Wooden case for PC. Should I?

Page 2 - Seeking answers? Join the Tom's Hardware community: where nearly two million members share solutions and discuss the latest tech.
Archived from groups: alt.comp.hardware.homebuilt (More info?)

On 27 Aug 2005 23:01:34 -0700, rjerry@myway.com wrote:

>ok. I've thought about it some more and this is my current design.
>
>I will use wood only as asthetics, but the main material of the
>computer will be out of aluminum sheets(not sure how thick). That is
>what a computer case is typically made out of, correct? The wood, if
>desired, will be added on top of the aluminum material. My only problem
>is that "I've never done this before", well...with aluminum that is.
>Great learning experience, but *could* be a b--- to pull off. But
>simplicity is the key here. My main question is:
>
>1.> HOW do I weld aluminum together? I've thought of either using a
>torch or maybe better yet that mixing compound that you usually find at
>auto part stores. But I don't know how strong that stuff will be. I
>mean if I drop this casing will it break? My experience with those
>compounds is that they do not hold up well when it comes to two metals
>twisting against each other. I know what ya'll are thinking..."DON'T
>TWIST THE CASE!". But seeing how little torque I applied to the two
>metals...makes me wonder how good of a adhesion there will be for a
>casing. Perhaps you know of a good compound or would torching work
>better?
>And to give you an idea of my casing design, just imagine square sheets
>of aluminum...welded together to form a box. I know...revolutionary!

I'd give up on the idea of welding ali...about the best you can do I
soft-solder it using a specialised solder and a rather poisonous flux.
It won't be a neat job either, or cheap.
Using a resin compound isn't likely to work either, not without
substantial bracing.
Unless you have access to an accurate bending machine, I'd advise
sticking to flat plates.

>
>2.> anybody know of a place where I can buy affordable riser cards?
>Some of the risers are pretty pricey!
>
>3.> if I put a monitor over this case, will thick aluminum be enough to
>hold it up? Or should I put a beams underneath to spread out the
>pressure from the monitor?

At a rough guesstimate you'd need 5mm ali unsupported.
Best bet would be to include cross beams ( you'll need beams anyway to
form the inner matrix of the case.
>
>I hope these are the only issues that I need to alert myself to. I want
>this project to be very simple. I've been taking my pretty time here
>about how I want to build this computer and what I want for a computer.
>And now I have the image of the custom case, in my head, and I've made
>the design so utterly simple. But now I need to know about the 3
>concerns that mentioned above. comments?

You know what I'd be inclined to do?
Adapt a case design for something else.
I'm sitting here looking at an old Akai cassette deck - the case is in
brushed ali, and all that would be needed to mod it would be to remove
and replace the front and rear.
The rear would be easy - just rip the back off an old desktop pc.
The front could be cut from wood - and about the only other mod
required would be a beam across the top to stiffen it.

If you looked long enough you might even find one of those old 70's
decks that were lined with solid mahogany ( Pioneer made a few of
these ).
They're not much smaller than a modern mini-tower though...

Regards,



--
Stephen Howard - Woodwind repairs & period restorations
www.shwoodwind.co.uk
Emails to: showard{whoisat}shwoodwind{dot}co{dot}uk
 
Archived from groups: alt.comp.hardware.homebuilt (More info?)

Conor wrote:

> In article <11h4gie7mnshca@corp.supernews.com>, David Maynard says...
>
>
>>I'm not thinking of fan failure and other 'normal' failures as those are
>>catered for. It's the 'unexpected' kind that make for unforeseen problems.
>>
>
> In that case, all computers are in the same boat.

In that something can go wrong with any of them, yes. Never said otherwise.
 
Archived from groups: alt.comp.hardware.homebuilt (More info?)

Conor wrote:

> In article <njt3h1hvg5q3mpbfeiq8n21c738fq0fhou@4ax.com>, Mxsmanic
> says...
>
>
>>I agree. But having some sort of protection in the CPU is an extra
>>layer of security. All fans fail eventually, and I assume the MTBF of
>>a CPU is greater than even that of the best fans (?).
>
>
> However it has been proven in the P4 that if you run an app that has a
> high CPU utilisation such as some 3D games, the CPU will throttle
> itself when it gets hot and bring the system to a crawl.
>

Only if you have inadequate cooling to begin with. In which case 'crawling'
is better than crashing.
 
