Transcript Safety Board Review Rev D

Safety Board PCB Review
(Rev D)
Eric Warmbier
Safety Board Review (Rev D)
Jan. 15, 2009
1
Overview
• Board is in use and well established, so the
functional details will be skipped.
• Safety Board needs a new layout
– Current board has errors in design
• Modified via wire and additional components to this point
– Grounding / noise issues cause inadvertent (and
annoying) watchdog errors
• Goal is to mainly look for errors in circuits on
schematics and that all new features were added
– Darryl and Eric will be responsible for looking over
signals and off board connections.
Safety Board Review (Rev D)
Jan. 15, 2009
2
Documentation: Files & Format
• There are 3 documents required for this review
– “TCS3-2070_Safety_Board_Schematics_Rev_C”
• Original Rev B with hand wired modifications = Rev C
– “TCS3-2070_Safety_Board_Schematics_Rev_D”
• New schematics for this review
– “Safety_Board_Rev_D_mathcad.pdf”
• Equations explaining/documenting why value
Safety Board Review (Rev D)
Jan. 15, 2009
3
Issues & Resolution
Issue
Resolution
+/- 15V Comparator has -15V logic low output going
directly in CPLD (it clamps).
Optocouplers perform logic level conversion.
Loading on comparator circuit circuits. Uneven +/trip points.
Buffer amp and selecting correct resistor.
Loading on references.
Changed design to use undivided reference.
Not enough test points, especially GND.
Added many test points.
Mechanical support for board when inserting 50 pin
connectors is too low.
Machine a plate.
(see slide on this issue)
Current board routing crosses digital / analog split.
Design uses optocouplers. Not an issue now.
Names on schematic don’t indicate active low or
high logic (George’s comment).
Good point. Signals should now indicate active
logic level.
No watchdog timer on CPLD clock.
(clock gone = most error detection gone)
Add a watchdog timer to this clock.
Watchdog timers trip inadvertently.
Optocouplers used to separate them. Currently
in use at summit via small PCB & working.
No redundancy on brake relay control circuitry.
Redundancy added.
Safety Board Review (Rev D)
Jan. 15, 2009
4
Added Features
Feature
Comment
On board push button for testing watchdog timers.
Makes it easy to test.
Lab/summit selection jumpers are now switches,
with a status to TCS3 software.
Won’t allow anybody to inadvertently install
board up at summit in wrong configuration.
Added damped LC filters to board for power
filtering.
Probably not necessary, but they’re there if
needed.
Board will be laid out to allow for a cover that can
be mounted to it with screws to the bottom plate.
Won’t initially make cover. It’s probably not
required, but it’s available.
Trip points for comparators will have resistor paths
or can use pots.
Resistors are preferred for signals that are
almost never adjusted. Pots good for lab.
Additional power connector has more pins. IRTF
PCB connector still installed in parallel.
Allows for 2 pins for each supply voltage.
Reduces connection issues.
Mostly surface mount components used.
Smaller, more available, etc. We also have
complete resistor and cap kits for 1206 size.
Safety Board Review (Rev D)
Jan. 15, 2009
5
Mechanical Mounting
• General concept – not to scale or precisely drawn
power
50 pin
50 pin
Optional Cover
Bottom Mounting Plate
(top view)
(top view)
16 pin
power
50 pin
50 pin
(bare copper)
PCB
Stack up
(side view)
(top view)
(bare copper)
16 pin
Safety Board Review (Rev D)
Jan. 15, 2009
6