Matronics Email Lists :: View topic - Diodes vs. Bi-Directional Zeners for Contactors

Matronics Email Lists
Web Forum Interface to the Matronics Email Lists

Get Email Distribution Too! FAQ Search Memberlist Usergroups Register

Profile Log in to check your private messages Log in

Diodes vs. Bi-Directional Zeners for Contactors

Matronics Email Lists Forum Index -> AeroElectric-List

View previous topic :: View next topic

Author Message

nuckolls.bob(at)aeroelect
Guest

Posted: Mon Oct 12, 2009 6:32 am Post subject: Diodes vs. Bi-Directional Zeners for Contactors

At 06:20 AM 10/12/2009, you wrote:

  	  Quote: 	
		  I'm building a new power grid for a customer and the research on contactors ended up with some digging on the diodes that we use across the coils of the contactors, and whether they're necessary.  We're using a couple Tyco Kilovac EV200s and since they have pigtail wires instead of terminals, it's tougher to install the diodes.  Tyco says we don't need "back EMF protection" with the EV200s. 	


    EV200's have internal electronics that take care

    of the mag-field collapse spike from the contactor's

    coil.



  	  Quote: 	
		   But we are also using the LEV100 contactors, which do need the back EMF protection.



 Tyco has a couple documents that explain the various methods to handle the back EMF.

  

   http://relays.tycoelectronics.com/appnotes/app_pdfs/13c3311.pdf

  

   http://relays.tycoelectronics.com/appnotes/app_pdfs/13c3264.pdf

  

 Eric at Perihelion Design sells SnapJacks, and claims that they are superior to using diodes.  Eric's claims match up with the information in the Tyco documents.  Both Eric and Tyco claim that bi-directional zeners have all the advantages of protecting the switch without slowing down the contact separation speed, which decreases contactor life. 	


    Which is all demonstrable BS. Even the authors of the

    Tyco documents correctly observe that "tight" coil

    suppression of a plain diode causes contactor OPENING

    DELAY and then incorrectly extend that observation

    into a reduction of CONTACT SPREADING VELOCITY which

    would indeed increase the duration of arcing during

    opening.



    Plain diodes have been used by the hundreds of

    thousands on TC aircraft for over 40 years and

    they are still meeting the design goals of the

    system designers today. Further, papers like

    those on the Tyco site speak to relay life

    on the order of hundreds of thousands of cycles.

    Their research into contact life vs. coil suppression

    techniques will be based on laboratory

    cycle testing (Hey Joe! Lookit here! That evil coil

    suppression technique kicked our service life

    in the head . . . instead of 110,000 cycles we

    only got 95,000 cycles.)



    The contactors in the average light plane

    won't get 5,000 cycles over the LIFETIME

    of the airplane.



    Use what ever coil suppression technique

    gives you warm fuzzies but know too that the least

    expensive and simplest technique is fine.



  	  Quote: 	
		  One concern I have is the failure modes of diodes and bi-directional zeners.  If they fail open, no big deal, but if either has a failure mode that would create a short, that could cause problems. 	


    What problem? Suppose your contactor fails to

    close for any one of many other reasons? Are any of

    these contactors used in a manner where failure

    puts outcome of flight at risk? If so, you need to craft

    a Plan-B to deal comfortably with that failure. May I suggest

    that your worries about forestalling failure by

    judicious selection of coil suppression is a

    distraction. You should ASSUME that the contactor

    will be unavailable to you at some time in the

    future . . . now, what alternatives are in place

    to deal with that failure? See chapter 17 of the

    'Connection.



  	  Quote: 	
		    I read somewhere that high voltage tends to result in a short failure, but high amperage results in an open failure.  There's no circuit protection on these coils (except for the starter). 	


    The ENERGY dissipated in ALL coil suppression

    is very short duration (milliseconds per flight

    cycle) and perhaps 1,000th the rated capability

    of the device. They don't fail except for reasons

    of manufacturing defect or installation error . . .

    and particularly NOT from having selected the wrong

    rating for the device. When a coil suppression device

    is working, the VOLTAGE and CURRENT impressed upon

    it is trivial compared to its ratings. 

