One String Reads 80 V Low, Current Normal: What Changed?

Correct answer: A) Roughly two modules’ worth of series voltage has stopped contributing — bypass diodes conducting, or modules internally shorted

In a series string the voltages add and the current does not. So a deficit that appears in voltage while the current is untouched tells you that series elements have stopped contributing voltage, and that the current path itself is still continuous end to end. The normal short-circuit current is the half of the measurement that rules things out.

Then do the division, because it names the fault. 800 − 720 = 80 V. Each module in this string contributes 40 V of open-circuit voltage, so 80 V is exactly two modules. A bypass diode usually spans one third of a module, which makes this six substrings: it could be two modules fully bypassed, or six substrings spread across several modules. Either way the cells behind those diodes are out of the voltage sum while the rest of the string carries current past them undisturbed, which is precisely what a bypass diode exists to do.

The O&M point is the last sentence of the scenario rather than any of the numbers. A commissioning crew measuring 720 V has nothing to compare it against except a calculation from the datasheet, and a calculation cannot tell them whether the string was always like this. You have years of these three strings tracking each other, so you know this is a change, and roughly when it was not yet happening. Turning a measurement into a diagnosis usually takes a second measurement, and in service the second one is very often the system's own history. That record is the most useful thing an O&M technician owns that the crew who built the site never had.

Why the other options are wrong

B) Soiling or partial shading on that string

Irradiance moves current almost in proportion to itself and moves open-circuit voltage only logarithmically, so losing a large share of the light costs a few percent of Voc rather than ten percent. The measurement rules it out directly anyway: short-circuit current came back normal, and soiling or shade heavy enough to cost 80 V would have shown there first. Worth being precise about why, though, because shading and this fault are related: shade does pull bypass diodes into conduction and does lower Voc while it lasts. These strings were measured in full sun, so nothing is shaded at the moment of the reading, which means whatever is bypassed is bypassed for a reason of its own.

C) A high-resistance connection somewhere in that string

A high-resistance connection is invisible to this particular test, and that is worth knowing as a fact about the test rather than about the fault. Open circuit means no current is flowing, and a series resistance with no current through it drops no voltage at all. A bad joint will not move a Voc reading by a single volt. It shows up under load, where it takes its share of the voltage, and it shows up on a thermal camera, where it turns that share into heat.

D) That string was built with one module fewer than the other two

The arithmetic rules this out and the records rule it out twice over. One module short would put string 3 exactly 40 V below its neighbours, and the gap is twice that. More to the point, a string built one module short is wrong from the day it is energised, and these three matched at handover and have tracked each other since. A build error does not wait years to appear.

References

• NABCEP OMAT JTA v.2026.5 — Domain 2, Task 3.G: "Series and parallel circuits"
• No external standard is cited for the correct answer. It follows from how voltage and current behave in a series circuit and from what a bypass diode does. The module open-circuit voltage used in the arithmetic is a property of the modules on that site, read from their datasheet, not a figure from a standard.