Is the beQuiet System Power 2 good?

[NunoLava1998]

I want to change my system from a bulky Micro-ATX one to a SFF thin one (Silverstone ML07B).

SFX power supplies are really hard to find for cheap, however I found a System Power 2 400W on amazon.de for only 59,99€.

From what I've seen on the internet and specifications, this seems to be a really good power supply. I do want to double check if this is good though.

[NunoLava1998]

Actually it's not System Power 2, it's SFX Power 2 (400W, BN227).

I'm going to post some specifications here so you don't have to look it up:

• 70W on 3.3V, 120W on 5V, 396W on 12V. Rated for 50ºC.
• 80+ Bronze, 89%
• 1x 24-pin, 2x SATA, 1x legacy HDD power, 1x FDD, 1x 4 pin CPU, 1x PCIe 6+2 pin
• Active PFC, OCP, OVP, UVP, SCP, OPP
• Noise 20dBa until 60% load (240W), 30dBa at 350W
• 80mm fan
• 100-240VAC, 50-60Hz

In my opinion it looks like a good power supply and @LukeSavenije should probably add it to the tier list.

[OrionFOTL]

There are no reviews for SFX Power 2, so it's unknown whether it's a good power supply or not.

 

9 minutes ago, NunoLava1998 said:

In my opinion it looks like a good power supply and @LukeSavenije should probably add it to the tier list. 

Can't add something if we don't know if it's good or not.

[NunoLava1998]

 

1 hour ago, OrionFOTL said:

There are no reviews for SFX Power 2, so it's unknown whether it's a good power supply or not.

 

Can't add something if we don't know if it's good or not.

Eteknix did a review on it.

https://www.eteknix.com/quiet-sfx-power-2-300w-power-supply-review/

It looks like a pretty good power supply.

[OrionFOTL]

2 hours ago, NunoLava1998 said:

 

Eteknix did a review on it.

https://www.eteknix.com/quiet-sfx-power-2-300w-power-supply-review/

It looks like a pretty good power supply.

Eteknix actually does one of the worst reviews on the internet, they are riddled with issues. It's only just one step above those hilarious reviews where they just put it in a normal PC and read voltages from BIOS.

 

On their "testing methodology" page, they incorrectly describe +12V ripple limits as 150mV:

Quote

We use the same time scale and horizontal millivolt scale on our oscilloscope for all ripple tests, that is a 20ms T/DIV (horizontal) and a 0.02 V/DIV (vertical) meaning the scale is from -80mV to +80mV, ATX spec dictates that the 12v rail must fall within 150mv of ripple and the 3.3/5 within 50mv so that scale allows us to include both 150 and 50mV peaks. (Some older PSU reviews use different scales which were later ditched as the visual representation they give is inadequate, in these reviews written measurements are provided only).

 

Then the voltage regulation test is a disaster:

Quote

To test voltage regulation we load the power supply to five different load scenarios that give an equal spread of load across every single rail. So that means 20% on all rails, 40% on all rails and so on. We then calculate the average deviance of each rail from its expected voltage.

 

First of all, we have no idea what is the load on each rail, and that is absolutely crucial to judge if the load is distributed correctly or if there's too much or too little load on minor rails.

They give "an equal spread of load across every single rail". What exactly does that mean? If they want to load it to 120W, do they "equally spread it across" and put 24W on +12V, 24W on +5V, 24W on  +3.3V, 24W on +5VSB and 24W on -12V? Obviously not, because that wouldn't make sense.

 

In next sentence they try to clarify: "So that means 20% on all rails, 40% on all rails and so on", but it still doesn't make sense. So if they want to load the PSU to 60% capacity, do they load the 12V rail to 60% of *its* capacity, 5V rail to 60% of its capacity, and so on? That wouldn't make sense either, because if you take 60% of SFX Power 2's 12V capacity, 60% of 5V, and 60% of 3.3V, (even ignoring the other two rails), that sums up to 281W, while they got 180W. So what exactly is the load on all the rails? How many amps?

