CPU Install question

[Flasherly]

You don't want to run the Corsair fans at 100%, because they're
"vacuum cleaner fans". They're like the old Dell fans, designed
to work into a high resistance load, and when spooled up,
would be obnoxious. (Things would be different, if Corsair
selected a larger fin spacing on the cooler, but that
in turn reduces cooler performance. The vacuum cleaner
design idea, is for high performance systems.)

Right.
CORSAIR Hydro Series H105 Extreme Performance 240mm Liquid CPU Cooler,
CW-9060016-WW

Sweet setup. Shouldn't be awhile before considering "modifications,"
were those Dynatrons exhibit wear (at $109US, holy cow, awhile...or
more). Its self-contained water pump (not adjustable) might be
another matter, tho.

Believe I may have bought a Dynatron fan before, or at least heard of
them.

That intricate of modern modular system, when retrospectively looking
at just about a whole case, in itself, to contain some of the older
water systems, of course, PWM and advancements being what they are...

Anything less than that pump, though, I'd be tempted or think to
purchasing it with a mind to swapping stock voltage controlled fans
(for finding one with a suitable rating). Only the H105 isn't so much
about that;- looking over that product reminds me of higher,
higher-end passive heatwick pipes for that grade of processor, its
intent, nor would come especially cheap, either...well within or near
to H105 costs.

Saying earlier in some unrelated context I'd buy, offhand, a
self-contained waterblock CPU cooler, all else being equal (to my
thinking, provided quite a few steps below in processors, considering
the H105's intents, for roundabout sub-$50 purposes).

The H105, however, I'd be busy for some time collecting information
and making notes from reviews. Quite. I haven't even have half a
mind for a suitable processor, in the first place, to place the class
of cooling with. (Being all I've ever used, I'd probably be skewed to
air;- Piped heatwicking, withstanding, water in any volume would be a
different perspective, to say the least.)

 

[Bill]

One more question please.

The fans on the H105 are "PWM". Do you think it is normal for 1 fan
of the 2 radiator fans on the H-105 to run while the other sits and
maybe just "jiggles" a little (at low speeds). That is what drew my
attention to it in the first place, is that I could hear some very,
very light sounds (of the 2nd fan sort of trying).
They both run together at slightly higher speeds/heat.

I'm rewriting the following part:

The two fans are co-joined by a "Y-cable" (a term I picked up from
Paul), and are attached to the 4-pin CPU_OPT header. I'm not sure
about the number of pins of the fans (the picture Paul posted
indicates they are 4-pin).

So far, I'm assuming that the fans are slightly out of
sinchronization with sensitivity to voltage.
You both provided me with plenty of reading to think through (I
haven't done that part yet)!

Thank you!
Bill

 

[Bill]

If it was my rig:

1) Pumps runs at constant speed. The three pin header for it
likely doesn't have voltage adjustment in hardware anyway,
so this should be easy to meet. Verify, using Speedfan,
that the pump is in the right RPM range. If too high, there
could be air in the liquid loop. If too low, it could be that
the header isn't delivering +12V.

2) If I had the headers, I'd run the two fans separately.
But that would complicate fan control. It isn't that
common to have control software that couples two headers
in terms of settings and makes them do the same thing.
I expect that's why Corsair provided that Y cable. You
can use the Y cable if you want, but just make sure the
fan runs properly.

I have the H105 pump connected to CPU_FAN, and it runs very consistently
at about 1850 RPM invariant of the conditions (clearly pumping the
water through the radiator faster won't improve cooling very much!).

The two Corsair PWM fans are connected via a Y-cable to the CPU_OPT
connector.
The fans definitely both go to 0 RPM at idle sometimes (this was much of
the appeal of the cooler to me in the first place--that is was PWM
managed, unlike all, or at least most, of it's Corsair siblings).

Even if the 2nd fan isn't hiccupping trying to start, I've observed that
it still pulses sometimes at lower speeds. Using FanSpeed, I observed
the first fan running at 668 RPM, while the 2nd fan hiccupped. The RPM
specifications for the fan is 800-2700 RMP (+/- 10%). This is outside
that range.

