1. This forum section is a read-only archive which contains old newsgroup posts. If you wish to post a query, please do so in one of our main forum sections (here). This way you will get a faster, better response from the members on Motherboard Point.

Where would you protect this control system?

Discussion in 'Embedded' started by Chris Carlen, May 23, 2006.

  1. Chris Carlen

    Chris Carlen Guest

    Hi:

    We have a 4 channel motion control servo system for electrohydraulic
    actuation of four engine valves. The basic components in reverse from
    the business end are as follows:

    1. voice coil motor (to actuate hydraulic valves)
    2. PWM H-bridge power amplifier (to drive voice coils, with fuses to
    protect coils from pegged amplifiers)
    3. TI 6711 DSP with 16 channels each of 100kHz 16-bit A/D and D/A to
    digitally implement closed-loop control algorithms
    4. LVRT position sensors (for precise control) and proximity (for
    monitoring and for a backup check on control)
    5. absolute encoder on engine camshaft to inform DSP where valves
    should be (position reference data)
    6. extra incremental encoder on crankshaft (with one index mark blanked
    by a hall sensor on the cam) feeds to DSP to give a backup check on cam
    encoder alignment.

    Additionally:

    1. low voltage power +5V and +/-15V for DSP board and it's analog IO
    system.
    2. two 48V 600W switchers to power PWM amps.
    3. PC connected via USB to DSP to implement GUI. This link need not be
    active for the DSP to do its job.

    The main safety issue is that we must prevent the valves from
    accidentally crashing into the piston. Due to the complexity of the
    research engines (with optical access) involved, the cost in repair time
    which would result from a failure of the valve control would be huge.
    This must be avoided in any fault modes where the mitigation cost/repair
    cost ratio is <= 1.

    In analyzing possible failure modes of this system which could lead to
    this undesirable occurance, I have enumerated the following
    possibilities in decreasing order of likelyhood and indicate whether the
    problem has been mitigated. If not, in the next paragraph I describe
    what is being planned to mitigate an unsolved problem.

    POSSIBLE PROBLEMS (feel free to comment on the prioritization order):

    1. position sensor failure (mitigated by use of redundant sensors)
    2. DSP software or hardware failure (see below)
    3. encoder alignment slippage (mitigated by use of redendant encoders)
    4. line power failure (mitigated by putting system on a UPS)
    5. internal DC power supply failure (see below)
    6. power amplifier failure (see below)

    FIXES:

    re: #2. To deal with the DSP hardware or software going bonkers, I plan
    to utilize a watchdog timer (WDT). The DSP board has one built into its
    FPGA, but it is of little use because it doesn't have an adjustable
    timeout. Rebooting the DSP takes a long time, much too long to wait to
    regain control. Thus, an external custom watchdog is being designed.

    I plan to use the simplest possible discrete logic to build a WDT using
    some one-shots and flip-flops plus a few glue gates. It will detect the
    absence of a 20kHz pulse train from the DSP (from the control loop
    iteration) and if the pulses stop, will first apply a withdraw pulse to
    the valves, then disable the PWM amplifiers. Thus, the WDT must include
    4 pairs of analog multiplexers to switch the inputs of the PWM amps
    from the DSP to a constant voltage source of appropriate polarity and
    magnitude to force the valves to close quickly.

    re: #5. To deal with a power failure of one of the DC power supplies
    (either high or low power supplies), I am considering using redundant
    supplies with a diode network to allow them to be paralleled. This
    should mitigate the problem to an extremely low likelyhood of causing a
    disaster. Furthermore, the WDT circuit could include power monitoring
    comparators to signal the DSP if a supply was malfunctioning.

    re: #6. I had considered a means of dealing with power amplifier
    failure. Solid state relays could be employed to build a multiplexer
    for the voice coils that could switch the coils to a constant withdraw
    voltage in the event of the detection of a serious control discrepancy
    by the DSP. However, the differential output of the H-bridges combined
    with the fact that the motor currents are bidirectional makes this
    tricky. I have concluded that the cost of fixing vs. the probablility
    of occurence makes this fault mode something we will just have to hope
    doesn't happen.

    The cost of fixing this might be less than the cost of a failure, but at
    present we require that the system be put into service shortly in
    prototype configuration, and there isn't enough time to deal with this.
    It is still possible to include this capability later when the system
    is being finalized.

    Would you consider solving the faults in the methods described?

    Would you agree with the decision to give up attempting to mitigate the
    possibility of amplifier failure while the system is being used for a
    few months before packaging and finalizing?

    Thanks for input.

    --
    Good day!

    ________________________________________
    Christopher R. Carlen
    Principal Laser&Electronics Technologist
    Sandia National Laboratories CA USA

    NOTE, delete texts: "RemoveThis" and
    "BOGUS" from email address to reply.
     
