As an Amazon Associate, Akermin earns from qualifying purchases. Product links on this page are affiliate links.
An end point tester visits two operating states, open terminals and shorted terminals, and reports the voltage and current it finds there. A curve tracer visits every state between them and reports the path. That distinction decides what each can detect: the end points will catch a string that is miswired, short of a module or reversed, while only the path reveals a fill factor that has collapsed, a bypass diode carrying current or a leakage path across the cells, because none of those three moves either end of the trace very far.
An end point tester samples two states, a tracer samples the whole sweep
Consider what each instrument physically does once the leads are on. An end point tester presents a very high impedance to read voltage with no current flowing, then closes an internal shorting path through a current shunt to read current with the voltage collapsed. Two states, two readings. A maximum power point tester adds a third behaviour: it steps an electronic load until the product of current and voltage stops increasing, and reports that peak.
A tracer does something categorically different. It puts a continuously varying load across the array, most often by charging a capacitor, so the operating point drifts from one end to the other while the instrument samples current and voltage pairs throughout. The output is not three numbers but a locus of several hundred coordinates captured before the sun has had time to change. Everything the instrument can tell you that an end point tester cannot comes from the coordinates in between.
Both classes appear in the same toolbags and the roundups treat them together, which is why choosing a PV tester turns on what you intend to prove rather than on feature counts. A straightforward field unit such as the one covered in the Allmeter panel tester review answers a different question from a sweep instrument, and answers it faster.
What a pair of end points can still tell you
It would be wrong to treat the simpler instrument as a toy. A great deal of what goes wrong with a new array shows up cleanly at the two ends. A string with one module omitted reads an open circuit voltage short by one predictable step, and counting that step is often enough to find the gap. A string wired with a module reversed reads short by two steps, because the missing contribution is subtracted rather than absent. A blown string fuse reads full voltage and no current at all.
Comparing strings against each other is just as productive. Identical strings in the same plane under the same sky should report open circuit voltages within a percent or so of one another and short circuit currents within a few percent. An outlier in voltage points at module count or a connection; an outlier in current points at shading, soiling or a string that is not actually in the same plane. None of that needs a sweep, and all of it is faster without one.
The limitation only bites when the end points agree with expectation. That is the condition under which the simple instrument has nothing left to say, and it is also the condition under which the most expensive faults hide.
Faults that only appear between the end points
Series resistance is the clearest case. Added resistance in connectors, solder joints or cable drops no current at open circuit and almost none at short circuit, so both end readings stay where they were while the knee of the curve leans over and maximum power falls. A module can shed a tenth of its output this way and pass a voltage and current check without raising a question. Chasing it means looking at the shape, then confirming with a resistance measurement at the joints.
A conducting bypass diode is the second case. It removes a block of voltage at a current the rest of the module still carries, which produces a stair in the plateau. The open circuit voltage does fall, but if the string is long the loss is a small fraction of the total and can be mistaken for temperature. The stair is unmistakable. A shunt path is the third: it tilts the plateau downward as voltage climbs, and again barely touches either end.
These are exactly the faults that commissioning standards put into the second tier of testing, and the instruments built for that tier say so. The Fluke SMFT-1000 solar tools kit states an all in one solution meeting IEC 62446-1 for Category 1 and Category 2 tests, open circuit voltage at the module or string up to 1000 V direct current, short circuit current up to 20 A direct current, and on location comparison of measured curve results against manufacturer curve data on the analyser screen.
How the two instrument classes describe themselves
| Question being asked of the array | Two end points | Full sweep |
|---|---|---|
| Is the string connected with correct polarity? | Yes, immediately | Yes, but slower |
| Is a module missing or reversed? | Yes, from the voltage step | Yes |
| Is a string fuse open? | Yes, current reads zero | Yes |
| Has fill factor fallen from series resistance? | No, both ends look normal | Yes, the knee is visibly rounded |
| Is a bypass diode conducting? | Rarely, only on a short string | Yes, a stair in the plateau |
| Is there a shunt leakage path? | No | Yes, the plateau slopes |
| Does the string match the manufacturer curve after translation? | Partly, at two points | Yes, across the whole trace |
The published descriptions follow that split closely. The HT Instruments Solar I-Ve states tracing of modules and strings up to 1500 V and 40 A, front and rear irradiance measurement for bifacial modules, open circuit voltage and short circuit current measurement, and a module performance database to compare against. The FLIR PV48 states instant maximum power and temperature measurement with graphical curve analysis up to 800 W per panel, a high contrast display readable in direct sunlight and a built in lithium battery; its voltage and current limits are not stated in the listing bullets, which matters because 800 W per panel is a power ceiling rather than a string rating.
Kit composition is the other axis along which these products differ. The Fluke SMFT-1000 Pro kit states the same Category 1 and Category 2 coverage plus TruTest Advanced software for solar asset management and a set of MC4 clamp test leads, which is a documentation and fleet management difference rather than a measurement one. The tracer roundup sorts that class out in more detail.
Choosing between a sweep and two numbers for the work in front of you
Match the instrument to the claim you have to make. If the claim is that the array is wired as designed and every string is alive, two end points per string, compared against each other and against the design, is the efficient and sufficient answer. If the claim is that the array performs as specified, or that a string which looks healthy really is, the sweep is the only evidence that covers the question, and it needs irradiance and temperature recorded alongside it.
The same reasoning appears whenever a device is characterised at its terminals. A battery judged on terminal voltage alone looks fine until it is asked to deliver current, which is why a discharge profile rather than a resting reading is the useful measurement, and why capacity testers exist as a separate instrument class. End points describe what a device is at rest. A sweep describes what it does under load, and under load is where the money is.
Frequently Asked Questions
Can a maximum power point tester substitute for a tracer?
It gets closer than a plain voltage and current check, because it hunts for the load that maximises the product and therefore reports a peak power figure. What it still does not give you is the path taken to reach that peak, so a module with a soft knee and a module with a conducting diode can report the same deficit for entirely different reasons. The peak tells you how much is missing; the shape tells you where it went.
How does a tracer sweep a string so quickly?
The common method charges a capacitor across the array through the instrument. As the capacitor fills, the operating point walks itself from short circuit toward open circuit while the instrument samples current and voltage pairs continuously, often several hundred of them inside a few tens of milliseconds. Doing it fast matters because irradiance and cell temperature must not change appreciably during the sweep, or the trace mixes two different conditions.
Does a tracer need an irradiance measurement to be useful?
For shape, no, and shape alone finds most field faults. For any comparison against a nameplate or against a sweep taken on another day, yes, because translating a measured curve to standard conditions needs irradiance in the plane of the array and a module temperature. Without those two inputs the comparison carries more uncertainty than the defect being hunted, and a healthy string can be condemned on a hazy afternoon.
Is a tracer safe to connect to a live string?
Only an instrument designed for the task and rated for the string voltage, connected through the proper leads with the string isolated at the combiner before the connection is made. Direct current arcs do not self extinguish at a zero crossing, so making or breaking a loaded connection on a string is the dangerous moment, not the measurement itself. The instrument should do the switching internally once it is connected.
Why do commissioning documents ask for both?
Because they answer different questions and the cheap one answers the common question. Voltage and current checks on every string catch wiring errors, missing modules, reversed polarity and blown fuses, which are the faults most likely to exist on handover day. The sweep is reserved for strings that look wrong, for contracts that specify it, and for the baseline record that makes later comparison possible.