Gadgetory

okay let's see if the great wall power supply blows up let's build the wall to prevent these power supplies from being sold well take a Mongol horde to make this great wall crumble looks like jion wants to shift to a new studio and so blow up their old one that the efficiency will be under 80% and ripple above 120 these are the types of comments that surfaced when we posted teasers of our power supply testing for the great wall power supply used in the wall Mart system and to be fair everything about it does look like a cheap power supply but great wall actually is a supplier and makes PS use for Corsair for instance as discussed in our Walmart case review it's uncommon to find great wall power supplies unbranded in the western market and this one didn't even have the maximum 12-volt capabilities listed so the unit did attract criticism from the community what we're here to do is test whether or not it's deserving of all of that criticism using our new power supply testing setup to benchmark efficiency ripple and over current protections before that this video is brought to you by the gigabyte Z 390 ARS master motherboard which comes equipped with one of the more powerful Z 390 V RMS for heavier overclocks on the new 9th gen intel cpus the ARS master is also one of the few motherboards with a really heatsink this generation featuring a mix of high surface area fins and looks oriented hover blocks oh and it's also got updated RGV illumination learn more at the link below starting with the very basics first just like in liquid coolers there are typically only a couple of suppliers that actually make the core of the product before the big brands that you know do something a bit custom with them so liquid coolers are a great example because Corsair NZXT EVGA plenty of other people they all ultimately kind of sell sort of the same product in that the base of it is likely a Sutekh and there are some exceptions to that of course your side that are new but typically it's someone like haze attack or is ASA TAC and they might have the same Gen 5 pompe Gen 6 pump they have slightly different tube lengths they have different fans that are really important they have different fan profiles but the pump itself the radiator and the tube material it's all gonna be the same the end of the day there is a supplier behind things and great wall is one of those is something mentioned in our Walmart case review where we were talking about suppliers like Jones Beaufort the case market just to give a list of a couple of the more popular power supplies that Great Wall does make Great Wall has made things like the corsair CX 500 the TX 850 they've done the rio toro enigma 850 OC z ZX 1000 and some of these are a bit better than others the CX series not really the best the enigmas from rio toro ocz is most mostly dead at this point but then there are other ones like the Corsair SF 600 which actually has really good voltage ripple and the CX 450 is a cheaper power supply it's been made by both great wall and channel while technology the cs5 50 M green label power supply that's great wall great wall makes a lot of these and to course has credit of course they do things to improve them as well as they might aid in some of the engineering but the platform buy a great wall is the heart of the product in those instances and although this one is is not rebranded it is still a great wall product and that list kind of shows that you may have even used one in the past before we get into the testing today power supply testings new for us so we're still learning we are still advancing our methods there's a lot we know we can improve already there's a lot we've already improved after this video goes live but we have enough data here to do some basics and the basics will be efficiency testing at different loads using a load generator that's the sun/moon 8800 we have a Sun Moon 220 for measuring that load and then we also have a couple of other tools like a no scope for measuring ripple and we are using a new setup that's not in for this test today for fans speed and noise testing for power slice that's a really difficult problem to tackle and we've solved it pretty well but it won't be showing up here today because we're building some of the setup for that and instead of going through all of the details of our new power supply testing methodology here what we're gonna do is finish the fan and noise test inside of PSU reviews and then make one video or article talking about how all the testing methodology will work but today the basics of what you need to know is that we're using an SM 220 SM 8800 and OU scope and we'll have some more information the article linked in the description below you want to check out some information on the loads that we used on this great wall power supply and on this EVGA power supply just as a baseline branded competitor made by HEC for this one we're plotting the Great Wall 500 watt nd + power supply in the Walmart PC versus an EVGA 500 watt 80 + model from about 2014 EVGA a power supply is made by HEC while great wall as the supplier of its own unit we'll start with the great wall power supply first version 1.4 - of the ATX desktop PSU design guide will move toward standardizing a 2 percent load number but most power supplies still fail two percent efficiency testing as the guidelines are relatively new at 2 percent load great wall operates at 55 point two percent efficiency with an input of 18 watts and an output of about 10 watts loading the three point three and five