Energy Efficiency and 5V/12V Devices

I need to power about 20 devices that run off AC adapters that provide 5VDC or 12VDC to the device. As I understand it, a lot of efficiency is lost
stepping down from 208V/220VAC to 120VAC and then again in the conversion to 12VDC. What is the most energy efficient way to power these devices?
I am guessing that there is a rectifier that would take 208VAC input (thus avoiding the step down to 125VAC) and then efficiently (90% efficiency?) convert 208VAC to 12VDC or 5VDC, for multiple end devices.
Alternately, I could identify individual AC to DC converters/adapters that do the conversion from 208VAC to 12VDC or 5VDC efficiently.
What is the most energy efficient path to follow there?
nish
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
nish was thinking very hard :

Where are you getting 208 volts from ? It is not a common distribution voltage anywhere in the world.
Just go with the current lol system as anything else is too cumbersome to contemplate. :/
--
John G Sydney.

Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

We did some tests a few years ago, and found auto-ranging switched mode computer power supplies were more efficient when fed near the top of their range than at the bottom. For 120-240V autoranging supplies, difference was typically around 8% higher power consumption, but in some cases it was 15% higher. (This made me think some "80+" PSUs were unklikely to be "80+" when run on 120V, but we didn't measure the absolute efficiency, only the change in power draw between running on 120V or 240V.)
This wasn't really a surprise - the I2R heating losses in the rectifier, power FETs, and transformer primary winding are going to be 4x higher at half the supply voltage.
Some of the server power supplies had to be derated when run at 120V, and in some cases this meant you lost redundancy of dual power supplies because both were needed to power the server.
This work was triggered by stats which showed we were getting 10x more mains wiring accessory burnouts in the US data centres than anywhere else in the world. Some of this was put down to servers using 120V outlets, but the incidents were still significantly more common even on US 240V circuits than on 220-240V circuits in other countries.
It was difficult to compare data, but we also suspected we were getting more PSU failures in systems running at the lower end of the input voltage range.
We never did collect data on hold-up times over short brown-outs, but I suspect that as the storage capacitors store 4x more energy at twice the supply voltage, systems would survive at least 4x longer power interruption without going down.

I steal 12V supply from by PC to power the ethernet switch and the WiFi access point. These items came with wall warts which got particularly hot implying very inefficient. This was a while back, and such inefficient PSUs are no longer permitted in the EU. Nowadays, wall wart PSUs have to be efficient at a level only achievable by good switched mode PSUs and have to have very low consumption when there's no load (i.e. they stay stone cold, and are usually < 0.1W).
One place you do want a separate isolated supply is powering anything which interfaces your phone line, such as a modem (even though most are well isolated from the line anyway).
--
Andrew Gabriel
[email address is not usable -- followup in the newsgroup]
  Click to see the full signature.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
On Thu, 31 Dec 2015 12:07:31 -0000 (UTC), snipped-for-privacy@cucumber.demon.co.uk (Andrew Gabriel) wrote:

As long as you have good surge protection with a single point ground system, this is not that big a problem. Here in Florida, we got pretty good at this.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
208Y/120 is a very common distribution systems in the United States. The relationship comes from Y-connected system and the square root of 3 (1.73 * 120 = 208). The voltage between any phase and ground will be 120 and the voltage between any two phases will be 208.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
snipped-for-privacy@ponce.com has brought this to us :

Yes but the fact is very few domestic instalations in the USA have THREE phase distribution and only have 120 and 240.
--
John G Sydney.

Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
On 12/14/2015 6:47 PM, John G wrote:

Generally quite true, and I don't think the OP has 208.
An apartment complex may be powered by 208/120V wye, with 2 phases fed to apartment panels. Those panels have 208V, and an electric range may be powered by 208V.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
I don't believe I suggested that the OP had a three phase installation, but he does likely have single phase 208.
My only real intent was to counter the reply that suggested the OP was askew in saying he had a 208VAC supply available and that it wasn't a common distribution voltage anywhere in the world.
I do agree with several posters that the OP is unlikely to do any better with efficiency or reliability than just purchasing a commercially available switching PS.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

If you want to do it all with one power supply, a "wide mouth" PC supply (100-250v input) will give you both voltages and plug right into your 208v supply.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

Why step down to 120V? Most switching power supplies are happy from ~90V to 250V. Switching supplies will be the most efficient, as well.

Universal power supplies should be pretty much the same for small currents.

Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

Polytechforum.com is a website by engineers for engineers. It is not affiliated with any of manufacturers or vendors discussed here. All logos and trade names are the property of their respective owners.