GOOD powersupply for under $40?

[w_tom]

Specs don't guarantee that a power supply meets them.
However a manufacturer that provides specs can be sued if he
does not meet those specs. Much easier not include
specifications so as to sell to 'computer experts' who only
understand one spec - price. Therein lies the indicator of
inferior supplies. They cannot be sued if they don't claim to
meet any specs. Then they can forget to include essential
functions such as EMI/RFI. Most 'clone assemblers' don't even
understand the concept. Most clone assemblers only understand
price.

100,000 or 1,000,000 units - makes little difference in the
price. Some years ago the same Chinese labor working for less
money was making supplies that cost $120. Why did labor costs
go up and power supply price go down?

It is an MBA myth that 'economy of scale' or cheaper labor
cut costs. The only thing that really cuts costs is
innovation. If 'economies of scale' actually existed, then GM
cars would be the least expensive to build. GM cars cost more
to build than Mercedes Benz. GM costs are higher than most
every other automaker. GM cars cost more to build than
equivalent Japanese models when Japanese labor (due to foreign
exchange values) cost 20% more. Why? GMs stifles
innovation. Then blames labor, government, unfair traders,
education system, tax system, pension plan - everyone but GM
management that does not even have driver's licenses.

The only thing that lowers costs (other than forgetting to
include essential functions) is innovation. MBAs don't like
to hear that. It means they cannot be good managers without
getting their hands dirty. It means they cannot blame
everyone else.

Power supply prices drop when innovation occurs - which is
why the $100 supply now sells for $80.

 

[LRW]

And where is the PFC? This Antec appears to be a new design
intended to go for the $40 market. The supply appears to
retail list for something like $60-$70. $43 would be a street
price.

PFC? "Power Factor Coeficient"? Or something else?
Well, the three top name brand PS at 350 watt for around $40 don't have
"PFC".
Is that a problem?

I've narrowed it down to this Sparkle:
http://www.newegg.com/app/ViewProdu...-103-486&catalog=58&manufactory=BROWSE&depa=1
Antec:
http://www.newegg.com/app/ViewProdu...atalog=58&manufactory=BROWSE&depa=1&section=1
and Fortron:
http://www.newegg.com/app/viewProductDesc.asp?description=17-104-966&depa=0
Of these, all the numbers are the same except the Sparkle has 10watts less
on the 3.3+5v lines than the other two, and the Antec lists MTFB at 50k
while the other two are 100k.
Considering how the MTFB is simply estimated, is that an issue?
I'm thinking the Fortron is looking like the best one... opinions?

Thanks!!
Liam

 

[w_tom]

Power Factor Correction is not an issue in some cases. One
way to cut costs is forget this function - which BTW is
required everywhere in the EU.

A standard $80 retail list (or whatever that list price is
at this year) supply does provide PFC. This function is
essential in commercial areas. And it may be necessary for
some homes with same utility power problems. However that
Sparkle supply, as listed, out rightly said it would not meet
any Intel specs. Anyone who recommended that power supply
needs to first learn basic power supply concepts.

That Fortron as listed is missing many requirements.
Functions may be there. Just that the retailer did not list
them. Specs as listed are insufficient.

Of course most suspicious is that Antec cannot even provide
input voltage and frequency? These specs don't look
trustworthy. Better see if manufacturer provides real specs.

MTBF is a ballpark number. A lower MTBF could be due to
less components. But 5 years continuos operation verse 10
years continuous - which is really something like 15 years
verse 30 years operation - well then who cares? Don't give
MTBF much worry. EMI/RFI and overpower protection is more
important.

 

[Ruel Smith (Big Daddy)]

And where is the PFC? This Antec appears to be a new design
intended to go for the $40 market. The supply appears to
retail list for something like $60-$70. $43 would be a street
price.

This Antec may be the $80 supply - a new design. For
example the $80 supply was previously selling for $100. Then
new innovations were implemented. It is possible that the new
wave of $80 supplies will now *list retail* for $65. However
and again, where is the PFC in this supply? Every top of the
line manufacturer at one time or other tries to dump a low
cost knock off onto the market - to go after the lower end of
the market. Antec tends to meet minimal specs - a major
accomplishment for any power supply manufacturer when the
customers are so technically ignorant. $80 retail list number
had remained a benchmark. Supplies selling for $40 routinely
did not include essential functions.

