X1000 CPU cooling

@Deniil

Didn't realize pipe coolers depended on cpu temp to fire them up, thanks for the info.


@Razel

I'd be very interested on how the new cooler works out for you. Monitor the temperature without a fan first. I've done this with the current heatsink, to hot for me without the fan.

@sundown

I had the very same idea after ordering, i'll do that in idle and get back to you.

@Deniil
taken from wikipedia (heatpipe article)
Quote:

So, for our system a "simple" Copper/Water based pipe would be more than sufficient...

@tonyw

Sorry for the late reply.

The pictures requested, can be found in in each MAGMAbreath close up pictures list on my website.

In regards to the way I made it, please allow me to keep this as a secret :)
But I think sooner or later the community will find what a good quality I deliver, so that should not play any role anymore.


Sincerely
Marcus

@Raziel

Thank you Raziel,

You received a PM already from me.

BTW, a first customer did leave a report for the MAGMAbreath on my Website.
I hope many more will follow!

www.marcus.computer/MAGMAbreath and click on the five stars.
(he did leave it on English only, so use BING or GOOGLE translator)


Sincerely
Marcus

@MARCOM

I wrote the report in GERMAN :)

Anyway, here is a brief summary:

I bought the MAGMABreath Super Silent 80 and replaced the original cooler and fan!
Now, I have a super silent X1000, which is so silent that you barely can hear her anymore :)


For those of you interested in details:

I have a Commodore Ckin Case with 1x 120mm fan at the front and 1x 120mm fan in the rear of the case, and additionaly 1x 190mm fan at the side door. My GFX card is a fanless RadeonHD 4850 passively cooled one.

I have done a series of tests with my X1000 and X1kTemp.docky to record temperatures of CPU, Core1, Core2 and Case. The following maximum values, I could observe with AOS4 running for several hours at 100% CPU load:


With side door fan enabled, but rotation reduced by half by controlling voltage:
Case: 30°C
CPU: 37°C
Core1: 43°C
Core2: 39°C

For comparision: the values with original cooler and fan and the rotation of the side fan not reduced at all, are all 10-11C higher!

With side door fan enabled, but with maximal reduced rotation:
Case: 33°C
CPU: 39°C
Core1: 45°C
Core2: 41°C

Side door fan disabled:
Case: 37°C
CPU: 43°C
Core1: 48°C
Core2: 44°C


For more detailed report, you can find here a report in German (click on "Bewertungen"):

MAGMABreath Super Silent 80


P.S. I should have mentioned that I have heaters in the floor and my X1000 is positioned on the floor as well. Hence, I am very satisfied by these cooling results despite the extra heat from below!

@nexus

After removing the cooling element do you need to add some thermal paste on the CPU before installing the new cooling element ? Could one replace the fan only and not replace the cooling element ?

@nexus

So where are the figures for *before* the conversion? You said "10-11C higher", but we can't compare them unless you give us the figures.

Your figures for *after* the conversion are all very consistent for the rise above (case) ambient:

Case 30, CPU +7, Core 1 +13, Core 2 +9
Case 33, CPU +6, Core 1 +12, Core 2 +8
Case 37, CPU +6, Core 1 +11, Core 2 +7

Those temperature rises are all pretty close.

But what were they before the conversion?

@tonyw

And here's a partial score from the Super Bowl:

Seahawks 24

@LyleHaze
That's great! Glad to see they are performing so well!!




er...




is there another team?

@tonyw

Nope, they beat themselves on the pitch

@Raziel

Quote:

I tested a heatpipe once (designed for a gfx-card) by holding it in my hand, expecting the other end to reach hand-temperatur pretty quickly, but that never happend. The heat transfer was obviously very poor at 35degC.

To make a pipe most efficient, it would have to cool something that is hotter than the boiling point, and cooled by something that is less that the boiling point. Only then will you get the maximum effect through the evaporation and condensation of the liquid/gas and fast transfer of the heat in the form of gasified medium.

Maybe water still is efficient with a diff between 40degC (CPU) and 20degC (ambient), I don't know, but it doesn't feel like you would get enough flow without a mechanical pump in this case. Having vacuum in the tube would lower the boiling point from 100degC to something lower, but how low?

