[Beowulf] Re: What is the right lubricant for computer rack sliding rails?

Robert G. Brown rgb at phy.duke.edu
Fri Feb 6 08:37:08 PST 2009


On Fri, 6 Feb 2009, Gus Correa wrote:

> I am talking about telescoping rails, metal on metal,
> no ball-bearing mechanism,
> which I think match Bob Drzyzgula description.

Truly no ball-bearings?  Usually even cheap drawer rails for use in the
kitchen have either rollers or bearings.  There is sometimes only a
single ball bearing in a small cage at the torque/stress point where one
rail is pressed against another, but straight metal-on-metal -- well,
there is your problem.  If they aren't lubed to the point where there is
a relatively thick glide-layer of grease, they'll rub.  If you overload
them, they'll just scrape off the grease.

> Visual inspection of the rails didn't show any signs of
> them being warped or twisted.
> With no load, they work properly.
> The rails support a heavy 5U RAID file server.
> When loaded, sliding is not smooth.
> The telescoping stages don't always move symmetrically,
> or as expected (e.g. the inner stage may come
> out after the middle stage, etc).

Sure.  Metal on metal rails with no bearings are going to have to bear
all the torque on the corners of the rails.  Every time you slide in or
out, you gouge the corner -- all the force is exerted by a tiny fraction
of the supporting area (unless spread out by grease).  The force exerted
at the torque points is probably many times the weight of the server --
each -- when the rails are fully extended.

Good rails will spread that out on several rollers or a line of
bearings, which will also ROLL instead of SCRAPE.

So IMO, you're best solution is to buy better rails, at least for your
heavy servers.

For your nodes it probably doesn't matter.  As I said, you pull them in
or out maybe once or twice in their lifetime.  Who cares if they're
sticky, as long as they don't wedge.

    rgb

>
> I saw similar problems before,
> mostly with telescoping rails on bulky multi-U
> RAID servers and heavy rackmount UPSs.
> Mind you, I didn't assemble all of those rails.
> And to date, I have never used any lubricant on them.
> But in the case at hand, I can see no alternative to
> a proper lubricant.
>
> Less often, even the higher quality ball-bearing rails that
> support 2U nodes may get a bit hard to move.
> However, this may be less of a rail problem than
> caused by small asymmetries on the rack mounting holes, I suppose,
> as the bottom and top of a node chassis drag along the neighbor
> chassis.
>
> The particular rack I am using is old,
> salvaged not from the armed forces,
> but from our oceanographic equipment of our old decommissioned ship.
> Although the rack is in good shape, it may have larger asymmetries
> on the mounting holes than modern racks do.
>
> Anyway, it's been a lovely and instructive debate, I must say.
> Please, advise!
>
> Many thanks,
> Gus Correa
> ---------------------------------------------------------------------
> Gustavo Correa
> Lamont-Doherty Earth Observatory - Columbia University
> Palisades, NY, 10964-8000 - USA
> ---------------------------------------------------------------------
>
>
> Bob Drzyzgula wrote:
>> One thing that Gus didn't explain to us was what sort of
>> rails he's talking about. If they are ball-bearing slides,
>> or even the kind with small wheels, it is actually kind of
>> surprising that they would need any lubrication at all. Is
>> it possible that these are just metal-on-metal telescoping
>> slides without any sort of sophisticated bearing system? I
>> know those can get pretty sticky. In any event I'll
>> join those who say a light application of some sort of
>> non-volatile grease would be best. Among other issues,
>> any oil or aerosol is too likely to get into the general
>> environment and on other surfaces in the vicinity of the
>> rack, and will likely will not do well after long periods
>> of disuse at the modertately warm temperatures and high
>> airflow of a machine rack.
>> 
>> I will also especially join those who argue against
>> WD-40. To the extent that it acts as a lubricant, this is
>> almost a side effect. I think that it used to be that they
>> wouldn't even call it a lubricant -- the word certainly
>> didn't used to be listed on the can -- but I see from
>> their website that they do now use the word "lubricates",
>> in number 4 (out of 5) on the list of things that it can
>> be used for.  The downsides -- especially its volatility
>> -- pretty well outweigh its advantages for something
>> like this.
>> 
>> --Bob
>> 
>> On 05/02/09 11:49 -0500, Gus Correa wrote:
>>> Dear Beowulfers
>>> 
>>> A mundane question:
>>> 
>>> What is the right lubricant for computer rack sliding rails?
>>> Silicone, paraffin, graphite, WD-40, machine oil, grease, other?
>>> 
>>> Thank you,
>>> Gus Correa
>>> ---------------------------------------------------------------------
>>> Gustavo Correa
>>> Lamont-Doherty Earth Observatory - Columbia University
>>> Palisades, NY, 10964-8000 - USA
>>> ---------------------------------------------------------------------
>>> _______________________________________________
>>> Beowulf mailing list, Beowulf at beowulf.org
>>> To change your subscription (digest mode or unsubscribe) visit 
>>> http://www.beowulf.org/mailman/listinfo/beowulf
>
> _______________________________________________
> Beowulf mailing list, Beowulf at beowulf.org
> To change your subscription (digest mode or unsubscribe) visit 
> http://www.beowulf.org/mailman/listinfo/beowulf
>

Robert G. Brown	                       http://www.phy.duke.edu/~rgb/
Duke University Dept. of Physics, Box 90305
Durham, N.C. 27708-0305
Phone: 1-919-660-2567  Fax: 919-660-2525     email:rgb at phy.duke.edu





More information about the Beowulf mailing list