The SL350 is the same power supply Antec includes in all their Performance
and Performance II line cases like the SX835BII and SX635II. It's not a
budget minded power supply at all. It's simply their older line of power
supplies that have long been highly rated in the custom built computer
market. This isn't something new. Their, now, high-end cases get the newer
TruPower power supplies instead of these models.

http://www.antec-inc.com/specs/sl350_spe.html



--
Big Daddy Ruel Smith

My SuSE Linux machine uptime:
9:43pm up 45 days 6:28, 2 users, load average: 0.63, 0.30, 0.10

My Windows XP machine uptime:
Something less...

 

[Ruel Smith (Big Daddy)]

Specs don't guarantee that a power supply meets them.

This is similar to the days I was into car stereos. Many people would
purchase equipment based on stated THD, signal/noise, etc. but that never
meant that equipment with better specs actually performed better. It's easy
for manufacturers to "fudge" specs. Look around for some reviews to see how
they actually PERFORM.



--
Big Daddy Ruel Smith

My SuSE Linux machine uptime:
9:56pm up 45 days 6:41, 2 users, load average: 0.44, 0.22, 0.14

My Windows XP machine uptime:
Something less...

 

[larrymoencurly]

Airflow on heatsink (in LFM) determines it thermal conductivity.
It takes almost no airflow (air moving so slow that hand cannot
feel it) to make a major heatsink improvement.

I saw for myself how even a little air flow could make a big
difference. I placed a hard drive horizontally, several inches above
a table, and after a few hours of continuous reads the hottest chip
reached 68C. Then I repeated this with the drive sitting vertically
on the table, that chip cooled to 53C. Ambient was 23C all the time,
no ventilaton (closed garage).

The first Sparkle 350W supply is inferior. They say so
right in the specifications:

But the $39 price should alone be a dead giveaway that
something is wrong with that supply. Neither supply even
tries to claim EMI/RFI protection.

I've seen expensive Fortron/Sparkle PSUs with those specs and some
very cheap models without them (models not listed at Fortron website),
and they looked very similar inside, with the same Fairchild KA3511
main regulator chip and identical-looking EMI filters. IOW I don't
see how that first Sparkle 350W is necessarily inferior in performance
just becuase it lacks specifications.

 

[jeffc]

PFC? "Power Factor Coeficient"? Or something else?
Well, the three top name brand PS at 350 watt for around $40 don't have
"PFC".
Is that a problem?

Not necessarily. It's Power Factor Correction, and it isn't essential.

 

[jeffc]

Power Factor Correction is not an issue in some cases. One
way to cut costs is forget this function - which BTW is
required everywhere in the EU.

A standard $80 retail list (or whatever that list price is
at this year) supply does provide PFC. This function is
essential in commercial areas. And it may be necessary for
some homes with same utility power problems.

Yes, but of course not if someone is already using good surge protection for
their whole computer, which means PFC would be redundant.

 

[jeffc]

Can't remember - street price I believe.

Sub $40 retail list or sub $40 street price?

 

[jeffc]

This is similar to the days I was into car stereos. Many people would
purchase equipment based on stated THD, signal/noise, etc. but that never
meant that equipment with better specs actually performed better. It's easy
for manufacturers to "fudge" specs. Look around for some reviews to see how
they actually PERFORM.

Ain't that the truth. It's true in home stereos too by the way

 

[R. Anton Rave]

100,000 or 1,000,000 units - makes little difference in the
price. Some years ago the same Chinese labor working for less
money was making supplies that cost $120. Why did labor costs
go up and power supply price go down? It is an MBA myth that
'economy of scale' or cheaper labor cut costs. The only
thing that really cuts costs is innovation.

Your own example of higher prices when Chinese labor was cheaper
supports the contention that economies of scale do matter. But I
agree that innovation is the main reason why productivity is has
generally improved 1,000% since the start of the 20th century. In
manufacturing, it would be hard for a company 10 times as large as
another to not have at least 10-20% lower costs, although plenty of
MBAs manage to get around that advantage.

