submerged pc
48 views
Skip to first unread message
Matt Wittmann
Oct 30, 2012, 10:40:29 AM10/30/12
to milwaukee...@googlegroups.com
Is anybody interested in a group project to submerge an pc in oil . We could showcase it in the space or get attention for the space. Just laying some ideas down. Just think we could add leds ooooooohhh.
Joseph Bozarth
Oct 30, 2012, 10:42:41 AM10/30/12
to milwaukee...@googlegroups.com
Yes. I have been wanting to do one of these for a while I could just never justify the cost being a poor college student up until 6 months ago.
--
Joseph Bozarth
On Tue, Oct 30, 2012 at 9:40 AM, Matt Wittmann <mfran...@gmail.com> wrote:
Is anybody interested in a group project to submerge an pc in oil . We could showcase it in the space or get attention for the space. Just laying some ideas down. Just think we could add leds ooooooohhh.
--
Joseph Bozarth
Kevin Crowley
Oct 30, 2012, 11:04:18 AM10/30/12
to milwaukee...@googlegroups.com
Don't forget the overclocking and the extra cooling. Dismantled room air conditioner would work.
--
Brent Bublitz
Oct 30, 2012, 11:06:08 AM10/30/12
to milwaukee...@googlegroups.com
Kevin, the point of doing an oil submersion is to provide near perfect
cooling. The oil has a much greater thermal wicking than air and fans.
Just have to keep the oil circulating somehow.
> --
>
>
>
cooling. The oil has a much greater thermal wicking than air and fans.
Just have to keep the oil circulating somehow.
>
>
>
Brent Bublitz
Oct 30, 2012, 11:08:24 AM10/30/12
to milwaukee...@googlegroups.com
You would need a oil tight structure, a few gallons of mineral oil and
a pump of some kind.
The hard drive should stay dry and any fans have to get removed, but
since mineral oil is not conductive, you could just dunk the PC in at
that point.
a pump of some kind.
The hard drive should stay dry and any fans have to get removed, but
since mineral oil is not conductive, you could just dunk the PC in at
that point.
Matt Wittmann
Oct 30, 2012, 11:09:49 AM10/30/12
to milwaukee...@googlegroups.com
Whos in.
--
Joseph Bozarth
Oct 30, 2012, 11:11:15 AM10/30/12
to milwaukee...@googlegroups.com
I am. Any yeah. it is really not that hard of a thing to do. it just becomes a gigantic pain if you need to change a part.
Kevin Crowley
Oct 30, 2012, 11:31:47 AM10/30/12
to milwaukee...@googlegroups.com
Cooling the oil still increases the dissipation of heat from the components. I realize my experience for this sort of thing was from the early 90s but a 386 in a small aquarium maintained an ambient oil temp of ~85F I don't remember the exact temp of the CPU but it was considerably higher than that. When the oil was cooled to ~40F the CPU temp dropped proportionally.
--
Have Blue
Oct 30, 2012, 12:01:06 PM10/30/12
to milwaukee...@googlegroups.com
The thermal conductivity of the oil
will still be just about the same whether cooled or ambient - the
only way that the amount of heat removed from the components might
vary is if convection loops are more vigorous in the cooled oil
state.
--
Brent Bublitz
Oct 30, 2012, 12:20:42 PM10/30/12
to milwaukee...@googlegroups.com
I'll add that cooling the oil will generally increase it's viscosity,
making proper circulation more difficult. I'm sure there is a sweet
spot in oil temperature/viscosity/convection for proper cooling.
> --
>
>
>
making proper circulation more difficult. I'm sure there is a sweet
spot in oil temperature/viscosity/convection for proper cooling.
>
>
>
Tom Gralewicz
Oct 30, 2012, 12:39:10 PM10/30/12
to milwaukee...@googlegroups.com
Rather than sticking it in a big tank of oil, why not make a custom acrylic case with less than 1" of oil around the boards.
An impeller in the oil magnetically coupled through the acrylic can provide circulation.
Kevin Crowley
Oct 30, 2012, 12:42:19 PM10/30/12
to milwaukee...@googlegroups.com
Mineral oil at ~40F will still flows pretty readily. It get a bit gooey and thick in the freezer.
--
Kevin Crowley
Oct 30, 2012, 12:44:34 PM10/30/12
to milwaukee...@googlegroups.com
I like the impeller idea. A lot.
--
Matt Wittmann
Oct 30, 2012, 2:46:09 PM10/30/12
to milwaukee...@googlegroups.com
I was going to make it out of lexan i have.
--
Kevin Crowley
Oct 30, 2012, 3:57:13 PM10/30/12
to milwaukee...@googlegroups.com
How much and how thick?
--
Matt Wittmann
Oct 30, 2012, 6:33:04 PM10/30/12
to milwaukee...@googlegroups.com
Enough. I think.
--
Shane
Nov 2, 2012, 12:34:14 AM11/2/12
to milwaukeemakerspace
How 'bout we build one of these: http://www.popularmechanics.com/technology/how-to/build-pc/4318067
> >>>> >> wrote:
>
> >>>> >>> Whos in.
