• Welcome to TechPowerUp Forums, Guest! Please check out our forum guidelines for info related to our community.

TechPowerUp is Hiring a Power Supply (PSU) Reviewer

Joined
May 26, 2023
Messages
109 (0.19/day)
Thermal energies make it easier to overcome the diode drop, and so increase efficiency.

voltage - Why does temperature modify the characteristics of a diode? - Electrical Engineering Stack Exchange

View attachment 371151

So, a hot diode has higher efficiency.
Your attention is focused on wrong quadrant :roll:. Look at 3rd one ( anti clockwise counted) - you have way higher reverse current for hot diode. At almost two orders of magnitude higher reverse voltage. Sure - this part of loss is not as high as losses in first quadrant but they are going blatantly in opposite direction. And keep in mind real doiodes have their conductive connectors too. So total loss is a sum of few components. And their ohmic loss increasing for sure along the temperature rise.

Your blatantly stupid sentece was a wrong generalization for all hot semiconductors. You cant say the same about power mosfets for sure.
I'm too lazy to teaching you all rudiments here.
 
Joined
Mar 21, 2021
Messages
5,098 (3.76/day)
Location
Colorado, U.S.A.
System Name CyberPowerPC ET8070
Processor Intel Core i5-10400F
Motherboard Gigabyte B460M DS3H AC-Y1
Memory 2 x Crucial Ballistix 8GB DDR4-3000
Video Card(s) MSI Nvidia GeForce GTX 1660 Super
Storage Boot: Intel OPTANE SSD P1600X Series 118GB M.2 PCIE
Display(s) Dell P2416D (2560 x 1440)
Power Supply EVGA 500W1 (modified to have two bridge rectifiers)
Software Windows 11 Home
And yet

A hot bridge rectifier has higher efficiency.
 
Last edited:
Joined
May 26, 2023
Messages
109 (0.19/day)
And yet

A hot diode has higher efficiency.
No way if reverese losses is becoming dominant :laugh:
You are definitely unable to catch - even simple semiconductor like a diode is not acting just like pure ideal model. It is still complex device for modeling.
And you are totally focused on one aspect of this one device,
So even your changed sentence has very limited validity.

And definitely diode efficiency is way lower than modern mosfet efficiency. But for sure mosfet needs to be driven so it implies additional costs of course.
But we are living in 21 century now so we have a choice - for what solution we want paying.
If efficiency is our concern so solution based just in diodes is like a car on wooden wheels.
 
Joined
Mar 21, 2021
Messages
5,098 (3.76/day)
Location
Colorado, U.S.A.
System Name CyberPowerPC ET8070
Processor Intel Core i5-10400F
Motherboard Gigabyte B460M DS3H AC-Y1
Memory 2 x Crucial Ballistix 8GB DDR4-3000
Video Card(s) MSI Nvidia GeForce GTX 1660 Super
Storage Boot: Intel OPTANE SSD P1600X Series 118GB M.2 PCIE
Display(s) Dell P2416D (2560 x 1440)
Power Supply EVGA 500W1 (modified to have two bridge rectifiers)
Software Windows 11 Home
No way if reverese losses is becoming dominant :laugh:

You do realize, I hope, that quadrant 3 was not to scale? and that the reverse current is much smaller than the forward current.

diode-IV.gif
 
Last edited:
Joined
May 26, 2023
Messages
109 (0.19/day)
You do realize, I hope, that quadrant 3 is not to scale? that the reverse current is much smaller than the forward current.

View attachment 371184
Sure I do.
But you dont understand diode is intrisically inefficient against modern power mosfet. So it doesnt matter at all if high temperature help a bit or not.
And keep in mind in general case load is not constant so in some cycles when diode rectifying very short spikes or nothing at all reverse loss will be dominat and higher for hot diode than for colder one.
So your generalization is intrisically wrong.
You should provide complex simulation to declare valid conclusions for limited range of cases.
Generalization like yours even after changes is blatantly wrong.
 
Joined
Mar 21, 2021
Messages
5,098 (3.76/day)
Location
Colorado, U.S.A.
System Name CyberPowerPC ET8070
Processor Intel Core i5-10400F
Motherboard Gigabyte B460M DS3H AC-Y1
Memory 2 x Crucial Ballistix 8GB DDR4-3000
Video Card(s) MSI Nvidia GeForce GTX 1660 Super
Storage Boot: Intel OPTANE SSD P1600X Series 118GB M.2 PCIE
Display(s) Dell P2416D (2560 x 1440)
Power Supply EVGA 500W1 (modified to have two bridge rectifiers)
Software Windows 11 Home
OK, here I bug out.
 
