Connect with us

diode for collapsing magnetic field

Discussion in 'General Electronics Discussion' started by Blank Stare, Oct 10, 2012.

Scroll to continue with content
  1. CDRIVE

    CDRIVE Hauling 10' pipe on a Trek Shift3

    4,960
    651
    May 8, 2012
    I realize that.


    Chris
     
  2. BobK

    BobK

    7,682
    1,688
    Jan 5, 2010
    CDIRVE,

    I really don't see what you are getting at. If the back EMF dominates when you remove power from the motor, the diode would do nothing since it would be reverse biased. Are you suggesting that the diode should be the other way around?

    Bob
     
  3. CDRIVE

    CDRIVE Hauling 10' pipe on a Trek Shift3

    4,960
    651
    May 8, 2012
    I explained that in my first reply, in the Edit comment. But my primary concern is what's destroying his relay contacts. Windshield wiper motors and power windows can approach 10A and I don't think flyback diodes are employed. You can bet his tractor starter motor draws More than this 20A motor. Yet, the starter solenoid he tried fried in a heartbeat. I'm questioning if there's something else going on here. I'm fishing, so to speak. Which is why i asked him if he's measured the coil voltage to insure those contacts are being closed hard.. For certain, I'm not saying it's definitely not BEMF.

    Chris
     
  4. BobK

    BobK

    7,682
    1,688
    Jan 5, 2010
    Hopefully, the OP will try out the diode and report back. Then we will know more.

    Bob
     
  5. CocaCola

    CocaCola

    3,635
    5
    Apr 7, 2012
    That is a good point, I wonder if his 'remote' switch is not fully switching or holding the relay, thus he is getting 'chatter' on the relay increasing the likelihood of excess arcing...
     
  6. CDRIVE

    CDRIVE Hauling 10' pipe on a Trek Shift3

    4,960
    651
    May 8, 2012
    I realize that this is circumstantial evidence but Blank Stare said that the starter solenoid failed faster than any of the other relays. I thought that quite odd and really begged an explanation. When you consider that starter solenoids pull coil currents far greater than typical relays it does beg questions about the remote control circuit that powers the coil..

    Chris
     
  7. Blank Stare

    Blank Stare

    48
    0
    Oct 10, 2012
    The air pump is made by a company by the name of "Viaire" (sp?). The pump serves only one purpose - to fill a 2-1/2 gallon tank to 125 psi. The pump is rated 100% duty . It is rated 20 amps, 12 volts.

    The pump is controlled by a pressure switch, like you find on a contractor's compressor, or the air bladder on a water well. There is also a pressure regulator, and bleed valve, in the cab of the truck. So it is essentially the same as a contractor's portable compressor.

    The pressure switch is powered (switched) by the relay, which is activated/deactivated by a remote switch on the dash of the truck. It's a simple push-button, on/off switch.

    There is also an over-ride switch, which allows me to take the pressure up, in cases where the pressure has not fallen to the lower end of the pressure switches range, where it would normally activate the pump. Even when not using that over-ride switch, the relays have failed.

    All of which, of course, hooks into the fuse box, under the hood. Yes, is has it's own fuse, as well.

    The air from the tank fills the air shocks, feeds an air horn, and is used to fill tires, and to blow dust and other "crap" off of tools and parts, such as when I take my chainsaw apart in the field.

    Everything about this setup is custom. With the help of some folks from another forum (which I can no longer find a link to,) I designed the system, and assembled the electronics on a piece of breadboard, with a surface area of about the size of 2 large paper clips. Obviously that doesn't include the switches, relays, fuses, etc...

    Now on to the concepts at hand.

    • 1. Either the pump, or the relay or both are having issues with a collapsing EMF (I think?)
    • 2. The diode on the pump should eliminate EMF from the pump
    • 3. The diode on the relay goes on the low power side, because that is the side that actually creates/removes the EMF that operates the high power side of the relay.

    Which leaves me confused: Do we think that the relay is arcing on the low power side, the high power side, or both?

    Alright, how do I measure the coil voltage? Simple multi-meter? Set on DC? Am I looking for volts, amps, or something else?

    I only know one place that I can reliably order diodes. Mouser.com. The problem is, they'll charge me 8 bucks for shipping, for these .25 cent parts. I guess I'll go do a search for something more reasonable, before I grab my ankles, and take the punishment. If anyone knows a good source for "onsie-twosie" orders, I'd love to know about it.

    OK, all that said, here's the diagram I used to build the system., and picture of the system as seen from inside the truck (the items in the 2nd oval are no longer connected.) I see that it says 30 amp on the pump, but I think that's a typo, and should actually be 20 amp. Either way, it blew 40 amp relays, and even a 50 amp relay I bought, that was nearly the size of a Klondike Bar...not to mention the solenoid from a lawn tractor.

    Thanks,

    ~ Blank Stare
     

    Attached Files:

  8. CDRIVE

    CDRIVE Hauling 10' pipe on a Trek Shift3

    4,960
    651
    May 8, 2012
    See that statement in bold? One can't help but wonder if something happened causing the first failure after it had cycled, trouble free, for a few months. Hey, I could be chasing my tail but from a trouble shooting standpoint it's hard to ignore that statement.

