{"id":19567,"date":"2026-09-24T05:08:54","date_gmt":"2026-09-24T08:08:54","guid":{"rendered":"https:\/\/rtmedical.com.br\/tmp-en-1790237333821\/"},"modified":"2026-09-24T05:09:01","modified_gmt":"2026-09-24T08:09:01","slug":"pacs-workstation-power-consumption","status":"publish","type":"post","link":"https:\/\/rtmedical.com.br\/en\/pacs-workstation-power-consumption\/","title":{"rendered":"PACS Workstations Draw Power Even When No One Reads"},"content":{"rendered":"

A PACS reading workstation draws an average of 238 W<\/strong> in use and still pulls 175 W<\/strong> when the computer is put to sleep, according to a study published on September 22 in Cureus<\/em>. Across an academic department with 105 workstations, that adds up to 25 kW<\/strong> with everything active and still 18.4 kW<\/strong> with the machines idle. Only a full PC shutdown brings the load down to 2.4 kW. The work, covered by AuntMinnie, is by Anna Luisa K\u00fchn, Alexander A. Bankier and colleagues.<\/p>\n

\"Chart
Sleep mode saves little. The real drop comes when the PC is actually shut down.<\/figcaption><\/figure>\n

How the study measured consumption<\/h2>\n

The authors started from a simple observation: the literature on radiology’s carbon footprint concentrates on CT and MRI, and almost no one has methodically measured the consumption of reading workstations, even though they vastly outnumber scanners. Two attending radiologists took the measurements between December 2024 and January 2025, using an external energy meter placed between the wall outlet and the workstation.<\/p>\n

Three real setups were assessed: a workstation in the cardiothoracic imaging reading room, one in mammography, and a remote-reading station in one radiologist’s home. All three used the same computer (HP Z440) with different combinations of Barco diagnostic displays, general-purpose Lenovo monitors and a dictation microphone. Each station was measured in five modes: active, reading and dictating; PC asleep with monitors on standby; PC fully shut down with monitors on standby; and the two transitions back to ready-to-read status. Each combination was repeated three times, for 45 measurements in total.<\/p>\n

Static modes were measured over 10-minute windows and extrapolated to average power. Variability between repetitions was low and not statistically significant, and including or excluding the mammography station did not change the result either. Mammography was nonetheless left out of the departmental calculation: at the institution, those stations cannot be shut down overnight because they undergo remote quality inspections.<\/p>\n

The numbers per workstation and per department<\/h2>\n

Per station, mean values were 238 W active, 175 W asleep and 23 W with the PC shut down. Waking from sleep took about 1.5 minutes at an average of 290 W. Booting from a full shutdown took 5.7 minutes on average, with wide variation between measurements. Scaled to the department’s 105 stations, the figures reach 24.99 kW at full operation, 18.37 kW asleep and 2.41 kW shut down. Active mode draws 10.35 times the power of shutdown; sleep draws 7.61 times as much.<\/p>\n

To make the magnitude intuitive, the study benchmarks the load against household appliances: one hour of the department at full operation equals just over eight hours of a 3 kW air conditioner.<\/p>\n

The most useful figure for daily practice is the break-even time<\/strong>: how long a station must stay in a low-power mode for the savings to offset the energy needed to bring it back. The authors calculated 6.67 minutes for sleep and 5.86 minutes for full shutdown. In other words, any break longer than about seven minutes already justifies dropping the workstation into a lower-power state, and a full shutdown pays off within practically the same interval.<\/p>\n

Technical reading: sleep is not really sleeping<\/h2>\n

What stands out is the small gap between active (238 W) and sleep (175 W). In both low-power modes the monitors are on standby, so the 152 W difference between sleep and shutdown comes essentially from the computer. This is our interpretation, not a conclusion of the paper, but a PC in true suspend (the ACPI S3 state) typically draws only a few watts. A value around 150 W suggests the machine never reaches real suspend, which is common on enterprise workstations because of IT-locked power policies, professional graphics cards that stay powered, remote management agents or wake-on-LAN settings. The abstract itself describes this mode as “logged off,” reinforcing the idea that logging out is not the same as saving energy.<\/p>\n

As an illustrative exercise of our own using the study’s figures: if all 105 stations spent 16 hours a day asleep rather than shut down, the difference would be about 16 kW, roughly 255 kWh per night and more than 90 MWh a year. The real value depends on how many stations truly sit idle and on the shift schedule, but it shows why the item belongs in a department’s energy inventory.<\/p>\n

The finding connects with our look at how diagnostic displays are now ecosystems, not just monitors<\/a>: DICOM calibration, remote QA and management agents expand what the workstation does and also what it consumes after hours.<\/p>\n