Joined
May 26, 2023
Messages
109 (0.19/day)
OK, here I bug out.
I'm not sure you did.
Keep in mind diode can rectifying very short spikes and be reversed most of the rest of the cycle. It is especially valid in synchronous rectification when mosfet bypassing most of the forward pulse. In such setup diode working acively very shortly at both ends of the forward pulse. But dissipating energy by all time of reverse pulse. In such setup reverse loss could be dominant for very high temperatures. So hotter definitely doesnt mean better or more efficient in such case.
 
Joined
Apr 24, 2007
Messages
305 (0.05/day)
Location
Bay Area, CA
thermally conductive?
Yes. And that potting material is VERY expensive. I've looked into it to create a fanless PSU option, and the current solution of "just sell a 1200W PSU as an 850W PSU" is actually a much cheaper solution.

If a PC uses 300W, the amount of energy it pumps into a room is 300W, regardless of fan speed or cooling efficiency (because 99.9999% of the energy a PC uses is converted to heat - sound and light are very very very low-energy by comparison)

The temperature of that airflow depends on the rate of airflow. 300W dissipated into a small amount of air (low rate of airflow) will inevitably produce higher exhaust air temperatures, but higher temperatures at a lower rate is the same amount of energy per second being dumped into the room. That's the energy conservation law.
You're not wrong. But we're talking about the PSU specifically. And yes... if there is 100W of AC not converted to DC, that's 100 units that needs to be exhausted. :D
@Chrispy_
Guess I phrased that poorly. What I am talking about is subjective human perception here, not physics. Which you have indeed described correctly. I definitely feel the exhaust heat out from my Torrent far more than from my old Define S, even though they both houses the exact same components. The Torrent is just far more effective at removing the hot air from the case. The Define was slower and subjectively it FELT like it wasn’t breathing out as much heat, even though objectively, of course, it was. The Torrent both has more fans and a freer exhaust area in the back, which IS noticeable.
Zero RPM fan mode was never my thing. Putting the knob on the back of the current RMx was. :D

(Stating the obvious here) I find that if you set the fan to a lower RPM, and the desirable RPM may be different for different people because the audible frequency of the fan varies and some people are more sensitive to certain frequencies than others, that fan will run longer at a lower RPM. Whereas, Zero RPM fan mode will wait until there is so much heat to dissipate, the fan kicks on at 400W and the air coming out of the back can be used to melt heatshrink tubing.

BTW: Someone on Discord posted this and I LOL'd:
1731277894689.png

Of course, eventually that fan WILL have to turn on even at <400W load. But I still find it funny.
 
Joined
Mar 21, 2021
Messages
5,098 (3.76/day)
Location
Colorado, U.S.A.
System Name CyberPowerPC ET8070
Processor Intel Core i5-10400F
Motherboard Gigabyte B460M DS3H AC-Y1
Memory 2 x Crucial Ballistix 8GB DDR4-3000
Video Card(s) MSI Nvidia GeForce GTX 1660 Super
Storage Boot: Intel OPTANE SSD P1600X Series 118GB M.2 PCIE
Display(s) Dell P2416D (2560 x 1440)
Power Supply EVGA 500W1 (modified to have two bridge rectifiers)
Software Windows 11 Home
Yes. And that potting material is VERY expensive. I've looked into it to create a fanless PSU option, and the current solution of "just sell a 1200W PSU as an 850W PSU" is actually a much cheaper solution.

But then are you not relying on the case for ventilation (as with Seasonic)? or did you have in mind a sealed case with the hot components using the case as a heat-sink (as with SilverStone)?

Or maybe both ideas combined as with the SilverStone Nightjar? I would hate potted as I like to replace components.

SilverStone Nightjar.jpg
 
Last edited:
Joined
Apr 24, 2007
Messages
305 (0.05/day)
Location
Bay Area, CA
But then are you not relying on the case for ventilation (as with Seasonic)? or did you have in mind a sealed case with the hot components using the case as a heat-sink (as with SilverStone)?

Or maybe both ideas combined as with the SilverStone Nightjar? I would hate potted as I like to replace components.

View attachment 371209
You do realize that the Nightjar and fanless Seasonic are EXACTLY what I am describing when I say, "the current solution of 'just sell a 1200W PSU as an 850W PSU' is actually a much cheaper solution"? Because that is exactly what they are doing and why their cost is so high.
 
Joined
Mar 21, 2021
Messages
5,098 (3.76/day)
Location
Colorado, U.S.A.
System Name CyberPowerPC ET8070
Processor Intel Core i5-10400F
Motherboard Gigabyte B460M DS3H AC-Y1
Memory 2 x Crucial Ballistix 8GB DDR4-3000
Video Card(s) MSI Nvidia GeForce GTX 1660 Super
Storage Boot: Intel OPTANE SSD P1600X Series 118GB M.2 PCIE
Display(s) Dell P2416D (2560 x 1440)
Power Supply EVGA 500W1 (modified to have two bridge rectifiers)
Software Windows 11 Home
One advantage of potted would be to eliminate coil whine which might otherwise ruin the silence of no fan.

Anyhow, it is really great to have you so active on this forum.

I find that if you set the fan to a lower RPM, and the desirable RPM may be different for different people because the audible frequency of the fan varies and some people are more sensitive to certain frequencies than others, that fan will run longer at a lower RPM. Whereas, Zero RPM fan mode will wait until there is so much heat to dissipate, the fan kicks on at 400W and the air coming out of the back can be used to melt heatshrink tubing.

I can't speak for others, but it is the variation that drives me nuts. I replaced a CPU heatsink with one that ran at the same temperature but had a copper slug and that eliminated the constant speed hunting.

I also wonder about the baffle next to many power supply fans and how this affects both flow rate and noise (yes, I know why the baffle is there).

jonnyGURU said:
Thing is, those push pull 80s don't have to be loud and will cool much better. But customer perception is so distorted now.....

Maybe go back to 80mm?
 
Last edited:
Joined
Feb 20, 2019
Messages
8,310 (3.93/day)
System Name Bragging Rights
Processor Atom Z3735F 1.33GHz
Motherboard It has no markings but it's green
Cooling No, it's a 2.2W processor
Memory 2GB DDR3L-1333
Video Card(s) Gen7 Intel HD (4EU @ 311MHz)
Storage 32GB eMMC and 128GB Sandisk Extreme U3
Display(s) 10" IPS 1280x800 60Hz
Case Veddha T2
Audio Device(s) Apparently, yes
Power Supply Samsung 18W 5V fast-charger
Mouse MX Anywhere 2
Keyboard Logitech MX Keys (not Cherry MX at all)
VR HMD Samsung Oddyssey, not that I'd plug it into this though....
Software W10 21H1, barely
Benchmark Scores I once clocked a Celeron-300A to 564MHz on an Abit BE6 and it scored over 9000.
Modular interface gets in the way, but maybe an 80mm exhaust on a Gold or Platinum. Submit it to Cybenetics to see if we can get a A+ rating. :D
Yeah, the other problem these days is PSU basement shrouds that only have cutouts/ventilation for a large 120mm+ fan in the top. With an 80mm rear fan and the modular interface at the front, the obvious vent locations in the PSU sides would be flush against the wall of the PSU basement in many case designs :(

Actually, I'm overthinking it - they don't have to be rear/side vented like they were 20 years ago - the vents could still probably be on the top of the PSU towards the back - in a location where case manufacturers have made venting holes.
 
Joined
Apr 24, 2007
Messages
305 (0.05/day)
Location
Bay Area, CA
Yeah, the other problem these days is PSU basement shrouds that only have cutouts/ventilation for a large 120mm+ fan in the top. With an 80mm rear fan and the modular interface at the front, the obvious vent locations in the PSU sides would be flush against the wall of the PSU basement in many case designs :(

Actually, I'm overthinking it - they don't have to be rear/side vented like they were 20 years ago - the vents could still probably be on the top of the PSU towards the back - in a location where case manufacturers have made venting holes.
Or.. Just vents all over... and an aluminum housing. :D
 

W1zzard

Administrator
Staff member
Joined
May 14, 2004
Messages
27,869 (3.71/day)
Processor Ryzen 7 5700X
Memory 48 GB
Video Card(s) RTX 4080
Storage 2x HDD RAID 1, 3x M.2 NVMe
Display(s) 30" 2560x1600 + 19" 1280x1024
Software Windows 10 64-bit
Joined
May 26, 2023
Messages
109 (0.19/day)
Maybe my opinion wont be popular but i think as a consumer we dont need a lot of detailed reviews. What we need in first place is splitting false tinanium from real titanium and similiar job with platinum grade/advertised PSU-s as well. We can assume gold rated PSU are obsolete from technical point of view - but sure they can fill the mainstream for many years still.
Only very few extended size PSU graded lower than Platinium are worth of closer inspection imho. So more than review - detailed comparisions would be much more helpfull for buyers.
 
