditions can be very erratic and not predictable to design for. Now virtually all my power supplies have gone through some event and survived, here in Florida it is mostly brown outs, routinely, sometimes a couple a day followed by a month of no power loses. Bright to low lights -> voltage variation is also common. Now I was foolish in buying a cheap PS, Vampire, it did not shutdown on a short, maybe it was not low enough resistance. I had one of the dim plugs going to a switch which would turn on a second power supply. I did not mount the switch correctly and it was shorting out to the metal base which was grounded, the power supply did not shut down and the cable melted (fire hazard), I caught it because of the smell, if I was not in the room I am certain that cable would have caught on fire. In this case that PS did not protect against a ground enough to melt a dim power cable. About a week later the power supply failed, most likely with a big arc, I wasn't in the room at the time, the power supply power rail on its board basically looked like it exploded.
 

DukenukemX

Supreme [H]ardness
Joined
Jan 30, 2005
Messages
5,875
The fallacy of the real pros working at newspapers and magazines has been put to rest long ago.
No kidding.
No, they didn't. They do not have a test program in place. People love them for it because people like to join groups and complain loudly about things they don't know anything about.
They need a test program to test? Consumers don't know anything about PSU's which is why tests are needed. Admittedly I'm disappointing that there was no discussion about the quality of the components themselves in the Gigabyte failing PSU's. Like what are the mosfets rated for? How robust is the OPP so my overpriced GPU isn't fried? In regards to the OPP on the Gigabyte PSU's, it's a fire hazard waiting to happen and GN proved that more than adequately. You couldn't pay me to use one of those in my PC unless you pay me exactly what my PC is worth in the current market. Did your review catch the failure of OPP on these units?
I would not be on year 15 if people didn't pay attention to my reviews.
Who are you again?
UnlawfulHeartyChanticleer-max-1mb.gif
 

hititnquitit

[H]ard|Gawd
Joined
Apr 16, 2018
Messages
1,384
Yeah, I think someone earlier in this thread got the email. Good on them. At least Newegg is trying to fix it, and I guess Gigabyte is doing something, assuming the problem is limited to those serial numbers.
Right but they only offered him a replacement! They may actually be monitoring forum/social media chatter haha. Being that everyone has been saying the same thing, to just give people refunds.
 

[Spectre]

[H] Admin
Staff member
Joined
Aug 29, 2004
Messages
18,082
They need a test program to test? Consumers don't know anything about PSU's which is why tests are needed. Admittedly I'm disappointing that there was no discussion about the quality of the components themselves in the Gigabyte failing PSU's. Like what are the mosfets rated for? How robust is the OPP so my overpriced GPU isn't fried? In regards to the OPP on the Gigabyte PSU's, it's a fire hazard waiting to happen and GN proved that more than adequately. You couldn't pay me to use one of those in my PC unless you pay me exactly what my PC is worth in the current market. Did your review catch the failure of OPP on these units?

There is no specification for OPP to be able to test OPP properly. I only test things which can be properly tested and documented. If I wanted to, I could kill every single power supply I get by testing things that do not have a specification to follow.

Who are you again?
UnlawfulHeartyChanticleer-max-1mb.gif
If you can't figure that out that is more a you problem than a me problem.
 

noko

Supreme [H]ardness
Joined
Apr 14, 2010
Messages
6,644
How a PSU shouldn't explode.


lol, CYA video. Where is the data for all the other power supplies in that 5 months of waiting period? Samples of 10+ for different models etc.? Similar to the number of Gigabyte one's sampled to ensure testing procedure reflected accurately a weak power supply. Do appreciate someone actually taking the time to test the OPP. Would like other checks from thermal, short etc. tested as well.
 

MaZa

2[H]4U
Joined
Sep 21, 2008
Messages
3,383
Another PSU test, this time a cheap POS Amazon PSU with problems in being truthful about their specs, but unlike Gigabyte it did not explode after OPP protection test. So I guess this automatically makes it a better buy. 🤔

 

GoodBoy

2[H]4U
Joined
Nov 29, 2004
Messages
2,186
Because I can make any unfair test fail any unit I want. That is not a legitimate or ethical way to do my job. It also would not allow me to ever draw a legitimate conclusion or compare units. Some of us have a job to do and to do it correctly.

