Speaker
Description
Summary
I. POWER DISTRIBUTION SCHEME FOR THE LIQUID ARGON DETECTOR
The amount of power required by the front end crates of the Liquid Argon Detector
of the ATLAS experiment imposes the presence of a power supply in their vicinity.
As the magnetic field limits the use of power transformers, a front end power
supply based on modern DC/DC converters was chosen. AC/DC converters that sit in a
control room located 100 meter away of the detector provide the bulk power to the
front end power supplies.
The long DC power link faces several electromagnetic compatibility issues that are
specific to this configuration:
• Stability of the power link.
• Common mode and differential mode noise propagation along the cable.
• EMI emissions of the power cable.
II. STABILITY OF THE DC POWER LINK
When both AC/DC and DC/DC converters chained together, the resulting transfer
function involves the ratio between the output impedance of the AC/DC converter and
the input impedance of the DC/DC converter. The stability of the system must be
insured by proper matching of the converters impedances.
The impedance of the long cable used to link the converters is the dominant factor
in the transfer function. The resulting impedance seen by the DC/DC converters
towards the back end power supply is considerable, and the gain and phase margins
become critical for the overall stability.
III. NOISE PROPAGATION ALONG THE LINK
The long DC power link is modelled as a multiconductor transmission line. As the
impedances at both ends of the cable are low, the CM current gets amplified at some
resonance frequencies as it propagates along the cable. The resonance frequencies
must be known to make sure they do not match an eventual resonance of the front end
DC/DC converter input filter.
IV. EMI EMISSIONS OF THE DC POWER LINK.
The power converters are a source of CM and DM noise. The noise source is
identified as inductive in the near field region.
The DM noise is caused by the back end converter filtering, and by the front end
switching device. It is a source of electromagnetic interferences at low
frequencies.
The common mode noise is contributed by the switching devices of both converters.
As it returns through the ground, it is a potential source of strong EMI emission.
The CM noise is the dominant source of EMI by several orders of magnitude when
compared to the DM source.
In sake of a healthy electromagnetic environment of the experiment, the EMI
emissions caused by CM and DM noise must be minimised. This is achieved by
shielding the link; the shield is preferably grounded on both ends to provide and
adequate return path for the CM currents in a minimised loop.
V. CONCLUSION
The DC power link used to feed the front end electronics of the Liquid Argon
Calorimeter brings specific electromagnetic compatibility issues. The critical
stability requirements are analysed and measured. The noise propagation along the
line is measured to identify the resonance frequencies of the power link. The
converters input and output filter must be such that they do not resonate at the
link critical frequencies. The optimal shielding method that minimises the EMI
emissions of this setup is last achieved by connecting the shield on both ends.
