As an Amazon Associate, Akermin earns from qualifying purchases. Product links on this page are affiliate links.
DANOPLUS · Solar Power Meter Digital Solar Meter, Solar Irradiance Meter for Solar Energy Testing, Solar Transmission Meter with Accuracy Drift<1.5%/Year&Btu/(ft2-h)& W/m² Units for Meteorology Sun Measurement
View on AmazonPrice and availability are shown on Amazon and can change at any time.
A handheld solar irradiance meter reading 0.1 to 1999.9 watts per square metre, with the same span selectable in British thermal units per square foot hour, a published spectral response of 340 to 1100 nanometres, a hold button, a sampling figure given as 2.5 per second, a large display reading to 19999 counts and a stated drift of less than 1.5 percent a year. The bullet headed High Accuracy contains that drift figure and no accuracy figure, and the two describe different properties. The spectral range is the line that tells you most, and it is also the line that contradicts one of the applications.
Drift and accuracy are different specifications
The first bullet opens by saying the meter is accurate and then supports it with a drift figure of less than 1.5 percent a year. The product name carries the same figure. Nowhere does an accuracy specification appear.
These are not interchangeable. Accuracy answers how far today a reading can be from the true value, and it is normally written as a percentage of the reading plus a fixed term, because the two sources of error behave differently. Drift answers something else entirely: how much the calibration moves as the instrument ages, through sensor degradation, window discolouration and component tolerance shift.
The gap between them is easy to see with an example. An instrument reading 15 percent low, and continuing to read 15 percent low for years, has excellent stability and poor accuracy. Drift under 1.5 percent a year says the number you get next summer will closely resemble the number you get this one. It does not say either of them is right.
That said, stability is worth publishing and few listings in this category do. It tells you how long a calibration is worth, which matters because photodiode based instruments do age. For repeated measurements of the same array over seasons, which is a large part of what these meters are bought for, stability is arguably the more useful of the two. Our note on what sensor drift is explains how to read a figure like this and what recalibration intervals it implies.
340 to 1100 nanometres is the most useful line here
Buried in the third bullet, between an operating note and a sampling figure, is the specification that defines this instrument: a measuring wavelength range from 340 to 1100 nanometres.
That band identifies the technology without naming it. Silicon responds to light from roughly the near ultraviolet out to about 1100 nanometres, where the photon energy falls below what the material can absorb. Publishing 340 to 1100 is publishing a silicon photodiode.
Now compare that with what arrives from the sky. Solar radiation at ground level carries useful energy from around 300 nanometres in the ultraviolet through the visible and out past 2500 nanometres in the infrared. A sensor covering 340 to 1100 therefore samples a portion of the incoming spectrum, not the whole of it.
Yet the meter reports watts per square metre, a figure that means total energy. It can only do so by taking its partial measurement and scaling it, using an assumed relationship between the band it sees and the energy it does not. That assumption is baked in during calibration under a reference sky. When the sky resembles the reference, the scaling is sound and the reading is good. When the spectrum shifts, the scaling is quietly wrong and the display gives no hint.
Publishing the band is what lets you reason about all of this, which is why it is the best line in the listing. Most competing listings publish a range in watts per square metre and never mention wavelength at all. Our primer on what solar irradiance is covers how the quantity is defined and why the spectrum matters to measuring it.
Why that band suits a solar panel and not a weather station
Here is the pleasing part. A crystalline silicon photovoltaic module responds over very nearly the same wavelengths as a silicon photodiode, for exactly the same physical reason. Both are limited at the long end by where silicon stops absorbing.
So this meter sees sunlight approximately the way the panel sees it. Energy out beyond 1100 nanometres, which a panel cannot convert into anything but heat, is also invisible to the sensor. A full spectrum thermopile instrument is more correct about total energy arriving and therefore, for the specific question of how much convertible light is reaching this array, arguably less representative.
That is a real argument in favour of a silicon meter for photovoltaic work, and it holds whether or not the seller understood it well enough to make it.