  	  Quote: 	
		   

 I'm interested in comments on the relative merits of using diodes vs. bi-directional zeners.  Also, the failure modes of both devices.  And how to size the bi-directional zeners to minimize the chance of a failure, especially a short failure.

 	


    You're worrying waaaayyyy too much about it.

    This topic has been discussed and many platinum

    plated coil suppression techniques have been offered

    up as the path to Nirvana. But when it gets down

    to the physics of contactor and power relay

    control, it's easy to demonstrate.



  http://www.aeroelectric.com/Pictures/Curves/704-1OpeningTimeNoDiode.gif 



  http://www.aeroelectric.com/Pictures/Curves/704-1OpeningTimeWithDiode.gif 



  http://www.aeroelectric.com/Pictures/Curves/704-1DelayWithDiode.gif

 

   http://www.aeroelectric.com/Pictures/Curves/7041DelayWithDiode.gif

 

  http://www.aeroelectric.com/Pictures/Curves/CH_w_1n5400.gif



  http://www.aeroelectric.com/Pictures/Curves/CH_w_2x18v_Transorbs.gif

 

   http://www.aeroelectric.com/Pictures/Curves/CH_w_120_Ohm.gif



  http://www.aeroelectric.com/Pictures/Curves/S704-1_Drop-Out_Delay_with_Diode.jpg

 

   http://www.aeroelectric.com/Pictures/Curves/S704-1_Drop-Out_Delay_without_Diode.jpg 



 

    These are but a few of dozens of traces recorded by

    myself where we looked at drop-out-delay, contact

    transition velocity, and contact bounce vs. various coil

    suppression techniques (including simple placement

    of a 120 ohm resistor across the coil!).



    Bottom line is that anything you want to do will

    be fine . . . make it easy on yourself and be wary

    of pronouncements and dire warnings from folk who

    not been there nor done that. Do a failure mode

    effects analysis and craft architectures that offer

    a Plan-B for ALL uncomfortable failures. The road

    to comfortable termination of flight is not paved with

    prophylactics AGAINST failure . . . rather a PLAN for

    dealing comfortably with any failure.



    Bob . . .    [quote][b]

- The Matronics AeroElectric-List Email Forum -

Use the List Feature Navigator to browse the many List utilities available such as the Email Subscriptions page, Archive Search & Download, 7-Day Browse, Chat, FAQ, Photoshare, and much more:

http://www.matronics.com/Navigator?AeroElectric-List

mikeeasley(at)aol.com
Guest

Posted: Mon Oct 12, 2009 7:16 am Post subject: Diodes vs. Bi-Directional Zeners for Contactors

Bob,

  

 So I believe there's a decent argument to do nothing.  If the spike is small, and the switches are robust enough to handle far more than the spike, and the current through the contactors is far below the contactor rating, and the number of cycles of the switch and the contactor are far below the advertised life...

  

 ...then leaving off the diodes is a viable alternative.

  

 Mike Easley

  

  

  

 In a message dated 10/12/09 08:33:46 Mountain Daylight Time, nuckolls.bob(at)aeroelectric.com writes:

  	  Quote: 	
		     .aolmailheader          {font-size:8pt; color:black; font-family:Arial} a.aolmailheader:link    {color:blue; text-decoration:underline; font-weight:normal} a.aolmailheader:visited {color:magenta; text-decoration:underline; font-weight:normal} a.aolmailheader:active  {color:blue; text-decoration:underline; font-weight:normal} a.aolmailheader:hover   {color:blue; text-decoration:underline; font-weight:normal}  At 06:20 AM 10/12/2009, you wrote:

  	  Quote: 	
		  I'm building a new power grid for a customer and the research on contactors ended up with some digging on the diodes that we use across the coils of the contactors, and whether they're necessary.  We're using a couple Tyco Kilovac EV200s and since they have pigtail wires instead of terminals, it's tougher to install the diodes.  Tyco says we don't need "back EMF protection" with the EV200s. 	