>
> >>>> >>> On Oct 30, 2012 10:08 AM, "Brent Bublitz" <phot...@gmail.com>
> On Oct 30, 2012 2:57 PM, "Kevin Crowley" <caoha...@gmail.com> wrote:
>
>
>
>
>
>
>
> > How much and how thick?
>
>
>
>
>
>
>
>
> > How much and how thick?
>
> > On Tue, Oct 30, 2012 at 12:46 PM, Matt Wittmann <mfrankwi...@gmail.com>wrote:
>
> >> I was going to make it out of lexan i have.
> >> On Oct 30, 2012 11:39 AM, "Tom Gralewicz" <m...@ieee.org> wrote:
>
> >>> Rather than sticking it in a big tank of oil, why not make a custom
> >>> acrylic case with less than 1" of oil around the boards.
>
> >>> An impeller in the oil magnetically coupled through the acrylic can
> >>> provide circulation.
>
>
> >> I was going to make it out of lexan i have.
> >> On Oct 30, 2012 11:39 AM, "Tom Gralewicz" <m...@ieee.org> wrote:
>
> >>> Rather than sticking it in a big tank of oil, why not make a custom
> >>> acrylic case with less than 1" of oil around the boards.
>
> >>> An impeller in the oil magnetically coupled through the acrylic can
> >>> provide circulation.
>
> >>> On Tue, Oct 30, 2012 at 11:20 AM, Brent Bublitz <phot...@gmail.com>wrote:
>
> >>>> I'll add that cooling the oil will generally increase it's viscosity,
> >>>> making proper circulation more difficult. I'm sure there is a sweet
> >>>> spot in oil temperature/viscosity/convection for proper cooling.
>
> >>>> On Tue, Oct 30, 2012 at 11:01 AM, Have Blue <haveb...@airsoldier.com>
>
> >>>> I'll add that cooling the oil will generally increase it's viscosity,
> >>>> making proper circulation more difficult. I'm sure there is a sweet
> >>>> spot in oil temperature/viscosity/convection for proper cooling.
>
> >>>> wrote:
> >>>> > The thermal conductivity of the oil will still be just about the same
> >>>> > whether cooled or ambient - the only way that the amount of heat
> >>>> removed
> >>>> > from the components might vary is if convection loops are more
> >>>> vigorous in
> >>>> > the cooled oil state.
>
> >>>> > On 10/30/2012 10:31 AM, Kevin Crowley wrote:
>
> >>>> > Cooling the oil still increases the dissipation of heat from the
> >>>> components.
> >>>> > I realize my experience for this sort of thing was from the early 90s
> >>>> but a
> >>>> > 386 in a small aquarium maintained an ambient oil temp of ~85F I
> >>>> don't
> >>>> > remember the exact temp of the CPU but it was considerably higher
> >>>> than that.
> >>>> > When the oil was cooled to ~40F the CPU temp dropped proportionally.
>
> >>>> > On Tue, Oct 30, 2012 at 9:11 AM, Joseph Bozarth <boza...@gmail.com>
> >>>> > The thermal conductivity of the oil will still be just about the same
> >>>> > whether cooled or ambient - the only way that the amount of heat
> >>>> removed
> >>>> > from the components might vary is if convection loops are more
> >>>> vigorous in
> >>>> > the cooled oil state.
>
> >>>> > On 10/30/2012 10:31 AM, Kevin Crowley wrote:
>
> >>>> > Cooling the oil still increases the dissipation of heat from the
> >>>> components.
> >>>> > I realize my experience for this sort of thing was from the early 90s
> >>>> but a
> >>>> > 386 in a small aquarium maintained an ambient oil temp of ~85F I
> >>>> don't
> >>>> > remember the exact temp of the CPU but it was considerably higher
> >>>> than that.
> >>>> > When the oil was cooled to ~40F the CPU temp dropped proportionally.
>
> >>>> wrote:
>
> >>>> >> I am. Any yeah. it is really not that hard of a thing to do. it just
> >>>> >> becomes a gigantic pain if you need to change a part.
>
> >>>> >> On Tue, Oct 30, 2012 at 10:09 AM, Matt Wittmann <
> >>>> mfrankwi...@gmail.com>
>
> >>>> >> I am. Any yeah. it is really not that hard of a thing to do. it just
> >>>> >> becomes a gigantic pain if you need to change a part.
>
> >>>> >> On Tue, Oct 30, 2012 at 10:09 AM, Matt Wittmann <
> >>>> >> wrote:
>
> >>>> >>> Whos in.
>
> >>>> >>> On Oct 30, 2012 10:08 AM, "Brent Bublitz" <phot...@gmail.com>
> >>>> wrote:
>
> >>>> >>>> You would need a oil tight structure, a few gallons of mineral oil
> >>>> and
> >>>> >>>> a pump of some kind.