The answer is that, as I have said before, there is no OPP defintion or testing procedure. So, I can make up whatever I want to fail a unit. We can't test what is not a standard becuase I can manipulate the test any way I want to fail a unit. And that is not ethical even though you guys want some pitchforks to raise.
So your implication is, that GN is making up 'whatever they want' to make the unit fail?

Have you supervised their testing procedure and methodology?

Tested something that has no specification to test. With no specification to test, no test can be proper.


I am not defending Gigabyte, I gave the unit a poor review when I reviewed it. Youtube is not the be all end all of anything. Most of us who know how to do this work still mostly work in print because we aren't trying to generate clickbait youtube videos. Make sure to like, follow, and subscribe for more on how to test powers supplies. Oops. My bad.

Sure, testing can be proper. You've already stated that you do not want to review a feature that has no industry defined testing methodology as you fear getting sued. GN has no such fear, reminds me [H] of old, before the vasectomy.

Clickbait youtube videos? Sour grapes much? It's clear that you are biased against GN because they are a youtube reviewer, and youtube reviews in general now foreshadow written [H] style reviews, and because the written page driven solely by ads it no longer a viable business model.

... I will be taking this into account when reading any of your replies.

... You just don't want to listen to people who know what they are doing but instead want your beliefs to be reaffirmed no matter the truth and you are upset that not everyone will rally to your need for confirmation bias.
Confirmation bias?? You reading minds now? You might try reading the above sentence and apply it to your own responses.
... it is a silly argument for you to try to make when I have seen the backend for 15 years on the numbers. Power supply reviews don't usually generate huge day 1 impressions like a new GPU or CPU. They numbers take time to aggregate. The one good side is there are very few of us who do the work and power supplies stay relevant for longer so a 3 or 4 year old article will still get pulled by someone searching out a new purchase.
You tested power supplies for 15 years, so there's no way any of us can have any relevant opinion eh? And you most certainly know more about testing than anyone else, certainly more than a 'damn youtuber'!

We get it.
No, they didn't. They do not have a test program in place. People love them for it because people like to join groups and complain loudly about things they don't know anything about.
Again, so you know how GN tests power supplies? You are making claims about something you have no way of knowing.
I would not be on year 15 if people didn't pay attention to my reviews.
I'm sure I have read a few over the years.
 

cybereality

[H]F Junkie
Joined
Mar 22, 2008
Messages
8,490
At this point [Spectre] is going to have to come up with some proof. This thread is going around in circles and seems more about disparaging GN then providing any actual evidence of malice.

GN has produced hours worth of video on the topic, including the EVGA video which shows how a PSU is supposed to work. I trust them.
 

[Spectre]

[H] Admin
Staff member
Joined
Aug 29, 2004
Messages
18,082
So your implication is, that GN is making up 'whatever they want' to make the unit fail?

Have you supervised their testing procedure and methodology?

I don't have too. The feature described is not covered by a specification. So, there is no way to properly test it to a standard. You can not even test it to the manufacturers/OEMs design without them giving your the design paramteres which they did not.

Sure, testing can be proper. You've already stated that you do not want to review a feature that has no industry defined testing methodology as you fear getting sued. GN has no such fear, reminds me [H] of old, before the vasectomy.
No, at [H] we never tested anything with power supplies that did not have an industry defined specification. You need to go back and actually look at what we did.

Clickbait youtube videos? Sour grapes much? It's clear that you are biased against GN because they are a youtube reviewer, and youtube reviews in general now foreshadow written [H] style reviews, and because the written page driven solely by ads it no longer a viable business model.

... I will be taking this into account when reading any of your replies.
No, clickbait is used because someone is complaining about something that has no specification or standard. By definition, that is click-bait.

Confirmation bias?? You reading minds now? You might try reading the above sentence and apply it to your own responses.

You tested power supplies for 15 years, so there's no way any of us can have any relevant opinion eh? And you most certainly know more about testing than anyone else, certainly more than a 'damn youtuber'!
No. I still test power supplies professionally.

Yes, I do know more about testing them than someone who makes clickbait journalism videos. They were not even aware that there was no specification for what they complained about. That pretty much seals it.
We get it.

Again, so you know how GN tests power supplies? You are making claims about something you have no way of knowing.
I know how they tested OPP and it was not to any standard that is accepted. That is not a valid test.

I'm sure I have read a few over the years.
Sounds like you haven't....
 