@Deniil

That depends how hard your vaccuum is, to get it to boil at Room temperature you'd need to take the pressure down to 0.5 PSI

Based on this info

http://www.jbind.com/pdf/Cross-Reference-of-Boiling-Temps.pdf

@all

Quick test of the new MAGMACooler heatsink

@tonyw

Quote:

Sorry for late reply. I only can remember the temperature for Core1 and Core2 and with the following setup:

"With side door fan enabled, but rotation reduced by half by controlling
voltage"

With old cpu cooling element, the results were:

Core1: 49-50°C
Core2: 53-54°C

These numbers were the regular ones when my X1000 ran for a while. I never tried to measure the "maximum" numbers, so they might be even higher when cpu is permanently at 100% load.

As a reminder, the following were the numbers with the new MAGMABreath silent 80 running for at least 1h with maximum cpu load
and

"With side door fan enabled, but rotation reduced by half by controlling
voltage"

Case: 30°C
CPU: 37°C
Core1: 43°C
Core2: 39°C

So, indeed around 10°C less. Actually, I did not replace the fan and cooling element because of the heat. I replaced them to get a silent system. So, with not useing the side door fan, I have and indeed *silent* system and the temperatures are still a few degree lower than they were before.

Cheers,
nexus

The MAGMAcolosseum...may it be helpful for anyone interested


Thanks to Macrom for the first passive cooling solution on a NEMO.




This feedback references a Fractal Design Define R3 case.

First off...as you can see on the pictures, this thing is huge.
That comes with pros and cons.

Lets start with the disadvantages.

1) You can't reach the two top fans anymore without some fiddling.
The shielding plates are thin enough to still be installed and deinstalled,
but Fans in sizes of 120mm and 140mm are not (the only two sizes one can use
without creating an "air loop" around the fans), as you won't able to place the
space in the ceiling from either side.

The way to do it is to (ab)use one of the front 5,25" bays to reach a fan
through and install it by pushing it gently over the MAGMAcolosseum.
The Fans, given a "normal" height of 2.5cm max, will touch the
MAGMAcolosseum slightly but will not bend it in any direction, thus not
performing any more force to the mainboard and they also won't be limited in
it's airflow (the wings can freely rotate).

You should think about installing or removing those fans before you place and
secure the MAGMAcolosseum.

HINT:
Spare them, that thing does the cooling well enough to drop those fans.

2) It says in the installation manual to only use original ram modules, those
that hasn't been tampered with in any kind, no heatspreaders or other funny
stuff. The reason for this is that the height of those changed modules will
prevent the heatsink from actually making contact with the CPU at all.

ATTENTION:
If you are installing the MAGMAcolosseum without having the mainboard lying
flat in front of you, you won't be able to check if there is actually contact!
This is due to the cramped space once the heatsink is in place)

Scary Side-Story:
I installed the MAGMAcolosseum with Heatspreader RAMs (didn't read the manual
close enough) and missed that the heatsink didn't touch the CPU (there was at
least 1mm of air above the CPU).
The CPU still didn't go above 70°C in the CFE Idle Test.
(Awesome Design by P.A. Semi - Repect!)

3) Due to the sheer size of the MAGMAcolosseum you won't be able to use the
first PCI-E slot.

According to the X1000 manual the seond slot only uses 8 of the 16 lanes.

Despite the tests with GPU hogging games and programs (lol) i wasn't able to
encounter any slowdowns or drawbacks.
Even the direct comparision with a similar card as mine and the 2D Benchmark
Test didn't show any differences.
Purists and those you seek raw power might be put off by this fact.

An inquiry to the developer of the graphics card driver (Hans de Ruiter) got me
the following answer:

"It would halve the RAM <=>VRAM bandwidth. That will affect how fast data
is transferred to/from VRAM, so it will affect video playback and CPU
rendering. Whether it's noticeable in anything but benchmarks remains to
be seen.

It might also slow down submission of commands to the GPU, but that's
already limited because we're not using GART/DMA, so I doubt that this
will be noticeable."
(Thanks to Hans de Ruiter for his time)

4) After the installation it is not possible to replace or add any RAM modules
due to missing space between RAM modules and heatpipes.

Now for the advantages.