   EV200's have internal electronics that take care

   of the mag-field collapse spike from the contactor's

   coil.



  	  Quote: 	
		   But we are also using the LEV100 contactors, which do need the back EMF protection.



Tyco has a couple documents that explain the various methods to handle the back EMF.

 

http://relays.tycoelectronics.com/appnotes/app_pdfs/13c3311.pdf

 

http://relays.tycoelectronics.com/appnotes/app_pdfs/13c3264.pdf

 

Eric at Perihelion Design sells SnapJacks, and cla  ims that they are superior to using diodes.  Eric's claims match up with the information in the Tyco documents.  Both Eric and Tyco claim that bi-directional zeners have all the advantages of protecting the switch without slowing down the contact separation speed, which decreases contactor life. 	


   Which is all demonstrable BS. Even the authors of the

   Tyco documents correctly observe that "tight" coil

   suppression of a plain diode causes contactor OPENING

   DELAY and then incorrectly extend that observation

   into a reduction of CONTACT SPREADING VELOCITY which

   would indeed increase the duration of arcing during

   opening.



   Plain diodes have been used by the hundreds of

   thousands on TC aircraft for over 40 years and

   they are still meeting the design goals of the

   system designers today. F  urther, papers like

   those on the Tyco site speak to   relay life

   on the order of hundreds of thousands of cycles.

   Their research into contact life vs. coil suppression

   techniques will be based on laboratory

   cycle testing (Hey Joe! Lookit here! That evil coil

   suppression technique kicked our service life

   in the head . . . instead of 110,000 cycles we

   only got 95,000 cycles.)



   The contactors in the average light plane

   won't get 5,000 cycles over the LIFETIME

   of the airplane.



   Use what ever coil suppression technique

   gives you warm fuzzies but know too that the least

   expensive and simplest technique is fine.



  	  Quote: 	
		  One concern I have is the failure modes of diodes and bi-directional zeners.  If they fail open, no big deal, but if either has a failure mode that would create a short, that could cause problems. 	


   What problem? Suppose your contactor fails to

   close for any one of many other reasons? Are any of

   these contactors used in a manner where failure

   puts outcome of flight at risk? If so, you need to craft

   a Plan-B to deal comfortably with that failure. May I suggest

   that your worries about forestalling failure by

   judicious selection of coil suppression is a

   distraction. You should ASSUME that the contactor

   will be unavailable to you at some time in the

   future . . . now, what alternatives are in place

   to deal with that failure? See chapter 17 of the

     'Connection.



  	  Quote: 	
		    I read somewhere that high voltage tends to result in a short failure, but high amperage results in an open failure.  There's no circuit protection on these coils (except for the starter). 	


   The ENERGY dissipated in ALL coil suppression

   is very short duration (milliseconds per flight

   cycle) and perhaps 1,000th the rated capability

   of the device. They don't fail except for reasons

   of manufacturing defect or installation error . . .

   and particularly NOT from having selected the wrong

   rating for the device. When a coil suppression device

   is working, the VOLTAGE and CURRENT impressed upon

   it is trivial compared to its ratings. 

  	  Quote: 	
		  

I'm interested in comments on the relative merits of using diodes vs. bi-directional zeners.  Also, the failure modes of both devices.  And how to size the bi-directional zeners to minimize the chance of a failure, especially a short failure.

	


   You're worrying waaaayyyy too much about it.

   This topic has been discussed and many platinum

   plated coil suppression techniques have been offered

   up as the path to Nirvana. But when it gets down

   to the physics of contactor and power relay

   control, it's easy to demonstrate.