    Chris Carlen, May 23, 2006
    #1
    1. Advertisements

  2. I'd leave the controller h/w as simple as possible and build a redundant
    hydraulic actuator that is as separate from the main system as possible.
    Have a secondary hydraulic valve system that, when energized, will drive
    the valves to a safe position (closed) regardless of the primary system
    commands. Using a separate crank angle sensing system that provides a
    signal when the piston approaches a 'minimum safe clearance' position
    around TDC and separate valve position sensors that provide a signal
    when the valve is open. ANDing each 'valve open' signal with the 'crank
    near TDC' signal drives the backup hydraulic controls to shut the
    valves.

    If the valves close under spring force, even a complete power failure
    will save the engine as long as the default (de-energized) backup
    hydraulic valve position is to allow the valves to close.
     
    Paul Hovnanian P.E., May 24, 2006
    #2
    1. Advertisements

  3. Chris Carlen

    Ken Smith Guest

    I have used just about exactly this sort of watchdog. I suggest you make
    it so that the watchdog timer defaults to the disabling condition at power
    on and require that the DSP write so some port or something before it gets
    control.

    Doing this somewhat protects you against the DSP waking up in the middle
    of the code because the reset cirduit didn't work etc.
    If you have the room, you can put great huge capacitors on the supply and
    fire the watchdog if the input drops out. If the capacitors are big
    enough, you can get things safe just using the energy in them.
     
    Ken Smith, May 24, 2006
    #3
  4. What about mechanical stops, and a mechanical design such that the valve
    _cannot_ contact the piston. Even if that means an additional cam...

    This is a research engine, so that sounds like very cheap insurance...

    Anything else is going to struggle with arbitary loss of power while
    the engine is spining, or an arbirary loss of phase information due
    to Software crashes.... ?

    -jg
     
    Jim Granville, May 24, 2006
    #4
  5. 7. Loss of system hydraulic pressure (use pressure sensors)

    Paul
     
    Paul Keinanen, May 24, 2006
    #5
  6. When using a redundant system, put the hydraulic valves in _series_
    when driving towards the 'active' position, thus both systems must
    agree on driving to the 'active' state.

    Put the hydraulic valves in _parallel_ when driving towards the 'safe'
    state, thus either system alone would drive to the 'safe' state. This
    would eliminate the tug of war situation if the failed systems drives
    towards opposite direction.

    Alteratively a triple redundant system could mechanically vote in a
    tug of war situation.

    If instead of a fully redundant system an accurate main system and a
    crude security system is used, put the security system 'active' state
    on/off valve in series with the main system 'active' control valve and
    open the security system valve as soon as it is safe and close as late
    as possible. Let the main system control valve do the actual timing
    within the window.

    On the 'safe' side, put the safety system on/off valve in parallel
    with the accurate main system control valve and activate the safety
    system 'safe' state valve only during times, when there could be a
    catastrophic failure during each cycle. Thus, the complex fine control
    system would only be allowed to actually control within some safe
    timing windows.

    Of course, this assumes that the hydraulic pressure is present and
    there are no leaks.

    Paul
     
    Paul Keinanen, May 24, 2006
    #6
  7. Chris Carlen

    Chris Carlen Guest

    Not possible to alter the hydraulics.
    It's not that simple. The voice coil position determines whether or not
    the hydraulics can vent or not, so unless the voice coil is retracted,
    the valve spring (which is present) can't help. There is a spring which
    forces the voice coils into the "retract valve" position when unpowered,
    but this won't close the valves fast enough when the engine is running.

    The only other way for the valves to retract is if they are pushed with
    greater force than the hydraulic force, which would occur if hit by the
    piston. So at least they won't catastrophically burst something, but we
    don't know if the forces involved which would also involve some lateral
    force on the valve guides would damage things or not. We assume damage
    would occur and so must be avoided.

    Thanks for the input.

    --
    Good day!

    ________________________________________
    Christopher R. Carlen
    Principal Laser&Electronics Technologist
    Sandia National Laboratories CA USA

    NOTE, delete texts: "RemoveThis" and
    "BOGUS" from email address to reply.
     
    Chris Carlen, May 24, 2006
    #7
  8. Chris Carlen

    Chris Carlen Guest

    Interesting addition. I'll consider that.
    Yes, I haven't quite decided whether to rely on caps or redundant
    supplies. The caps might be easier and tolerable space-wise. Some
    testing will need to be done...

    Thanks for the input.


    --
    Good day!

    ________________________________________
    Christopher R. Carlen
    Principal Laser&Electronics Technologist
    Sandia National Laboratories CA USA

    NOTE, delete texts: "RemoveThis" and
    "BOGUS" from email address to reply.
     