I have observed that they are plenty of H105 owners with "pulsing fans".
Strangely, it gets mentioned more in forums than in online reviews. One
person went through 3, and they all shared the same issue.

I looked in the BIOS for a way to set a minimum fan speed (and didn't
find one--it does show the fan speed 1336 RPM at boot). I'll keep
working on this. I regard it more of a 2nd tier problem than a 1st tier
problem, i.e. I'm not going to let it keep me from getting my work
done. I'll keep looking for a software solution. Gigabyte provides ways
to speed up their motherboard, so maybe their is something handy for
fans (I wish to increase the minimum fan speed)!

My 2 case fans (front and back) are actually controlled by a 3-option
switch (for a 4-fan controller) on the top of the case and are not
attached to the motherboard at all. The high, medium, low settings are
not so much different. Of course, I could connect them to the
motherboard if I wanted to. I'm just a novice, so please let me know if
I'm messing up anywhere. I.e., I am Not speaking as an authority!

--------------------------
I enjoyed playing with, I mean working with, the Prime95 program. Of
course it trys to run all 8 cores at 100%. I believe I chose what was
called the "balanced" configuration. Using a i7-4790K CPU with the H-105
case:

The (core) temperatures quickly rose to the 51-56 range-with Fan RPM of
2000, and then a few minutes later increased to a max of 73 (mean
71)-with RPM of 2500, and within a few more minutes settled to 50-55
again, and pretty much stayed there.

I'm guessing the hardware was smart enough to "back off a little" for
the sake of the system. This makes me that much more satisfied.

Cheers,
Bill

 

[Paul]

I have the H105 pump connected to CPU_FAN, and it runs very consistently
at about 1850 RPM invariant of the conditions (clearly pumping the
water through the radiator faster won't improve cooling very much!).

The two Corsair PWM fans are connected via a Y-cable to the CPU_OPT
connector.
The fans definitely both go to 0 RPM at idle sometimes (this was much of
the appeal of the cooler to me in the first place--that is was PWM
managed, unlike all, or at least most, of it's Corsair siblings).

Even if the 2nd fan isn't hiccupping trying to start, I've observed that
it still pulses sometimes at lower speeds. Using FanSpeed, I observed
the first fan running at 668 RPM, while the 2nd fan hiccupped. The RPM
specifications for the fan is 800-2700 RMP (+/- 10%). This is outside
that range.

I have observed that they are plenty of H105 owners with "pulsing fans".
Strangely, it gets mentioned more in forums than in online reviews. One
person went through 3, and they all shared the same issue.

I looked in the BIOS for a way to set a minimum fan speed (and didn't
find one--it does show the fan speed 1336 RPM at boot). I'll keep
working on this. I regard it more of a 2nd tier problem than a 1st tier
problem, i.e. I'm not going to let it keep me from getting my work
done. I'll keep looking for a software solution. Gigabyte provides ways
to speed up their motherboard, so maybe their is something handy for
fans (I wish to increase the minimum fan speed)!

My 2 case fans (front and back) are actually controlled by a 3-option
switch (for a 4-fan controller) on the top of the case and are not
attached to the motherboard at all. The high, medium, low settings are
not so much different. Of course, I could connect them to the
motherboard if I wanted to. I'm just a novice, so please let me know if
I'm messing up anywhere. I.e., I am Not speaking as an authority!

--------------------------
I enjoyed playing with, I mean working with, the Prime95 program. Of
course it trys to run all 8 cores at 100%. I believe I chose what was
called the "balanced" configuration. Using a i7-4790K CPU with the H-105
case:

The (core) temperatures quickly rose to the 51-56 range-with Fan RPM of
2000, and then a few minutes later increased to a max of 73 (mean
71)-with RPM of 2500, and within a few more minutes settled to 50-55
again, and pretty much stayed there.

I'm guessing the hardware was smart enough to "back off a little" for
the sake of the system. This makes me that much more satisfied.

Cheers,
Bill

You could try the "blipping" fan by itself,
with the good fan disconnected, and using just the one
fan to cool the radiator.