http://www.aeroelectric.com/Pictures/Curv  es/704-1OpeningTimeNoDiode.gif 



http://www.aeroelectric.com/Pictures/Curves/704-1OpeningTimeWithDiode.gif 



http://www.aeroelectric.com/Pictures/Curves/704-1DelayWithDiode.gif



http://www.aeroelectric.com/Pictures/C  urves/7041DelayWithDiode.gif



http://www.aeroelectric.com/Pictures/Curves/CH_w_1n5400.gif



http://www.aeroelectric.com/Pictures/Curves/CH_w_2x18v_Transorbs.gif



http://www.aeroelectric.com/Pictures/Curves/CH_w_120_Ohm.gif



http://www.aeroelectric.com/Pictures/Curves/S704-1_Drop-Out_Delay_with_Diode.jpg



http://www.aeroelectric.com/Pictures/Curves/S704-1_Drop-Out_Delay_without_Diode.jpg 

   These are but a few of dozens of traces recorded by

   myself where we looked at drop-out-delay, contact

   transition velocity, and contact bounce vs. various coil

   suppression techniques (including simple placement

   of a 120 ohm resistor across the coil!).

<  BR>   Bottom line is that anything you want to do will   be fine . . . make it easy on yourself and be wary

   of pronouncements and dire warnings from folk who

   not been there nor done that. Do a failure mode

   effects analysis and craft architectures that offer

   a Plan-B for ALL uncomfortable failures. The road

   to comfortable termination of flight is not paved with

   prophylactics AGAINST failure . . . rather a PLAN for

   dealing comfortably with any failure.



   Bob . . .  	  Quote: 	
		  

	


	
  

 

   [quote][b]

- The Matronics AeroElectric-List Email Forum -

Use the List Feature Navigator to browse the many List utilities available such as the Email Subscriptions page, Archive Search & Download, 7-Day Browse, Chat, FAQ, Photoshare, and much more:

http://www.matronics.com/Navigator?AeroElectric-List

Colyn Case

Joined: 12 Oct 2009
Posts: 13
Location: Thetford Center

Posted: Mon Oct 12, 2009 10:00 am Post subject: Re: Diodes vs. Bi-Directional Zeners for Contactors

Bob,



I agree with you about being prepared for any contactor to fail and not to be overly concerned about that.  A failure that results in not being able to operate the contactor should not in itself be a hazard to flight.



However, a failure that results in smoke, although in other ways benign, is undesirable.    I would be interested in what happens to the coil circuit as a function of the failure mode of the suppression device.  If it fails open, so what, you lost some suppression.  If it fails closed I think you get a dead short on the coil circuit, in other words the wire that goes to your starter key and to ground.   



On my airplane, I don't have that circuit fused, reasoning that the coil itself is going to melt down before the wire does.   but if you put a device in parallel with the coil and it shorts, then depending on the characteristics of the device you've got smoke in the cockpit.



So  either you have to choose a device which you are pretty sure is going to fail open, or you need to fuse that circuit, right? (I don't see a fuse shown in the diagrams in the appendix).   



Taking the devices:



resistor - likely fail open

diode - ?

zener diode - fail closed

snapjack - ?



thanks,



Colyn

- The Matronics AeroElectric-List Email Forum -

Use the List Feature Navigator to browse the many List utilities available such as the Email Subscriptions page, Archive Search & Download, 7-Day Browse, Chat, FAQ, Photoshare, and much more:

http://www.matronics.com/Navigator?AeroElectric-List

_________________
Colyn Case
colyncase@earthlink.net

Eric M. Jones

Joined: 10 Jan 2006
Posts: 565
Location: Massachusetts

Posted: Tue Oct 13, 2009 5:56 am Post subject: Re: Diodes vs. Bi-Directional Zeners for Contactors

Bob N., you are truly to be admired for the limitless energy you have put into this subject. I don't know how you do it.



I sell "SnapJacks" bi-directional zeners for putting across the coils of relays, inductors, and especially contactors--as recommended by Tyco, Potter and Brumfield, Megavac, Kilovac, and virtually all other relay engineers who have ever looked at the problem.



But hey, you could be right! One lone voice in the wilderness. I admire you.



"Every act of conscious learning requires the willingness to suffer an injury to one's self-esteem...." 