 

Now the biggest and most hilarious fuckup is the very way they calculate voltage regulation. Just like the table describes, they literally take all individual deviances of each load for a given rail, and average it! So for 12V1, they do (0.13+0.13+0.11+0.09+0.07)/5 = 0.106 / 12 * 100% = 0.88%. That's a COMPLETELY wrong way to do that! The point of testing voltage regulation isn't to find out how much it deviates on average, but to find out to what degree it deviates!

 

For example, imagine you test a PSU and you get the following results: 12.60, 12.30, 12.00, 11.70, 11.40. Such results are absolutely horrible and they literally hang right on the ATX limits (+12.60 max, 11.40 min). The deviance at the first test is +5%, at last test it's -5%, and the load regulation is 10%. But according to Eteknix, this PSU's voltage regulation score would be (0.6+0.3+0-0.3-0.6)/5 = 0%!!! Perfect voltage regulation, better than AX1600i!

 

 

I used to have an even more detailed post about this on Jonnyguru forums, but unfortunately it got lost after the backend upgrade. Eteknix tests are worthless and we still don't know anything about this PSU, except the internal photos which are the only plus of that site.

[NunoLava1998]

42 minutes ago, OrionFOTL said:

im sorry if this pinged 3x or some shit im just trying to fix the wall sorry

From the beQuiet! specs (and whatever the shit Eteknix provides) though, it still does look like a pretty good power supply. I'd place it at Tier A on the PSU tier list, maybe B.

@jonnyGURUWhat do you think about the PSU? Eteknix is garbage but they did provide internals, which are crucial:

https://www.eteknix.com/quiet-sfx-power-2-300w-power-supply-review/3/

 

[Guest]

9 hours ago, NunoLava1998 said:

SFX power supplies are really hard to find for cheap

Because doing that kind of density without screwing up is hard to do on the cheap.

 

2 hours ago, NunoLava1998 said:

 

@jonnyGURUWhat do you think about the PSU?

[NunoLava1998]

9 minutes ago, jonnyGURU said:

Because doing that kind of density without screwing up is hard to do on the cheap.

 

lol

Does look like a good power supply though (except for the non-modularity and the noise at 60%+ load) and i've already ordered it. 70€. 

[LukeSavenije]

i'll add it when i got more info... that's what has been holding me back for a while

[NunoLava1998]

11 hours ago, LukeSavenije said:

i'll add it when i got more info... that's what has been holding me back for a while

Eteknix did a review on it, and while their way of doing reviews definitely isn't good, they do have internal pictures and they still did measure i.e. the PSU ripple, etc.

Plus beQuiet! obviously has a page with some info on it too

[LukeSavenije]

okay... I've taken a look into it with Stefan, it'll likely be a tier c as it is close to the pure power 11 bronze platform, group regulated. this is not final yet, but you have an idea of where it'll be.

[NunoLava1998]

Just now, LukeSavenije said:

okay... I've taken a look into it with Stefan, it'll likely be a tier c as it is close to the pure power 11 bronze platform, group regulated. this is not final yet, but you have an idea of where it'll be.

Tier C but rated for 50ºC and 11 years at 25C?

 

[LukeSavenije]

2 minutes ago, NunoLava1998 said:

Tier C but rated for 50ºC and 11 years at 25C?

 

well... i have to keep myself on the notes

Quote

Groupregulated units will not be placed above Tier C.

 

[NunoLava1998]

16 minutes ago, LukeSavenije said:

well... i have to keep myself on the notes

I do know; I just think it's quite strange a group regulated PSU is rated for 50ºC.

Also looks like there are more than 2 coils which doesn't mean group regulation, but i'm not sure if these are coils at all:

 

Plus according to the Eteknix's review, the average voltage deviation is about 1%. That's better than the Corsair CXM at about 1.3-1.7%.

