<html>
<head>
<meta content="text/html; charset=utf-8" http-equiv="Content-Type">
</head>
<body bgcolor="#FFFFFF" text="#000000">
<p>On 08/19/2016 12:10 PM, Stu Midgley wrote:<br>
</p>
<blockquote
cite="mid:CAEM1RsUzanFquzeU3j72Gs4d7vnJ9ZucFwqCemd411Bg3S_gXg@mail.gmail.com"
type="cite">
<div dir="ltr"><br>
<div class="gmail_extra"><br>
<div class="gmail_quote">
<blockquote class="gmail_quote" style="margin:0 0 0
.8ex;border-left:1px #ccc solid;padding-left:1ex">I think
direct contact liquid cooling (DCLC)solutions from Asetek
and CoolIT are much more practical, as are immersion
cooling systems using 3M Novec engineered fluid. Novec is
really appealing because of the heat transfer efficiency
of the change of state. I believe it was originally
designed as an electronic parts cleaner, too. It
evaporates at room temp like other electronic cleaners, so
when you pull hardware out the liquid, it's dry in a
matter of seconds, and very clean, too! Much more
practical than mineral oil in that regard. 3M claims it
has a very low global warming factor, too.<br>
<br>
Personally, I think DCLC since any leaks in the data
center should be smaller than if a tank-like chassis
springs a leak, and there one is scared of the
environmental impact of water, even when antifreeze and
other water treatment chemicals are added to it, as would
be needed in this use case.<br>
</blockquote>
</div>
<div><br>
</div>
<div><br>
</div>
<div>2 phase solutions are sexy but ultimately impractical.</div>
<div><br>
</div>
<div> * what happens when your recondensing loop fails? and
you've evaporated all the fluid off...</div>
</div>
</div>
</blockquote>
I agree that's a problem, but some failures can be mitigated with
redundancy, like redundant circulating pumps or including the pumps
on UPS and having proper failsafes incorporated into the cooling
system. <br>
<blockquote
cite="mid:CAEM1RsUzanFquzeU3j72Gs4d7vnJ9ZucFwqCemd411Bg3S_gXg@mail.gmail.com"
type="cite">
<div dir="ltr">
<div class="gmail_extra">
<div><br>
</div>
<div> * try and get a permit to hold large volumes of it in a
computer room</div>
</div>
</div>
</blockquote>
<blockquote
cite="mid:CAEM1RsUzanFquzeU3j72Gs4d7vnJ9ZucFwqCemd411Bg3S_gXg@mail.gmail.com"
type="cite">
<div dir="ltr">
<div class="gmail_extra">
<div> * try and find a computer room that will allow it in</div>
</div>
</div>
</blockquote>
<blockquote
cite="mid:CAEM1RsUzanFquzeU3j72Gs4d7vnJ9ZucFwqCemd411Bg3S_gXg@mail.gmail.com"
type="cite">
<div dir="ltr">
<div class="gmail_extra">
<div> * the above two are mostly due to the low boiling point
of the fluid</div>
</div>
</div>
</blockquote>
I understand that. That's why they are usually closed systems. I
would like to point out at SC, several booths had open tanks of the
stuff on the exhibitors floor. I believe IBM says it's safe, but I
wouldn't want to take my chances breathing it in every day. <br>
<blockquote
cite="mid:CAEM1RsUzanFquzeU3j72Gs4d7vnJ9ZucFwqCemd411Bg3S_gXg@mail.gmail.com"
type="cite">
<div dir="ltr">
<div class="gmail_extra">
<div><br>
</div>
<div> * expensive (~ an order of magnitude)</div>
<div><br>
</div>
<div>AND you don't actually need a 2 phase solution to get the
heat off the cpu's etc. A single phase solution works just
fine.</div>
</div>
</div>
</blockquote>
That's true, which is why water DCLC is a viable option, as well as
some other thermally conductive liquids that don't rely on boiling
(besides mineral oil), too, BUT multiphase cooling is much more
efficient in terms of absorbing/releasing energy. <br>
<blockquote
cite="mid:CAEM1RsUzanFquzeU3j72Gs4d7vnJ9ZucFwqCemd411Bg3S_gXg@mail.gmail.com"
type="cite">
<div dir="ltr">
<div class="gmail_extra">
<div><br>
</div>
-- <br>
<div class="gmail_signature" data-smartmail="gmail_signature">Dr
Stuart Midgley<br>
<a moz-do-not-send="true" href="mailto:sdm900@sdm900.com"
target="_blank">sdm900@sdm900.com</a></div>
</div>
</div>
</blockquote>
<br>
</body>
</html>