Alias_Go
New Member
- Total Posts : 7
- Reward points : 0
- Joined: 2021/04/24 12:46:34
- Status: offline
- Ribbons : 0
Has anyone built an enclosure to flood with nitrogen(gas) for subzero overclocking to eliminate condensation?
|
Flint 1760
Omnipotent Enthusiast
- Total Posts : 8295
- Reward points : 0
- Joined: 2009/04/26 15:44:26
- Status: offline
- Ribbons : 45
Re: Nitrogen Atmosphere for long duration Subzero Overclocking?
2021/07/28 15:13:04
(permalink)
Are you talking about nitrogen purging?
|
Alias_Go
New Member
- Total Posts : 7
- Reward points : 0
- Joined: 2021/04/24 12:46:34
- Status: offline
- Ribbons : 0
Re: Nitrogen Atmosphere for long duration Subzero Overclocking?
2021/07/28 15:30:18
(permalink)
|
Flint 1760
Omnipotent Enthusiast
- Total Posts : 8295
- Reward points : 0
- Joined: 2009/04/26 15:44:26
- Status: offline
- Ribbons : 45
Re: Nitrogen Atmosphere for long duration Subzero Overclocking?
2021/07/28 17:04:09
(permalink)
It could be done as you are just making a pressure vessel. However, all cables, wiring, remote system controls (which you will need) monitoring leads, etc., would have to be sealed. Your access system would require a hermetic seal. Then you have the purge valves and you'll have to maintain a positive pressure. The weak point will be your LN2 pot, which would be like a chimney sticking through, because you need direct access to it. You would run through a lot of cylinders of nitrogen gas and probably LN2. All in all, I think it be a real pain to use and just not worth it. The cost of building it would greatly exceed the cost of a whole lot of motherboards, GPUs, and other components. Just my thoughts.
post edited by Flint 1760 - 2021/07/29 06:47:36
|
Alias_Go
New Member
- Total Posts : 7
- Reward points : 0
- Joined: 2021/04/24 12:46:34
- Status: offline
- Ribbons : 0
Re: Nitrogen Atmosphere for long duration Subzero Overclocking?
2021/07/28 18:18:48
(permalink)
All good points,
Though it doesn’t need to be a pressure vessel, just gas tight at slightly above 1atm.
I’m thinking about an acetone loop rather than LN2 Pot. However LN2 pots could be sealed in with inflatable seals as long as you had a the right pot design.
Cables and wiring are fairly easy to make gas tight through connections for and there are several IR on/off reset switch options… but also physical switch pass throughs that are easy enough to seal.
I like the concept of being able to control the atmospheric temp for say air flow to memory modules while having cold fluids for GPU + CPU cores.
I like to run GPUpi for CPU and other GPUPI versions which often take quite some time to run so LN2 isn’t super practical, but dry ice is cheap and if I run an acetone loop should work for long durations.
I’m generally happy at subambient but I want to try some new ideas and colder temps but without the mess of a dry ice pot or expense of LN2.
Building the enclosure is not technically difficult just curious if anyone did it yet or not.
After all the “ice box” coolers that were built 10 years ago or so I figured someone would have already tried this and have experience to share. I hadn’t messed about with over clocking since GTX280 was a thing but had some time in the pandemic to kill and have gotten back interested again… not so interested in current Gen hardware, just second hand bargains that can cheaply push until dead or a new bargain comes up.
|
Jyoussi
New Member
- Total Posts : 44
- Reward points : 0
- Joined: 2021/06/13 22:48:47
- Status: offline
- Ribbons : 0
Re: Nitrogen Atmosphere for long duration Subzero Overclocking?
2021/08/07 20:39:18
(permalink)
mahn yah would gain a slim overall gain... It's the card's limits that are in place, not so much heat. Yah can get overclocked toa certain point. stuff and programs
Just fail to run even if ur not running a hot gpu, .. The cards topout so to say.
|
Flint 1760
Omnipotent Enthusiast
- Total Posts : 8295
- Reward points : 0
- Joined: 2009/04/26 15:44:26
- Status: offline
- Ribbons : 45
Re: Nitrogen Atmosphere for long duration Subzero Overclocking?
2021/08/08 03:44:01
(permalink)
^OP was talking about extreme overclocking.
|
the_Scarlet_one
formerly Scarlet-tech
- Total Posts : 24581
- Reward points : 0
- Joined: 2013/11/13 02:48:57
- Location: East Coast
- Status: offline
- Ribbons : 79
Re: Nitrogen Atmosphere for long duration Subzero Overclocking?