[NunoLava1998]

1 hour ago, NunoLava1998 said:

wall of le text

@LukeSavenije Looks like it is group regulated (contacted bequiet) unfortunately, but other than that it looks like a good quality unit honestly. It's Tier A-B except for crossload as even the normal voltage regulation is excellent

[LukeSavenije]

1 hour ago, NunoLava1998 said:

@LukeSavenije Looks like it is group regulated (contacted bequiet) unfortunately, but other than that it looks like a good quality unit honestly. It's Tier A-B except for crossload as even the normal voltage regulation is excellent

again, notes says it'll be tier c...

 

then again, it's not final, and if you're able to convince @Stefan Payne...

[NunoLava1998]

1 minute ago, LukeSavenije said:

again, notes says it'll be tier c...

 

then again, it's not final, and if you're able to convince *...

ik it'll be tier C

im fine with it tho since despite being group regulated the voltage regulation shouldn't go ut of spec in crossload

[RejZoR]

Can one even really buy a "bad" PSU these days from a brand name? Unless you buy a no name Chinese PSU, any will be perfectly good.

 

Hell, we had crappy LC Power PSU's back in the day and most ran problem free for years.

[LukeSavenije]

5 minutes ago, RejZoR said:

Can one even really buy a "bad" PSU these days from a brand name?

for sure, how many do you want me to mention? cooler master elite, mwl (in some cases), corsair vs (in some cases), seasonic m12ii/s12ii (in some cases), fsp aurum (in some cases)

 

heck, even something like a seasonic focus can be a bad idea with some systems

[Guest]

Why do so many of these reviews not show the PCB? Nor do they seem to do safety tests, look at separation between mains and output, track layout of the primary side and how close some of it can be to the casing. I've opened many PSUs in the past and some of the findings safety wise are shocking, no pun intended.

[NunoLava1998]

44 minutes ago, RejZoR said:

Can one even really buy a "bad" PSU these days from a brand name? Unless you buy a no name Chinese PSU, any will be perfectly good.

 

Hell, we had crappy LC Power PSU's back in the day and most ran problem free for years.

You can buy bad power supplies. See: SeaSonic ST-SF (Bronze SFX). Shit quality, but atleast it runs at a quiet 60dB during max load (for the 450w). At 30ºC ambient.

LC Power still makes power supplies.

And they have 80+ Platinum and server power supplies, wtf?

 

The SFX Power 2 is a group-regulated power supply, but from the eteknix review and be quiet's specs it's one of the most good performing group regs i've seen. @jonnyGURU would probably be surprised at a group regulated power supply having (not on crossload) 0.5-1% voltage offset on the 5V and 12V rail, and being rated for 50ºC.

 

8 minutes ago, Phill104 said:

Why do so many of these reviews not show the PCB? Nor do they seem to do safety tests, look at separation between mains and output, track layout of the primary side and how close some of it can be to the casing. I've opened many PSUs in the past and some of the findings safety wise are shocking, no pun intended.

Eteknix did OPP test and showed the internals. Nothing else.

 

 

[RejZoR]

People seem to confuse shit PSU's with PSU's that are nothing special. Today, everyone throw them into the same bin. Unless it has ridiculous tolerances, modular cables and titanium rating it's considered "bad". I've frankly never heard of any brand name PSU actually being bad to a point it would have screwed up out of spec tolerances or straight up blowing up after few years. There were times when PSU's were abd and there were bunch of crappy generic models, but PSU industry changed dramatically several years ago and it's really hard to actuall get an objectively bad PSU from a brand name.

[Guest]

12 minutes ago, NunoLava1998 said:

 

 

Eteknix did OPP test and showed the internals. Nothing else.

 

 

Yep, and that was my point. To me, safety tests are important. While many would not know a spark gap if it bit them in the harris, at least give a safety rating and show the PCB for those with a bit more of a clue. Obviously without unwinding the transformer we can never know the quality of that, how good the insulation is etc. Good review sites could remove and check those.