I do not recommend changing fan configuration while
a system is running. Always shut down, if adding or removing
the Y cable, or plugging or unplugging fans. I tried a "hot swap"
once here, and the inrush of motor current, crashed the computer.
So it's not really all that safe, to fiddle with those on
a hot system.

You would test the single fan, to see if the problem is related
to two fans being on the same cable. Maybe a "pulse" of current
drawn by one fan, causes the rail voltage on the second fan
to change enough to cause the brushless DC motor to malfunction
(not fire at the right point).

I feel the electrical response curve for the fan invites this
problem. If Corsair rates the fans for 800-2700 RPM, then with
the PWM set to 0%, the fan should spin at 800. Rather than zero.
What you're basically doing, is running the fan response curve
through the point at which the motor won't start. Which is why
it is blipping. You would need a motor with some sort
of hysteresis in response, to not attempt to start until
the PWM mediated motor voltage was above the recommended
minimum internal value.

You'd need whatever is adjusting the PWM pulse width, to go
from 0% PWM to say 30% PWM, all in one jump, so the motor
always starts cleanly. Which requires a level of control
over the fan speed control system.

I don't have a lot of experience with the fan adjustments
(BIOS automated or via Speedfan). I use Speedfan once in
a while for measurement, but not for adjustment. And I
hardly ever enable BIOS fan speed control. In this case,
I don't know what is adjusting your setup. Maybe it's a
BIOS thing.

Paul

 

[Bill]

You could try the "blipping" fan by itself,
with the good fan disconnected, and using just the one
fan to cool the radiator.

I do not recommend changing fan configuration while
a system is running. Always shut down, if adding or removing
the Y cable, or plugging or unplugging fans. I tried a "hot swap"
once here, and the inrush of motor current, crashed the computer.
So it's not really all that safe, to fiddle with those on
a hot system.

You would test the single fan, to see if the problem is related
to two fans being on the same cable. Maybe a "pulse" of current
drawn by one fan, causes the rail voltage on the second fan
to change enough to cause the brushless DC motor to malfunction
(not fire at the right point).

I feel the electrical response curve for the fan invites this
problem. If Corsair rates the fans for 800-2700 RPM, then with
the PWM set to 0%, the fan should spin at 800. Rather than zero.
What you're basically doing, is running the fan response curve
through the point at which the motor won't start. Which is why
it is blipping. You would need a motor with some sort
of hysteresis in response, to not attempt to start until
the PWM mediated motor voltage was above the recommended
minimum internal value.

You'd need whatever is adjusting the PWM pulse width, to go
from 0% PWM to say 30% PWM, all in one jump, so the motor
always starts cleanly. Which requires a level of control
over the fan speed control system.

I don't have a lot of experience with the fan adjustments
(BIOS automated or via Speedfan). I use Speedfan once in
a while for measurement, but not for adjustment. And I
hardly ever enable BIOS fan speed control. In this case,
I don't know what is adjusting your setup. Maybe it's a
BIOS thing.

Paul

Paul, You've provided me with a lot of useful information. Thank you!
You understand my issue better than I do! %-)

You wrote: If Corsair rates the fans for 800-2700 RPM, then with
the PWM set to 0%, the fan should spin at 800. -- If that happened, I
don't
think I would have any problem at all. Corsair would probably blame the
motherboard maker (Gigabyte)! : )

Thank you too for mentioning other details about "hot swapping" fans.
I would not have have been so "prudent".

The "reviewer" who went through 3 units, ended up with a pair of these
(Noctua NF-F12-PWM):

http://www.newegg.com/Product/Produ...5608026&cm_re=notua_nf-_-35-608-026-_-Product

He said his system ran a few degrees cooler with them. But I have to
take one review with a grain of salt. Do you have an opinion whether
that is a reasonable substitute? I notice it's rpm range only goes up to
1500 RPM, but I suppose the m^3/h might be a better basis for
comparison--I just "guessed" what m3/h means! Learning in progress here!
I note that's it's not cheap (about $60). I know Corsair has other fans
people "upgrade" to as well (usually for quietness, I thought). I
haven't done my homework on this. I'm still "configuring software" on
two computers, this one and the one that is moving on to a new owner.
For the first time, I'm trying this time to make proper use of admin and
user accounts, something I've neglected doing in the past. I don't
quite have all of the subtleties down yet! : )