    Chris
     
  9. Blank Stare

    Blank Stare

    48
    0
    Oct 10, 2012
    Yeah, I've been down that train of thought. I've checked all my wiring, and found no damage or suspect wires or connections. Since it works for a varying number of cycles, each time I install a new relay, I don't know where else to look, but the collapsing field.

    A collapsing field would have a random amount of arcing each cycle, right? Add to that the (lack of) quality control in the manufacture of the relays, I am guessing that the relay(s) could/would last a few, or many cycles.

    Still, given that the wiring seems fine, and indeed functions correctly until the relay takes a crap, how can I trouble shoot other parts of the system?

    BTW, I am planning to measure the coil voltage, as soon as I understand what to do.

    Thanks,

    ~ Blank Stare
     
  10. CDRIVE

    CDRIVE Hauling 10' pipe on a Trek Shift3

    4,960
    651
    May 8, 2012
    You must measure the coil voltage physically at the coil. This is to insure that you're measuring any possible voltage drop in either the 12V or ground leg, I would place the probes directly across the coil terminals.

    Are these crimped solderless terminals?

    Chris
     
  11. KrisBlueNZ

    KrisBlueNZ Sadly passed away in 2015

    8,393
    1,271
    Nov 28, 2011
    Arcing in relay contacts can be worsened if the relay's contacts cannot move apart quickly; this can be caused by the relay's magnetic field being unable to collapse quickly due to current flow through the diode connected across the coil to suppress the back-EMF from the coil. I'm not sure whether you're using a diode across the coil in this application, but if you are, have a look at http://relays.te.com/appnotes/app_pdfs/13c3264.pdf
     
  12. CDRIVE

    CDRIVE Hauling 10' pipe on a Trek Shift3

    4,960
    651
    May 8, 2012
    Kris, he has a diode in series with the coil. Did you see the last schematic he posted?

    Chris
     
  13. KrisBlueNZ

    KrisBlueNZ Sadly passed away in 2015

    8,393
    1,271
    Nov 28, 2011
    Yes, but that doesn't give the full story. When the voltage on the anode of the diode disappears, the inductance of the relay coil will try to pull the diode voltage negative. Whether that is suppressed or not then depends on the circuitry connected to the diode's anode, which is just shown as a black box. That's why I referred him to the article; it's up to him to figure out whether it's relevant or not.

    I just referenced that article since it's well-written and raises a consideration that he may not be aware of. I'd been designing with relays for years before I found out about it. It's actually important to know, I think, because fast and clean opening of relay contacts is important in some applications.
     
  14. CDRIVE

    CDRIVE Hauling 10' pipe on a Trek Shift3

    4,960
    651
    May 8, 2012
    Kris, I read and saved your pdf. Good stuff!! ;)

    Chris
     
  15. CDRIVE

    CDRIVE Hauling 10' pipe on a Trek Shift3

    4,960
    651
    May 8, 2012
    I just took a peek at the schematic again. Too bad the components aren't numbered. It would make this post easier. As I see it there are multiple current paths that the coil's BEMF is going to find when the pressure switch opens. It will find its way to gnd through both 1n4001's, through a bi-color LEDs anode resistor and through the LEDs when reverse breakdown is reached. There is other path that finds there way back to the battery through another led anode resistor..

    Whether any of this is related to contact dropout I don't know. I just thought I'd mention it.

    Blank Stare, Is this circuit a commercial product? Is it sold as a kit? If either, did you search for any feedback by others that have experienced your problem?

    Chris
     
  16. Blank Stare

    Blank Stare

    48
    0
    Oct 10, 2012
    MIA

    Sorry I have been out of touch.

    Doctor says I have Pneumonia.

    I should be back in a few days or so.

    In the meantime, thanks for all the help to date. :)

    ~ Blank Stare
     
  17. CDRIVE

    CDRIVE Hauling 10' pipe on a Trek Shift3

    4,960
    651
    May 8, 2012
    Rest and get well.

    Chris
     
  18. Blank Stare

    Blank Stare

    48
    0
    Oct 10, 2012
    Thanks for the well wishes. I have periods of feeling good, and periods of feeling like doo-doo...

    Let's see if I can answer questions asked, and ask questions I need to move forward.

    CDRIVE,

    Crimped, yes - every connection in the system, except where I used a trailer wiring harness to bring certain items through the firewall. Those are mostly LEDs, which all seem to work properly, when the relay is working.

    When you say "I would place the probes directly across the coil terminals." is that on the low power side? That's where the "coil" is right?

    Please be specific, and feel free to talk "down" to me. Which component is a "black box"?. I'll be happy to do my best to remove any ambiguity.

    That is the point - this appears to be the problem I am having, but being a newb hobby-ist, I came here first for advice and guidance.

    I tried to read that PDF - way over my head. Still, I appreciate the effort.

    I do? Oh boy, now I am completely lost. Are you sure?

