ECN Electrical Forum - Discussion Forums for Electricians, Inspectors and Related Professionals
ECN Shout Chat
Recent Posts
Any women electricians here?
by HotLine1 - 01/27/22 02:44 PM
12 Lead Motor Connection Weird
by emolatur - 01/19/22 02:40 PM
Well There's your problem.
by gfretwell - 01/08/22 12:17 AM
Happy New Year!!
by HotLine1 - 01/04/22 09:50 AM
Sangamo s309.2
by Robski - 12/31/21 04:54 PM
New in the Gallery:
240/208 to a house
240/208 to a house
by wa2ise, October 9
Now you know.
Now you know.
by Tom_Horne, September 7
Who's Online Now
0 members (), 50 guests, and 14 robots.
Key: Admin, Global Mod, Mod
Previous Thread
Next Thread
Print Thread
Rate Thread
#144085 10/05/05 08:32 AM
Joined: Sep 2005
Posts: 97

I have a data centre which contains about 30 UPSs. I am worried about electric shock risks for the fire brigade. Currently if the fire alarm is triggered to full fire and releases the extinguishant, it cuts the 250 Amp 3 phase supply off but all the UPS's then run on battery.

The UPSs have no EPO terminal as they are all about 1000-3000VA apart from a few 10KVA ones. I want to know what voltage I could get if I measure between the live on one UPS and the live on another. I heard somewhere that I could get up to 575V. If this is true, is there some way of keeping it down to 230V?


I took my time, I hurried up, The choice was mine, I didn't think enough
#144086 10/05/05 11:23 AM
Joined: Aug 2001
Posts: 7,520
The absolute voltage between one UPS output and another will vary with the phase difference between them.

If they're set for 230V nominal, then at 180 degrees out of phase you'd get 460V RMS. As each unit is free running, the phase differences will drift over time, varying the actual voltage difference between zero and the full level.

The only way I can think of to eliminate this would be to synchronize them all together.

#144087 10/05/05 11:59 AM
Joined: Jul 2002
Posts: 8,412
Likes: 3
Hi there John,
As a Fire Officer here in NZ, I have to say the thought of this occuring never actually crossed my mind.
Not that I've ever attended a fire in a computer room or the like.
I'm not sure how things are in the UK, but over here we're required to wear rubber Live Gloves around anything too serious (electrically).
Generally in a fire of this type we'd be using Carbon Dioxide extinguishers anyway or as a last resort Dry Powder.

#144088 10/05/05 04:47 PM
Joined: Dec 2004
Posts: 161
Many data-centre operators are rather unhappy with the idea of water being sloshed around their precious disks, so install their own auto-extinguishers using CO2. But then you need emergency breathing sets, a strict entry control system...

#144089 10/06/05 03:07 AM
Joined: Sep 2005
Posts: 97
Thanks for all the responses, sorry it has taken so long to reply. I suppose you would have to be really unlucky to come between 2 seperate upss so it might not be as bad. The fire brigade are aware of what the room is for anyway. If there is no water you would actually have to go and grab at them to get contact with anything live.

We have 160Kilos of HFC227ea extinguishant and then we also have some very large hand held CO2 extinguishers as well as some dry powders in case there is a battery fire.

we keep entry controlled with digital locks and trained staff. There is also a status panel to check and the system is put into manual when the lights are turned on.

Do we need to put those stickers that warn of 400V between <-- --> on the sockets that are close and on different phases?


I took my time, I hurried up, The choice was mine, I didn't think enough
#144090 10/06/05 05:10 AM
Joined: Sep 2002
Posts: 1,498
Likes: 1
C-H Offline
Do we need to put those stickers that warn of 400V between <-- --> on the sockets that are close and on different phases?

No. To get between two phases fed from two different sockets would require the breach of four safety barriers:

1. Basic insulation on appliance one.
2. Earthed casing or outer insulation on appliance one.
3. Earthed casing or outer insultation on appliance two.
4. Basic insulation on appliance two.

I leave it to you to figure out how plausible this is. [Linked Image]

To label them with which circuit they're on is practical.

How do you get the figure 575V? I can guess:

Three phase 400V would give you 565V if you rectify it: root 2 * 400 = 565V DC
This could be the battery voltage on a 3-phase unit. It could easily be 575V to allow for some voltage drop.

#144091 10/06/05 03:54 PM
Joined: Sep 2005
Posts: 97

I guess I was being a little paranoid about shock risk! I only got 575V from what someone told me.

Can I measure the Live-Live voltage safely with my multimeter? obviousely with it set on volts. Loud bangs arent pleasurable anymore [Linked Image]

All of the circuits are labelled and on RCBOs I would hope the UPS internal breaker would trip should it be soaked.

Thanks for all your help, I can sleep easy now [Linked Image]

I took my time, I hurried up, The choice was mine, I didn't think enough
#144092 10/06/05 04:24 PM
Joined: Aug 2001
Posts: 7,520
Can I measure the Live-Live voltage safely with my multimeter?

Sure you can, but with the UPS oscillators not being synchronized a reading taken at any one instant will not tell you very much.

With no synchronization between the oscillators in each UPS, the phase angle between any two outputs will vary over time due to minor differences in the running frequencies.

Your meter reading will vary from zero volts (indicating in phase) to twice the nominal RMS output of one supply (indicating 180 degrees out of phase). Between zero and 180 degrees, you'll get a reading between those two extremes.

#144093 01/22/06 07:49 AM
Joined: Jun 2005
Posts: 17
As I have experiance of a data centre I can answer your question from the kit that we operate here - You have to be careful with any UPS derived power supply and as you pointed out you can get when you lose mains power up to twice the phase-neutral voltage across two conductors. In theory this could equal not far off 700v - the workings ? simple rms equations. the rms value of a single phase 240 circuit gives around 345v now put in two UPS units and worst case 180 degrees out then you have a voltage of 690v.

Where you can find people getting caught out with this is in simple dual fed items of kit - If you have say one side of the dual fed kit running on UPS and the other on mains of UPS then you have the possibility of having this sort of voltage kicking about. This is when people forget just what they are working with and need to be amde aware that there is a chance that they may encounter higher voltages than expected. Some test equipment may not be geared up for going across this sort of voltage if it has surge protection inbuilt and the person working with unfused probes then you could get a pretty nasty fright.

Any equipment with dual feeds and UPS sourced supplies you really need to mark up and ensure that the operator knows this, also if you are working with kit that has integral UPS units then you need to mind this if you have emergency power offs in the room - this will not kill the power and you may have a risk if someone gets connected to the power side of things and then somoene else hits the power kill button - the circuit may remain live and dangerous !


Link Copied to Clipboard

Tools for Electricians
Tools for Electricians

* * * * * * *

2020 Master Electrician Exam Preparation Combos
2020 NEC Electrician
Exam Prep Combos:
Master / Journeyman


Member Spotlight
CA, Inland Empire
Posts: 43
Joined: September 2013
Top Posters(30 Days)
Popular Topics(Views)
288,257 Are you busy
220,326 Re: Forum
206,314 Need opinion
Powered by UBB.threads™ PHP Forum Software 7.7.5