  • Like
Reactions: noko
like this

[Spectre]

[H] Admin
Staff member
Joined
Aug 29, 2004
Messages
18,082
At this point [Spectre] is going to have to come up with some proof. This thread is going around in circles and seems more about disparaging GN then providing any actual evidence of malice.
I already provided the proof earlier. There is no specification for what they claimed to have tested.

GN has produced hours worth of video on the topic, including the EVGA video which shows how a PSU is supposed to work. I trust them.
I can produce crap for hours on video too. Doesn't mean that it is valid. And no, they showed a unit that acted the way they expected. Not the way that is uniformly accepted by a standard. I can do the same thing for any unit. I can make one pass or I can make one fail if I work in the space of things that are not set down in specifications.
 

Ebernanut

[H]ard|Gawd
Joined
Dec 15, 2010
Messages
1,557
I already provided the proof earlier. There is no specification for what they claimed to have tested.
Dying much less blowing up when pushed too hard is pretty much a pass/fail test and you still haven't given any substantiative reasons as to why this should be the expected outcome when tested in this manner, actually I think you danced around the issue of whether this even should be expected which is the only good reason I can think of for disregarding the results entirely out of hand like you have.

I have no doubt that you're knowledgeable on the subject but from my perspective your argument doesn't seem to be based on this knowledge nearly as much as your clear bias against the source. I can't blame you for having some animosity and I don't like the trend towards video reviews myself because they favor flash over substance among other reasons but there was enough substance in that second video to convince me that these units are most likely garbage at best and dangerous at worst, nothing you've posted makes me question that conclusion.
 

DukenukemX

Supreme [H]ardness
Joined
Jan 30, 2005
Messages
5,875
I don't have too. The feature described is not covered by a specification. So, there is no way to properly test it to a standard. You can not even test it to the manufacturers/OEMs design without them giving your the design paramteres which they did not.
You sound bought.
No, at [H] we never tested anything with power supplies that did not have an industry defined specification. You need to go back and actually look at what we did.
Just because a specification wasn't given doesn't mean one isn't there. Imagine if an auto manufacturer didn't give a specification for fuel pressure and as a mechanic you decided not to test it even though you totally could because of this reason. Your excuses are lazy at least and incompetent at most.
No. I still test power supplies professionally.

Yes, I do know more about testing them than someone who makes clickbait journalism videos.
This is highly debatable. In my experience a person who claims to be a professional and then tells me how many years they've been doing this, usually knows nothing. This is where I step in and fix the problem they couldn't fix. Just cause you've been doing something for years doesn't mean you're good at it, and claiming to be a professional isn't an excuse for incompetence. You failed to perform a simply OPP test on this PSU and claim not to because of lack of a specification. The whole point of OPP is to prevent a catastrophic failure and you don't need a specification to test this. You push the PSU until it shuts down and then see if it survives as well as the rest of the PC. How is this hard for you to understand?
They were not even aware that there was no specification for what they complained about. That pretty much seals it.

I know how they tested OPP and it was not to any standard that is accepted. That is not a valid test.
You need to crawl back into the hiding place you came out of. Nobody cares about OPP specification when it clearly failed spectacularly. If you can make any PSU go BOOM then it failed. The specification at that point is irrelevant.
 

jfreund

[H]ard|Gawd
Joined
Sep 3, 2006
Messages
1,271
I didn't watch the Gigabyte testing vid, but I did watch part of the EVGA testing. Now I'm seeing [Spectre]'s point on methodology. "We're going to from 20% to 60% to 100% load at 1 minute increments, then over power by 10% at a time every 10 seconds." Why? What's the point or rationale behind giving 1 minute at in spec voltages, and cranking up to the next step every 10 seconds at out of spec voltages? Documentation may say what OPP is supposed to do, but there is no documentation spelling out HOW to determine what it does.

This doesn't mean these Gigabyte power supplies aren't shit, and I don't see anybody here saying they aren't shit. I still say Gigabyte's and Newegg's handling of the whole situation is shit, from bundling these shitty power supplies with video cards to not recalling both models. I can see the argument, though, that GN's testing isn't the way to prove these power supplies are shit.
 
