MTA Instrumentaion
University of Illinois (UIUC) Debbie Errede Mike Haney Zack Conway Fermilab Robert Goodwin Mike Kucera Mike Shea
Feb 22, 2003
What’s Going On?
• Instrumentation
– Hardware – Data Collection – Data Processing
• Measurements
4/28/2009
2
Hardware
• In Urbana we have:
– FISO bus system.
• Fiber optic light conditioner.
– LakeShore 218 temperature monitor.
• Diode/resistive temperature sensor monitor.
– IRM(Internet Rack Monitor)
• Fancy ADC connected to the Fermilab Ethernet.
– “Fast” ADC with 8 differential channels
• PC card.
4/28/2009 3
FISO BUS SYSTEM
Part 1
• Currently we have 4 signal conditioners.
– Can take data at 1kHz. – Room for 8 total(4 more) signal conditioners.
• We have temperature, pressure, and strain gauges.
– Each signal conditioner can illuminate a separate sensor.
4/28/2009 4
FISO BUS SYSTEM
Part 2
• Gauges are not certified below 250K.
– The gauges have been used to 35K but never below(H. Kirk @ Brookhaven). – Working with FISO technologies Inc. to modify the internal calibration curves.
• By mid-March we will be able to simulate many aspects of the LH2 absorber in Urbana for sensor testing.
4/28/2009 5
LakeShore 218 Temperature Monitor
• 8 channel temperature monitor.
– Supports two sets of 4 channels.
• Each set of channels can be calibrated for either diode or resistive temperature sensors. • 1.4K-325K temperature range for each channel. • Temperature resolution of ~10mK at 20K in a rad free environment.
• Maximum data collection rate on all channels in 16 Hz.
4/28/2009 6
IRM(Internet Rack Monitor)
• IRM comes with:
– – – – 64 differential A-D input channels. 8 bytes of digital I/O channels. 8 bytes of D-A output channels. Ethernet LAN interface.
4/28/2009
7
PCI-MIO-16E-1 ADC
• 16 channel ADC(8 differential channels)
– 12 bits per channel
• 1.25 Msamples/s
– Each ADC channel built into an IRM has a maximum operating rate of 100 ksamples/s.
• Internal PC device.
4/28/2009
8
Data Collection
• All hardware can communicate with external devices. • The process repeats itself at regular intervals(15Hz???). • Using Labview as a user interface between the PC and the hardware.
– Labview comes with built in functions supporting hardware communication:
• GPIB, Serial, UDP, TCP/IP, ect.
4/28/2009 9
Labview
• Very modular
– Each instrument has its own independent code. – Labview can run each instrument’s code concurrently.
• Labview code is very easy to understand.
4/28/2009
10
Collection Control
• At this time:
– One master configuration file.
• All of the data will be saved to the PC’s hard drive. • There are size issues when writing to the IRM.
– ~1kB every 66 milliseconds is the current goal. – What should be posted on the IRM?
4/28/2009 11
IRM Data Collection
• Operates repetitively at a constant rate.
– In Urbana we are running at 12.5Hz. – Can definitely handle 15Hz operation.
• Faster???
• Each cycle includes:
– The revision of the analog and binary data tables. – Answering/generating messages for other IRMs, ACNET, our PC, ect….
4/28/2009 12
IRM
Part 2
• University of Illinois IRM:
– Has room for 1000 virtual channels.
• Can be upgraded to 2000 virtual channels. • All channels are ACNET compatible.
– 2MB of free memory.
• Not as structured as the channel data.
4/28/2009
13
Data Processing
• The PC can handle almost any data type sent to it at high rates. • The IRM
– Refresh rate of data is limited. – Amount of data refreshed per cycle is limited. – Data must be formatted.
• Each reading occupies 2 bytes of memory. • This leaves 65535 steps to record each data point.
4/28/2009 14
IRM Data Processing
• Take an ammeter.
– Let’s say it needs to measure a current from 0 to 200mA. – So the maximum resolution the IRM data can have is:
• 200mA/65535=0.0031mA
– In other words
• IRM Reading = (1/0.0031mA)*Current Hardware value.
4/28/2009 15
Hardware Data
• All sensor readings use different units. • Each sensor requires its own calibration curve. • The current plan:
– Save the non-calibrated data to disk. – Post the adjusted data for ACNET.
4/28/2009
16
Wrap up
• U of I @ Urbana-Champaign is focusing on the data collection and hardware for the LH2 absorber. • The hardware works but sensor calibration remains a high priority.
4/28/2009
17