1) The installation is, like it was with it's little brother MAGMAcooler,
extremely well explained and very easy to do and should be able to be performed
by anyone.
The modifications done are well thought through and matches the needs of the
Nemo board perfectly. Furthermore are the used materials of a great quality.
(Metal, not some cheap plastic)

2) Due to the massive and superb manufacturing (by NoFan of South Korea) you
only need one fan which should be placed somewhere high up and in the back of
the case to support the natural convection of the MAGMAcolosseum.

HINT:
In conjunction with a passively cooled graphics card there is nothing stopping
you building a "nearly" completely passively cooled Amiga.

3) The possible fan in the side door can still be easily installed, as the
MAGMAcolosseum ends a little bit above it.

HINT:
Spare it

4) The cooling performance is awesome and it's passive.

Comparision figures:
(The first value corresponds to the measurement of the MAGMAcolosseum with only
*one* 120mm fan installed at the backside. The other two values were measured
with *five(!)* fans. (Reason? It got two hot for my taste!)
(2x 140mm Top, 120mm Backside, 140mm Side Door, 120mm on the CPU with a
size adaptor)
The whole setup is in a room with a temperature of approximately 21°C.

(The first braces hold the values measured with the MAGMAcooler)
(The second braces hold the values measured with the original heatsink/fan)

CFE (Idle):

Local: 29°C (31°C) (30°C)
Remote1:35°C (44°C) (46°C)
Remote2: 33°C (41°C) (43°C)
Remote3: 31°C (39°C) (41°C)

Gain:
MAGMAcollosseum --> Original heatsink/fan: ~23%
MAGMAcollosseum --> MAGMAcooler: ~20%

Workbench (Idle):

Case: 33°C (33°C) (34°C)
CPU: 35°C (41°C) (45°C)
Core1: 39°C (46°C) (50°C)
Core2: 37°C (43°C) (47°C)

Gain:
MAGMAcollosseum --> Original heatsink/fan: ~22%
MAGMAcollosseum --> MAGMAcooler: ~14%

Workbench (100% CPU Usage):

Case: 34°C (34°C) (35°C)
CPU: 36°C (43°C) (47°C)
Core1: 41°C (49°C) (54°C)
Core2: 39°C (46°C) (50°C)

Gain:
MAGMAcollosseum --> Original heatsink/fan: ~23%
MAGMAcollosseum --> MAGMAcooler: ~16%

The MAGMAcooler already adds a lot to the cooling, but it's still beaten by
the superior MAGMAcolosseum. And it's passive!

Additionally to the fact that case fans are more or less obsolete with this
solution and that it adds a whole deal to the lower noise gain, it's just a
feast for the eyes if you are a modder.

The Conclusion:
Thumbs way up for a clear recommendation to buy!

@MARCOM

finally got around to installing my MagmaBreath in my X1K and after 48 hours I'm very happy with the results in both sound reduction heat reduction so big thanks to Marcom Computer

Prior to the upgrade after 48 hours of constant use numbers were:
CPU- 47 deg.C
Core1-52
Core2-50

with MagmaBreath installed after 48 hrs I was pleased to see a significant decrease:

CPU- 38 deg.C
Core1-43
Core2-40

for those interested in the install I updated my Man Cave Ramblings blog

I also would like to know the CPU heatsink style. After finally coming across the rarely for sale x1000, and needing to repair damage caused by shipping company, I'd like to get a nice big cold heatsink kit.

It seems the magmabreath and magmacolluseum I had on my list for a long time have ceased to exist.

Any suggestions?

@billt

i believe Amiga On The Lake is US bought out the reminding stock of the CPU coolers so check them out...as per my post above and related blog post I would highly recommend them !

@billt

http://www.youtube.com/watch?v=ArJjI0F32rU

Would be nice to know what mounting holes the X1000 and X5000 uses

@Helloworld

you can't use that, no on-board graphic card.
and you need pci express ports so no...

What I did was transfare the dimensions using a marker pen, drilled holes in mounting bracket, got some bracket that a friend of my did not use for old Pentium. I'm sure you can brackets that can be modified.

I guess if have passive cooler like that you bend the motherboard, if stands up right, I think this might not be so good idea.