MantisBT

View Issue Details Jump to Notes ] Issue History ] Print ]
IDProjectCategoryView StatusDate SubmittedLast Update
0000138Hall D OnlineMagnetspublic2011-09-21 15:582012-04-04 15:06
Reporterwolin 
Assigned Towolin 
PrioritynormalSeveritymajorReproducibilityN/A
StatusresolvedResolutionfixed 
PlatformOSOS Version
Summary0000138: Determine whether Danfysik quench detector is adequate for the full solenoid
DescriptionThe Danfysik QD can only work if there are good inductance balances between coil pairs. If not we'll need to design and build our own.
Additional InformationFloyd will get the inductances given the new coil tap locations. With that I can determine whether the Danfysik QD will work.
TagsNo tags attached.
Work by outside group for Hall DNo
Estimated man-weeks1
Percentage complete0
Actual man-weeks0
Attached Files

- Relationships

-  Notes
(0000162)
wolin (administrator)
2011-09-21 15:59

Floyd will not get a chance to work on this until late Fall.
(0000168)
wolin (administrator)
2011-10-12 14:52

Hai is looking into off-the-shelf FPGA boards with A/D's on the front end that might work if the Danfysik QD is not suitable.
(0000206)
wolin (administrator)
2011-11-15 11:17

Floyd might not get to this in time for the test Fall 2012 (4 months lead from start to delivery plus 1 man-month EE time needed), so we are considering building a new one anyway, and soon. Further, even if the simulation says the Danfysik is ok, the typical 10% error on the simulation may mean it won't work in practice.

An important consideration is whether a home-built version will pass a safety review and whether it can be a digital device (micro-controller- or FPGA-based). Some of these questions may be answered by the FMEA exercise. If we build one we'll need EE help from outside.
(0000216)
wolin (administrator)
2011-11-22 17:24

George will look into getting something going on the coil simulation. I think I may have misunderstood our QD problems...it appears that the QD can handle a fairly large inductance imbalance if adjusted properly. Paul Brindza got me in contact with Alexander Elkiaer, designer of the QD. They have a new version that makes adjustment easier.
(0000217)
wolin (administrator)
2011-11-29 11:45

Becomming more convinced Danfysik QD can handle large inductance imbalances. Will test in the lab.
(0000221)
wolin (administrator)
2011-12-02 17:11

Lab tests show Danfysik QD works exactly as advertised. Simply changing the input resistors will allow us to handle any reasonable inductance imbalance. Didn't see any asymmetry when bias set to around 80% of full range, but a few more checks are needed.
(0000223)
wolin (administrator)
2011-12-07 09:28

Almost certain QD will work for full solenoid, just have to adjust input voltage divider to compensate for inductance imbalance. From Floyd's note the max imbalance we should expect is around 30% so modifying the QD for a 50% imbalance should work. Will modify and test.

Also, Fernando's -IN bias circuit is needed to ensure equal sensitivity for the two coils when the QD is biased to compensate for an inductance imbalance. This will be tested as well.

First guess indicates that replacing the 100k resistors with 20k should accommodate inductance ratio of 1.5.
(0000235)
wolin (administrator)
2011-12-20 13:25

Mark reports max 5 mA to keep equipment at Class 1. When George tells us the maximum voltage during a quench we can size the currently limiting resistors to keep things below 5 mA. At that point we can determine the resistors needed in the QD. We may have to change the potentiometer as well, depends on the max voltage and what the QD designer says.
(0000333)
wolin (administrator)
2012-04-04 15:06

We believe the device is adequate as long as the external current limiting resistors are taken into account when the on-board resistors are tuned. George thinks the PLC can act as a redundant QD, so another device is not needed.

- Issue History
Date Modified Username Field Change
2011-09-21 15:58 wolin New Issue
2011-09-21 15:58 wolin Work by outside group for Hall D => No
2011-09-21 15:58 wolin Estimated man-weeks => 1
2011-09-21 15:58 wolin Percentage complete => 0
2011-09-21 15:58 wolin Actual man-weeks => 0
2011-09-21 15:59 wolin Note Added: 0000162
2011-09-21 15:59 wolin Assigned To => wolin
2011-09-21 15:59 wolin Status new => acknowledged
2011-10-12 14:52 wolin Note Added: 0000168
2011-10-19 14:42 wolin Severity block => major
2011-11-15 11:17 wolin Note Added: 0000206
2011-11-22 17:24 wolin Note Added: 0000216
2011-11-29 11:45 wolin Note Added: 0000217
2011-12-02 17:11 wolin Note Added: 0000221
2011-12-07 09:28 wolin Note Added: 0000223
2011-12-20 13:25 wolin Note Added: 0000235
2012-04-04 15:06 wolin Note Added: 0000333
2012-04-04 15:06 wolin Status acknowledged => resolved
2012-04-04 15:06 wolin Resolution open => fixed


Copyright © 2000 - 2020 MantisBT Team
Powered by Mantis Bugtracker