Cheers,
Bill

 

[Paul]

Paul, You've provided me with a lot of useful information. Thank you!
You understand my issue better than I do! %-)

You wrote: If Corsair rates the fans for 800-2700 RPM, then with
the PWM set to 0%, the fan should spin at 800. -- If that happened, I
don't
think I would have any problem at all. Corsair would probably blame the
motherboard maker (Gigabyte)! : )

Thank you too for mentioning other details about "hot swapping" fans.
I would not have have been so "prudent".

The "reviewer" who went through 3 units, ended up with a pair of these
(Noctua NF-F12-PWM):

http://www.newegg.com/Product/Produ...5608026&cm_re=notua_nf-_-35-608-026-_-Product


He said his system ran a few degrees cooler with them. But I have to
take one review with a grain of salt. Do you have an opinion whether
that is a reasonable substitute? I notice it's rpm range only goes up to
1500 RPM, but I suppose the m^3/h might be a better basis for
comparison--I just "guessed" what m3/h means! Learning in progress here!
I note that's it's not cheap (about $60). I know Corsair has other fans
people "upgrade" to as well (usually for quietness, I thought). I
haven't done my homework on this. I'm still "configuring software" on
two computers, this one and the one that is moving on to a new owner.
For the first time, I'm trying this time to make proper use of admin and
user accounts, something I've neglected doing in the past. I don't
quite have all of the subtleties down yet! : )

Cheers,
Bill

That's a pretty fancy fan, as computer cans go.
It makes a lot of claims about design, but it's coming
from a computer fan maker.

The airflow flat out is 93.4 m³/h.

So that's hours, so I divide by 60 to get minutes. 1.56

Next, I need to convert cubic meters to cubic feet (for CFM).

(39.37/12) ^ 3 = 35.3 cubic feet/cubic meter

1.56*35.3 = 55CFM

Which seems a little high for a fan of that size,
depth, and rotational speed. I would have guessed
35CFM. And the noise rating is also suspiciously
low for a fan like that. 55CFM should be over 30dBa.

One reviewer on the Newegg page, seems to think
they don't have the rated static pressure. So they're
not "vacuum cleaner fans". Again, the 22dBa rating
suggests "wimps". If the current setup really needs
all the fans can give, these won't be as good
from a performance perspective. But they will be
quiet (if the specs can be believed).

The claim is, it has a 300RPM minimum speed (which I
assume is at 0% PWM). So it follows the Intel recommended
design. But, Noctua doesn't provide a simple fan speed
curve to reinforce my observation.

Considering the price of the original cooler, that's
a relatively expensive upgrade. I suppose you could view it
as giving you some fans for a future project, if the
cooler (pump) ever wears out or leaks.

*******

Purely as a joke, if you want a fan, *this* is a fan. What
gives it the edge, is it is 37.5mm deep instead of 25mm.
So it's thicker than your average fan. That's what helps
the static pressure as well. They even make fans at 15mm
thick, but those would have practically no pressure at all.

http://www.circuittest.com/cfa1212038mb-12vdc-ball-bearing-120mm-fan.html

I have one of those. It was an impulse purchase at my local
electronics store. I bought it, because at the time it
was made of metal. So less smelly plastic to deal with.
But boy, is it loud! And it has a static pressure rating
of 5.6 mm H2O. So it's in the vacuum cleaner class. If you
had two of those, you'd have to leave the room

I run that fan off 7V, to make it liveable. And at 1 amp
current draw, you'd need a pretty nice rheobus to control
it. I wouldn't run that off a motherboard header. In my
case, it's powered off a Molex, and I've got a little
circuit to give 7V. Since it's not PWM, I don't have to
worry quite as much about running it at a lower voltage.

I don't think those Noctua fans would be a bad purchase,
but they may not give quite the same level of performance
as the originals. To me, the specs don't add up, and
common sense tells you they have to be less performing
than the originals.