Archived from groups: alt.comp.hardware.homebuilt (More info?)

kony writes:

> Both manufacturers sold retail and oem (bare) product. Buy
> an OEM Intel CPU and you don't get a fan either- same diff.

Well, the cheapo PCs I bought had an AMD processor, but the fan looked
generic, and the heatsink might have been as well. The boxed Intel
processors I bought to build replacements for the cheapos included
fans.

--
Transpose mxsmanic and gmail to reach me by e-mail.
 
Archived from groups: alt.comp.hardware.homebuilt (More info?)

Conor writes:

> ROFLMAO. 6502 and Z80 CPUS didn't need fans or heatsinks and they
> regularly overheated in arcade machines.

The arcade machines were poorly ventilated.

> So do AMD Retail.

I don't know what SonBook used to build their cheapo PC, but it had a
very cheap generic sleeve-bearing fan that failed within a week or so
(as did its replacement).

--
Transpose mxsmanic and gmail to reach me by e-mail.
 
Archived from groups: alt.comp.hardware.homebuilt (More info?)

Conor writes:

> However it has been proven in the P4 that if you run an app that has a
> high CPU utilisation such as some 3D games, the CPU will throttle
> itself when it gets hot and bring the system to a crawl.

Would you rather that it overheat and suffer permanent damage instead?

--
Transpose mxsmanic and gmail to reach me by e-mail.
 
Archived from groups: alt.comp.hardware.homebuilt (More info?)

In article <11h64vvho2plfad@corp.supernews.com>, David Maynard says...

> > However it has been proven in the P4 that if you run an app that has a
> > high CPU utilisation such as some 3D games, the CPU will throttle
> > itself when it gets hot and bring the system to a crawl.
> >
>
> Only if you have inadequate cooling to begin with.

Actually, with Intels own heatsink and fan.


--
Conor

"You're not married, you haven't got a girlfriend and you've never seen
Star Trek? Good Lord!" - Patrick Stewart, Extras.
 
Archived from groups: alt.comp.hardware.homebuilt (More info?)

Conor wrote:
> In article <11h64vvho2plfad@corp.supernews.com>, David Maynard says...
>
>
>>>However it has been proven in the P4 that if you run an app that has a
>>>high CPU utilisation such as some 3D games, the CPU will throttle
>>>itself when it gets hot and bring the system to a crawl.
>>>
>>
>>Only if you have inadequate cooling to begin with.
>
>
> Actually, with Intels own heatsink and fan.
>
>

Actually, no. Unless you have inadequate cooling.
 
Archived from groups: alt.comp.hardware.homebuilt (More info?)

In article <tfg6h1ptnqbfn59598ii8r1ascnlapos95@4ax.com>, Mxsmanic
says...
> Conor writes:
>
> > ROFLMAO. 6502 and Z80 CPUS didn't need fans or heatsinks and they
> > regularly overheated in arcade machines.
>
> The arcade machines were poorly ventilated.
>
REally? 12" square PCB mounted on a board in a cabinet that is 2ft
square by 5ft high and only a monitor in mounted about a foot above it.


--
Conor

"You're not married, you haven't got a girlfriend and you've never seen
Star Trek? Good Lord!" - Patrick Stewart, Extras.
 
Archived from groups: alt.comp.hardware.homebuilt (More info?)

Conor wrote:

> In article <tfg6h1ptnqbfn59598ii8r1ascnlapos95@4ax.com>, Mxsmanic
> says...
>
>>Conor writes:
>>
>>
>>>ROFLMAO. 6502 and Z80 CPUS didn't need fans or heatsinks and they
>>>regularly overheated in arcade machines.
>>
>>The arcade machines were poorly ventilated.
>>
>
> REally? 12" square PCB mounted on a board in a cabinet that is 2ft
> square by 5ft high and only a monitor in mounted about a foot above it.
>
>

If they overheated, then yes.
 
Archived from groups: alt.comp.hardware.homebuilt (More info?)

In article <7ig6h1pgv269udnbforou5l8ic1jovu8d7@4ax.com>, Mxsmanic
says...
> Conor writes:
>
> > However it has been proven in the P4 that if you run an app that has a
> > high CPU utilisation such as some 3D games, the CPU will throttle
> > itself when it gets hot and bring the system to a crawl.
>
> Would you rather that it overheat and suffer permanent damage instead?
>
Athlon64 runs alot cooler than a P4.

