Improved PV system reliability results from surge evaluations at Sandia National Laboratories Page: 2 of 4
This article is part of the collection entitled: Office of Scientific & Technical Information Technical Reports and was provided to Digital Library by the UNT Libraries Government Documents Department.
The following text was automatically extracted from the image on this page using optical character recognition software:
incremented in steps of 1000 Vdc from 1 to 6 kV. As the
charge voltage is increased it is probable that a protecting
device will suddenly be activated. This may result in less
coupled energy at higher voltages than at lower voltages
where the protective device is still inactive. Thus the EUT
may prove to be more susceptible to a lower or intermediate
voltage than to a higher voltage.
b. Dual Polarity Testing. Both negative and positive pulses
c. Repetitive Pulses. High voltage pulses can incrementally
weaken components with no apparent damage after the
initial pulse. Thus for an inverter which survives all voltage
increments three pulses are applied at the highest charge
voltage level. Repetitive pulses are at least one minute
d. Unpowered Evaluations. When external power is
provided to the EUT there is the potential for greater stress
to components. Breakdowns due to surges can be
exacerbated by the large amount of energy available from
either the dc or ac lines. This extra current, supplied by the
equipment under test, is referred to as "follow" current.
Since the stress is lower in the unpowered EUT case,
unpowered evaluations may be completed prior to powered
e. Powered Evaluations. Powered evaluations are necessary
because of the likelihood of "follow" current and because
of the need to evaluate the survivability of anti-islanding
features of grid-tied inverters. Because of the possibility of
latent damage to these critical circuits, an anti-islanding test
should always be conducted after each significant voltage
increase in a surge test. Filters are required between the
EUT and other power sources to protect the power source
and to present a high impedance to the surge.
f Evaluation of Transfer Functions. A transfer function
that defines the current that passes through the protection
circuitry is useful for designers. Care must be taken to
avoid significantly changing the circuit configuration or
damaging instrumentation. This is accomplished by
monitoring the signal with a current probe. The monitored
wire is wrapped in mylar to prevent arcing and centered in
the current probe, to minimize capacitive coupling. The
signal from the current probe may be fed to a battery-
operated oscilloscope and thus isolated from ground.
The transfer function also provides a means for detecting
flashover or breakdown in the applied signal. Initially the
signals are applied at low voltage where no possibility of
flashover exists. As the charge voltage is increased, the
coupled signal envelopes will be identical and will scale in
magnitude unless a nonlinear effect occurs. Thus a change
in the coupled signal envelope or a failure to scale linearly
implies that a nonlinear event has occurred.
3. Evaluations of Surge Mitigation Devices
The evaluations of a typical rhetal-oxide varistor (MOV)
and a silicon spark gap are presented below. Note that the
ring wave, which has a faster rise time, rises to a higher
voltage prior to being clamped. As the charge voltage is
increased a higher amplitude of peak voltage is coupled past
the MOV to potentially vulnerable electronics. This
particular MOV was pulsed more than 12 times with no
Figure 3. Pulse Clamp voltage versus Charge Voltage
for an MOV
Ring wave has a 3.5 microsecond nsesmee
More voltage is coupled through.
- - Rmng Wave
Pulse has a 8 microsecond n.setime
Prove des more time to clamp to a lower voltage
0 2000 4000
A Delta Model LA 302-RG was also evaluated.
This unit has been used in many PV applications; it is not
clear that the users understood its proper application.
Delta's description defines the clamping voltage but does
not define the initiation voltage. The initiation voltage, the
voltage required to start an arc is usually much larger than
the clamping voltage. The LA 302RG had no effect at surge
voltages up to 6 kV. A physical examination of the arrestor
revealed that it is a silicon-filled (sand) spark gap.
Current Coupled Past MOV
* *~ - -
Figure 4. Test Data from the Omnion Inverter
This clamping effect is clearly seen in data from a pulse
test of an Omnion inverter that uses an MOV. In Figure 4
the 6000 volt 8 x 50 p second output of the surge generator
is limited to a peak of 700 volts. This is quickly clamped to
a voltage of about 400 volts. The current spike (current into
the inverter) is limited to 29 amps peak and is of very short
duration. The current is measured after the MOV and its
shape is affected by other inverter components.
The surge testing of power electronics and surge
protection devices at SNL will continue with emphasis on
providing practical guidelines for manufactures and users.
[I] IEEE Std C62.41 91, IEEE Recommended Practice on
Surge Voltages in Low-Voltage AC Power Circuits.
 IEEE Guide on Surge Testing for Equipment Connected
to Low-Voltage AC Power Circuits:
Peak Vol = 700 volts
- -. _ . Voltage Pulse at Inverter _
- _ - _ Input.
Peak I = 29 amps
Here’s what’s next.
This article can be searched. Note: Results may vary based on the legibility of text within the document.
Tools / Downloads
Get a copy of this page or view the extracted text.
Citing and Sharing
Basic information for referencing this web page. We also provide extended guidance on usage rights, references, copying or embedding.
Reference the current page of this Article.
Bonn, Russell H. & Gonzalez, Sigifredo. Improved PV system reliability results from surge evaluations at Sandia National Laboratories, article, April 11, 2000; Albuquerque, New Mexico. (digital.library.unt.edu/ark:/67531/metadc711682/m1/2/: accessed January 23, 2019), University of North Texas Libraries, Digital Library, digital.library.unt.edu; crediting UNT Libraries Government Documents Department.