rfi - Current "Leading Edge" Power Supply Innovations

[Crossy_2000]

Hi to all,
after lurking for years on your great forum (being an IT consultant in Vienna) I decided to register in order to join your discussion.
Let me start that your spirited exchange of PSU-centric information has provided a fountain of knowledge on advanced technical
matters on the topic so far: PSU design, pitfalls & perils, advantages and drawbacks of particular platforms & specific models currently
sold.

Specifically, I'd like to inquire on current "true" Leading Edge Power Supply Innovations. Not merely the current "lege artis" stuff,
(LLC resonator in the primary part of the PSU coupled with DC/DC on the secondary) which imho should form the baseline for
modern designs but the "EXTRA" stuff only to be found in select contenders. Being a computer junkie, I love to understand
the building blocks of tomorrow's power supply.

So far I already made out a couple of these (inferring from your spirited discussion)

1) GaN (and other compound semiconductors) solutions in the PSU-circuitry
To my knowledge GaN FETs have been around (for RF power applications ?) for a decade or longer. Now Corsair's AX-1600i
features "a very unique bridgeless totem-pole front end with GaN transistors". This caught my eye. I love compound semiconductors.
Opto-electronics would be nonexistant without them. Telecom would still be using "metallic" T1 and T3 without them and we
would still doling out a fortune for communications cost (1980ies like). Now I do know there are other compound semis used
in power applications, such as SiGe or also "Carbon Diamond" applications (see Dynex Semiconductors for the latter)

2) The "Bridgeless Totem-pole" front end from the AX-1600i

3) Digital (MCU like) active control of the PSU (not just an analog IC or A/D converter that allows you to read out your PSUs measures
(like for example the Thermaltake Smart DPS)

4) "Burst Mode" in the APFC circuit (?) such in the Gospower built Cooler Master MWE v2 series to further increase
PSU efficiency at low loads

... are there others ??
Just attempting to gain insight and some candidates for my own "next" power supply. It should incorporate such "Holy Grail" technology,
(sort of as a tech demonstration) but Wattage isn't top (600-750W should be more than enough for my own main system). Definitely not
a gamer, only single mid to high end grafcard (right now using an Nvidia GTX-1660)


kind regards from Vienna/Austria,
Crossy_2000

[mdk777]

So,

This is where I get triggered.
Once you get into the mid 90% efficiency....where exactly do you think the next major leap in advancement is going to come from?
A quick internet search show that 94% efficiency PSU were developed 8 plus years ago and Seasonic has been selling them to consumers for nearly 5 years now.

94%....that is simply insane...

I understand that you are interested in technology for the sake of technology...I get it.

I just wonder if your focus could be better spent on other areas that actually could benefit from optimization.

For example.

Here in the States we hear daily from the religion of electric cars.
Yet, for some reason, I have not read a single article in the mainstream media describing the nearly 60% loss in transmission from source to consumer in the electric grid.

Wait....you mean my super eco friendly TESLA actually waists 60% of the energy before it is even charged???? regardless of the "carbon free/carbon neutral/ renewable/plant saving energy source?"

That is just the start.

What is the efficiency of solar power cells?

Have they gotten above the low single digits yet(so, yeah low double digits I guess...learn something new every day)?

Sooo.....you do your best to trap as much solar energy as possible, using every advancement of technology....and you get something like 5%(EDIT...looked up and claims of 20%+ are made today for cells themselves...I'd need to look at conversion losses to get you your appliance) conversion to electricity.
But wait, where does all the rest of that energy go?

Could it be that you are trapping a huge amount of thermal energy to get that 20% electricity?
Could it be that the net effect is that you are actually CONTRIBUTING to global warming by the net amount of thermal energy you are trapping in those solar cells?

Again, for some reason, this energy equation is not mentioned in the 24/7/365 religious genuflection of the PC environmental media.

So, my point again?

There are some issues in energy use where ENORMOUS gains can be made through application of appropriate technology.

I just think PSU optimization is not that area anymore, the gains have already been made, further optimization is physically impossible...(there is only 6% potential left...you can't get above 100%)

[ashiekh]

One must be careful with efficiency; one can extract energy from the 25C temperature difference between the surface of the ocean and 300m deep, with a theoretical maximum efficiency of 7% (so much less in practice).

But this does not make it a source we should not explore; same argument for solar cells, as if not used we would be throwing it away.

This does not mean we should not look for efficiency, just that there are exceptions.


