the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 24 Apr 2025
Solar power systems for polar instrumentation: why night consumption matters
Abstract. Autonomous instruments, powered using solar panels and batteries, are a vital tool for long-term scientific observation of the polar regions. However, winter conditions, with low temperatures and prolonged lack of sunlight, make power system design for these regions uniquely challenging. Minimising winter power consumption is vital to successful operation, but power consumption data supplied by equipment manufacturers can be confusing or misleading. We measured the night consumption (power consumption in the absence of sunlight) of 16 commercially available solar regulators and compared the results to the manufacturers' reported values. We developed a simple model to predict the maximum depth of discharge of a battery bank, for given values of regulator and instrument power consumption, solar panel size, location, and battery capacity. We use this model to suggest the minimum battery capacity required to continuously power a typical scientific installation in a polar environment, consisting of a single data logger (12 mW power consumption) powered by a 12 V battery bank and 20 W solar panel, for eight different models of solar regulator. Most of the tested solar regulators consumed power at or below the manufacturer's reported values, although two significantly exceeded them. For our modelled scenario, our results suggest that the mass of the battery required may be reduced by a factor of 26x by exchanging a solar regulator with high night consumption for a more efficient model. These results demonstrate that a good choice of solar regulator can significantly increase the chances of successful year-round data collection from a polar environment, eases deployment and reduces costs.
- Preprint
(637 KB) - Metadata XML
- BibTeX
- EndNote
Status: final response (author comments only)
- RC1: 'Comment on egusphere-2025-1529', Anonymous Referee #1, 06 May 2025
- RC2: 'Comment on egusphere-2025-1529', Rolf Hut, 13 May 2025
-
AC1: 'Authors response to reviewers', Michael Prior-Jones, 11 Jul 2025
We thank the authors for their reviews and helpful comments.
We will incorporate the vast majority of suggestions given by the reviewers, and provide a tracked changes version of the manuscript highlighting our changes in response to the minor amendments requested.
There are however, two points to which we disagree with the suggestions:
- Incorporating Figures 3-5 into a single plot. Since the Y-scaling differs, we will instead merge into a multi-panel figure.
- Some of the comments made by Reviewer 1 do not match with our experience of polar fieldwork and our understanding of the facts around power system performance. Our position is supported by Reviewer 2. We outline our reasoning here:
Their initial statement that the work is of limited value because no scientist would deploy equipment in the Arctic without thorough prior testing does not fit with our experience. Part of the inspiration for this paper was seeing how scientists based in university departments, especially those without access to electronic engineering technician support, make poorly-informed choices when it comes to the engineering of their equipment. Reviewer 2 supports us on this point.
Their statement “l. 51, it is wrong that the load changes the capacity of the battery. That will always be the same. The sentence should be changed to say that increasing the load will reduce the record span, the time over which a battery of a given capacity can supply power to a system” is incorrect, as the reference cited on that line makes clear. Usable battery capacity is a function of load, as described by Peukert’s equation (Ioannou et al., 2016; Peukert, W, 1897). With a smaller load, more energy can be extracted from the battery. We will add some additional clarification on this point in the revised paper.
Table 2: Reviewer 1 suggests that this ‘should provide an estimate of the record length rather than the “minimum battery size required”’ – this is a misunderstanding of the model and so we will clarify this in the text. The model is calculating the minimum battery size required for the installation to operate continuously, day and night, for a complete year – i.e. the “record length” is infinite as the cycle simply repeats in subsequent years. We could add a calculation of the “number of days of autonomy”, which is the number of days that the battery can power the load with no solar input, if that is felt to be useful information.
In line 239 they doubt the effectiveness of insulation in battery boxes. These are commonly used in the Antarctic – one such example from EarthScope (formerly IRIS/PASSCAL) is available on the web here https://epic.earthscope.org/content/polar/equipment/year-round/station-enclosure
They also find the suggestions and recommendations unnecessary. We beg to disagree – whilst some of this may appear obvious and trivial to an experienced engineer, a scientist without an engineering background may well find these pointers helpful. We ask the editors for their judgement as to whether this is appropriate.
References
Ioannou, S., Dalamagkidis, K., Stefanakos, E. K., Valavanis, K. P., and Wiley, P. H.: Runtime, capacity and discharge current relationship for lead acid and lithium batteries, in: 2016 24th Mediterranean Conference on Control and Automation (MED), 2016 24th Mediterranean Conference on Control and Automation (MED), 46–53, https://doi.org/10.1109/MED.2016.7535940, 2016.
Peukert, W: Über die Abhängigkeit der Kapazität von der Entladestromstärke bei Bleiakkumulatoren, Elektrotechnische Zeitschrift, 156–158, 1897. Available electronically from https://babel.hathitrust.org/cgi/pt?id=mdp.39015084594855&seq=178&q1=peukert
Citation: https://doi.org/10.5194/egusphere-2025-1529-AC1
Data sets
Data and plots for figures 1-5 Michael R. Prior-Jones, Lisa Craw, Jonathan D. Hawkins, Elizabeth A. Bagshaw, Paul Carpenter, Thomas H. Nylen, and Joe Pettit https://doi.org/10.5281/zenodo.15114766
Model code and software
Spreadsheet model M. Prior-Jones and J. Hawkins https://doi.org/10.5281/zenodo.15115132
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 405 | 57 | 17 | 479 | 17 | 32 |
- HTML: 405
- PDF: 57
- XML: 17
- Total: 479
- BibTeX: 17
- EndNote: 32
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
In the manuscript I reviewed, the authors compare the power consumption of sixteen commercially available solar regulator devices to evaluate their impact on the energy design of instrumentations in polar regions, where the long polar night poses the challenge of long time dark and cold conditions. The study design is simple but adequate for the narrow, envisioned goal. The overall study is relevant and timely. It might be a proper contribution to the journal and of help especially for people who wish to put together an autonomous system without extensive testing, benchmarking and optimising, although it is clearly not a rocket science paper. That said, I shall note that the overall depth of the study – apart from the laudable goal of summarising the specifics of commercially available solar regulators and maybe putting together some specifics of lead battery models from one manufacturer – is rather poor. Many of the figures are redundant, poor in content and sometimes convoluted in their way of delivering information. Hence, in principal this manuscript is eligible to become published, if the editors decide that sufficient scientific depth is given and some improvements on the presentation are implemented.