Joined
Mar 21, 2021
Messages
5,098 (3.76/day)
Location
Colorado, U.S.A.
System Name CyberPowerPC ET8070
Processor Intel Core i5-10400F
Motherboard Gigabyte B460M DS3H AC-Y1
Memory 2 x Crucial Ballistix 8GB DDR4-3000
Video Card(s) MSI Nvidia GeForce GTX 1660 Super
Storage Boot: Intel OPTANE SSD P1600X Series 118GB M.2 PCIE
Display(s) Dell P2416D (2560 x 1440)
Power Supply EVGA 500W1 (modified to have two bridge rectifiers)
Software Windows 11 Home
For me one of the biggest issues is fan noise (and coil whine) and as jonneyGURU points out, zero RPM fan mode is a bit of a crock.

Zero RPM fan mode will wait until there is so much heat to dissipate, the fan kicks on at 400W and the air coming out of the back can be used to melt heatshrink tubing.

I am one who can't stand fan variations more than just fan noise and even in the full-blown reviews this tends not to get covered.
 
Last edited:
Joined
Nov 18, 2010
Messages
7,568 (1.47/day)
Location
Rīga, Latvia
System Name HELLSTAR
Processor AMD RYZEN 9 5950X
Motherboard ASUS Strix X570-E
Cooling 2x 360 + 280 rads. 3x Gentle Typhoons, 3x Phanteks T30, 2x TT T140 . EK-Quantum Momentum Monoblock.
Memory 4x8GB G.SKILL Trident Z RGB F4-4133C19D-16GTZR 14-16-12-30-44
Video Card(s) Sapphire Pulse RX 7900XTX. Water block. Crossflashed.
Storage Optane 900P[Fedora] + WD BLACK SN850X 4TB + 750 EVO 500GB + 1TB 980PRO+SN560 1TB(W11)
Display(s) Philips PHL BDM3270 + Acer XV242Y
Case Lian Li O11 Dynamic EVO
Audio Device(s) SMSL RAW-MDA1 DAC
Power Supply Fractal Design Newton R3 1000W
Mouse Razer Basilisk
Keyboard Razer BlackWidow V3 - Yellow Switch
Software FEDORA 41
Did you done calcs i asked you ?? No. Instead you are comparing aplles to oranges. Laptop chargers have typically way higer Vout than 12V in ATX standard.
So their Iout is way lower at the same Pout levels as for desktop PSU-s. :eek:

Why should I? You think I would bait on such childish bait? Unlike you I deal daily with those, especially with failed units from RMA channel. Nor your complaint actually proves something as I said, people do not care about the specifics, they never will. There is a black box that does the power thingy, and leave it there. We are not discussing engineering design choices with economic tradeoff Vs performance gains, and when we can state that the platform is over engineered. The reviewers also must steer away from it, just as making stupid badges.

Most mainstream laptop chargers are type c and are PD capable these days, they support granular voltage steps at very high loads, does output voltage differences make some drastic changes using modern designs? No... You have to thank Qualcomm and other makers investing to quit the shitshow intel does, that's pretty much it.

If you look into high power bricks, like Delta does, they are usually bleeding edge designs unlike PSU's, more space, more budget to cheap out, there are flops, there is no denying about that, but I cannot share the details.

I am completely with Johnny about PSUs being passive, that's masochism, best of any worlds is hybrid, that ramps up still after a time, like 20 minutes to cool heat buildups from caps, not completely passive rule binded by current consumption, that's crazy, I have never understood how the hell it is allowed.
 
Joined
Mar 21, 2021
Messages
5,098 (3.76/day)
Location
Colorado, U.S.A.
System Name CyberPowerPC ET8070
Processor Intel Core i5-10400F
Motherboard Gigabyte B460M DS3H AC-Y1
Memory 2 x Crucial Ballistix 8GB DDR4-3000
Video Card(s) MSI Nvidia GeForce GTX 1660 Super
Storage Boot: Intel OPTANE SSD P1600X Series 118GB M.2 PCIE
Display(s) Dell P2416D (2560 x 1440)
Power Supply EVGA 500W1 (modified to have two bridge rectifiers)
Software Windows 11 Home
Is it the caps that cause heat? I imagined it was mainly the semiconductors.
 