As far as I know, there is only one fan design, that made
an attempt at a departure from standard design. And that
seems to have disappeared from the market. Its claim to fame,
was controlling turbulence near the fan blade, by the
smooth uniform air acceleration. The blade had a bit of a
"cone" theme to it, as the design of the blade
was defined by how it was to accelerate the air.
Those fans were expensive too, but partially because
they were made in Germany. I haven't seen any knockoffs
or attempts to copy them.

Paul

 

[Bill]

*******

Purely as a joke, if you want a fan, *this* is a fan. What
gives it the edge, is it is 37.5mm deep instead of 25mm.
So it's thicker than your average fan. That's what helps
the static pressure as well. They even make fans at 15mm
thick, but those would have practically no pressure at all.

http://www.circuittest.com/cfa1212038mb-12vdc-ball-bearing-120mm-fan.html

I have one of those. It was an impulse purchase at my local
electronics store. I bought it, because at the time it
was made of metal. So less smelly plastic to deal with.
But boy, is it loud! And it has a static pressure rating
of 5.6 mm H2O. So it's in the vacuum cleaner class. If you
had two of those, you'd have to leave the room

Yes, that's a fan.. : ) Those hardware "extremes" help keep the
matter interesting.
The "heartiest" thing I use my computer for is for running compilers or
math software similar to
Mathematica, or Matlab. So I may run my computer hard for 10-30 seconds
at a time.
Once upon a time, I ran simulation programs that I would let run for a
day or two, but not lately.

I have been experimenting recently with virtualbox and virtual machines
(which is an easy way to put Linux, for instance, on a Windows desktop),
so I secured more RAM than usual (16GB). The price difference between
8GB and 16GB was not so much. I'm in a situation now where I can't
predict exactly what I'll be doing with my computer for in the next few
years, but I think my new one should be plenty adequate--fan hiccups or
not!

I upgraded from an Intel i7-860 (2.8 Ghz) to Intel i7-3740K (4.0 Ghz).
I'm not sure whether it's the new mouse driver or the clock speed, or
something else, but I can definitely "feel" an improvement in the way
that my newsreader/email-client works. It's practically, "smooth as
glass". I don't even particularly like, and maybe can't afford anyway,
"responsiveness" in a car, but I enjoy it in a computer. "Quiet" is a
desirable attribute for a computer to have too, IMO. I may build a
little partition, a pseudo-wall, so that I don't interrupt it if it
wants to read my e-books! ; )

Cheers,
Bill

 

[Bill]

I upgraded from an Intel i7-860 (2.8 Ghz) to Intel i7-3740K (4.0 Ghz).

Opps, the 3740 probably hasn't been invented. I'm sure you know what I
meant.
I was off by 1050K!

 

[Paul]

Opps, the 3740 probably hasn't been invented. I'm sure you know what I
meant.
I was off by 1050K!

Yes. There aren't too many that do 4GHz as the stock value.

This has a higher clock, but benches lower. And it also uses
a lot of power.

http://products.amd.com/en-us/DesktopCPUDetail.aspx?id=840

http://www.cpubenchmark.net/high_end_cpus.html

Passmark
4790K 11,251
FX-9590 10,220

Paul

 

[Bill]

Yes. There aren't too many that do 4GHz as the stock value.

This has a higher clock, but benches lower. And it also uses
a lot of power.

http://products.amd.com/en-us/DesktopCPUDetail.aspx?id=840

http://www.cpubenchmark.net/high_end_cpus.html

Passmark
4790K 11,251
FX-9590 10,220

Paul

Ah, AMD. I had an AMD Athlon XP 2100+, when it was "hot"! Memory Lane!

http://en.wikipedia.org/wiki/List_of_AMD_Athlon_XP_microprocessors

I was surprised (in looking at the link) just how many different models
were made.

My current build was my 4th one. Thanks again for your help! Flasherly,
Thank you too!

Bill

 

[Flasherly]

Flasherly,
Thank you too!

And thank you for that listing. My first Athlon may even not be
within Wiki's scope.

A slotted Athlon that took a dedicated CPU slot (alongside ISA/Memory
slots), and, shortly eclipsed by a socketed same model. May have not
even been an Athlon, per se;- came out around/after the K6s (350Mhz).