volt rails to zero point three amps and five BSB to one point five this isn't a high end power supply so as load tapers in either direction efficiency Falls hard we can give it a pass at two percent load as this is still a new enough specification that it's not being applied yet to these types of power supplies but it's something to consider going forward at 10 percent load the great wall unit is at seventy seven point seven percent efficiency which isn't great but 80-plus doesn't require any ten percent efficiency up until the titanium ratings the recent titanium radians that were added it's twenty percent that starts to matter where the great wall power supply is at eighty three point one percent efficiency that clears the requirement for an eighty plus white label which is eighty percent efficiency as the floor 30 percent load has us at eighty four point six percent efficiency while forty percent load has us at eighty five percent efficiency this is the peak of the curve as 50 percent stays roughly the same with eighty four point nine percent efficiency eighty plus certification requires eighty percent efficiency at 50 percent load which this clears it misses 80 plus bronze but it's close enough to error margins that we can give it a pass if it clears it 100 percent load validation plus we test maybe slightly differently than plug load would anyway at sixty percent we dipped to eighty four point four percent efficiency than eighty three point four percent at seventy percent eighty two percent at 80 percent and eventually hit eighty point three percent efficiency at 100 percent load despite there being no like report for validating this power supply at the time of writing and despite its lack of any official 80-plus certification and branding which does cost money it looks like this unit is at least 80 plus white efficient or the base label and is bordering on bronze efficiency although the plugs load group will test differently than we do let's look at the EVGA power supply next this one is 56% efficient at two percent load 76 percent efficient at 10 percent load and is exactly tied with the great wall power supply at 20 percent load or 80 3.1 percent efficiency with the efficiency trailing off after that the EVGA power supply hits 84 percent efficiency at 50 percent load before dipping towards 80 percent at 100 percent load we also included overload measurements where the great wall power supply does 77.7% efficiency when loaded to 615 Watts output or 792 Watts input which is really pretty damn good all things considered we'll talk over CT and current figures in a minute the EVGA power supply at the same 615 watt load with the same current distribution so 75 point 4 percent efficiency at 123 percent load again not bad both of these power supplies can sustain more than they advertised which is always a great thing to see this wasn't measured on the great wall unit after a 40 minute load period to allow the power supply to reach steady state temperatures internally but for the EVGA unit it did eventually reach a shutdown point and we would have had to step down the current on the 12 volt rail in order to sustain operation put these percentage numbers into perspective here's the input versus output power applauded for just the great wall power supply this chart really helps visualize the efficiency curve with real numbers this is power in versus DC out on the power supply DC out would be what your PC receives and likely requests while the input is what is required in the AC to DC conversion process in terms of percentages this power supply is its most efficient at 40% to 50% load and trails off toward the end of testing 2% load is the least efficient of all but because we're talking percentages versus absolutes in the previous chart was this one that isn't necessarily obvious on this kind of chart the previous chart would be better for that but this one helps to visualize the actual power input and output numbers in watts voltage ripple is one of the most important metrics for determining the quality of power delivered to a system a PSU can be efficient while still having bad ripple characteristics which would in system instability or reduce overclocking Headroom we are using a regal 10:54 scope connected to the SMA 8800 via BNC for this testing veteran VPP for ripple we've measured average voltage ripple on the 12 volt rails at about 39 millivolts under 20 percent load 43 millivolts under 50 percent load and 59 millivolts under 100% load before peaking at 89 millivolts under 120% load I expect calls for under 120 millivolts of rip hole for example with the absolute best $500 power supplies nearing 10 to 15 millivolts of ripple just for example with 43 millivolts at 50 percent and 59 millivolts at 100% the great wall power supply is simply fine it isn't amazing it isn't offensive making for acceptable overall ripple on a dirt cheap power supply on the 12 volt line the other rails are less important the worst of which is 5 vs B towards the back end of testing 12 volt ripple is better than one might anticipate of going solely based on the amount of comments calling for the explosion of this power supply again it's not a great part but it's not going to cause instability issues as a result of ripple at least on this kind of system that it was installed with and in future power spot reviews this is really just a preview of what we're working on we'll be including some charts of the actual voltage ripple measurements so you can see how clean the signal is and then we might be looking at transient response and things like that along with fan and noise testing in the future so last one here OCP even though we said in our Walmart case review that great wall is a pretty common supplier most of the concerns we're still talking about the power supply quote blowing up or catching on fire as a result of over drawing current we tested OCP on this unit great wall says that this power supply should be able to take at least 40 amps on the 12 volt rail so we set the 3.3 and 5 volt rails at 3.4 amps PSB to 1.5 then push the 12 volt rail until it hit 40 amps the power supply ran inefficiently of course but it didn't shut down or overheat this was for a minimum of 30 to 40 minutes to reach steady state and we switched our SM 8800 into amps mode and increased the current until we hit a breaking point for the great while 500 watt unit that breaking point was approximately 56 amps demanding 672 Watts down three 12-volt cables alone or if 707 watts total against all of our settings VRMs fell to 111 point nine five volts efficiency was seventy seven point seven percent and ripple was about 92 millivolts on the ETF's header and in the 80s for EPS and 12-volt headers but overall it did survive and that's 123 percent load so really it's I mean it's not supposed to run that high anyway the power supply ended up tripping OCP and safely shutting itself down when pushed this hard at 50 amps or 123 percent total power a continued operation and did not trip any protection so it was not at any major risk of catastrophic failure because it was capable of handling it once we went over 50 amps that's when it started just tripping OCP almost immediately concluding this one the Great Wall 500 watt power supply which is it's got a name of GW 6000 in the model and it does technically say 80 plus on it although there's no sticker it's just part of the model name so that's what we're working with here and it's actually okay it's inoffensive it is not an exciting power supply it is not a high quality power supply but it's not a really tremendously low quality power supply either it's just fine the voltage ripple is acceptable especially for a 500 watt ultra budget power supply and we've spoken with some sources in the industry and these are incredibly cheap because Walmart could have put a branded power supply in their system like a lot of the system builders do for example Iowa power typically uses the sort of branded power supplies that you were all familiar with they could have done that or they could have had more margin and they chose more margin and this one it's it's probably I mean it's it's probably better than some of the other low end power supplies on the market that are branded by companies that you all know because it's just it's fine so not really the most exciting it didn't blow up if that's what you were waiting for I'm sorry it did not do that but it also didn't post the best ripple numbers in the world didn't post the best efficiency numbers in the world it's simply okay so Walmart for their part or their their supplier om as part their system builders part the power supply is not Marvel the rest of the computer the part selection like the motherboard for example really weird USB type-c selection the case is terrible but the power supply is actually okay so this is the one part of that build unless you count silicon like the CPU this is the one part of that bill that's actually somewhat redeeming so we would still absolutely not recommend buying one of the over powered pcs especially because if you look at our case review maybe we can put one of those charts on the screen the Delta for the internal component temperature by removing the front panel of that case that glass panel it's it's was it 40 degrees it's over 30 degrees whatever it was it's a huge Delta and that case is just suffocating everything in it so that's bad and then the price was really bad and they did drop it a little bit after all the videos and coverage but it's still just not a good deal services is not good in the experience we've had other youtubers have posted content like Kyle bit wit and Linus tech tips and they've had issues with cables being disconnected on arrival so still wouldn't recommend it but strictly speaking of power supplies it's actually okay now you could complain that the cables aren't pretty you could complain the box isn't pretty the cables you have a better point because these are visible inside the case if you're paying over $2000 for the high-end system and you've got this kind of mess inside the case yeah you deserve a better power supply for that kind of money absolutely if only for the looks but functionally it's acceptable so that's it for this one thank you for watching power supply stuffs gonna get really interesting coming up so make sure you subscribe to catch more of that you can go to store documents access net to pick up one of our mod mats or shirts and patreon.com slash gamers Nexus tops out directly where we'll also be posting some extra behind-the-scenes stuff on power supply testing as we further iterate it be posted one maybe two or three weeks ago already so and then one final note we do have a lot of cool things we're doing with this type of testing that didn't make it into this video because it just wasn't ready yet but working on it so definitely check back soon because we should have some oh scope shots will have fan speed versus power response will have noise testing thermal testing all that stuff so thank you for watching I'll see you all next time