I work for a company that doesn't hire MBAs, at least not for their
MBA degrees, except accountants, because scientists and engineers can
do whatever MBAs can, while the converse is rarely true. Also they
seem to be nicer people, overall, probably because they deal so much
with the natural world rather than the political world of MBAs. That
may also explain why there's long been much better income parity by
race than in other fields.

If 'economies of scale' actually existed, then GM cars would be
the least >expensive to build. GM cars cost more to build than
Mercedes Benz. GM costs are higher than most every other automaker.
GM cars cost more to build than equivalent Japanese models when
Japanese labor (due to foreign exchange values) cost 20% more.
Why? GMs stifles innovation.

That information is out of date because GM is now a lower cost
producer than Ford or DamlierChrysler, mostly thanks to those
competitors stumbling but also because its plants are more up to date
and the company is now lead by an actual "car guy" instead of a pencil
pusher. As for M-B's costs, the last time they were cheaper than GM's
was back when international exchange rates were fixed and the dollar
overvalued. But nobody is as efficient as the better Japanese
companies, who probably need 20% fewer hours to produce their vehicles
than anyone else, thanks to having the best plants, utilization, and
design, and their cost advantage may be even greater because they
don't have union labor in most of their U.S. plants (Mazda, NUMMI the
exceptions). On the other hand, GM has a higher percentage of union
labor than anyone (outsources less) yet still beats Ford and Chrysler,
which outsource more of their work.

Power supply prices drop when innovation occurs - which is
why the $100 supply now sells for $80.

Today's PC power supplies are almost the same, mechanically and
electrically, as the original IBMs in design, complete with the same
lousy 70% efficiency, except the metal is thinner and you see
MOSFETs,IGBTs, or power factor correction. I don't see how they're
cheaper to build, so I have to conclude that competitive pressures and
economies of scale are the reasons for the much lower prices.

 

[w_tom]

Please provide one specification from any "good surge
protection" that even CLAIMS power factor correction. How
does any surge protector - using MOVs - modify a phase angle
between voltage and current? How does it adjust this angle
for varying conditions? The simple answer is .... neither
"good surge protector" nor a typical plug-in UPS does any
power factor correction - nor claims such abilities.

How to make transient damage to a computer - powered on or
off - easier? Plug it into an adjacent "good surge
protector". It does not even claim protection from the
destructive type of surge. AND it does no Power Factor
Correction. But then prove me wrong. Post the specifications
of a "good surge protector" that claims power factor
correction as required required even in every European Union
country.

 

[w_tom]

Big difference between retail list price and street price.
Explains confusion over an $80 price benchmark - confusing
street price with "retail list".

 

[w_tom]

GM has long remained the world's highest priced
manufacturer. Their vehicles still required 6 cylinder
engines to do what most others do with 4 cylinder. GM
assembly hours per car still remain higher than most other
manufacturers, as they have been for decades. Point remains -
there is no such thing as "economies of scale" once a minimal
size is achieved. That lesson also applies to power supplies.

Mazda, a tiny auto manufacturer could earn a profit on one
model by only selling 13,000 per year. GM with existing
'economies of scale' could not earn a profit unless a model
sold at least 50,000. This applies to power supplies as
well. Virtually every ATX power supply manufacture has
achieved sufficient production that 'economies of scale' are
irrelevant.

Lower costs do not come from 'economies of scale' nor from
lower wages. It comes only from innovation. Innovation is
the missing factor in GM which is why they still don't have a
70 Hp per liter engine in every vehicle and therefore must
sell V-8s that only equal the competition's V-6. 70 Hp per
liter technology was ready for 1975 GM products and stifled by
anti-innovation GM MBAs - as the Vice President of drive train
development bluntly stated in the early 1990s. GM costs
remain some of the world's highest even though every 5 years
they output propaganda to the contrary.

Don't you get tired every five years that GM says, "we were
bad but now we are better". The sky is falling!