In very general terms, I think it is too simplistic to assume one would pick a solar regulator, put together a sensor-logger system and place it in the arctic. What people will usually do is – as suggested in the manuscript – a dry run and logging of the energy consumption. So, there will be extensive information on the power consumption already available. Nevertheless, it may be helpful to have an overview of what is available on the market of solar regulators. In any way, the authors should down tone their way of framing the study a bit, to get closer to a scientist’s daily work reality.
The text sometimes struggles to provide quantitative information but instead stays vague and descriptive. I list some of these issues in the details below. I strongly encourage the authors to check their text and add quantitative information where possible as this would increase the weight of arguments and help readers to retrieve key information.
I find the term “night consumption” not really much more helpful than the other terms listed by the authors. Ideally, we are interested in the combined effect of “night and day” consumption. It is fair to stick to “night” conditions only for the specific case of polar deployments over winter time, but a more realistic case will be “day” deployments with fuzzy to cloud cover reduced sunlight, too. Hence, if I were to buy solar regulators, I would be keen on knowing the power consumption also during “day” conditions when incoming solar power – however weak it may be – needs to be managed, too. It would be good to at least mention this aspect and justify why, in the narrow context of polar nights, it is not considered by the study.
In the title, it would help to add “solar regulators” to be specific and give the implication that the study indeed focusses on that one item rather than on power consumption of the entire system.
l. 13, remove “uniquely”, no need to emphasise
l. 22, “factor of 26x” reads cumbersome, replace by something like “reduction to xx %”. In addition, converting power consumption to battery weight reduction is not wrong but also it dilutes the crispness of the message. I suggest to report on the percent power reduction and then, in a next sentence, report on several of the consequences of that, which includes reduction in battery capacity, costs and weight.
l. 23, “good choice”, remove “good”
l. 42, “3x” replace by “three times”
l. 51, it is wrong that the load changes the capacity of the battery. That will always be the same. The sentence should be changed to say that increasing the load will reduce the record span, the time over which a battery of a given capacity can supply power to a system.
l. 55, it would be good to provide some graph here instead of arbitrary listing some charge numbers for some temperatures. In line 59, there are references to such graphs. I really would like to see the relationship between temperature and charge as this relationship is non-linear and would help readers to better understand the problem of cold conditions and energy design.
l. 65, please give an overview of the range of power consumption of such LVD units, so that readers can get a feeling for the burden arising from them.
l. 79-92, this section about solar regulator architecture is overly long and clearly off scope. I encourage the authors to shorten it by at least 50 %. All what is needed is a basic understanding of the two designs (PWM and MPPT) and the consequences.
l. 110, please add that (I assume this was done in the test) the battery was always recharged before each test sequence. Add information on ambient temperature. Add information on Multimeter specs. Add information on the current measurement interval. Check, there are maybe further test specifications that need to be delivered to make the study reproducible.
l. 135, personally I am not a big fan of bullet points as they disrupt (intentionally) the flow of text and semantics. Make sure this is what you want to do with this style element in a “flow text piece of scientific communication”.
l. 148, there is some copy-paste fragment in this line. Please revise.
l. 156, not really meaningful in my opinion. Usually I would charge a battery fully full before dumping it in a hostile environment for many months. You may want to justify the assumed 50 % discharge level.
l. 179-180, two lines of text are a fair bit from enough to make a paragraph or even a chapter. Revise structure to be more meaningful.
All figures: remove the title from the plots as the figure caption is about to give that information.
Fig.1, I would combine fig.1 and fig. 2 into a two-panel figure and discuss these issues together. This might also solve the above comment of mine.
Fig. 3-5, these are arbitrary split visualisations of the same type and content. The data should be moved to a single figure, perhaps separated by thin vertical lines between the suggested classes. Although, I do not see a need for these classes (4.5, 7 and 9 A). If these are to be kept they need to be justified and explained.
I rather would like to see more figures of other aspects on the data. These could include: 1) A_rated versus A_measured as scatter plots (A classes as different symbols, and a 1:1 line) to focus explicitly on the difference of the two metrics rather than indirectly via bar plots; 2) price versus percent deviation of measured from rated Ampere values, 3) metrics separated by technology (PWM and MPPT), or further explicit tests and illustrations.
Fig. 4, X-axis label is not meaningful, revise.
Table 2 should also provide an estimate of the record length rather than the “minimum battery size required”.
l. 235, give information on which devices have temperature compensation as column in table 1
l. 236, define “LVD” (again, since it is a long time since its last usage)
l. 239, well but better insulation also prevents sunlight from warming up the casing, and I doubt that any insulation is good enough to prevent a battery from cooling under arctic conditions for many months. So, this argument does not make sense to me in the specific scope of the study.
l. 241-244, 256-257, these topics are really trivial information, I would suggest to remove them
l. 245-246, 249-255, these topics are way off scope. Consider removing.