Bergoz MX-BPM, Руководство пользователя

Страница: 44 / 49

application was transverse feedback and beam interlock systems near insertion devices in
the ALS (Advanced Synchrotron Light Source at the Lawrence Berkeley National
Laboratory).
Our BPM signal processor uses the same basic concept and architecture, but
its centre frequency is programmable anywhere between 100 MHz and 1 GHz. It tracks
input signals over a limited frequency range (±200 kHz, extension possible up to ±1
MHz). Optional features are fast signal gating and a limited capability for single turn
measurements. Low cost, reliability and simplicity of production were the other key
requirements.
The central function of the signal processor is a receiver in the
VHF/UHF frequency range. It was therefore a good idea to have a closer look at modern
telecommunication equipment before starting our project. High-Tech consumer products
like cellular telephones and GPS (Global Positioning System) satellite receivers have
started a revolution in modern rf technology. Complex systems are now being built on
very small circuit boards, using novel integrated circuits functions and miniature passive
components . They do not need heavy shielding enclosures.
Taking inspiration and components from those designs allowed to shrink the
processor module to a single height Eurocard (100 ¥ 160 mm). A build-in pre-
programmed frequency synthesizer for the down converter makes the unit completely self
contained and independent of any external control or timing system. In its default
operating mode, the user has only to connect the pick-up signals and the power supplies
and obtains the normalised x and y outputs. All intermediate coaxial connectors were
eliminated by using on-board miniaturized band pass filters, micro strip low pass filters
and adjustable attenuators. This improves the reliability of critical interconnections and
lowers the component cost significantly.
Operating principle
The BPM signal processor has 4 parallel input channels for the 4 input signals
A to D. The input multiplexer switches sequentially from one input channel to the next. A
single receiver measures the signal for each input channel in turn and stores its value in 4
corresponding analog memories. The voltages (VA to VD) in each of these memories are
therefore proportional to the power levels of the original input signals. To normalize the
readings (to make them independent of changing beam current), automatic gain control
(agc) is used to keep the sum VA+VB+VC+VD constant.
Sequential scanning and the use of a single receiver simplifies the circuits and
has as major advantage, that the gain for all channels is always identical. Gain changes
due to temperature effects or component aging are eliminated by the automatic gain
control. The measurements are only valid if input conditions do not change during the
scan. In practice, this is a very good method for beam position measurements of the
average closed orbit.
The X and Y coordinates are calculated with a matrix of sum & difference