| Table of Contents |
|---|
General Instrument
...
Information and Commissioning Instructions
Mercury iPS
The power supply for e1039 is an Oxford Instruments Mercury iPS (intelligent power supply), the manual for which is available here:
...
At this point, a standard instrument power cable may be connected to the back of both power supplies, and the switches on the back of both units set to on. The Master may then be switched on by pressing the button on the front. It takes several minutes for the power supply to initialize at this stage, so please wait until the front panel displays readings for current and voltage and the like before attempting to do anything else.
To make sure the power supply is in remote mode, be sure the button in the lower left corner of the screen displaying a single letter is lit up Orange. If it is not, it must be pressed to change the power supply into remote operation mode.
USB Relay
The other piece of equipment necessary to commission is the USB Relay board which allows the computer to control the switch heaters for the shim coils, which is a small metal box with alligator clips on the top labeled "USB Relay." This box may be connected to the computer via a provided USB-B to USB-A cable. The shim coils should then be connected to the alligator clips at the top of the USB Relay, with both ends of the circuit for the shim 1 switch heater connected to the clips next to the "1" label, and both ends of the circuit for the shim 2 switch heater connected to the clips next to the "2" label.
...
https://www.sainsmart.com/products/4-channel-5v-usb-relay-module
...
Labview Control
The computer control for the power supply and shims is a labview VI that uses the library of functional subVI's provided by Oxford Instruments for use with the Mercury iPS. Once properly commissioned, the control works to prevent dangerous situations from occurring with respect to the power supply. It also contains several controls for the shim coils, and possible use as a level meter, which will be described below.
...
The state of the shim coils is also changed automatically by using the other features of the power supply control. When the Switch Heater is turned on, both Shim 1 and Shim 2 are also turned on, and when the main Switch Heater is turned off, both Shim 1 and Shim 2 are both turned off. Further, to prevent moving the field without both shim coils being on, if the user switches the power supply to "To Set" or "To Zero" while the switch heater is on, Shim 1 and Shim 2 will both be turned on, and locked from being turned off until the power supply is back in Hold mode. This means the shim coils may only be off at any point while the power supply is in Hold mode, and otherwise will be automatically turned on.
Safety Features
The control contains several safety features, in addition to those already described in the Switch Heater and Shim Control sections.
[?]: The ? button in the bottom left will display a small box of instructions and information about the VI for quick reference, if the user clicks and holds on the button.
Control Lock: The grey switch under the title will lock or unlock all controls from being used. If it is toggled on, all controls will be greyed out and unusable until the control lock is turned off. This is to prevent accidentally sending commands to the power supply, and should be used when the control is not actively in use.
Emergency Stop: In an emergency situation where the integrity of the VI is not trusted, the EMERGENCY STOP button in the top of the screen can be clicked to abort any current processes. On clicking this button the power supply will be switched to Hold mode, and then the Labview control will be instantaneously aborted from running, preventing further commands. This should primarily be used if the user believes that the VI itself is doing something unsafe. If the power supply itself is doing something unsafe and sending the hold command regularly is not doing anything, pressing the Emergency Stop button should have no effect either.
Password: If the Admin password is entered into this field correctly, several safety features of the labview control will become disabled or editable. Primarily, while the password in the box is correct, the user can type any ramp rate and current into the target current and ramp rate boxes, even if these values exceed the guidelines given in that section. Additionally, the timer built in for "Safety Mode" will become editable, appearing on the left of the Safety Mode button (see below.) The password box will be cleared automatically ten minutes after the user enters it, and must be entered again if it is necessary to exceed the safety values of the control for longer than ten minutes.
Special Modes
The labview module contains support for several special modes which may be toggled on or off as the user wishes. The box containing these special modes may be closed by pressing the Special Modes button itself, to make accessing them require more clicks and prevent accidentally activating them.
Baseline Mode: This button exists for the purposes of quickly taking a baseline once the magnet is at field. On baseline mode being turned on, the current will ramp to 1 A below the set point at the fastest safe rate possible. If the switch heater is off, it will be turned on and then ramp 1 A down, if it is on, the power supply will simply ramp 1 A down. On baseline mode being toggled off, the current will ramp back to the set point-- again, if the switch heater is off, it will first be turned on. While baseline mode is on, the "Output Current" box flashes, and a large warning reading "Baseline Mode" appears over the color dial on the left, so the user cannot mistake the 1 A lowered current for the actual current at the set point.
Safety Mode: This mode is on by default, but can be toggled off by pressing the Safety Mode button. While Safety Mode is on, an internal timer is started, which resets any time an action is taken within the labview control. After 20 minutes of inactivity, if Safety Mode is on, the Switch Heater will be automatically turned off to switch the magnet into persistent mode, providing conditions are safe for it to be disabled. This 'timeout' duration may be changed to any number of minutes if the Admin password is entered.
As when toggling the switch heater directly, a 30 second timer will activate if either of these modes has to change the state of the switch heater.
Level Monitor
Lastly, the power supply and the control support additional use as a level meter. A nitrogen and helium level monitor card is installed inside the master unit of the power supply, so if connected to a level probe, the iPS will send the resulting measured level to the tanks on the main screen of the control. This feature has been mostly untested but may provide an additional level monitor if needed.
Transient Timer Issue
Over its lifetime being tested by e1039 collaborators, the power supply has mostly behaved well, but had two isolated incidents of odd behavior, one in January 2018 during a UVA cooldown, and on in August 2018 during a UNH cooldown.
The January 2018 issue is documented as:
"The power supply suddenly and quickly ramped down from field to a value close to zero, and then shortly after, began to ramp up very quickly until it passed the set point, causing a quench."
While the August 2018 issue is described as:
"During a ramp down, the power supply suddenly jumped from an output current of 17 A to an output current of 2 A when the switch heater cable was momentarily disconnected. Due to a field probe in the magnet itself, we could tell the current was actually still 17 A. Both the power supply reading and field probe reading were slowly approaching a current of zero, as directed by the control. After 10 minutes ramping down, when there was approximately 1 A in the coils, the magnet suddenly ramped down to a value of -16 A with no user input, at which point the power supply and field probe agreed, and the switch heater was turned off. At this point we were able to ramp to zero without a quench."
We currently believe that both of these issues are as a result of a strange value set for the transient timer. Upon testing the power supply after the August 2018 issue, it was determined that if the switch heater cable becomes momentarily disconnected during a ramp, the power supply becomes confused about the actual value of the current and displays an erroneous value. This was the cause of the August 2018 issue, and may be related to the January 2018 issue, but the common factor between both appears to be the power supply's behavior after detecting a problem. The switch heater cable being disconnected would normally cause the power supply to detect a quench and begin internal safety procedures, as would any unsafe situation which occurred in January. However, the power supply's behavior is such that on measuring dangerous conditions, it logs the current, and then waits the duration of an internally configured "Transient Timer" before acting to try and fix the situation, to prevent false positive quenches. Further, it acts based on the initially measured current, which may lead it to ramp to values below zero or above the set point, if the output current has been changed at all by the user since the transient timer started. We believe this is consistent with the issue observed in August, and may also be consistent with the issue observed in January.
For both incidents, the value of the transient timer was set to 10 minutes, even though the manual advises a maximum value of 3 seconds. This allowed for a very long time for the current to change drastically from what it was when quench conditions started. We have tested and been able to replicate the issue from the August cooldown with a dummy setup, and showed that it cannot occur once the transient timer is set to a safe value of 3 seconds.
The transient timer has been currently set to the safer value of 3 seconds. This has hopefully resolved the root cause of both incidents.