Dell publishes data centre cleaning guidance, suggests hiring pros to disinfect enterprise kit Dell has developed guidance for cleaning its data centre products, on the off-chance you feel like giving them a once-over during these complicated times. The Register raised the prospect of data centre hygiene earlier this week when we pondered if it’s a great idea to walk into warm buildings that are full of fans and plenty …
Just by the bye: for max.effectiveness as disinfectant, you have to dilute isopropanol -- the water acts as catalyst re membrane penetration. 75% is peak effectiveness, so pour out 1/4 and top up with distilled/deionised water.
Having said that, you need to soak surfaces for about a minute to kill 50% of most coronaviruses, and alcohols evaporate much faster, so you could apply the rat's arse rule. Or bleach.
Interestingly, dry handrubbing done properly works slightly better than using gels or disinfectantsoap (page 38, WHO Guidelines on Hand Hygiene in Health Care), which matches rather a lot of researching showing that proper+lengthy handwashing beats magic potions + 3 seconds gesturing.
The peak effectiveness of 70/30 IPA/water is for disinfecting surfaces of bacteria, since water is required to penetrate the cell membranes. For many viruses, neat IPA is an effective steriliser.
Using acetone will do more than discolour equipment. It's okay on nylon, but will dissolve ABS and cause polycarbonate to crack.
See Table II in my link immediately below for effectiveness of all the usual suspects.
Both of these are effective but sodium hypochlorite looks more so: 0.5% hydrogen peroxide kills 60% (log(4)) in 1 minute, whereas 0.2% bleach kills 60% in 30 seconds.
Note that a common ingredient of disinfecting soaps, chlorhexidine digluconate, is NOT very effective.
Sorry, I was referring only to coronavirus effectiveness. Yes, 70% is best for bacteria and bacteria is the major problem in most hospitals so this is the standard formulation.
Versus viruses, however, and specifically coronaviruses, 75% typically shows better effectiveness than 70%. I noticed it when plowing thru tables of research results. Eg, you'll notice here: {Journal of Hospital Infection: Persistence of coronaviruses on inanimate surfaces and their inactivation with biocidal agents}, that 75% is 20% more effective than both 70% and 100% vs coronaviruses : log(4) vs log(3.3).
You'll also note from there that ethanol seriously dominates isopropyl alcohol. So maybe get the meths out of the fridge and use that instead.
One problem with 100% is that a crust can form around/over the virus, protecting and leaving live virus underneath it.
Try a vets or horse tack shop it's where I source my 99% iso from. Used on horses dunno what for
People weight about 80kg, a horse will weigh something like 500kg. Therefore, for a horse to get the proper effect of booze, it needs to have around 6x the same amount. Most people having 3 pints can't afford 18 pints for the horse.
It's possible cough droplets get sucked into it by the fans, I guess.
Still, working on them for a while with gloves on should mitigate it enough I think. It doesn't seem to live for very long outside of the body.
I'd be much more worried about datacenter door handles, shared cage keys, and those KMM trolleys people use there.
There was a time when most circuit boards were washed in de-ionized water to clean the yucky acidic soldering flux off (if it was the water-soluable type). Today's parts increases the risk water gets trapped under the chips and then the EPA finds you and starts banging on the doors.
Dell's guidance is most likely aimed at: Do not apply liquid when things are turned on or connected to the mains.
That mains-connected power supply was not designed to go snorkeling while plugged into the wall. Snorkel-ready supplies need bigger cooling fans.
Server motherboards have a number of voltages ranging from 12V to 0.9V core voltages. Never shall the two meet on the same conductor or, a large percentage of your chips, including that $2K chip with the Intel logo will explode with a tiny "poink!" sound into the heatsink covering it.
If this was a laptop you just poured a coffee into, that LED backlight runs off of up to 48V.
Leave it to the professionals unless you're about to scrap it.
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2024