The same reasoning turns against the last bullet, which offers solar energy research, building energy performance assessment and meteorology among the applications. Meteorological irradiance measurement is about total incoming radiation, which is a different quantity, and it is conventionally made with a classified thermopile instrument. No classification, no directional response and no temperature coefficient appears here. The meteorology claim is the one application this specification does not support, and our comparison of solar irradiance meters covers where that line sits.
The vehicle window claim argues with the spectral range
The final application in the final bullet is measuring light intensity through vehicle windows. Set that next to the third bullet and the two do not sit comfortably together.
Automotive glazing is engineered to change the solar spectrum. Laminated windscreens cut ultraviolet heavily, and a great deal of modern glass and film is designed specifically to reject near infrared, which is where a lot of the heating energy sits. That is the entire point of the product category.
A sensor that sees 340 to 1100 nanometres and scales its reading on an assumed spectral shape is therefore being handed exactly the situation its calibration does not describe. The light has had both ends of the spectrum selectively removed before it arrives.
What you get is still useful, provided you read it for what it is. The reading will be repeatable, so comparing one pane with another, or the same pane with and without a film, or a front screen against a rear one, is a sound relative measurement. What you cannot do is treat the resulting figure as watts per square metre in the sense the instrument prints, because the scaling behind that unit no longer applies. Relative comparison good, absolute figure not.
Ranges, sampling and the rows left empty
| Specification | What the bullets publish |
|---|---|
| Measuring range | 0.1 to 1999.9 watts per square metre |
| Alternative unit | 0.1 to 1999.9 British thermal units per square foot hour |
| Spectral response | 340 to 1100 nanometres |
| Stability | Drift under 1.5 percent a year |
| Sampling | Given as 2.5 t/s, read as two and a half samples per second |
| Display | Large liquid crystal, maximum displayed value 19999 |
| Data retention | Hold button |
| Setup required | None, direct measurement without adjustment |
| Accuracy of a reading | Not stated in the listing bullets |
| Resolution | Not stated in the listing bullets |
| Angular or cosine response | Not stated in the listing bullets |
| Temperature coefficient | Not stated in the listing bullets |
| Instrument classification | Not stated in the listing bullets |
| Logging or stored readings | Not stated in the listing bullets |
Two of those ranges deserve a comment each. The top of scale at 1999.9 watts per square metre gives generous headroom, since full sun on a clear day at a sensible angle lands near 1000 and the standard condition panels are rated at is exactly 1000. Reflected light from snow or a bright roof can push a tilted reading above that, so headroom is sensible rather than wasted.
The 19999 count display is a specification that does not correspond to either published range. Both ranges stop at 1999.9, which uses five digit positions but not the five digit span. Either the display is shared with another model or the range could be wider than stated. It is a small inconsistency on a page that otherwise publishes real figures.
The absent angular response is the omission with the most practical weight. Irradiance on a surface depends on the angle between that surface and the sun, and a sensor whose response does not follow the cosine of that angle correctly will read differently simply for being tilted. For measuring in the plane of an array, which is the job, that error goes straight into the result and the listing says nothing about it. The practical defence is to hold the sensor flat against the module glass every time, so whatever the angular error is, it stays the same across your readings.
Use it to compare panels, not to classify a site
This is a good instrument for relative solar work. Checking whether an array is getting the light you assumed. Comparing two roof planes before deciding where panels go. Confirming a module that reads low is in the same sun as the one beside it. Watching how irradiance falls through an afternoon or under thin cloud. Judging shading from a tree at different times. For all of that the silicon response is well matched to a silicon panel, the published drift says the readings will be comparable from one season to the next, the hold button lets you take a figure at arm length on a roof and read it afterwards, and the unit selection suits whichever convention your paperwork uses.
It is the wrong instrument for anything that has to be defensible. No accuracy figure is published, so no tolerance can be quoted on a reading. No angular response is given, so a tilted measurement carries an unknown error. No classification is stated, so the meteorology and building assessment applications in the last bullet are not supported by anything on the page. And through treated glass, the one application that directly conflicts with the published spectral range, use it for comparison only.