    Chris Carlen, May 24, 2006
    #8
  9. Chris Carlen

    Chris Carlen Guest

    Unfortunately, such mechanical solutions are impossible. The engine is
    a mid-size diesel, which must retain realistic compression ratios and
    combustion chamber structure. The point is to research real engine
    conditions with real engines. But the engines are modified with
    extensive instrumentation and optical access.
    But it does seem like a software crash can be detected quite readily and
    dealt with quickly enough. Also, power loss can be dealt with. Control
    needn't be retained in these circumstances, only rapid valve retraction
    is essential. Going open-loop to accomplish that is acceptable.

    Thanks for the input.


    --
    Good day!

    ________________________________________
    Christopher R. Carlen
    Principal Laser&Electronics Technologist
    Sandia National Laboratories CA USA

    NOTE, delete texts: "RemoveThis" and
    "BOGUS" from email address to reply.
     
    Chris Carlen, May 24, 2006
    #9
  10. Chris Carlen

    Chris Carlen Guest

    Oh yeah, that's a big one! Thanks, I don't know if this has been
    addressed by the folks on the mechanical side of things.




    --
    Good day!

    ________________________________________
    Christopher R. Carlen
    Principal Laser&Electronics Technologist
    Sandia National Laboratories CA USA

    NOTE, delete texts: "RemoveThis" and
    "BOGUS" from email address to reply.
     
    Chris Carlen, May 24, 2006
    #10
  11. Chris Carlen

    Nico Coesel Guest

    A bit far fetched: What if you could have an electrical simulation of
    the fly-wheel (using a servo motor system). Doing this makes it
    possible to stop the engine almost instantly. A system which keeps
    track of valve and piston position can choose to stop the engine when
    it gets critical. Besides, an electronic fly-wheel can be told to stop
    pushing beyond a certain force preventing damage when things do go
    wrong.

    By the way: only 4 valves? I assume a 2-stroke turbo charged diesel
    engine?
     
    Nico Coesel, May 24, 2006
    #11
  12. Chris Carlen

    Rich Grise Guest

    FWIW, on two separate occasions, with two different cars, I've lost the
    timing chain/belt. In both cases, there was no valve damage because they
    were "clearance engines", i.e., even open, the piston wouldn't hit the
    valves.

    That would probably lower the effeciency, because you couldn't use as
    high a compression ratio, but you're virtually guaranteed not to
    break any valves or pistons. :)

    As an aside, it somehow pleases me to see someone actually working
    on electric valve actuators - saves HP for the camshaft and all that
    schtuff. :)

    Good Luck!
    Rich
     
    Rich Grise, May 24, 2006
    #12
  13. I was thinking more along the lines of a hydraulic safety system that
    would be inactive during normal operations, but override the operating
    controls in the event of a fault. The philosophy is to keep as much of
    the backup system separate from the control loop as possible.

    But, it appears as though mods to the hydraulics cannot be made.
     
    Paul Hovnanian P.E., May 24, 2006
    #13
  14. I'd still pursue the goal of keeping as much of the safety system out of
    the main control loop as possible. In this case, you're stuck with a
    shared path (the voice coils). So I'd design the voice coil drives with
    the capability of 'crowbaring' the voice coils into the 'close valve'
    position when commanded by the safety. Then, I'd keep as much of the
    crowbar circuitry separate from the control drive.
     