    Other than that, I am lost in the whole anode/inductance discussion.

    I am somewhat sure that the speed of opening/closing of the relay is a factor here, since it keeps coming up, and makes sense, even to a newb like me.

    CDRIVE,

    Sorry, it never occured to me to number the components. could you refer me to a good example of a simple schematic that is numbered? I don't have a problem updating the schematic to make it easies to work with, but I am not sure what the nomenclature is.

    This is a completly non-commercial application of commercial parts, gathered together from a variety of sources. To my knowledge no one sells a kit like this, ready to install, or already installed. The closest things to it are air pillows on newer cars, and air lockers on big rigs, that have air brakes.

    ~~~~~~~~~~~~~~~

    I'll get outside and put a multi-meter on that relay, as soon as I am feeling better. It occurs to me that I can check both sides, after I replace the relay (again...)

    Here are some pictures - the pressure switch is a close approximation - I could not find an exact match to the one I installed. I think that schematic should help with understanding what I used to control the pressure in the "air locker".
     

    Attached Files:

    Last edited: Oct 16, 2012
  19. KrisBlueNZ

    KrisBlueNZ Sadly passed away in 2015

    8,393
    1,271
    Nov 28, 2011
    My comments relate to the connections to the coil of your big relay.

    In your circuit, the coil is energised by current flowing through one of two diodes. The diodes have their cathodes tied together and feeding one end of the relay coil; the other end of the relay coil is grounded.

    The relay coil is an inductor. When a steady DC current is flowing in an inductor and the source of that current suddenly disappears, a phenomenon called "back EMF" occurs: as the magnetic field in the relay collapses, the relay coil generates a brief voltage spike of the opposite polarity to the voltage that had been applied. The voltage of this spike may be hundreds or even thousands of volts.

    It will occur each time the source of current to the relay coil is interrupted; this will be when the "automatic pressure switch" (the black box I was referring to) or the SPDT switch or the SPST switch is opened (whichever contact breaks last).

    So whenever the supply current to the relay coil disappears because one of those switches opens, the voltage at the cathodes of those two diodes will briefly spike negative by several hundreatd volts. You can see the spike if you connect a diode, a resistor (e.g. 10k) and an LED in series across the coil, with the cathodes pointing away from ground. The LED will blink briefly each time the coil is de-energised.

    This back EMF voltage from the relay coil will flow through the relevant diode and will cause sparking in the contact that opens last, and may eventually damage the contacts. (Not the contacts of the relay; the contacts of the switches that drive the coil.) It could also damage the LEDs, which do not like reverse voltage.

    The traditional way of dealing with back EMF from relay coils is to connect a diode directly across the coil, with (in this case) its anode to the grounded end. When the coil tries to produce its back EMF voltage, its non-grounded end tries to go below (negative with respect to) ground, and this causes the diode to conduct, so the voltage never actually goes more than about 1V negative. This is what your friends were referring to with their comments in your first post.

    This "back EMF suppression diode" is a convenient way to suppress the back EMF generated by the relay coil upon turn-off, but it can cause a problem, because it greatly slows the rate at which the magnetic field in the relay can collapse, which slows down the separation of the relay contacts, which can cause arcing and welding of the contacts. This is the subject of the PDF I linked to.

    I raised this issue because you mentioned you were having trouble with damage to the relay contacts. I imagine that's due to the inductive nature of the load (the pump), which itself can cause back EMF when your relay contact tries to open and interrupt the flow of current to it, causing arcing between the contacts of your big relay. Steve suggested reverse-connecting a diode across the load, I think, to try to suppress this back EMF. I agree this would be worth trying as well. But it needs to be rated for at least as much current as the motor draws during operation.

    As for the relay coil circuitry, the PDF I linked to recommended a zener diode in series with a diode, to clamp the back EMF voltage to a reasonable value such as 100V or so. This is a good compromise - it allows the magnetic field to collapse fairly quickly and cleanly, and it still prevents wild uncontrollable voltage spikes being fed back into the driving circuitry. Each LED can be protected from the -100V spike with a series diode with a high enough voltage rating, for example 1N914.

    A better solution here would be a transistor or MOSFET driving the coil, with the diode and zener suppression circuit as described in the PDF. Let me know if you want to try that. Personally I would try suppressing the back EMF from the load first, as Steve suggested.
     
  20. CDRIVE

    CDRIVE Hauling 10' pipe on a Trek Shift3

    4,960
    651
    May 8, 2012
    I mentioned this before but maybe not as clearly.

    I don't think that the back emf of the relay coil ever comes close to what it would be if it was an open circuit when the pressure switch opens. I say this because the 1N4001 diode on the right provides a current path through two resistors and back to the battery. There's also a path to ground through a bi-color led when rev breakdown is reached.

    Here's a couple of questions to the OP:
    I asked before but didn't get an answer. What's the make and model number of the motor?
    Who created this schematic?

    Based on this ....

    I know you didn't design and draw this circuit, so where's the link to where you got it from?

    Chris
     
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.
Electronics Point Logo
Continue to site
Quote of the day

-