--
Conor

"You're not married, you haven't got a girlfriend and you've never seen
Star Trek? Good Lord!" - Patrick Stewart, Extras.
 
Archived from groups: alt.comp.hardware.homebuilt (More info?)

On Mon, 29 Aug 2005 19:14:01 +0200, Mxsmanic
<mxsmanic@gmail.com> wrote:

>Conor writes:
>
>> However it has been proven in the P4 that if you run an app that has a
>> high CPU utilisation such as some 3D games, the CPU will throttle
>> itself when it gets hot and bring the system to a crawl.
>
>Would you rather that it overheat and suffer permanent damage instead?


The underlying theme here should be that not *everyone*
should be designing systems... and no, I'm not pointing
fingers at any particular newsgroup participants, rather a
sweeping declaration that any *desktop* system with a P4
that ends up throttling needs rethought, sound not be left
running beyond the first moment the design is tested and
shown to be ineffective of sustaining a full load running
state.
 
Archived from groups: alt.comp.hardware.homebuilt (More info?)

Conor writes:

> REally? 12" square PCB mounted on a board in a cabinet that is 2ft
> square by 5ft high and only a monitor in mounted about a foot above it.

Sounds poorly ventilated to me. How many fans, and where, and what
was the path and volume of the airflow inside?

--
Transpose mxsmanic and gmail to reach me by e-mail.
 
Archived from groups: alt.comp.hardware.homebuilt (More info?)

Conor writes:

> Actually, with Intels own heatsink and fan.

The processor should never overheat with its own heatsink and fan if
ambient temperature is within spec and ambient air can easily get to
the CPU fan. Obviously if you install it in a hermetically sealed box
it will overheat with or without a fan.

--
Transpose mxsmanic and gmail to reach me by e-mail.
 
Archived from groups: alt.comp.hardware.homebuilt (More info?)

Conor writes:

> Athlon64 runs alot cooler than a P4.

The processor has to run well within spec; the actual temperature
isn't very important.

--
Transpose mxsmanic and gmail to reach me by e-mail.
 
Archived from groups: alt.comp.hardware.homebuilt (More info?)

On Tue, 30 Aug 2005 01:55:03 +0200, Mxsmanic
<mxsmanic@gmail.com> wrote:

>Conor writes:
>
>> Athlon64 runs alot cooler than a P4.
>
>The processor has to run well within spec; the actual temperature
>isn't very important.


That's true, BUT there's also a practical maximum amount of
heat that can be removed from a small core with an
air-cooler. Since air-coolers are still the cheapest way
known to get the job done, we see Intel's P4 pushing the
limits, requiring systems to be set up with more attention
to actual cooling requirements... an attention often not
provided.
 
Archived from groups: alt.comp.hardware.homebuilt (More info?)

kony writes:

> That's true, BUT there's also a practical maximum amount of
> heat that can be removed from a small core with an
> air-cooler. Since air-coolers are still the cheapest way
> known to get the job done, we see Intel's P4 pushing the
> limits, requiring systems to be set up with more attention
> to actual cooling requirements... an attention often not
> provided.

Systems have had poor cooling for years, as the example of arcade
games shows (Z-80s don't get that warm!). Conversely, a P4 is fine as
long as it's cooled properly.

I don't think the trend towards ever-increasing cooling requirements
is a good one, but since faster processors encourage software bloat,
and software bloat in turn requires faster processors, it's hard to
see how the trend will be stopped. If water cooling becomes
commonplace the cost of computers is going to go up and their
reliability is going to go down, and the average user isn't really
going to gain much in exchange for this.

--
Transpose mxsmanic and gmail to reach me by e-mail.
 
Archived from groups: alt.comp.hardware.homebuilt (More info?)

In article <vt77h158vn7ut7b5prk3ad8mdvfgr04807@4ax.com>, Mxsmanic
says...
> Conor writes:
>
> > REally? 12" square PCB mounted on a board in a cabinet that is 2ft
> > square by 5ft high and only a monitor in mounted about a foot above it.
>
> Sounds poorly ventilated to me. How many fans, and where, and what
> was the path and volume of the airflow inside?
>
Have you ever seen an 80's arcade videogame?


--
Conor

"You're not married, you haven't got a girlfriend and you've never seen
Star Trek? Good Lord!" - Patrick Stewart, Extras.
 
