The Transition Radiation Detector (TRD) has 18 SuperModules (SMs) with 5 stacks of 6 Read-Out Chambers (ROCs). Each ROC has 6 or 8 Read-Out Boards (ROBs), 1 Detector Control System (DCS) board, and 2 Optical Read-Out Interfaces (ORIs). Each ROB has 17 or 18 Multi-Chip Modules (MCMs). See figure below for ROC coverage with electronics.
Apart from its electronics, a chamber has an active volume with: a radiator stack, a drift region, layers of cathode and anode wires, and a pad plane. The pad plane has 144 pads in each column, covered by 8 data collection MCMs (see below) - hence 18 pads per MCM.
An MCM consists of a PreAmplifier and Shaper (PASA) chip and a Tracklet Processing (TRAP) chip. The PASA chip collects the pad signals from 18 pads and sends them to Analog-to-Digital Converters (ADCs). The TRAP chip contains the ADCs, followed by digital filters, event buffers, and the LTU.

How a ROC is covered by electronics:

• each box is a ROB - 6/8 per ROC;
• each filled square is an MCM - 16 per ROB for data collection plus 1 per ROB for data merging plus 1 in two ROBs for half chamber merging;
• 1 DCS board for the whole chamber;
• 2 ORIs (not shown in the picture).

The Local Tracklet Unit (LTU) does the hit detection and tracklet fitting in its 4 CPUs. The LTU results from the ROCs in one stack (6 chambers) are sent to the Global Tracking Unit (GTU) which finds the global TRD tracks used by the TRD trigger.