D. Managing ion chamber signal chains

In the 2023-3 and 2024-1 cycles, BMM went through a transition in how direct beam signals (I₀, I_t, I_r) are measured.

D.1. The old system

BMM has been using the same ion chambers that were used at X23A2 at the old NSLS. They are excellent ion chambers – robust, linear, easy to use. At BMM, we used a QuadEM electrometer to read the current signal from each ion chamber. The QuadEm is a four-channel device, so I₀, I_t, and I_r were fed into channels 1, 2, and 3. The fourth channel was occasionally used for an electron yield detector.

Voltage was supplied from the same Ortec high voltage supply used at X23A2. The Ortec device sits outside the hutch. The high voltage was fed via cabling through a roof labyrinth and patch panels.

An IOC for reading the QuadEM lives on xf06bm-ioc2 and communicates with the QuadEM device via a cat6 cable.

Fig. D.1 (Left) The old I_r detector. (Right) The quadem electrometer used to measure the three old-style ion chambers.

D.2. The new system

The new ion chambers are integrated devices. The low half of the enclosure is a gas-filled volume with capacitor plates made from printed circuit board. The plates are guarded at both ends and upright print circuit board pieces are used to guard the fringes of the fields. These upright pieces are strips of copper bridged by surface-mounted resistors acting as a sequence of voltage dividers. This keeps the field lines art the periphery of the capacitor plates nicely parallel.

The upper part of the enclosure has the full signal chain and a voltage supply. This includes an amplifier, an analog-to-digital converter, and a microprocessor running an embedded Debian system and a two-channel version of the QuadEM IOC. The voltage supply is fully adjustable up to 2000 V, but are run at 200 V.

The electronics package and the voltage supply are powered a 6 V DC power supply. A cat6 cable for each ion chamber allows the IOCs to communicate on the INST network.

The ion chambers also have an optical fiber input port so that data can be time stamped with the global timing signal from the accelerator. This will be used in the near future to implement continuous scanning on the monochromator.

Fig. D.2 The new I₀ detector with integrated electronics and voltage supply.

D.3. Configuration in bsui profile

I am trying to make it easy to configure bsui to switch easily between ion chambers and electrometers. This is a work in progress. Here I document the current, slightly awkward, state of affairs.

In BMM/user_ns/dwelltime three boolean parameters are set: with_ic0, with_ic1, and with_ic2.

These are used to synchronize setting integration times across the various signal chains via the LockedDwellTimes object.

Each new ion chamber in use needs to have its flag set to True.

The detectors themselves need to be configured correctly in this file using the flag values.

The QuadEM and each individual ion chamber will be configured if available on the network. If the network connection cannot be established, a noisy_det from ophyd.sim is created with the same name.

Notes:

1. If the ion chamber device can be made, it will be made even if its flag is set to False. This allows interaction with an ion chamber even if it is not expected to be used in a scan.

2. Once all 3 new ion chambers are in place and in use, the QuadEM device will still be made. This will allow use of electron yield and other detectors as well as measurement of any other current signals.