2021/08/08 06:20:21
(permalink)
Flint 1760
^OP was talking about extreme overclocking.
Jyoussi’s response is pretty much on par even for XOC. Even sealing everything to keep condensation out, the card is going to hit a limit that a sealing chamber won’t help with. Memory doesn’t like being too cold, so if you seal everything and fill it with nitrogen, the memory is still going to be a hard limit for the board, and now you have a sealed system that hasn’t really allowed much more headroom. The condensation is less of the XOC issue for experienced XOC’ers compared to the other hard limits that come up, so the gain versus the work/time wouldn’t be beneficial.
|
Alias_Go
New Member
- Total Posts : 7
- Reward points : 0
- Joined: 2021/04/24 12:46:34
- Status: offline
- Ribbons : 0
Re: Nitrogen Atmosphere for long duration Subzero Overclocking?
2021/08/08 09:35:23
(permalink)
The point of a sealed atmosphere is to control both the temperature and composition of the atmosphere….
I.e. you can control the atmosphere temp and circulate it for best memory performance while using Die blocks to circulate cryogenic fluid to cool the components you want to. Controlling the composition of the atmosphere means you can be water free so condensation free.
Without heating individual memory modules I don’t see a good way to get die temps to below zero but memory to “optimal” using aconventional XOC rig…
|
Flint 1760
Omnipotent Enthusiast
- Total Posts : 8295
- Reward points : 0
- Joined: 2009/04/26 15:44:26
- Status: offline
- Ribbons : 45
Re: Nitrogen Atmosphere for long duration Subzero Overclocking?
2021/08/08 10:02:16
(permalink)
^The way I understood what you were trying to do, was use nitrogen purging (gas), which you would have to keep at a positive pressure as well as recirculate. Your LN2 pot would go through the pressure vessel. As an engineering exercise, it could be done, but the cost would be very excessive and not really worth it from an end user standpoint of ease of use, time, and complexity. I'm old school, if the old way still works for me at a cost in both time and money that I'm comfortable with, then that is the way I'll go.
post edited by Flint 1760 - 2021/08/25 08:05:18
|
Alias_Go
New Member
- Total Posts : 7
- Reward points : 0
- Joined: 2021/04/24 12:46:34
- Status: offline
- Ribbons : 0
Re: Nitrogen Atmosphere for long duration Subzero Overclocking?
2021/08/08 12:07:20
(permalink)
There is not need to re-circulate the nitrogen gas... just contain it.
Also, i said that i'm wanting to use acetone and dry ice not 2, but that LN2 pots could be made to work if sealed correctly.
|
coinshark
Superclocked Member
- Total Posts : 105
- Reward points : 0
- Joined: 2021/08/15 15:38:09
- Status: offline
- Ribbons : 1
Re: Nitrogen Atmosphere for long duration Subzero Overclocking?
2021/08/25 06:15:48
(permalink)
As someone who has worked over 10 years in the cryogenics industry, I advise extreme caution when playing with cryogenic fluids and gasses. That being said, I've always questioned why no one has created a nitrogen pump block for a gpu. As long as you prevent condensation from forming you should be able to have a loop the flows the cryogenic fluid. All of the benchmark tests I have seen have been using a pot style system which is hard to fill and truly only a temporary testing solution. At those temperatures, it is extremely difficult to prevent condensation or "rine" of any sort however. Unless it is sealed in a vacuum chamber you will begin to ice up components over short periods of time due to the nature of the cryogen.
|
Alias_Go
New Member
- Total Posts : 7
- Reward points : 0
- Joined: 2021/04/24 12:46:34
- Status: offline
- Ribbons : 0
Re: Nitrogen Atmosphere for long duration Subzero Overclocking?
2021/08/28 14:21:38
(permalink)
Thanks coinshark,
This is one of the reasons i'm shooting for the "safer" cryogens and gasses.
Dry Ice acetone bath with a heat exchange coil closed acetone loop pumped with an acetone rated pump... with the electronic components all in a 1atm Nitrogen atmosphere... which is dry and therefor no condensation...
Stepping up to pumped liquid nitrogen/helium would be cool, but building a zero boil off solution with continuous circulation is more coin, time and power consumption than i'm prepared to go for right now...
|