Power Supply?

[Rod Speed]

Rod's in full-troll-mode. No stopping the bull.

Couldnt bullshit its way out of a wet paper bag even if
its pathetic excuse for a 'life' depended on it, as always.

 

[larry moe 'n curly]

news:[email protected]:

Thermal sensors are usually designed to have hysteresis in their
behaviour, to avoid switching 'chatter' problems. The result is that
there is often a significant difference between the cut-out trigger
temperature and the reset temperature, hence the time delay while the
system cools.

Is that really necessary for an ATX PSU, which turns off completely and
needs to be manually turned back on?

 

[larry moe 'n curly]

Few power supplys have thermal sensors and even less
shut down because of the dust that accumulates inside them.

I think when Tom's Hardware tested PSUs a few years ago, only about 3
out of 13 had thermal protection. The older Antec SmartPowers don't,
but the newer ones with the higher +12V amp ratings do and have a
second thermistor for this. Some Fortrons/Sparkles are probably
designed the same, but the second thermistor and certain surrounding
parts were left out of the ones I bought.

 

[CWatters]

Uh, CWatters is using the UK definition of torch = flashlight, not the
US definition of torch = welding apparatus.

Well I guess you could use a welding torch - that would probably shift the
dust ok!

 

[larry moe 'n curly]

To All Cognoscenti, Greetings!

My computer keeps shutting down as if the plug was yanked from the wall. The
power cannot be turned back on for a few minutes, after which it can be
powered up again. This has happened three times so far, with about a week or
two of normal operation in between.

I would check the PSU fans to see if they're spinning about right and
the air isn't blowing out much hotter than another comparable
computer's. If they're OK, try running the PC with the side of the
case removed, and if this helps, then you have a heat related problem.
According to PC Power & Cooling, power capacity can drop by 1/3 when
the surrouning temperature rises from 25C to 40C, but this shouldn't
matter unless your PSU is a cheapo that was marginal to begin with.
Bad electrolytic capacitors can also reduce capacity, not only of the
PSU but also of the voltage regulator(s) built into the mobo. See
www.badcaps.net for more information.

You may want to check if there's a very slight short between the mobo
and case, even if every mounting hole has a pair of washers for it
because they sometiems get pierced. Mobos and cases can flex enough to
cause the two to touch, so support the mobo at every corner, either
with standoffs (they dont't have to go into holes in the case) or
press-on rubber bumper feet stuck to the case).

 

[Rod Speed]

Rod Speed wrote

I think when Tom's Hardware tested PSUs a few years ago,
only about 3 out of 13 had thermal protection. The older Antec
SmartPowers don't, but the newer ones with the higher +12V amp
ratings do and have a second thermistor for this. Some Fortrons/
Sparkles are probably designed the same, but the second thermistor
and certain surrounding parts were left out of the ones I bought.

And few power supply shut down because of the DUST inside them.

 

[Rod Speed]

Rod Speed wrote

I think when Tom's Hardware tested PSUs a few years ago,
only about 3 out of 13 had thermal protection. The older Antec
SmartPowers don't, but the newer ones with the higher +12V amp
ratings do and have a second thermistor for this. Some Fortrons/
Sparkles are probably designed the same, but the second thermistor
and certain surrounding parts were left out of the ones I bought.

They're mostly just controlling the fan
based on temp, not a thermal shutdown.

And few power supply shut down because of the DUST inside them.

 

[McSpreader]

bought.

And few power supply shut down because of the DUST inside them.

Wriggle, wriggle.

Caught out bullshitting yet again, but still digging.

 

[McSpreader]

Is that really necessary for an ATX PSU, which turns off
completely and needs to be manually turned back on?

Maybe not, but the PSU designer would typically choose off-the-shelf
components that behave that way.

 

[Lil' Dave]

I would not rule it out completely, but you have a point.

Arno

Thanks.
I see this thread has taken its normal routine again. One person trying to
dominate his opinion over others. Instead of sharing information and
shutting up. Am not reading this person's posts due to their inciting
nature. Nor any subthread where this person is prevalent.
Was there a time that this particular newgroup was not dominated by such a
person?

 

[larry moe 'n curly]

larry moe 'n curly <[email protected]> wrote

Few power supplys have thermal sensors

They're mostly just controlling the fan
based on temp, not a thermal shutdown.

I wired a pot across each of the Antec's 10K ohm thermistors. With one
thermistor it varied the fan speed, but with the other it didn't seem
to do anything until I cut its resistance way down, and then the PSU
stopped. I wish I knew what parts to add to the Fortron/Sparkle to add
thermal shutdown to it.