  • Like
Reactions: noko
like this

DukenukemX

Supreme [H]ardness
Joined
Jan 30, 2005
Messages
5,875
I didn't watch the Gigabyte testing vid, but I did watch part of the EVGA testing. Now I'm seeing [Spectre]'s point on methodology. "We're going to from 20% to 60% to 100% load at 1 minute increments, then over power by 10% at a time every 10 seconds." Why? What's the point or rationale behind giving 1 minute at in spec voltages, and cranking up to the next step every 10 seconds at out of spec voltages? Documentation may say what OPP is supposed to do, but there is no documentation spelling out HOW to determine what it does.
Nothing is going to fall inline to any specification given by the manufacturer when it comes to OPP. Specification for OPP is just an excuse for Gigabyte. Why is 10 seconds intervals used to increase the load by 10%? My guess is that gives a more realistic scenario as to someone turning on their PC and then starting up a game. It also gives gigabyte a fighting chance instead of just going to 140% and waiting for the kaboom. Imagine someone buying a RTX 3090 along with a Intel 11900k and then you decide to overclock everything to the max. Then you could be going past this power supplies rated power and not trigger OPP but we know this PSU will fail before triggering OPP.
I can see the argument, though, that GN's testing isn't the way to prove these power supplies are shit.
Lets assume that GN didn't do it correctly. So what should have GN done instead? If no specification was given then what GN did was find it, which was about 140%. If 140% isn't the specification then why was the OPP set to 140%? If the specification is something like 110% and you have to wait 5 minutes to increase the load then go away. It's not a realistic use case and doesn't explain why it takes to 140% load to finally kick in.
 

D-EJ915

[H]ard|Gawd
Joined
Jan 31, 2003
Messages
1,511
I'm sure manufacturers are really smiling now that the general pop knows they can buy undersized PSUs and run them over spec just fine lol.

The GN testing though, it just reminds me of car audio guys hooking up shitty subs to huge amps and blowing them up for memes at comps. I mean it's cool but what does it really mean if it works fine at its rated condition? Issue with these psu seems to originally have been there was a long run of really bad components but then morphed into over power focus. I've never seen anybody specifically test to see how far they can push a psu and if it works or not after before at least from what I can recall.

It must suck to be a PSU brand since seasonic made a new revision of their PSU just because people were complaining theirs were too limiting with the over-current protection feature.
 
Last edited:
  • Like
Reactions: noko
like this

Ebernanut

[H]ard|Gawd
Joined
Dec 15, 2010
Messages
1,557
I'm sure manufacturers are really smiling now that the general pop knows they can buy undersized PSUs and run them over spec just fine lol.

The GN testing though, it just reminds me of car audio guys hooking up shitty subs to huge amps and blowing them up for memes at comps. I mean it's cool but what does it really mean if it works fine at its rated condition? Issue with these psu seems to originally have been there was a long run of really bad components but then morphed into over power focus. I've never seen anybody specifically test to see how far they can push a psu and if it works or not after before at least from what I can recall.

It must suck to be a PSU brand since seasonic made a new revision of their PSU just because people were complaining theirs were too limiting with the over-current protection feature.
The old JG reviews tested PSUs beyond spec in both wattage and heat. I don't know if they went that high but it wasn't that uncommon in the more extreme tests for them to trip protections, iirc doing that out of spec wasn't considered bad unless it killed the unit. The first thing that happens when you push a unit beyond spec is the efficiency drops and is often one of the factors that goes into a wattage rating, for instance a unit might pass gold efficiency requirements at a lower wattage while also being fine at a higher wattage but only meet silver or bronze requirements(sometimes efficiency falls off a cliff above rated wattage too). Beyond that protections should kick in including protection from frying itself by running at way too high of wattage however it should be noted that is a failsafe and shouldn't relied on to work 100% of the time, it also shouldn't fail 100% of time either though.

Your amp/speaker analogy would apply better to someone increasing the voltage on a CPU until it fries, you could make the argument that it's like throwing 4 ohm load on an amp that's only rated for over 8 ohms and getting the amp to fry but in that case you're tricking the circuitry that limits the power so I don't think that really applies either.
 

D-EJ915

[H]ard|Gawd
Joined
Jan 31, 2003
Messages
1,511
The old JG reviews tested PSUs beyond spec in both wattage and heat. I don't know if they went that high but it wasn't that uncommon in the more extreme tests for them to trip protections, iirc doing that out of spec wasn't considered bad unless it killed the unit. The first thing that happens when you push a unit beyond spec is the efficiency drops and is often one of the factors that goes into a wattage rating, for instance a unit might pass gold efficiency requirements at a lower wattage while also being fine at a higher wattage but only meet silver or bronze requirements(sometimes efficiency falls off a cliff above rated wattage too). Beyond that protections should kick in including protection from frying itself by running at way too high of wattage however it should be noted that is a failsafe and shouldn't relied on to work 100% of the time, it also shouldn't fail 100% of time either though.

