ECN Forum Forums General Topics of the Trade General Discussion Area Correct rotation, wrong sequence

Recently I was taking part in a project at the building where I work as a maintenance electrician. A contractor was hired to install a UPS on an emergency power riser that feeds critical equipment that would benefit greatly from not seeing power bumps, etc. Our electrical staff worked with the contractor at various times during the project including the installation of tie breakers to feed certain panels on one emergency riser from another to avoid interruption of service while the UPS was being installed. On the eve of the tie breakers installation the contractor had 3 groups at the three installation points where one panel from a 600A, 3 Phase, 120/208V riser would feed two panels on the 300A, 120/208V riser where the UPS would later be installed. Myself and another electrician co-worker observed and offered assistance as needed. There was a well laid out document of steps to follow including a contingency plan should certain breakers not reset after the shutdown of power. It appeared that everything went as planned including phase rotation testing and phase sequencing (zero voltage test). Once that stage was completed we would later come back to transfer the load from the 300A riser to the 600A riser and wait 24 hours to observe if any issues developed. To avoid another power shutdown the contractor, engineer, and others involved agreed to a “closed transistion” transfer. On the day of the planned transfer I was not involved but I was told that after the first panel was transferred successfully the second panel failed. Apparently five major breakers opened and a little chaos ensued. What went wrong? Over the next two days there was much speculation as to why the second panel’s load did not transfer. In the end the decision was made that the Teck cable and tie breakers between the panel from the 600A riser to the failed panel on the 300A riser would have to be tested again for proper phase rotation and sequencing. In the end it was discovered that a tie breaker had been improperly placed in the panel that failed to transfer load. Phase rotation was correct according to the meter but a zero voltage test showed the phase sequencing did not match. The contractor felt certain that the testing was performed and documented on the day of the breaker installation. I haven’t heard if this was proven. I was in a different location on the day of installation and wouldn’t be able to verify if it was done or not. I’ve never been involved in this type of power transfer before and despite it not going to plan it was a great learning experience.




How I suspect the mistake happened (I was at the panel on the 600A riser when the mistake was discovered but did not hear the specifics as to why the breaker may have been misplaced in the panel):

As a result of everything that happened we are now looking at a second shutdown to correctly place the breaker in the panel.

I have no idea what plans are being made for the timing of everything but when the day comes to finally install the UPS on the 300A riser (includes cutting into the existing MI cable) I wonder what the plan will be to deal with the neutrals. How do you break the neutral loop without another power shutdown? To be continued...

Second image is technically wrong - the 600A panel and should be on the right side:

Finally got a look at the panel where the breaker was misplaced - I was close in my guess at how it ended up in the wrong position. The panel was filled with single pole breakers and the 3 pole tie breaker was placed in the next available vacancy.

Phase rotation problems were so common in computer rooms most of our equipment after about 1980 had phase rotation detectors in it and a switch that you could flip 2 phases with when you installed it.

I'm going to plead ignorance here, but what is a tie breaker used for?
It may be called something else down here.

To me its a breaker which connects two sets of busbars which are normally supplied separately and independantly. That's from aerospace useage but looks the same here.

The term "tie breaker" confused me too the first time I heard it.

My first thought was about a race...and my second was a smart-aleck thought about a circuit breaker wearing something around its neck.

You see them a lot in factories where there are 2 incoming lines from the power company for reliability.

The incoming equipment has a place at each end of a lineup for each line to land as well as breakers for the outgoing loads. In the middle is a tie breaker that connects the left and right ends of the bus together. There are almost always keylocks installed so that only 2 of the 3 breakers (Main-Tie-Main) can be closed at a time. If either incoming line fails you just open the main on that side, close the tie breaker and the whole lineup is live again.

Re: Tie-breaker...

My diagram is not great but if you look at panels A, B, and C the tie-breaker is simply referred to as "Tie". The four tie-breakers were the added breakers in the 3 panels to allow for moving load off one power riser to the other in order to keep critical loads live during the installation of the UPS.

Most recent news is that in order to correct the problem of the misplaced breaker is to switch the conductors rather than physically move the breaker. This is to avoid another power shutdown. If work is ever done in that panel sometime in the future that requires a power shutdown then hopefully someone will remember to move the breaker at that time. In my opinion I would rather go through the pain of organizing a power shutdown now while the details of this mishap is still fresh in everyone's minds.

It's now been over a year since this project began and it looks like it may finally come to an end next week. The UPS is connected ahead of the 300A main breaker but that riser is still being back fed by the 600A riser. Next week we will preform a closed transition and the 300A riser will be powered through the UPS.
Just thinking about something that nobody involved seems concerned about (therefore, I must be overthinking the situation) regarding the neutrals. As shown in the diagram they are now connected so that there are two common points. Earlier today myself and another electrician measured 11A with a clamp-on ammeter on the neutral at the UPS. If there wasn't a second second common point then there wouldn't be a measurable load on the neutral at that location at this time - correct? Since there is essentially parallel neutrals I assume that the unbalanced load current is being distributed between them like any other parallel circuit? If this configuration remains after the UPS is brought online is this still acceptable?