    -Thomas Szasz

- The Matronics AeroElectric-List Email Forum -

Use the List Feature Navigator to browse the many List utilities available such as the Email Subscriptions page, Archive Search & Download, 7-Day Browse, Chat, FAQ, Photoshare, and much more:

http://www.matronics.com/Navigator?AeroElectric-List

_________________
Eric M. Jones
www.PerihelionDesign.com
113 Brentwood Drive
Southbridge, MA 01550
(508) 764-2072
emjones(at)charter.net

user9253

Joined: 28 Mar 2008
Posts: 1925
Location: Riley TWP Michigan

Posted: Tue Oct 13, 2009 8:58 am Post subject: Re: Diodes vs. Bi-Directional Zeners for Contactors

TYCO's document at  	  Code: 	
		  http://relays.tycoelectronics.com/appnotes/app_pdfs/13c3311.pdf 	


says, "It is the velocity of the armature that is most affected by coil suppression. If the suppressor provides a conducting path, thus allowing the stored energy in the relay's magnetic circuit to decay slowly, the armature motion will be retarded and the armature may even temporarily reverse direction."



TYCO's document seemed credible until the above paragraph.  How can a relay start to drop out, then reverse direction?  Where does the power come from to do this?  Using a diode for arc suppression, the relay will start to drop out when the voltage decays to some point less than 12 volts.  In order for the relay to reverse direction, the voltage will have to increase.  But how could it increase?  The voltage has already decayed to the point where the relay drops out, and the diode is still shorting out what little energy is left.  It is common knowledge that a relay requires a higher voltage to pull in than to drop out.  TYCO is suggesting the impossible.  Making that ridiculous statement makes one question the validity of the whole document.  The table showing the drop out times of various suppression devices does not make it clear exactly what time interval was measured.  Was it the interval from when power to the coil was first shut off to the time that the contacts were fully opened?  If so, that time is irrelevant to arcing of the relay contacts.  The only time interval of concern is from when the relay contacts first start to open until they are open far enough to quench the arc.  The important thing is, how much longer do relay contacts arc when using diode compared to using a zener?  And is that time difference significant in our application?

  I agree with Bob.  The TYCO article is BS.

Mike Easley and Colyn Case are concerned about the arc-suppression diode shorting out across the main contactor.  A 22awg wire can be used from the contactor fat terminal to the coil terminal, then a 18awg from the other side of the coil into the cockpit.  In case of an unlikely short, the 22awg wire will burn open first, keeping smoke on the engine side of the firewall.

Joe

- The Matronics AeroElectric-List Email Forum -

Use the List Feature Navigator to browse the many List utilities available such as the Email Subscriptions page, Archive Search & Download, 7-Day Browse, Chat, FAQ, Photoshare, and much more:

http://www.matronics.com/Navigator?AeroElectric-List

_________________
Joe Gores

nuckolls.bob(at)aeroelect
Guest

Posted: Tue Oct 13, 2009 10:42 am Post subject: Diodes vs. Bi-Directional Zeners for Contactors

I'm interested in comments on the relative merits of using diodes vs. 

bi-directional zeners.  Also, the failure modes of both devices.  And 

how to size the bi-directional zeners to minimize the chance of a 

failure, especially a short failure.

What would happen if either were fail shorted due to high voltage?



    Excellent question . . . and the answer goes directly

    to countless studies and adjustment of design goals

    for the purpose of producing an ACCEPTABLE reliability

    factor (generally stated in failures per flight hour).



    Any time we can deduce that the failure rate is better

    than 1 in 10 to the minus 6 failures per flight hour

    (i.e. 1,000,000 flight hours per failure), the guys

    who worry about such things relax a lot.



    Now, does such a determination say that all components

    under study will last 1,000,000 hours?  No, the

    testing and calculations tell us that the AVERAGE

    failure rate for a large population of identical

    components is 1,000,000 hours per failure or better.



    If you have a device fail in 500,000 hours, then

    other components of that population will have to

    do better than 1,000,000 hours for the average to

    hold.



    Experience shows us that robust and otherwise new

    components which fall victim to errors of manufacturing

    process will fail early (infant mortality) . . . especially

    if operated at or just above extreme stress ratings

    for that device. This process is called "screening"

    and is the basis for Mil-Std-883 qualified components

    favored in critical military and space programs. The

    screening process is designed to precipitate infant

    mortality failures.