Joined
Nov 18, 2010
Messages
7,568 (1.47/day)
Location
Rīga, Latvia
System Name HELLSTAR
Processor AMD RYZEN 9 5950X
Motherboard ASUS Strix X570-E
Cooling 2x 360 + 280 rads. 3x Gentle Typhoons, 3x Phanteks T30, 2x TT T140 . EK-Quantum Momentum Monoblock.
Memory 4x8GB G.SKILL Trident Z RGB F4-4133C19D-16GTZR 14-16-12-30-44
Video Card(s) Sapphire Pulse RX 7900XTX. Water block. Crossflashed.
Storage Optane 900P[Fedora] + WD BLACK SN850X 4TB + 750 EVO 500GB + 1TB 980PRO+SN560 1TB(W11)
Display(s) Philips PHL BDM3270 + Acer XV242Y
Case Lian Li O11 Dynamic EVO
Audio Device(s) SMSL RAW-MDA1 DAC
Power Supply Fractal Design Newton R3 1000W
Mouse Razer Basilisk
Keyboard Razer BlackWidow V3 - Yellow Switch
Software FEDORA 41
Is it the caps that cause heat? I imagined it was mainly the semiconductors.

If they do, then you are way past their life, it depends on the type, but of they do it, they explode.

But all devices have their life span and is influenced by temps. I will not submerge into ripple surpression working hours, why that is chosen in that place at that temperature, that is a question for designers, me as RMA guy seeing all those failed designs, I can only question, wtf dude, were you a student?
 

GabrielLP14

SSD DB Maintainer
Staff member
Joined
Aug 2, 2021
Messages
318 (0.26/day)
Location
São Paulo, Brazil
System Name Gabriel-PC
Processor Core i7-13700K (All Core 5.7GHz)
Motherboard MSI Z790-P PRO WIFI DDR4
Cooling NZXT Kraken X72 360mm
Memory 32GB Netac DDR4-3200 MT/s CL-16
Video Card(s) RTX 4070 Ti Super Colorful
Storage Memblaze P7940 7.68TB Gen5 (OS), Solidigm P44 2TB (Games) + 4x 4TB WD Black HD (Synology NAS DS1817)
Display(s) AOC G2460PF 144Hz 1ms (Kinda trash)
Case NZXT Phantom 820 Black
Audio Device(s) Motherboard onboard audio (good enough for me)
Power Supply Corsair RM1000X
Mouse Have no idea (Generic)
Keyboard Have no idea (Generic)
Software Windows 11 Pro 23H2 + Windows Server 2022 + Synology in NAS
I wish you guys good luck to find a person for this dificult position
 

Solaris17

Super Dainty Moderator
Staff member
Joined
Aug 16, 2005
Messages
27,012 (3.83/day)
Location
Alabama
System Name RogueOne
Processor Xeon W9-3495x
Motherboard ASUS w790E Sage SE
Cooling SilverStone XE360-4677
Memory 128gb Gskill Zeta R5 DDR5 RDIMMs
Video Card(s) MSI SUPRIM Liquid X 4090
Storage 1x 2TB WD SN850X | 2x 8TB GAMMIX S70
Display(s) 49" Philips Evnia OLED (49M2C8900)
Case Thermaltake Core P3 Pro Snow
Audio Device(s) Moondrop S8's on schitt Gunnr
Power Supply Seasonic Prime TX-1600
Mouse Razer Viper mini signature edition (mercury white)
Keyboard Monsgeek M3 Lavender, Moondrop Luna lights
VR HMD Quest 3
Software Windows 11 Pro Workstation
Benchmark Scores I dont have time for that.
Too much back and fourth arguing. Apply if you think you have something to prove. Thanks!
 
Joined
Mar 21, 2021
Messages
5,098 (3.76/day)
Location
Colorado, U.S.A.
System Name CyberPowerPC ET8070
Processor Intel Core i5-10400F
Motherboard Gigabyte B460M DS3H AC-Y1
Memory 2 x Crucial Ballistix 8GB DDR4-3000
Video Card(s) MSI Nvidia GeForce GTX 1660 Super
Storage Boot: Intel OPTANE SSD P1600X Series 118GB M.2 PCIE
Display(s) Dell P2416D (2560 x 1440)
Power Supply EVGA 500W1 (modified to have two bridge rectifiers)
Software Windows 11 Home
"TPU is looking to hire a PC enthusiast or professional to review PC power supplies (PSUs) full-time."

"The position is paid, part-time, remote, meaning you'll work from your own testing setup"


@W1zzard
which might it be? full-time or part-time
 
Last edited:

W1zzard

Administrator
Staff member
Joined
May 14, 2004
Messages
27,869 (3.71/day)
Processor Ryzen 7 5700X
Memory 48 GB
Video Card(s) RTX 4080
Storage 2x HDD RAID 1, 3x M.2 NVMe
Display(s) 30" 2560x1600 + 19" 1280x1024
Software Windows 10 64-bit
Top