 

[larry moe 'n curly]

news:[email protected]:

Thermal sensors are usually designed to have hysteresis in
their behaviour, to avoid switching 'chatter' problems. The
result is that there is often a significant difference between
the cut-out trigger temperature and the reset temperature,
hence the time delay while the system cools.

Maybe not, but the PSU designer would typically choose off-the-shelf
components that behave that way.

What thermistors have hysteresis? Isn't that in the feeback loop of
the comparator that connects to the thermistor? I've never seen a
Schmitt trigger chip being used with a thermistor in an ATX PSU (I
admit I've looked at very few ATX PSU circuits).

 

[Andy]

Thanks.
I see this thread has taken its normal routine again. One person trying
to
dominate his opinion over others. Instead of sharing information and
shutting up. Am not reading this person's posts due to their inciting
nature. Nor any subthread where this person is prevalent.
Was there a time that this particular newgroup was not dominated by such a
person?

I for one have so far seen nothing untoward in the contributions to this
thread; on the contrary, only a genuine desire to help, which I welcome and
appreciate. Honest differences of opinion are to be expected.

Andy

 

[Andrew Reid]

I would check the PSU fans to see if they're spinning about right and
the air isn't blowing out much hotter than another comparable
computer's. If they're OK, try running the PC with the side of the
case removed, and if this helps, then you have a heat related problem.
According to PC Power & Cooling, power capacity can drop by 1/3 when
the surrouning temperature rises from 25C to 40C, but this shouldn't
matter unless your PSU is a cheapo that was marginal to begin with.
Bad electrolytic capacitors can also reduce capacity, not only of the
PSU but also of the voltage regulator(s) built into the mobo. See
www.badcaps.net for more information.

You may want to check if there's a very slight short between the mobo
and case, even if every mounting hole has a pair of washers for it
because they sometiems get pierced. Mobos and cases can flex enough to
cause the two to touch, so support the mobo at every corner, either
with standoffs (they dont't have to go into holes in the case) or
press-on rubber bumper feet stuck to the case).

I'd like to add that when the side of the case is removed the air flow
is disrupted and hot spots can develop.

If your checking for a heat related problem by removing the side of
the case it's a good idea to use a large fan to blow air into the case
to ensure good air circulation during the test.

 

[McSpreader]

What thermistors have hysteresis? Isn't that in the feeback
loop of the comparator that connects to the thermistor? I've
never seen a Schmitt trigger chip being used with a thermistor
in an ATX PSU (I admit I've looked at very few ATX PSU
circuits).

You are correct. I should have used the term thermal cutout switch,
which includes a thermal sensor device such as a thermistor. I was
thinking more of chips that have the thermal sensing device on-
board and have a hysteresis characteristic.

 

[123]

I see this thread has taken its normal routine again.
One person trying to dominate his opinion over others.

Just rubbing the noses of the pig ignorant in the holes in their claims.

Instead of sharing information

Did that. You get to like it or lump it.

Reams of your puerile shit flushed where it belongs.

 

[Rod Speed]

Some gutless ****wit desperatey cowering behind
McSpreader <[email protected]> desperately
attempted to bullshit its way out of its predicament
and fooled absolutely no one at all, as always.

No surprise that it needs to cower.

 

[Rod Speed]

Rod Speed wrote

to shut the power supply down.

I wired a pot across each of the Antec's 10K ohm thermistors.
With one thermistor it varied the fan speed, but with the other
it didn't seem to do anything until I cut its resistance way down,
and then the PSU stopped.

That is why I used the word mostly, most dont have two, as you said.

I wish I knew what parts to add to the
Fortron/Sparkle to add thermal shutdown to it.

Why cant you work out that bit of the circuit diagram from the Antec ?

 

[McSpreader]

ks.com.au:

Just rubbing the noses of the pig ignorant in the holes in their
claims.


Did that. You get to like it or lump it.

Reams of your puerile shit flushed where it belongs.

I think Rod likes you.

 

[Paul Atreides]

To All Cognoscenti, Greetings!

My computer keeps shutting down as if the plug was yanked from the wall. The
power cannot be turned back on for a few minutes, after which it can be
powered up again. This has happened three times so far, with about a week or
two of normal operation in between. The electrical outlet is fine. Am I
right in suspecting an overloaded power supply? Could there be any other
cause?

Thanks for your suggestions in advance.

Andy

Hi,

I've had this exact same problem last week.
The power supply fan was dead.