    Even then, some folks paid to worry will sandbag their

    design goals by telling suppliers to DE-RATE their

    already screened 883 components by some factor. Resistors

    will be de-rated to 1/2 power. Capacitors operated at

    1/2 rated voltage, etc. etc.



    I've smiled when reading the purchase specs driven

    by such worries. Virtually every product failure

    over which I was given direction turned out to be

    an error in design or process, not a failure of component

    to meet requirements for which it was properly designed

    and qualified. It seems like the guy turning the wrenches

    is still the riskiest ingredient in the recipe. So let's

    potential for stress induced failure of a diode across the

    coil of a contactor. See:



http://www.fairchildsemi.com/ds/1N/1N4001.pdf



    For the most part, we're building 14v airplanes with

    a sprinkling of 28v. Okay, a 50v rated diode will have

    a 2x headroom in a 28v airplane, 3.5x in a 14v airplane.

    For a few cents more, one COULD install a 600V rated diode.

    Man, talk about de-rating!



    On the conduction side of the study, the typical 1A

    diode (1N400x series) is designed and qualified to

    conduct 1A at a max voltage drop of 1.1 volts. Okay,

    this is 1.1 watts of heat being dumped out in that

    diminutive glass or plastic package. Now, we never

    SEE that 1.1 volts under our anticipated usage.

    But let's take worst case.



    CAPACITORS charge to some voltage will deliver that

    SAME voltage at potentially HIGH current during

    the discharge cycle. INDUCTORS charged to some current

    will deliver that SAME current at potentially HIGH

    voltage during their discharge cycle.



    Now, here's an el-cheeso battery contactor running

    coil current of .6 to 1.0 amps. When the circuit

    is broken, we can expect a PULSE of current at no

    more than 1.0 amps. Further, the duration of that

    current is on the order of 200 milliseconds ramping

    down from 1.0 amps to zero. Let's assume an AVERAGE

    of 0.5 Amps over 200 milliseconds with a maximum

    voltage drop of 1.1 volts. Hmmm . . . that's a worst

    case value of 110 milljoules of energy over 0.2 seconds

    dumped into a device DESIGNED to dissipate 1100 millijoules

    of energy every second for an indefinite period of time.



    May I suggest that this too is an exceedingly conservative

    de-rating philosophy?



    So while we're discussing potential failure modes

    for becoming either shorted or open, let us not loose

    our grip on probability of failure based on the expected

    stresses versus stresses defined by the device's

    design. Devices used for coil spike suppression are

    VERY lightly taxed and therefore exceedingly low

    failure rates.



    Short answer is that "sizing" of these components

    based on their electrical characteristics is a

    exercise in no-value-added-worrying. I prefer to

    size them for mechanical robustness as illustrated

    many places in my writing and illustrating. The

    beefier 1N5400 series devices are less likely to

    be damaged by over-enthusiastic wrench turning.



    Whether you use zeners-diode combos, bi-directional

    Transorbs, resistors, diodes, etc will have no

    observable effect on system reliability based on

    ratings. But beating the little guys up with

    wrench or hammer is another matter. The more

    valid worry is not WHAT you use but HOW you

    use it.



        Bob . . .



         ---------------------------------------

        ( . . .  a long habit of not thinking   )

        ( a thing wrong, gives it a superficial )

        ( appearance of being right . . .       )

        (                                       )

        (                  -Thomas Paine 1776-  )

         ---------------------------------------

- The Matronics AeroElectric-List Email Forum -

Use the List Feature Navigator to browse the many List utilities available such as the Email Subscriptions page, Archive Search & Download, 7-Day Browse, Chat, FAQ, Photoshare, and much more:

http://www.matronics.com/Navigator?AeroElectric-List

Display posts from previous:

	Matronics Email Lists Forum Index -> AeroElectric-List	All times are GMT - 8 Hours
Page 1 of 1

Jump to:

You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum
You cannot attach files in this forum
You can download files in this forum