I think there is a lot more to gain by building low power consumption computers; seems Windows 10 runs well on an 8GB Raspberry Pi 4

[Crossy_2000]

@mdk777 - I started this thread to inquire into all avenues of "PSU related innovation"
not the motivations behind these efforts nor the pros and contras of these motivations

Indeed, the motivation here is "technology for technology's sake". But there's a twist.
I intend to LEARN from these innovations about the if's, why's, pros and contras.

And to follow your avenues, "efficiency" is just one parameter. To me it's not even the most imporant one.
For example "stabilitY" is much more important to me. Or "resiliency" or the absence of current components
that shouldn't be there (e.g. ripple) or the absence of "interaction" between the rails (cross-load etc...)

Of course, in every effort there's the law of diminishing returns.

In any respect - while I might be "brainwashed" by the spirit of innovation, I'm certainly not under the "spell"
of the "religious genuflection of the PC environmental media" as you just mentioned it. In fact, if you looked
a tad more closely at my post, I didn't even mention EFFICIENCY in my opening post <g>

In any respect the inquiry is still on...

are there any more "trends" or "innovations" out there in PC Power supply design that should be worth mentioning ???
I really would like to discuss these and find some neat implementations

best to you,
Crossy (from Vienna/Austria)

[ashiekh]

• Efficiency
• Stability
• Resilience

are all very important, but for me

• Longevity

is high on the list of desirables. Some people admire the way capacitors are so much more compact these day, but bemoan that they just don't last like they used to.

Something I totally missed was

• Reactivity

and by that I mean the speedy response to varying loads and the reason some power supplies and video cards just don't get on well with each other.

[mdk777]

Yeah, I understand.

Diminishing returns is exactly what I see....to the point of absurdity.

There is a reason this site and forum is nearly dead....the entire PC enthusiast community came to this conclusion some years ago.

Any real innovation is going to require a 're-think' of the computing model.
This has been going on in the HPC area continuously.

If it is going to fly, it will be proven there, well before it gets to the consumer PSU market.

[Crossy_2000]

@ashiekh
I hear you !

One of my clients in Vienna runs a smaller departmental server (Xeon E3 based) in his office premise.
The stock PSU went belly up years ago and I replaced it with an In-Win GreenMe 550W that certainly did its job.
When Covid19 broke out, this PSU started to emit strange noises. I feared for the worst and replaced the PSU.
In my own lab I later found out that the noises came from a bad fan, I think it was a ball-bearing type. Replaced
it with a better, more suited type spec wise (CFM, PSU wise) and voila - the PSU has been running in a custom
build silent machine smoothly ever since. So much about longevity ... not only concerning the PSU's design but also
the adjunctive componentry such as the PSU fan, the cables etc..

best to you,
Crossy

[ashiekh]

Yet, for some reason, I have not read a single article in the mainstream media describing the nearly 60% loss in transmission from source to consumer in the electric grid.

Wait....you mean my super eco friendly TESLA actually waists 60% of the energy before it is even charged???? regardless of the "carbon free/carbon neutral/ renewable/plant saving energy source?"

http://insideenergy.org/2015/11/06/l...and-your-plug/

Claims much lower losses in transmission

Could it be that you are trapping a huge amount of thermal energy to get that 20% electricity?
Could it be that the net effect is that you are actually CONTRIBUTING to global warming by the net amount of thermal energy you are trapping in those solar cells?

Again, for some reason, this energy equation is not mentioned in the 24/7/365 religious genuflection of the PC environmental media.

No, for two reason; firstly that the heat is received by the Earth if the solar cell is there or not, secondly, and perhaps more importantly, our waste heat does not significantly contribute to global warming. Global warming is due to the more efficient trapping of the Sun's heat by greenhouse gasses; one place where we really do not want to improve efficiency ;-)

[ashiekh]

In my own lab I later found out that the noises came from a bad fan, I think it was a ball-bearing type.

Interestingly ball bearings tend to be nosier than sleeve bearings so it may have been just fine; but I would also have replaced it.

[mdk777]

http://insideenergy.org/2015/11/06/l...and-your-plug/

Claims much lower losses in transmission

Yes, I was talking the total energy loss....say from a fuel...say compare a ICE auto to electric.

Your source claims 65% in generation, another "About 6% – 2% in transmission and 4% in distribution"

So we are up to 72 to 75 % loss according to your source...then we have a loss going from your home to charge the batteries on you car....then the loss from the batteries to the motor...

So, suddenly an ICE doesn't look so bad if we are looking at the conversion of fuel source to actually moving your car.