Your amp/speaker analogy would apply better to someone increasing the voltage on a CPU until it fries, you could make the argument that it's like throwing 4 ohm load on an amp that's only rated for over 8 ohms and getting the amp to fry but in that case you're tricking the circuitry that limits the power so I don't think that really applies either.
Cool, does reminds me a lot of PSU just barely squeak by on their efficiency ratings lol. My old ABS 1100W was like that. I had issues with that one cutting out so I replaced it with a 1250.
 

notarat

2[H]4U
Joined
Mar 28, 2010
Messages
2,348
People keep debating about testing methodology when Amazon reviews is full of DOA complaints

That's called "Changing the narrative".

The PSUs are more like POS but some will die before admitting it.
 

bigdogchris

Fully [H]
Joined
Feb 19, 2008
Messages
18,441
I don't have too. The feature described is not covered by a specification. So, there is no way to properly test it to a standard. You can not even test it to the manufacturers/OEMs design without them giving your the design paramteres which they did not.

OK, even if there is no industry specification, if they take a group of PSU's of different brands and all of them successfully shut down instead of pop, and only the GA model pops, doesn't that still tell you something about that specific model? They have tested a few other models and so far they have all passed their "flawed" testing, except the GA one.

I am guessing you agree that the PSU's should shut down instead of pop, right?

You just don't agree with how they caused them to fail?

If you do not like them testing something that does not have an industrial specification, do you think that it should not be tested at all?

If it should be tested, how would you do it differently?
 

[Spectre]

[H] Admin
Staff member
Joined
Aug 29, 2004
Messages
18,082
OK, even if there is no industry specification, if they take a group of PSU's of different brands and all of them successfully shut down instead of pop, and only the GA model pops, doesn't that still tell you something about that specific model? They have tested a few other models and so far they have all passed their "flawed" testing, except the GA one.
Well, that is the thing. No. A flawed test can single out a particular product to fail. If I set the test up a different way I could make another other model fail if I wanted to.

I am guessing you agree that the PSU's should shut down instead of pop, right?
Depends on the test and the parameters. It is not that simple as just shutdown always. There are events that are outside of normal design parameters that can not be accounted for that means that when they fail they do so in unexpected ways. But, again, I say this from a perspective of doing this professionally not sensationally.

You just don't agree with how they caused them to fail?

If you do not like them testing something that does not have an industrial specification, do you think that it should not be tested at all?

As I have said before, there is not a way to legitimately and repeatedly test a "feature" that does not have a specification. So, any test you do is inherently tainted. Should you run tainted tests? Ethically? No.

If it should be tested, how would you do it differently?

It can be once there is an industry standard that units are designed to accept.
 

cybereality

[H]F Junkie
Joined
Mar 22, 2008
Messages
8,490
But there is a specification. Someone designed the PSU, and the OPP is supposed to do something, even if it isn't an industry standard.

The specification is just not public, and may differ from brands, but it is there. In any case, the GN test seems fairly straight-forward and not tainted.
 

[Spectre]

[H] Admin
Staff member
Joined
Aug 29, 2004
Messages
18,082
But there is a specification. Someone designed the PSU, and the OPP is supposed to do something, even if it isn't an industry standard.

The specification is just not public, and may differ from brands, but it is there. In any case, the GN test seems fairly straight-forward and not tainted.

No there is not a specification. Full stop. Just because you have a feature from a vendor does not mean you have a standard. I could make one product line that acts one way and another that actas another way. And, they do it. I see docs all the time that the claimed OPP point is 150% and on other it is 130% but none of them dictate what the actual load is on each rail to get to that point. And, that matters in how the unit operates.
 

noko

Supreme [H]ardness
Joined
Apr 14, 2010
Messages
6,644
At this point [Spectre] is going to have to come up with some proof. This thread is going around in circles and seems more about disparaging GN then providing any actual evidence of malice.

GN has produced hours worth of video on the topic, including the EVGA video which shows how a PSU is supposed to work. I trust them.
I like they are extending testing to include Short Circuit Protection (which I think is very straightforward and very critical for a PSU to protect against), Over Power Protection and so on.