"Economies of scale" is also the myth that created some of
the most anti-American steel companies - formally known as
Bethlehem Steel, USX, etc. Again, companies so anti-American
as to stifle innovation and then run to government for
protection. Their MBAs feared electric arc furnace
'innovation'. They feared to innovate and instead advocated
the "economies of scale" myth for their higher costs. Classic
MBA solution: bigger obsolete technology blast furnaces.

What lowers costs in power supplies? Either innovation or
simply forgetting to install essential functions. "Missing
essential functions" is common in clone computers because
price - not value - is the customer's only specification.

Possible that list price of an "$80 power supply" benchmark
has dropped again due to innovation from those Asian power
supply manufacturers. We know more powerful transistors with
less cost exist. We know that switching power supply
frequencies are increasing. Innovations cut costs.
"Economies of scale" nor lower priced labor creates price
reduction in power supplies - no matter what MBAs claim.

 

[Ruel Smith (Big Daddy)]

GM has long remained the world's highest priced
manufacturer. Their vehicles still required 6 cylinder
engines to do what most others do with 4 cylinder. GM
assembly hours per car still remain higher than most other
manufacturers, as they have been for decades. Point remains -
there is no such thing as "economies of scale" once a minimal
size is achieved. That lesson also applies to power supplies.

This is getting way off topic but you're completely wrong. GM put 4 cyl DOHC
engines in the Cavalier Z24 and Sunfire GT producing 150 hp when all the
other manufacturers' so-called performace 4 cyl engines were producing less
than 130. Take a look back at 1995 and see how powerful the 4cyl engines
out of Japan and domestic companies were. You had Honda producing 127 HP in
the Civic, Ford producing 125 HP in the Escort, Toyota nowhere on the map,
etc. etc. I'm speaking of normally aspirated engines, here. Where you came
up with the information to make such a claim, I'll never know. Those
engines were left behind in recent years only due to GM's lack of attention
to that market. They simply abandoned it in favor of large trucks.

There is economies of scale with ever higher volumes, though they are
diminishing after some point. Comparing GM's cars to companies like Honda
and Toyota reveal that you pay a premium for a Honda or Toyota. Not only
are there few and, if at all, small rebates on the Japanese models, the
MSRP is higher for the content. Take for instance the Toyota Camry. A
top-end Camry with a V6, sunroof, and leather brushes the $30,000 mark. A
similar Pontiac Grand Prix is priced a few thousand dollars less. Go ahead
and shop for yourself and see. Honda prices are slightly lower than Toyota,
but they're still high. Both the Accord and Camry are not as roomy as the
Grand Prix. However, despite this, GM has remained the most profitable
manufacturer of automobiles. My info is just a couple of years old, but GM
was #1 of all manufacturers in profits on a per vehicle basis at
approximately $850 per copy.

Mazda, a tiny auto manufacturer could earn a profit on one
model by only selling 13,000 per year. GM with existing
'economies of scale' could not earn a profit unless a model
sold at least 50,000. This applies to power supplies as
well. Virtually every ATX power supply manufacture has
achieved sufficient production that 'economies of scale' are
irrelevant.

Mazda hasn't been a tiny manufacturer since the '80s. Where have you been?
Ford purchased them long ago. Ford employees even get the employee discount
on them, as well as Volvos.

Lower costs do not come from 'economies of scale' nor from
lower wages. It comes only from innovation. Innovation is
the missing factor in GM which is why they still don't have a
70 Hp per liter engine in every vehicle and therefore must
sell V-8s that only equal the competition's V-6. 70 Hp per
liter technology was ready for 1975 GM products and stifled by
anti-innovation GM MBAs - as the Vice President of drive train
development bluntly stated in the early 1990s. GM costs
remain some of the world's highest even though every 5 years
they output propaganda to the contrary.

Again, it was published just a couple of years ago that GM was the most
profitable auto manufacturer in the US market on a per vehicle basis (and
overall). You have some preconceptions that are literally wrong.

Don't you get tired every five years that GM says, "we were
bad but now we are better". The sky is falling!