What changes in practice for imaging departments<\/h2>\n

The first step is cheap: measure. A plug-in power meter costs little and lets a department repeat the protocol on one or two typical stations. The second is to work with IT so the power policy is real, with functional S3 sleep or scheduled after-hours shutdown and automatic wake before the first shift so that reading does not start late. The third is to separate the exceptions: mammography stations subject to remote QA, 24\/7 on-call stations and servers do not follow the same rule.<\/p>\n

For practices that run round-the-clock coverage or teleradiology, the rule has to follow the schedule: a reading-room station used only in business hours is an obvious candidate for overnight shutdown; an on-call station is not. Any change to shutdown routines should also be coordinated with the PACS architecture, prefetch queues and maintenance windows, as discussed in our piece on PACS integration, IHE profiles and disaster recovery<\/a>, and with the broader goal of an efficient PACS workflow<\/a>.<\/p>\n

Limitations and open questions<\/h2>\n

The authors list the constraints themselves. The study comes from a single academic center, measured only three stations that all shared the same PC model, and used 10-minute windows that may not capture load variation when opening large studies such as multiphase CT or tomosynthesis. Administrative and IT workstations were excluded, and the extrapolation to 105 units assumes all behave like the three measured. Finally, the paper reports power rather than annual kWh or carbon emissions, which depend on each region’s electricity grid.<\/p>\n

Even with those caveats, the study delivers what was missing: measured numbers, conversion factors between modes and a break-even time that is easy to communicate to the team. The underlying message is blunt: radiologists take breaks, but PACS workstations rarely do.<\/p>\n

Source:<\/strong> Cureus (K\u00fchn et al., 2026), via AuntMinnie<\/a><\/p>\n","protected":false},"excerpt":{"rendered":"

A study measured 238 W per active PACS workstation and 175 W asleep; 105 stations total 25 kW. Shutting down pays off for breaks over 7 minutes.<\/p>\n","protected":false},"author":1,"featured_media":19530,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"om_disable_all_campaigns":false,"_monsterinsights_skip_tracking":false,"_monsterinsights_sitenote_active":false,"_monsterinsights_sitenote_note":"","_monsterinsights_sitenote_category":0,"ngg_post_thumbnail":0,"_rt_cluster":"","fifu_image_url":"","fifu_image_alt":"","footnotes":""},"categories":[101,100,230],"tags":[],"class_list":["post-19567","post","type-post","status-publish","format-standard","has-post-thumbnail","category-pacs-en","category-radiology","category-software-en"],"aioseo_notices":[],"rt_seo":{"title":"","description":"A study measured 238 W per active PACS workstation and 175 W asleep; 105 stations total 25 kW. Shutting down pays off for breaks over 7 minutes.","canonical":"","og_image":"","robots":"index,follow","schema_type":"Article","include_in_llms":true,"llms_label":"PACS workstation power consumption","llms_summary":"K\u00fchn, Bankier and colleagues (Cureus, Sept 2026) measured three PACS workstations in five modes: 238 W active, 175 W with the PC asleep and 23 W with the PC shut down. Scaled to 105 stations: 24.99 kW active, 18.37 kW asleep, 2.41 kW shut down. Break-even times for low-power modes: 6.67 min (sleep) and 5.86 min (shutdown).","faq_items":[],"video":[],"gtin":"","mpn":"","brand":"","aggregate_rating":[]},"_links":{"self":[{"href":"https:\/\/rtmedical.com.br\/en\/wp-json\/wp\/v2\/posts\/19567\/"}],"collection":[{"href":"https:\/\/rtmedical.com.br\/en\/wp-json\/wp\/v2\/posts\/"}],"about":[{"href":"https:\/\/rtmedical.com.br\/en\/wp-json\/wp\/v2\/types\/post\/"}],"author":[{"embeddable":true,"href":"https:\/\/rtmedical.com.br\/en\/wp-json\/wp\/v2\/users\/1\/"}],"replies":[{"embeddable":true,"href":"https:\/\/rtmedical.com.br\/en\/wp-json\/wp\/v2\/comments\/?post=19567"}],"version-history":[{"count":1,"href":"https:\/\/rtmedical.com.br\/en\/wp-json\/wp\/v2\/posts\/19567\/revisions\/"}],"predecessor-version":[{"id":19569,"href":"https:\/\/rtmedical.com.br\/en\/wp-json\/wp\/v2\/posts\/19567\/revisions\/19569\/"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/rtmedical.com.br\/en\/wp-json\/wp\/v2\/media\/19530\/"}],"wp:attachment":[{"href":"https:\/\/rtmedical.com.br\/en\/wp-json\/wp\/v2\/media\/?parent=19567"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/rtmedical.com.br\/en\/wp-json\/wp\/v2\/categories\/?post=19567"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/rtmedical.com.br\/en\/wp-json\/wp\/v2\/tags\/?post=19567"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}