As for the AG850 Gold+ unit passing their OPP, that is a sample of 1, where they lucky? Now I don't remember if all of the Gigabyte samples failed OPP as in their goal for it in ability to operate afterwards, did they also have some lucky ones that passed? In other words if a sample of one Gigabyte 750w power supply was tested and it was the lucky one that passed their OPP testing, would it then be considered a good power supply if it also did reasonably well in everything else? I really don't have too much of a problem if they fail a power supply based upon their given test if consistent, standards explained so others could duplicate the tests on different samples. That means consistent input voltage, frequency, temperature and humidity if comparisons are going to be made. Also same load or fault conditions. This is not much different from a review stand point testing methods and variations employed on video cards.

The issue with power supplies I see from a manufacturer viewpoint is their liability and possible cause/blame for injury or death or loss of property. That becomes serious dealing with electrical shock hazards and fire hazards as well as component ejection from a power supply when it fails. If Gigabyte has 1000 samples submitted to UL laboratories or other testing labortory for testing and they all passed with flying colors so to speak and then a very poorly done review and test methodology is conducted giving a false impression of results leading them to a legal battle. It would not be the sales lost but more the legal liability that can result from a poor test that Gigabyte would have to face. On the flip side, if Gigabyte sold junk to the customers, which would put customers in danger, it is great that a reviewer exposes the bad product. Now if Gigabyte did have UL tests done and then changed the configuration or components that would fail UL testing (capacitors that emit poisonous gases if they overheat or would explode flammable gases and so on) and still keep the UL tag on, that I would say is fraudulent. UL certification only applies to original manufactured product and cannot know if any changes does or does not void the UL listing. GN did cover what they saw as changes to the components on Gigabyte Power Supplies.

As for GN, their explanations are cringy at best, for example ripple giving the ATX spec, that is great but why did they choose the numbers they choose? What tests says 30mv is the absolute good/bad point for something? Plus toting the ATX spec is not relevant or old and they know better, how would they even know other than picking numbers out of thin air. Plus explanation, maybe they will, what ripple can do -> shorten capacitors life, harder on voltage circuits trying to regulate etc. I will give them some BOD as they refine their testing. Then Steve rattling off a bad batch, my perspective from a manufacturing background in a number of companies, that normally is never allowed to happen due to frequent sampling during manufacturing and then warehouse/yard sampling of finished goods following by independent outside of the facility as in samples sent off to the company QC labs. Feedback from customers are immediately acted upon as well. Of course bad products still make it to market but normally that is result of bad QC more than anything else, not continually inspecting the manufacturing process and components. We used sensors/cameras that would identify imperfections constantly during production for example. A multi layer process to ensure the product meet all specifications plus also looked good and not be questioned in the end.
 
Last edited:

noko

Supreme [H]ardness
Joined
Apr 14, 2010
Messages
6,644
No there is not a specification. Full stop. Just because you have a feature from a vendor does not mean you have a standard. I could make one product line that acts one way and another that actas another way. And, they do it. I see docs all the time that the claimed OPP point is 150% and on other it is 130% but none of them dictate what the actual load is on each rail to get to that point. And, that matters in how the unit operates.
There are safety standards for PC power supplies that have to be meant. Short circuit protection, temperature (max allowed), over current and many others. Testing will vary on design, for example a current limiting circuit may not need a over current protection mechanism to shut it down but may incorporate that as a backup on another design. The new standard now being used in the US is IEC 62368-1 for PCs. I can see some power supplies OPP could just be a fuse that burns out and not a switching circuit. Unfortunately IEC 62368-1 cost $350 for a pdf file, now I am a super geek nerd when it comes to these types of manuals on standards and regulations (scary to be honest, abnormally so :D). Maybe I will hit a library to take a gander if the library has it.

https://www.cui.com/catalog/resource/power-supply-safety-standards-agencies-and-marks
https://www.cui.com/catalog/resourc...-new-safety-standard-for-ict-and-av-equipment
 
Last edited:

cybereality

[H]F Junkie
Joined
Mar 22, 2008
Messages
8,490
Now I don't remember if all of the Gigabyte samples failed OPP as in their goal for it in ability to operate afterwards, did they also have some lucky ones that passed?
To be fair, yes I think GN mentioned they tested about 10 units of the GB PSU and not all of them exploded, so I'll give Spectre credit there. I think 3 or 4 of them did explode, not 100%, but also not a great ratio either.
 
Top