Turn around for the world's largest employer isn't going to happen
overnight. Small companies can shift direction far more quickly than can
such a giant like GM. It's funny that the perception of GM's quality is so
low, but year after year when JD Power evaluates 5 year old vehicles for
reliablility, both Cadillac and Buick rank in the top 5. But they're just
soooo bad, aren't they? To the contrary, Mercedes, who enjoys almost
universal acclaim for quality, ranks miserably in reliability in both JD
Power studies and Consumer Reports surveys. Perception is the key.
Ex-Chairman for Chrysler Lee Iacocca once said that it amazed him that 3
cars go down the same assembly line with the exact same parts, but one gets
a Plymouth badge, one gets an Eagle badge, and one gets a Mitsubishi badge.
However, both public perception and journalistic reviews always proclaim
the one with the Japanese brand is a better car. Perception...

I snipped the rest of your argument because I'm tired of pointing out your
inaccuracies and I think your arguement doesn't hold water.

--
Big Daddy Ruel Smith

My SuSE Linux machine uptime:
9:53pm up 46 days 6:39, 2 users, load average: 0.34, 0.23, 0.13

My Windows XP machine uptime:
Something less...

 

[jeffc]

Please provide one specification from any "good surge
protection" that even CLAIMS power factor correction.

Power Factor Correction per se isn't the point. The point is that Power
Factor Correction doesn't buy you anything, except for the fact that active
PFC PSUs handle dirty main power better than passive ones, but a good
UPS/Surge Protector handles that as good or better anyway.

I'll tell you what - why don't YOU tell us what PFC is good for? In
practical terms - why do I want it for my PC PSU? Is it for efficiency - to
save on your electric bill? The only problem with PCs is that they're a
reactive load instead of a simple resistive load. But your power company
doesn't care about this for such a tiny amount of electricity, and you're
certainly not going to pay more for it. You're not going to be causing the
power company any problems with your tiny little reactive load, and you're
not going to be using less electricity. So why else would you want one on
your PC PSU? An active PFC PSU will handle spikes and surges, but anyone
who is going to pay for that is going to be foolish not to have a good
UPS/Surge Protector anyway. So what is the point exactly?

 

[jeffc]

Where you came
up with the information to make such a claim, I'll never know.

Tom's a good bluffer. He likes to push his weight around a little with a
few technical terms, start strong-arming people, and just hope they'll back
down. He fancies himself "on a higher plane" as a computer "builder" rather
than just "assembler", and is therefore "better" than the rest of us
enthusiasts. But at the end of the day he doesn't really get any more value
from his knowledge than we do.

 

[w_tom]

First power factor creates many problems - the least of
which is power efficiency. Power factor problems can cause a
computer to 'hickup' and can even create fires on 'interior
wall' neutral wires that normally would have been properly
sized. Just a few things that PFC solves. PFC solves enough
problems that EU requires it. It is a solution important in
commercial environments and not always necessary in
residential power (warning: some residential power really is
more 'commercial').

But again, where is this surge protector that solved power
factor problems. A question again because 1) what UPS/surge
protectors do and 2) what power factor is about are so basic
as to question whether you understand the concepts of PFC and
what a surge protector really does. To claim a surge protector
provides PFC is to either not understand what PFC is or to not
understand what a power strip protector does. Typical plug-in
UPSes don't even CLAIM power factor correction - since they
connect computer direct to AC mains when not in battery backup
mode.

In the meantime, there is no plug-in UPS that provides
effective surge protection - as made obvious by their own
specs. Specs that often don't even say which type of surge
they protect from - to leave the customer deceived. Another
point made here because basic electrical concepts were not
accurately represented.

No earth ground means no effective transient protection in
that plug-in UPS. Power factor correction does not exist in
any surge protector or plug-in UPS. And as posted earlier:

This function is essential in commercial areas. And it may
be necessary for some homes with same utility power problems.

PFC is considered important enough in the EU (where AC
electric standards are higher) as to be required in power
supplies. It buys us nothing but buys the EU something? How?
This is a discussion about reliable supplies verses cheap
supplies missing important functions. If PFC concepts were
understood and appreciated, then PFC correction would never
have been cited in power strip protectors and plug-in UPSes.

Rather than get upset, admit the concept was not understood,
why and where PFC is important, and reliability that PFC
brings to a computer. Then move on. Power factor correction
is clearly not as important as overvoltage and overpower
protection. But PFC is part of a reliable power supply -
especially in commercial environments.