> >>>> >>>> The hard drive should stay dry and any fans have to get removed,
> >>>> but
> >>>> >>>> since mineral oil is not conductive, you could just dunk the PC in
> >>>> at
> >>>> >>>> that point.
>
> >>>> >>>> On Tue, Oct 30, 2012 at 10:06 AM, Brent Bublitz <phot...@gmail.com
>
> >>>> >>>> You would need a oil tight structure, a few gallons of mineral oil
> >>>> and
> >>>> >>>> a pump of some kind.
> >>>> >>>> The hard drive should stay dry and any fans have to get removed,
> >>>> but
> >>>> >>>> since mineral oil is not conductive, you could just dunk the PC in
> >>>> at
> >>>> >>>> that point.
>
>
> >>>> >>>> wrote:
> >>>> >>>> > Kevin, the point of doing an oil submersion is to provide near
> >>>> perfect
> >>>> >>>> > cooling. The oil has a much greater thermal wicking than air and
> >>>> fans.
> >>>> >>>> > Just have to keep the oil circulating somehow.
>
> >>>> >>>> > On Tue, Oct 30, 2012 at 10:04 AM, Kevin Crowley <
> >>>> caoha...@gmail.com>
> >>>> >>>> wrote:
> >>>> >>>> > Kevin, the point of doing an oil submersion is to provide near
> >>>> perfect
> >>>> >>>> > cooling. The oil has a much greater thermal wicking than air and
> >>>> fans.
> >>>> >>>> > Just have to keep the oil circulating somehow.
>
> >>>> >>>> > On Tue, Oct 30, 2012 at 10:04 AM, Kevin Crowley <
> >>>> >>>> > wrote:
> >>>> >>>> >> Don't forget the overclocking and the extra cooling.
> >>>> Dismantled room
> >>>> >>>> >> air
> >>>> >>>> >> conditioner would work.
>
> >>>> >>>> >> On Tue, Oct 30, 2012 at 8:42 AM, Joseph Bozarth <
> >>>> boza...@gmail.com>
> >>>> >>>> >> Don't forget the overclocking and the extra cooling.
> >>>> Dismantled room
> >>>> >>>> >> air
> >>>> >>>> >> conditioner would work.
>
> >>>> >>>> >> On Tue, Oct 30, 2012 at 8:42 AM, Joseph Bozarth <
> >>>> >>>> >> wrote:
>
> >>>> >>>> >>> Yes. I have been wanting to do one of these for a while I
> >>>> could just
> >>>> >>>> >>> never
> >>>> >>>> >>> justify the cost being a poor college student up until 6
> >>>> months ago.
>
> >>>> >>>> >>> On Tue, Oct 30, 2012 at 9:40 AM, Matt Wittmann
> >>>> >>>> >>> <mfrankwi...@gmail.com>
>
> >>>> >>>> >>> Yes. I have been wanting to do one of these for a while I
> >>>> could just
> >>>> >>>> >>> never
> >>>> >>>> >>> justify the cost being a poor college student up until 6
> >>>> months ago.
>
> >>>> >>>> >>> On Tue, Oct 30, 2012 at 9:40 AM, Matt Wittmann
> >>>> >>>> >>> wrote:
>
> >>>> >>>> >>>> Is anybody interested in a group project to submerge an pc in
> >>>> oil .
> >>>> >>>> >>>> We
> >>>> >>>> >>>> could showcase it in the space or get attention for the
> >>>> space. Just
> >>>> >>>> >>>> laying
> >>>> >>>> >>>> some ideas down. Just think we could add leds ooooooohhh.
>
> >>>> >>>> >>>> --
>
> >>>> >>>> >>> --
> >>>> >>>> >>> Joseph Bozarth
> >>>> >>>> >>> boza...@gmail.com
>
> >>>> >>>> >>>> Is anybody interested in a group project to submerge an pc in
> >>>> oil .
> >>>> >>>> >>>> We
> >>>> >>>> >>>> could showcase it in the space or get attention for the
> >>>> space. Just
> >>>> >>>> >>>> laying
> >>>> >>>> >>>> some ideas down. Just think we could add leds ooooooohhh.
>
> >>>> >>>> >>>> --
>
> >>>> >>>> >>> --
> >>>> >>>> >>> Joseph Bozarth
> >>>> >>>> >>> 1(262)498-9762
>
> >>>> >>>> >>> --
>
> >>>> >>>> >> --
>
> >>>> >>>> --
>
> >>>> >>> --
>
> >>>> >> --
> >>>> >> Joseph Bozarth
> >>>> >> boza...@gmail.com
>
> >>>> >>>> >>> --
>
> >>>> >>>> >> --
>
> >>>> >>>> --
>
> >>>> >>> --
>
> >>>> >> --
> >>>> >> Joseph Bozarth
Kevin Crowley
Nov 2, 2012, 10:50:13 AM11/2/12
to milwaukee...@googlegroups.com
That would be a candidate. Might just a tad more plastic and aluminum.
--
Reply all
Reply to author
Forward
0 new messages