Surge electrolyzer has a rapidly oscillating minimal power usage that switches rapidly at idle between values of 0.0, 10.0, and 20.0 kW.

Oscillations come in packets. Within a packet they appear to be very regular to the extent the monitor's resolution allows to judge given that the spikes are well under a second each.

The packets, however, seem to have slightly irregular random duration, coming at slightly irregular intervals, such that the average power usage averaged over 5s seems to consistently fluctuate between values of 9.97, 10.0, and 10.03 kW.

Reproduces with electrolyzer 1 and 2, and regardless of whether the electrolyzer's fluid inputs/outputs are connected to the corresponding fluid pipelines or not.

On the screenshots by the link the case shown is a single electrolyzer 1, which is the only surge electrolizer in the entire power grid, is connected to the power grid but not fluid inputs / outputs, and neither it has ever been connected to fluid inputs or outputs, thus proving that it never actually worked and all of its power usage is the minimal power usage.

In cases when the electrolyzer is connected to fluid inputs, oscillations occur similarly but only when there is no excess power and thus the surge electrolyzer is not working. The issue does not occur with priority electrolyzers.

https://imgur.com/a/8UWrASd