 

[JAD]

UM TOM? please stay with the knowledge that you were blessed with, it wasn't car manufacturing.....Cars are designed wholly and
solely on customer response....Americans have decided that GAS MILEAGE doesn't matter anymore. ITS SIZE , you know that male adage,
Bigger is better size matters.....4cyls are no longer requested by American buyers,,,, brainwashing most likely. utter idiots...
oily bushes are runnin the show don't ya know!!!!!!!!!!!!!!!!!\

 

[w_tom]

The point remains - innovation is THE reaons for cutting
costs - which may have little to do with profits when other
games are played. Games such as GM still making profits when
they lose money on every car (which is why they so jealously
protect the 'truck as a car' loophole). (GM products remains
the most expensive to build. The source who says otherwise
invented his numbers.)

The only way to cut costs in power supplies is innovate - or
remove essential functions. How to sell a vehicle at up to
$5000 profit? Get government to not require essential
passenger car functions - and therefore no need for even 1990
innovation - a 70 Hp per liter engine. IOW eliminate defacto
world standard functions. Just like cheap power supplies.
Sell the supply at less cost and actually increase profits -
by forgetting to include essential functons.

It helps to first cite accurate numbers. GM will not put
both horsepower and liters on same sticker to avoid numerical
facts. 2003 Cavalier and Pontiac Sunfire - 2.2 liter outputs
only 115 HP (not a fictional 150) ... or 52 Hp per liter - low
performance - obsolete technology. Pay more money and get a
2.2 liter outputing 140 HP ... pathetic 63 Hp per liter.
Better. But base engine for world standard products - at no
extra money - is about 70 Hp per liter. A standard for at
least one decade and GM still does not meet it - which
explains their higher costs.

Superchargers should provide even higher performance.
Pontiac Bonneville - luxery with attitude - 240 HP from a 3.8
liter ... 63 Hp per liter. No better than an upscale
Sunfire! World standard supercharged engines output about 100
HP. Why does a supercharged Pontiac not even meet world
standard aspirated engines? Where is the innovation? Not
there once numbers are reviewed.

The point: numerical specifications are essential even in
selecting power supplies. Those low numbers explain why GM
cars still cost more to build than equivalent competition. GM
costs so high that missing functions in trucks resulting in a
$5000 per vehicle profit to mask losses on every car. That is
correct. World standard innovation is missing even in their
engines - which is why they don't put both HP and liters on
the sticker. They fear one might do some arithmetic.

Some have misrepresent facts to promote inferior products
such as the 2003 Pointiac Sunfire. Happens when specs are not
openly provided by the manufacturer. If manufacturer does not
openly provide specs, then myth purveyors will lie about the
numbers - such as 150 HP from a Pontiac Sunfire. Sticker on
that Pontiac Sunfire will not provide numerical specs when
their product does not even meet world standards. This is
common in selling power supplies - and yes - even plug-in
UPSes and power strip surge protectors.

The standard engine - at no extra money - is about 70 Hp per
liter. Those who talk kindly about GM often misrepresent the
facts to avoid GM's inferiority. Some who recommend power
supplies also may misrepresent a product that has no
overvoltage and no overpower protection. Reliable supplies
are why specs must be provided with the product.

It gets even more interesting. One has claimed a product
provides PFC when even the manufacturer does not make that
claim. Even worse, basic knowledge of what a surge protector
does make that PFC myth bluntly obvious. Myths are widespread
when manufacturer does not provide specs which is why
numerical specifications are necessary. Myth that a surge
protector provided PFC - example of how myths are promoted
when specs are not available.

BTW, it does not matter that you don't understand those
specs. Usually only 1% of the customers do. But those 1% are
why manufacturers who provide specs also meet those specs.
Some power supply manufacturers (and auto manufacturers) don't
provide those specs up front because their products don't meet
world standards. They few people like me who constantly go
for the juglar - the numerical specs. No specs for a power
supply? Specs are missing for good reason. No horsepwer and
liter provided up front by the manufacturer? Also for good
reason. Many GM products do a pathetic 52 Hp per liter. GM
and its fans would prefer you don't learn numbers.