Around a sub-1000MHz or slightly lower.

Memory Lane, alright, just can't remember, now, all the rooms in
Memory Motel I've frequented. Do recall that slotted one, though, it
was a horrible setup to seat, prone to lose connections and need
reseating before it would boot.

Oh, well - I suppose it's always good to know that would be X200
magnitude faster than an Intel 8088 4.77Mhz, or, an all-nighter to
convert archives, arc to zip, residing on perhaps half a 20Meg MFM
Seagate with its proprietary ISA interface controller.

 

[Bill]

My first Athlon may even not be within Wiki's scope. A slotted Athlon
that took a dedicated CPU slot (alongside ISA/Memory slots), and,
shortly eclipsed by a socketed same model. May have not even been an
Athlon, per se;- came out around/after the K6s (350Mhz). Around a
sub-1000MHz or slightly lower. Memory Lane, alright, just can't
remember, now, all the rooms in Memory Motel I've frequented.

My second computer was a CDC Cyber 172, which I programmed in Fortran
and assembly. From there I upgraded to a Digital (may they rest in
peace) Vax Cluster. Then to SUN (may they also rest in peace)
workstations... lol I did NOT anticipate Apple being successful with
computers at their onset (who would want them?), and this keeps my
"investor acumen" in check. Of course, at that point I had never heard
more than a monochrome beep from a computer, literally!

My first computer I dialed to with a phone modem from a dumb terminal,
in 1975, and I'm not sure what was on the other end. I saved my Basic
All-purpose Symbolic Instruction Code programs on paper tape.

But for all of my admittedly limited computer hardware experience,
everything else pales compared to watching a "card reader" work through
a big stack of cards!

Bill

 

[Flasherly]

My second computer was a CDC Cyber 172, which I programmed in Fortran
and assembly. From there I upgraded to a Digital (may they rest in
peace) Vax Cluster. Then to SUN (may they also rest in peace)
workstations... lol I did NOT anticipate Apple being successful with
computers at their onset (who would want them?), and this keeps my
"investor acumen" in check. Of course, at that point I had never heard
more than a monochrome beep from a computer, literally!

My first computer I dialed to with a phone modem from a dumb terminal,
in 1975, and I'm not sure what was on the other end. I saved my Basic
All-purpose Symbolic Instruction Code programs on paper tape.

But for all of my admittedly limited computer hardware experience,
everything else pales compared to watching a "card reader" work through
a big stack of cards!

Bill

Pretty much the same scenario here, except a little later;- tho' did
work a stint at an Evacuation Control Center with its
recording-keeping stored via cards. Came back, awhile later,
exchanging/updating a handheld calculator given to me. Found a
calculator with a subset of Basic interesting, a programming language
I'd never seen before;- took it home, scratching my head, but wouldn't
take no for an "answer" until figuring it out. Bought a couple
computers not long after, happenstance, in time immediately to also
exchange the first when visiting an electronics repair shop, upon
seeing my first IBM PC. So I bought a brandname Intel-powered PC. Sold
that within a year, by the time I was figuring how to update/augment
hardware on the thing, when I built my first from parts alone. Every
computer since I've built off parts.

Little later than the CDC's with a dual-core potential of 40Mhz and
provisions for a bank of peripheral processors. 8" glass harddisks
and all kinds of esoteric and prohibitively expensive setups. There's
a saying 'the revolution' was started in garages by tinkerers,
bringing computing out into a much broader front of popularity,
business, then, only could fractionally compare by sheer magnitude
ensuing. Little more than tinkerer, if to include myself and think
back to some of those crazy prices I would cough up for my digital
fix. Really. Got so bad for awhile I'd hustle them, just as quick as
I could see an update out of flipping one I'd just built.

Also did repair work on the side in a computer shop run by a former
NASA "assembler type," IT upgrades at local government offices, only
kept it at a certain distance, informally augmenting/expanding what I
was already doing for a few select individuals;- too interesting to
exchange for a stricter work disciple and a certain latitude of focus
to freely approach them. Computers have provided me certainly with
enough entertainment not to forget why.