[30639] in Hotline Meeting
Community Update on Electric Power
daemon@ATHENA.MIT.EDU (vsirianni)
Tue Oct 31 16:22:02 1995
Date: Tue, 31 Oct 95 15:35:13
From: sirianni@MIT.EDU (vsirianni)
To: aac-aoquery@MIT.EDU, plant-notify@MIT.EDU
Cc: plant-all@MIT.EDU, dorow@MIT.EDU
On July 11, 1995, I wrote to you in the aftermath of a series of severe
electrical outages that occurred on July 8 and July 10, 1995 across the
campus.
I would like to update you on the efforts completed and work underway
in the area of electric power since that time. This note is somewhat
long, but I believe the information warrants the detail.
INVESTIGATION OF THE EVENTS OF JULY 8 AND JULY 10, 1995
-------------------------------------------------------
We know now that both incidents were related and the outage that
occurred on July 10,1995 was caused by system disturbances from the
earlier cable fault of July 8, 1995. The 13.8KV transformer failure on
July 10, l995 not only caused a serious outage but it required that we
re-configure our electrical system by using rented diesel generators, etc.
Our investigation into the cause of the transformer
failure included examining the transformer windings at the factory where
the unit was built. They found that the unit was not defective, but
rather that the cable fault forty eight hours earlier had initiated a
sequence of events that caused the transformer to experience an impulse,
similar to, yet more severe than that of a lightning strike. We have
used the expertise of members of the Lab for Electromagnetic and
Electronic Systems (LEES) at MIT in researching the causes of the
transformer failure and they continue to work with us in this area.
REPAIR WORK DONE
-----------------
The manufactuer rebuilt the failed transformer at their plant in
Virginia. They retrofitted it, as well, with additional surge
protection devices on the high and low voltage terminals. This was
something the original designers felt was not necessary but that we
initiated based on previous research and with a view towards preventing
a reocurrence. We are also investigating reducing the time current
characteristics of the fuses as an additional means of protecting them
in case of a similar type cable fault.
Although we were confident that the transformer on the A side of the
distribution system had not been affected, we removed it and sent it to
a local engineering and testing firm. Their tests indicated that the
transformer was sound and had experienced no damage. This finding
strengthened the conclusions from the failure analysis of the sister
transformer. Following the tests, we also equipped this transformer with
additional protective devices, identical to those installed on the A
side. The electrical distribution system has since been re-configured
and is operating in its normal mode.
NEXT STEPS
----------
On September 16, 1995, our co-generation plant began to generate
electric power on a regular basis. We have thus reached an important
milestone since this project has been ten years in the making. There
are however a few lingering tests which will validate the robustness of
the control system operation and dynamic response of the generator. We
feel strongly that these tests should be done. I would like to outline
them for you and explain their importance.
Within the next month, we want to analyze the results from the following
three scenarios:
1. Test the systems response on opening the tie breakers to the
Utility company when there is a net zero interchange of power
between
MIT and the Utility.
2. Test the system response on opening the tie breakers to the
Utility company when there is a net EXPORT of power from MIT to
the
Utility.
3. Test the system response on opening the tie breakers to the
Utility company when there is a net IMPORT of power from the
Utility to MIT.
4. Test the new load shedding equipment when the MIT load is
greater than the generator capacity and there is a loss of power
from
the Utility Company.
Once we have completed these tests we can be assured of the generator's
capability to maintain stable power even in the major loss of our local
utility. While the chances of this are remote, we must nevertheless test
for the possibility.
YOUR SUPPORT
------------
These tests can only be conducted while we are operating with close to
normal load conditions. This means that we need your help in scheduling
the necessary outages and securing critical equipment. These tests
should take no longer than 4 hours on a weeknight between midnight and 6
a.m. We plan to complete this test before 23 November 1995 but the exact
date has not yet be determined. Physical Plant associates will
communicate further and specifically with you but I would like to
provide a list of buildings which will be affected.
Main Campus-------------
2, 4, 5, 7, 10, 12, 12A, 16, 17, 20, 24, 26, 31, 34, 36, 38, 56,
70
East Campus---------------
E15, E18, E19, E51, E52, E53, E55
West Campus---------------
W1, W2, W2A, W4, W5, W7, W8, W13, W15, W16, W20, W23, W31, W32, W33, W51,
W53, W61, W70, W71, W84, W85 and fraternities on Memorial Drive/Amherst
Alley
Even as I acknowledge the inconvenience of these tests to you who have
experienced so many instances of unstable power in the past, I must also
underline their importance. In order to establish a long-term stable
environment of generating and distributing electric power for the future
at the Institute, we must test for all possibilites no matter how remote
or how unlikely these occurrances might be. I look to you for continued
support and thank you for your patience.
Thank you
V. V. Sirianni
Director Physical of Plant