The listing earns genuine credit for publishing a wavelength band and a stability figure, which most of its competitors do not. It just labelled the stability figure as accuracy, and accuracy is the number that is missing.
Recommended Tools
Products shown because their documented specifications fit this guide. Price and availability are on Amazon and can change.
Frequently Asked Questions
Is drift the same as accuracy?
No. Accuracy is how close a reading is to the true value now, normally quoted as a percentage of reading plus a fixed term. Drift is stability: how much the calibration moves over time, here quoted as less than 1.5 percent a year. An instrument can be stable and wrong, which is to say it gives you almost the same incorrect figure for years. Drift tells you how often recalibration is worth doing. It tells you nothing about the number on the display today, and no accuracy figure is published anywhere in these bullets.
What does a 340 to 1100 nanometre response mean in practice?
It tells you the sensing element is a silicon photodiode, because that band is almost exactly where silicon responds. Sunlight at ground level carries meaningful energy from around 300 nanometres in the ultraviolet out past 2500 in the infrared, so a 340 to 1100 window sees part of the spectrum and not all of it. The instrument reports watts per square metre anyway, which it can only do by scaling its partial measurement using an assumed spectral shape. Under a sky resembling the one it was calibrated against that works well. Change the spectrum and the scaling is no longer right.
Why is that band well suited to photovoltaic work?
Because a crystalline silicon solar cell responds over very nearly the same range, for the same physical reason: the band gap of silicon sets where absorption begins and ends. So this meter sees light roughly the way the panel sees it. A full spectrum thermopile instrument is more correct about total energy and includes long wave infrared that a silicon panel cannot convert. For the question how much usable light is reaching this array, a silicon sensor is arguably better matched than the more expensive instrument.
Will it read correctly through a car window?
This is where the listing argues with itself. The last bullet offers measuring light intensity through vehicle windows as an application, and the third bullet publishes a 340 to 1100 nanometre response. Automotive glazing is specifically designed to cut ultraviolet and, in many modern cars, a good deal of near infrared as well, and that reshapes the spectrum the sensor is trying to scale from. You will get a repeatable number useful for comparing one pane against another, and not a trustworthy figure in watts per square metre.
Is this a pyranometer?
The second bullet calls it one, and the word is doing some work. In metrology a pyranometer usually means a thermopile instrument classified under ISO 9060, which absorbs across the whole solar spectrum and is categorised by its directional response, stability and temperature behaviour. This is a silicon photodiode meter covering part of that spectrum, with no class stated and no angular response published. Silicon devices are widely and legitimately used in solar work, but the absent class is why the meteorology application in the last bullet is a stretch.
What does a sampling time of 2.5 t/s mean?
The notation is the listing being careless rather than a real unit. Read in context it almost certainly means two and a half samples per second, which is a sensible update rate: fast enough to follow a cloud edge across a panel, slow enough that the display is readable. If it were intended as the interval between samples it would be 2.5 seconds, which would be unusually slow for this kind of instrument. Nothing in the bullets resolves the ambiguity.
Affiliate disclosure: Akermin earns a commission from qualifying Amazon purchases made through links on this page. Our editorial picks are based on documented specifications and owner feedback, not commissions.
Related Guides
UNI-T UT503PV Review: A DC 1000 V Headline on a CAT III 600 V Body
A solar insulation resistance tester that works on energised PV strings up to DC 1000 V, with test voltages of 125, 250, 500 and 1000 V, a range to 4000 megohms, automatic discharge, 1000 stored result sets and USB or Bluetooth transfer.
Read More→
ANCEL BA101 Review: The Exclusion List Is the Most Useful Bullet
A 12 volt battery, cranking and charging system tester covering 100 to 2000 CCA that reports state of health, state of charge, voltage, current and CCA, drawing its own power from the battery under test.
Read More→
Tenmars Solar Power Meter Review: One Accuracy Figure, Two Very Different Jobs
A handheld solar irradiance meter reading 0 to 1999 watts per square metre or 0 to 634 BTU per square foot per hour, with a selectable 199.9 or 1999 scale, maximum and minimum tracking, data hold and manual zeroing.
Read More→