Are computers (servers) considered continuous duty for t

Status
Not open for further replies.

Yuba

New member
Are computers (servers) considered continuous duty for the 80% rule for circuit breaker sizing?

Example: 40 servers in a rack at a data center.

Apologies if this has already been answered, I did a search and did not find a response.
 

iwire

Moderator
Staff member
Location
Massachusetts
Re: Are computers (servers) considered continuous duty for t

Originally posted by Yuba:
Are computers (servers) considered continuous duty for the 80% rule for circuit breaker sizing?
I would say sure they are continuous.

Servers run 24/7.
 

physis

Senior Member
Re: Are computers (servers) considered continuous duty for t

I'm not so sure.

Is the maximum current expected to continue for three hours?
 

iwire

Moderator
Staff member
Location
Massachusetts
Re: Are computers (servers) considered continuous duty for t

It's a point that could be argued about almost any equipment. :D

Say a water pump that runs 24/7 but pumps against varying pressures.

10 minutes every three hours it draws peak current.

Is that a continuous load?

[ June 30, 2005, 08:51 PM: Message edited by: iwire ]
 

don_resqcapt19

Moderator
Staff member
Location
Illinois
Occupation
retired electrician
Re: Are computers (servers) considered continuous duty for t

Continuous Load. A load where the maximum current is expected to continue for 3 hours or more.
There would be very few loads that meet the above definition. If the load varies any within the 3 hour time, it is no longer a continuous load. Is this another example of the code not saying what it means?
Don
 

George Stolz

Moderator
Staff member
Location
Windsor, CO NEC: 2017
Occupation
Service Manager
Re: Are computers (servers) considered continuous duty for t

Don't servers generally run all the time? I would think you'd have to consider the nameplate of the equipment: If the server pulls 3 amps, then you have 3 amps continuous load.

Whether it bumps up or down a tenth when it's accessing the hard drive, it's still essentially running at the same amperage.

I have a hard time imagining a scenario that fluctuates enough to disregard the nameplate rating, and go with readings taken in the field. :confused:
 
O

oliver100

Guest
Re: Are computers (servers) considered continuous duty for t

Apparently, these multiple servers (40) are part of a servers farm and they are engaged in a serious non stop operation. Tipically those machines have fault tolerant power supplies, asuring the non stop service. The disks are rotating constatntly and the heads, when not servicing data transfers are busy doing 'house maintenance'.

Therefor, each machine should be considered as a continuous load.
 

al hildenbrand

Senior Member
Location
Minnesota
Occupation
Electrical Contractor, Electrical Consultant, Electrical Engineer
Re: Are computers (servers) considered continuous duty for t

For a server to be a load, wouldn't the server have to be utilization equipment?
 

physis

Senior Member
Re: Are computers (servers) considered continuous duty for t

I'd be interested in how much power they actually draw compared to the name plate rating. I don't deal with these but my understanding is that the name plate will reflect what the contents of the cabinet could draw if it were fully populated and the actual current will be significantly lower.

If the name plate says 3.5 or 4 amps and it really only uses 2.5 amps do you still want to make the branch circuits 125%? Then again, one day the boxes might get filled up and draw the whole 3.5 or 4 amps.

If it were me I think I'd keep the actual load to 80% of the circuit because the things are a constant load and probably pretty consistant.

But it still doesn't meet the definition of continuous load.

I'm curious if "maximum current" is supposed to be in relation to the name plate or the actual current draw?
 
Status
Not open for further replies.
Top