Archived from groups: alt.comp.hardware.homebuilt (More info?)

In article <11h72lpov40qp48@corp.supernews.com>, David Maynard says...
> Conor wrote:
> > In article <11h64vvho2plfad@corp.supernews.com>, David Maynard says...
> >
> >
> >>>However it has been proven in the P4 that if you run an app that has a
> >>>high CPU utilisation such as some 3D games, the CPU will throttle
> >>>itself when it gets hot and bring the system to a crawl.
> >>>
> >>
> >>Only if you have inadequate cooling to begin with.
> >
> >
> > Actually, with Intels own heatsink and fan.
> >
> >
>
> Actually, no. Unless you have inadequate cooling.
>
>
Go look on the web...plenty of examples on reputable gamers sites.
--
Conor

"You're not married, you haven't got a girlfriend and you've never seen
Star Trek? Good Lord!" - Patrick Stewart, Extras.
 
Archived from groups: alt.comp.hardware.homebuilt (More info?)

Conor wrote:
> In article <11h72lpov40qp48@corp.supernews.com>, David Maynard says...
>
>>Conor wrote:
>>
>>>In article <11h64vvho2plfad@corp.supernews.com>, David Maynard says...
>>>
>>>
>>>
>>>>>However it has been proven in the P4 that if you run an app that has a
>>>>>high CPU utilisation such as some 3D games, the CPU will throttle
>>>>>itself when it gets hot and bring the system to a crawl.
>>>>>
>>>>
>>>>Only if you have inadequate cooling to begin with.
>>>
>>>
>>>Actually, with Intels own heatsink and fan.
>>>
>>>
>>
>>Actually, no. Unless you have inadequate cooling.
>>
>>
>
> Go look on the web...plenty of examples on reputable gamers sites.

So there are plenty of examples with inadequate cooling.
 
Archived from groups: alt.comp.hardware.homebuilt (More info?)

In article <0087h1dgmmqjlqirffu9oaffafnugki3js@4ax.com>, Mxsmanic
says...
> Conor writes:
>
> > Actually, with Intels own heatsink and fan.
>
> The processor should never overheat with its own heatsink and fan

Which is precisely why Intel tried to introduce the BTX form factor
when they realised what a pup they had with the 3GHz+ P4. Ever seen the
heatsink and fan arrangement for one of those?



--
Conor

"You're not married, you haven't got a girlfriend and you've never seen
Star Trek? Good Lord!" - Patrick Stewart, Extras.
 
Archived from groups: alt.comp.hardware.homebuilt (More info?)

In article <11h9p257iasrsc0@corp.supernews.com>, David Maynard says...

> So there are plenty of examples with inadequate cooling.
>
Nope. Plenty of examples with Intels own heatsink and fan working on a
test bench, i.e not even in a case.


--
Conor

"You're not married, you haven't got a girlfriend and you've never seen
Star Trek? Good Lord!" - Patrick Stewart, Extras.
 
Archived from groups: alt.comp.hardware.homebuilt (More info?)

Conor wrote:

> In article <11h9p257iasrsc0@corp.supernews.com>, David Maynard says...
>
>
>>So there are plenty of examples with inadequate cooling.
>>
>
> Nope. Plenty of examples with Intels own heatsink and fan working on a
> test bench, i.e not even in a case.

You making these arbitrary declarations doesn't mean any more than the last
reported sighting of Elvis.
 
Archived from groups: alt.comp.hardware.homebuilt (More info?)

On Wed, 31 Aug 2005 00:28:14 +0100, Conor
<conor.turton@gmail.com> wrote:

>In article <11h9p257iasrsc0@corp.supernews.com>, David Maynard says...
>
>> So there are plenty of examples with inadequate cooling.
>>
>Nope. Plenty of examples with Intels own heatsink and fan working on a
>test bench, i.e not even in a case.


Part of the problem with test benches is increased
recirculation of 'sink exhaust. Both Intel and AMD have
specified rear chassis exhaust fans behind the 'sink for
good reason. Even so, yes you do have to have all your
ducks lined up in a row to keep one cool enough, unlike
yesteryears' systems where the CPU stayed cool enough
despite poor system setup.
 
Archived from groups: alt.comp.hardware.homebuilt (More info?)

Conor writes:

> Have you ever seen an 80's arcade videogame?

Not the inside.

--
Transpose mxsmanic and gmail to reach me by e-mail.