Skip to content

Power Quality Analyses

About this page

This page was written by the New Horizons engineering team, which operated until December 2025. First-person references ("we", "our") are the original authors'. A number of those authors are now working with Phoenix GH, which maintains this material and stands behind it. Questions and comments are welcome — please get in touch.

Unevenness in the availability and quality of mains electricity can pose considerable challenges to low- and middle-income countries (LMICs), both in terms of economic development and implementation of essential public sector services such as health care delivery. This is particularly true in health care settings where alternating-current (AC) electrical power is required to support priority medical equipment such as vaccine refrigerators and freezers, diagnostic equipment, and oxygen concentrators.

Beginning in 2018, New Horizons – which, at that time, was operating within Global Good / Intellectual Ventures – analyzed voltage and other power measurements collected by mains-powered vaccine refrigerators operating in health facilities in Kenya and Nigeria to illuminate power availability and quality in these settings. These analyses should prove valuable for specifications- and standards-setting bodies, equipment designers, and equipment purchasers. Our team collaborated on the design of the Aucma CFD-50 vaccine refrigerators (hereafter referred to as devices) used in the study. In Kenya, 210 devices were installed in individual health facilities in 36 counties in collaboration with the National Vaccines and Immunization Program (NVIP), with 176 devices purchased by the Government of Kenya and the remainder purchased by GGF. In Nigeria, 93 devices were installed in 8 states and the Federal Capital Territory in collaboration with the National Primary Health Care Development Agency (NPHCDA), with all devices purchased by GGF. Data collection began as devices were installed beginning March 9, 2018 and continuing until August 31, 2019. All devices reported power data for at least 1.5 continuous months, with an average of 10 months of data reported per device.

The second linked document focuses exclusively on overvoltage conditions and represents a three-year period of data collection. The data underlying the overvoltage analysis was collected from the same devices as the main Power Quality analysis, but the duration of the data collection for the overvoltage study was twice as long.

Documents

Data

The data collected in Kenya is freely available to researchers for further analysis. This dataset extends from device manufacture, through the collection periods described above, and continues until June 2025.

  • Kenya power data in Apache Parquet format, partitioned by month. This dataset is generously hosted by nLine, Inc.
  • Kenya power data in tab-separated values (TSV) format, partitioned by quarter-year, and hosted at the Harvard Dataverse. This repository also contains the SQL queries used to analyze the data when it was stored in a relational database. They are intended to be informative since they are not directly applicable to the TSV-formatted data.