    Paul Hovnanian P.E., May 24, 2006
    #14
  15. This begins to sound like your organisation doesn't have anyone used to
    doing full risk assessments of the equipment to be controlled. We have a
    list of items that you intend to use in the control system at a stage much
    earlier than I would consider appropriate. Has anyone in the team
    understood the dynamics of the whole system yet?
    So, these are not the electrically driven engine valves like Aura do. They
    are hydraulically operated valves with pilot spool valves to control the
    hydraulic side of things.
    Depending on the powers involved you may find other means of power removal
    may work more dependably.
    Is a centralised controller the most appropriate solution for this system?
    You have to ask yourself such questions as by working through the
    consideration of such questions you may see the best solution structure
    emerge.
    Needs more explanation of where you are using these. Are they attached to
    the voice coils.
    Do you have enough information to describe the safe states of these valves
    and how the valves achieve the safe state. Do they need to be driven to the
    safe state or can they return to the safe state under their own power.
    I know that someone else pointed out the hydraulic system failure that you
    missed from the list. Some others that may impinge on the overall system,
    brought on by increasing complexity of the whole system, would be potential
    failure of the UPS. I know you added this to mitigate one of the failure
    potentials but the mere fact of adding it is a complication of the system
    and has to also be considered. It can have a bearing dependent on the way
    the UPS is implemented.
    Wise move.
    Be careful with the complexity of the design of this WDT. You want something
    ultra simple and dependable. You will also want to ensure that component
    failures in the WDT circuitry lead to achieving safe states.
    So, the valves are powered to the closed position. By switching the input of
    the PWM amplifiers you will not guarantee the valve closure if the PWM
    amplifiers themselves have a fault.
    There have been some incidence of such parallel power arrangements suffering
    if the protective diode network goes short circuit in one of the supplies
    which later fails, dragging the power rails down in the process. Attention
    to detail here will be required, particularly in how you guard the circuits
    that are meant to prevent such problems occurring.
    What have you against real relays.There are techniques which will help to
    speed up their operation so that they could quickly achieve a suitable
    state to divert the valve supply voltage to a withdraw power rail. Real
    relays also have the option of multiple contacts so one relay could divert
    all the valves power simultaneously. Then, of course, you need to ensure
    such a power rail was still active when needed.
    Pressures on time often lead to poor analysis and decision making. The
    decision on one real relay shouldn't hold you up that long anyway.
    What is the time cost of wrecking an engine with the prototype control
    system. How much is simulatable. Would the client prefer a simulated run of
    the prototype controller before they let you loose on an engine.
    Whether or not your chosen methods (or those implied in my response) are
    appropriate would require some more information about your system. Yes, you
    are getting into the realms of spending money on consultancy.
    No. I would spend more time understanding the failures and consequences of
    failures, doing a proper job of designing the system to ensure that you
    achieve the simplest approach commensurate with the risks posed. I would
    prototype areas of the system that are ill-defined or not well enough
    understood to gain the required knowledge to assist a proper design (not
    the whole thing).

    --
    ********************************************************************
    Paul E. Bennett ....................<email://>
    Forth based HIDECS Consultancy .....<http://www.amleth.demon.co.uk/>
    Mob: +44 (0)7811-639972
    Tel: +44 (0)1235-811095
    Going Forth Safely ..... EBA. www.electric-boat-association.org.uk..
    ********************************************************************
     
    Paul E. Bennett, May 24, 2006
    #15
  16. Even if you do not expect a loss of hydraulic pressure, you would
    still need a pressure sensor for the startup phase, since it does not
    make sense to try to control the system, until the hydraulic pressure
    is within specified limits.

    One practical note about any system containing multiple safeguard
    mechanisms. At the user interface, create one display screen that
    displays _all_ the signals at a single glance that may prevent the
    system from operating normally.

    While the state of all the disabling signals are available in separate
    displays or commands, finding the last signal that prevents operation
    may be quite time consuming, unless all disabling signals are readily
    visible at once.

    Paul
     
    Paul Keinanen, May 24, 2006
    #16
  17. Chris Carlen

    Chris Carlen Guest

    Wow, that's a different approach. But it really is much easier to
    withdraw the valves than do this. The engine actually is turned by a
    dynamometer, but from experiences with other labs it takes quite a few
    revs to stop the engine even with the dyno at 150% current.
    No, a single 4-valve cylinder of a 4-stroke diesel, running in HCCI
    combustion mode.

    Thanks for the input.


    --
    Good day!

    ________________________________________
    Christopher R. Carlen
    Principal Laser&Electronics Technologist
    Sandia National Laboratories CA USA

    NOTE, delete texts: "RemoveThis" and
    "BOGUS" from email address to reply.
     
    Chris Carlen, May 24, 2006
    #17
  18. Paul Hovnanian P.E. wrote:
    Good idea, maybe even to the point of small batteries at the voice
    coils ? then the system has to frequently ask for position,
    and no-ask => rapid retraction ?

    Still leave the issue of loosing phase sync (engine angle), and not
    knowing it...

    -jg
     
    Jim Granville, May 25, 2006
    #18
  19. Chris Carlen

    Chris Carlen Guest


    There are redundant encoders, so I think this has been addressed, no?



    --
    Good day!

    ________________________________________
    Christopher R. Carlen
    Principal Laser&Electronics Technologist
    Sandia National Laboratories CA USA

    NOTE, delete texts: "RemoveThis" and
    "BOGUS" from email address to reply.
     
    Chris Carlen, May 25, 2006
    #19
  20. I was thinking not so much of a hardware failure, but something along
    the lines of the Ariane Rocket.... - the SW _thinks_ it knows where the
    engine is - is this software part of the research ?

    if you have an absolute encoder, I'd be tempted to put a small CPLD in
    the VoiceCoil Driver(s), which can contain a simple drive over-ride and
    ROMs for 'not legal' areas. A digital index signal on the valve would
    complete this very simple 'sanity checker'.

    -jg
     
    Jim Granville, May 25, 2006
    #20
    1. Advertisements

Ask a Question

Want to reply to this thread or ask your own question?

You'll need to choose a username for the site, which only take a couple of moments (here). After that, you can post your question and our members will help you out.