TOV Tests Revisited

1 post / 0 new
Joined: 6 Jan 2011 - 19:02
TOV Tests Revisited

TOV Tests Revisited
From Richard Odenberg:
I reviewed the TOV submittals you provided [see agenda from upcoming 3.6.6 meeting].
NO TOV Tests. The Failure Mode Test is adequate.
Consider the Following:

1)The purpose of an AC/SPD is divided into two basic applications:
   a. Protection from exciding the “insulation withstand” of the power system.
       i. Protection against fire and electrical shock hazards.  To protect the electrical system has the SPD “Voltage Protection Level (VPL) higher then most TOV conditions. A TOV test would be realistic under this condition.
   b. Protection of Electronic systems and equipment.
    i. The SPD has a much lower voltage protection level (VPL), In General. Lower than a typical TOV. The need for this lower VPL is to eliminate the possibility of errors and malfunctions of the systems and electronic equipment.

If a TOV Test Is added, I would suggest that the SPD, under a TOV condition can either SURVIVE the TOV or FAIL IN A SAFE MANOR.
Remember, the prime purpose of the SPD is to protect the electrical and electronic systems from failure and malfunctions. Many SPD failures occur not from a true TOV but for selecting a Voltage Protection Level (VPL) to close to the normal power system voltage, without consideration of voltage regulation of the system.

Posted by Ron Hotchkiss for Richard Odenberg
1 Reply

Ron Hotchkiss Oct 24, 2009

        Dear Ron:
        I may understand you concern that it is desirable to have a face to face meeting on the TOV issue. But, I also believe that in order for such a meeting to be productive we all need to understand the issues that are to be discussed and to really understand the technology.
        As I see it there needs to be a clear understanding as to what is TOV and what is a permanent overvoltage (steady-state).  These two issues need to be separated and they also maybe technology related. All one has to do is look at the V-I curve for a metal oxide varistor and an avalanche diode. The first one has a slow increase in the the conductive current and then a knee point. At the knee point the rate of change of current with an increase in voltage across the varistor is change slope and the current increases slightly faster with a further increase in voltage, before thermal runaway occurs.
        In the case of an avalanche diode an increase in voltage above its maximum continuous operating voltage (MCOV) or Uc results in a very sharp rise in the current connected through the avalanche diode. It has almost no ability to withstand a overvoltage much above its rated protective level.
        What are the TOV conditions that need to be addressed? It is my opinion that there are two distinct issue that must be addressed:
        a) A TOV is one method that can be used to drive a SPD into catastrophic failure as is done in UL 1449. This method cause a puncture failure of the SPD and the resultant current results in destroying the SPD component. There is considerable debate as to whether the voltage used is a reasonable representation of a SPD end of live failure.
        b) A second type of TOV is one that is more likely to occur in the actual application. This type of TOV is very well described by Table 4-2, a copy of which is attached to this email. I believe what is generally consider a TOV is discribed by Item 2.0 Short Duration Variations. All other overvoltages would be covered by Item 3.3 Overvoltage - Long Duration Variations, where long duration is greater than 1 minutes. I do not believe that the voltages described could be called TOV, but are of the real world nature that would cause a fail or a SPD if allow to exist. (This would represent the UL 1449 Failure Test mode).
        This later TOV relates to the survivability of an SPD to withstand some value of over voltage above it MCOV. As can be seen from the my previous brief discussion of technology the amount of TOV that can be withstood by a component is going to be dependent upon the technology. But, the withstand to TOV is not just a measurement of the TOV withstand of the component, but the withstand (survivability) of the assembled SPD product at specific of RMS voltage and duration.
        A duration of any longer that 12s could be considered to be more or less steady-state. This selection of time is based on the TOV that can occur on a radial four wire multi-grounded common neutral (MGCN) distribution circuit for a phase to ground fault at that part of the distribution circuit which is most remote from the substation. IEEE C62.92.4 defines the value of he voltage at that point as 1.35 pu on an Overhead system and 1.45 pu on circuit composed primarily of cable. All most all utility distribution circuit's fault protection is supply by time-overcurrent relays that operate on inverse, moderately inverse or extremely inverse time-current curves.  At low phase to ground fault currents, such as might occur at the farthest point from the substation the clearing time of the fault can range between 6 and 12 seconds.  It is reasonable to expect that the SPD should be able to ride through this TOV and remain operational.
        This TOV survivability (withstand) would encompass all of the values in Item 1 of TAble 4-2 of IEEE 1159. If this is too severe than one could determine the probability of this scenario as being too improbable. The fall back would be 12s and 1.2 pu.
        I need the working group's help on this issue. As the US Technical Advisor to the IEC SC 37A Working Group and the IEEE Liaisons between IEEE SPDC and SC 37A/WG 5 I have to produce a value of TOV for IEC Standard 61643-1, 37A?216/CDV that is now out for vote. The US must provide a comment on this standard no later than mid January 2010.  
        You will note that based on this discussion that TOV must consist of two values: RMS value of Voltage Magnitude and Time Duration.
        I am sorry that I can not attend your meeting, but I believe that you will find it desirable to present my discussion.
        As many of you are aware I have much involvement in standards activities. Right now the most important standard topic is the identification of Smart Grid Standards which is being driven by NIST. I will be attending a meeting on this topic in Westchester, NY at the same time as the SPDC meeting.
        Joe Koepfinger
         Link to referenced excerpt: