OT Clamping diode on inductor
Close
Login to Your Account
Page 1 of 2 12 LastLast
Results 1 to 20 of 23
  1. #1
    Join Date
    May 2002
    Country
    UNITED STATES
    State/Province
    Oregon
    Posts
    692
    Post Thanks / Like
    Likes (Given)
    1
    Likes (Received)
    95

    Default OT Clamping diode on inductor

    How would one go about sizing a diode to be used as a clamping diode for an electric clutch coil? In this case, the clutch is on a garden tractor...unknown inductance, 12V DC electrical system. Switches that control the clutch are burning out fairly quickly, but the clutch seems to work just fine.

  2. #2
    Join Date
    Jun 2019
    Country
    UNITED STATES
    State/Province
    Colorado
    Posts
    12
    Post Thanks / Like
    Likes (Given)
    4
    Likes (Received)
    2

    Default

    The diode does not have to be as beefy as you might initially think. Without the diode, the voltage generated may be in the hundreds of volts (if not higher), but with the diode in place, the voltage that the diode sees is only ever Vfwd of the diode (e.g., ~0.6V).

    Also, even a tiny 1N4148 has a forward continuous current max of 300 mA. The peak forward surge current is 2A (!), but only for 1 uS; in my testing (on small 5V relays), it could take ~20ms for the energy in the inductor to dissipate, so you'd have to derate a lot from 2A. But, 2A at the extreme end and 300 mA on the low end gives you an idea of the range that can be handled safely by the diode.

    The hard part that I don't know the answer to, is what the actual forward current seen by the diode. I assume that the problem that the diode faces is one of heat dissipation. That's going to depend on the coil inductance, which for a 12V solenoid is likely going to be much higher than the puny 5V relays I mentioned. But, considering how "stout" at 1N4148 is, if you were to go to a diode a little bigger, it'd probably be fine.

    So my suggestion would be to start with the venerable 1N4148 and see if it pops--worst case, it pops. If it doesn't, problem solved. If it does, go to a fast switching panel mount diode, if such a thing exists. Otherwise, a rectifer panel mount diode is probably fine.

    HTH

  3. #3
    Join Date
    Jul 2007
    Country
    UNITED KINGDOM
    Posts
    4,613
    Post Thanks / Like
    Likes (Given)
    2394
    Likes (Received)
    2179

    Default

    might be just as cheap and rather more robust to use a 1N400<anything==1,4,7>. 1A continuous current, 30A peak and still throwaway prices.

  4. Likes sfriedberg, awander, andrewmawson liked this post
  5. #4
    Join Date
    Oct 2003
    Location
    Philadelphia, PA
    Posts
    886
    Post Thanks / Like
    Likes (Given)
    0
    Likes (Received)
    92

    Default

    Quote Originally Posted by nxd View Post
    The hard part that I don't know the answer to, is what the actual forward current seen by the diode.
    That's easy. The forward current seen by the diode is the same current that was flowing through the coil when the power was cut.

    That's the whole reason the diode is needed in the first place... The coil doesn't want the current to change value, and it'll do whatever it can (like producing a huge voltage) to keep that current flowing. The diode just gives the coil an easy place for that current to flow without producing a huge voltage.

    So, in theory, you need a diode capable of carrying a peak forward current equal to the clutch coil draw. It also needs to be able to carry that current long enough for it to decay to zero before the diode burns out.

  6. Likes andrewmawson, JST, Jmaks liked this post
  7. #5
    Join Date
    Feb 2009
    Location
    London, Ontario, Canada
    Posts
    242
    Post Thanks / Like
    Likes (Given)
    13
    Likes (Received)
    35

    Default

    Try a 1N5401,2,3,4 as they are about 4 amps continuous (depending upon lead length) and will effectively suppress the spike from the clutch coil. They will easily dissipate the coil surge while limiting the peak voltage. This is old school electronics often forgotten by later engineers.
    Last edited by Leadfootin; 06-08-2019 at 07:26 AM.

  8. #6
    Join Date
    Nov 2004
    Location
    Webster Groves, MO
    Posts
    7,029
    Post Thanks / Like
    Likes (Given)
    1742
    Likes (Received)
    3141

    Default

    Try an MOV instead of a diode. The MOVs resistance will dissipate energy while a diode simply keeps the current circulating until the coil resistance eats it up.

    Bill

  9. #7
    Join Date
    Nov 2003
    Location
    Beaumont, TX, USA
    Posts
    4,802
    Post Thanks / Like
    Likes (Given)
    164
    Likes (Received)
    1448

    Default

    The 1N4148 is a general purpose, silicone, SIGNAL diode. It is thought of as an updated 1N914 (a classic) and the two are somewhat interchangeable. For damping the spikes from an inductor I would suggest a power type diode as it would be a lot more robust. I have seen the 1N400X series used extensively for this. They are all rated at least at 1 Amp and the Voltages range from 50 Volts for the 1N4001 to 600 Volts for the 1N4005. They are fast enough for spike suppression. Almost any of them would be OK for your 12 Volt system. And they are quite inexpensive: I buy them by the hundred for $10 or less ($0.10 each). Probably double that in small quantities.

  10. #8
    Join Date
    May 2014
    Country
    UNITED STATES
    State/Province
    Massachusetts
    Posts
    111
    Post Thanks / Like
    Likes (Given)
    8
    Likes (Received)
    31

    Default

    I’ll also vote for a 1A 1N400x series (or a 3A 1B540x Series). Either are cheap from Mouser or Digital-Key.

  11. #9
    Join Date
    May 2017
    Country
    FINLAND
    Posts
    1,554
    Post Thanks / Like
    Likes (Given)
    442
    Likes (Received)
    721

    Default

    Quote Originally Posted by 9100 View Post
    Try an MOV instead of a diode. The MOVs resistance will dissipate energy while a diode simply keeps the current circulating until the coil resistance eats it up.

    Bill
    Shouldn't really matter in this application.
    Makes difference if you need really fast action like fuel injectors or reed relays switching in millisecond range but not much issue in garden tractor electric clutch.

  12. #10
    Join Date
    Mar 2012
    Location
    Geilenkirchen, Germany
    Posts
    2,248
    Post Thanks / Like
    Likes (Given)
    1376
    Likes (Received)
    1140

    Default

    I think you guys have it all wrong. The purpose of a clamping diode is to block back EMF from exceeding the the reverse bias voltage of the driver transistor. Since there is only a switch, this doesn't fix the problem of burning out contacts. The real answer is a very large cap. This gives the switch contacts adequate time to get enough dielectric air (space)between the contacts to prevent arcing.

  13. Likes andrewmawson liked this post
  14. #11
    Join Date
    May 2017
    Country
    FINLAND
    Posts
    1,554
    Post Thanks / Like
    Likes (Given)
    442
    Likes (Received)
    721

    Default

    Quote Originally Posted by steve-l View Post
    I think you guys have it all wrong. The purpose of a clamping diode is to block back EMF from exceeding the the reverse bias voltage of the driver transistor. Since there is only a switch, this doesn't fix the problem of burning out contacts. The real answer is a very large cap. This gives the switch contacts adequate time to get enough dielectric air (space)between the contacts to prevent arcing.
    Diode clamps the reverse voltage spike to less than 1 volt when the contacts open and there is not going to be any sparking.

    Capacitor has been used traditionally in ignition points either because it forms part of the LC resonant circuit or ignition is powered by ac source (magneto in small engines)
    Diode clamping here would stop the spark plug also from sparking and MOV or zener would end up with huge power loss and reduced spark energy at plug. But op's problem was not ignition points.

  15. #12
    Join Date
    Mar 2012
    Location
    Geilenkirchen, Germany
    Posts
    2,248
    Post Thanks / Like
    Likes (Given)
    1376
    Likes (Received)
    1140

    Default

    Quote Originally Posted by MattiJ View Post
    Diode clamps the reverse voltage spike to less than 1 volt when the contacts open and there is not going to be any sparking.

    Capacitor has been used traditionally in ignition points either because it forms part of the LC resonant circuit or ignition is powered by ac source (magneto in small engines)
    Diode clamping here would stop the spark plug also from sparking and MOV or zener would end up with huge power loss and reduced spark energy at plug. But op's problem was not ignition points.
    I disagree. There will be arcing. The cap acts as a small battery keeping the appliance energized long enough to create an isolating air gap. It's easy enough to test, don't take my word for it, test it if you have doubts. Consider for a moment a reversing motor as in an autopilot. In that application a diode could not work, but a large cap does.

  16. #13
    Join Date
    Sep 2002
    Location
    People's Republic
    Posts
    2,691
    Post Thanks / Like
    Likes (Given)
    194
    Likes (Received)
    1972

    Default

    25 pack 10 amp 1000 volt diodes will run you less than 10 bucks.

    when you are building a thousand tractors the price matters, one tractor, not so much.

  17. #14
    Join Date
    Jun 2018
    Country
    UNITED STATES
    State/Province
    New Mexico
    Posts
    36
    Post Thanks / Like
    Likes (Given)
    31
    Likes (Received)
    13

    Default

    It's truly amazing the electronic engineering knowledge available from the real machinists around. :-)
    ...lewie... (spent all my working life as an electronic engineer)

  18. Likes Mark Rand liked this post
  19. #15
    Join Date
    May 2017
    Country
    FINLAND
    Posts
    1,554
    Post Thanks / Like
    Likes (Given)
    442
    Likes (Received)
    721

    Default

    Quote Originally Posted by steve-l View Post
    I disagree. There will be arcing. The cap acts as a small battery keeping the appliance energized long enough to create an isolating air gap. It's easy enough to test, don't take my word for it, test it if you have doubts. Consider for a moment a reversing motor as in an autopilot. In that application a diode could not work, but a large cap does.
    No surprise there, can't use parallel diode with motor if you want to reverse the motor direction. But again this was garden tractor clutch, not autopilot motor nor fuel injector.

  20. #16
    Join Date
    Oct 2003
    Location
    Philadelphia, PA
    Posts
    886
    Post Thanks / Like
    Likes (Given)
    0
    Likes (Received)
    92

    Default

    Quote Originally Posted by steve-l View Post
    I disagree. There will be arcing.
    No, there will not be any arcing. You're putting the cart before the horse. The diode provides a low resistance path for the coil current to continue to flow after the switch has been opened so the coil never has to produce any significant voltage in the first place.

    Because it's a low resistance path, it doesn't take a lot of voltage to push that current through it. In fact, the coil only needs to produce enough voltage to forward bias the diode (less than one Volt). And since the voltage is so low, there will be no arcing.

    So when DOES arcing occur? It occurs when the only path for the inductor's current to continue to flow is a HIGH resistance path - like the air gap between the switch contacts. If that air gap is the lowest resistance path, the coil will produce a very high voltage to jump that gap.

    The point is... The diode gives the coil a completely different, easy to use, path for that current to flow so it doesn't have to jump the air gap in the switch.

    And I agree... Don't take my word for it. Test it. Start here:


  21. Likes MattiJ, Jmaks liked this post
  22. #17
    Join Date
    Jul 2014
    Location
    MI USA
    Posts
    865
    Post Thanks / Like
    Likes (Given)
    196
    Likes (Received)
    359

    Default

    Just went through this problem on a buddy's older model 25HP Scag zero turn mower. The electronic ignition timing module, safety interlock module and clutch switch were having longevity issues. Checking the service bulletins, it was a known issue, on machines with the Kohler engine. The engine manufacturer redesigned the entire ignition system to eliminate the service problem from reoccurring.

    Later, the machine manufacturer, came up with a diode clamping kit, to wire directly across the PTO clutch coil, to eliminate the high reverse voltage spikes. In later designs the clamping diode became standard equipment.
    The diode they chose was a 1N5408.

    imagejpeg_0_06-1-.jpg

    SAF Ω

  23. Likes digger doug liked this post
  24. #18
    Join Date
    Nov 2005
    Location
    Maryland
    Posts
    8,201
    Post Thanks / Like
    Likes (Given)
    746
    Likes (Received)
    1103

    Default

    Quote Originally Posted by SAF View Post
    Later, the machine manufacturer, came up with a diode clamping kit, to wire directly across the PTO clutch coil, to eliminate the high reverse voltage spikes. In later designs the clamping diode became standard equipment.
    The diode they chose was a 1N5408.
    Good choice.
    The 1N5408 is rated 3 amps at 1000 PIV.

    I use them for all similar clamping requirements for relays and such.

    - Leigh

  25. #19
    Join Date
    Jun 2001
    Location
    St Louis
    Posts
    17,766
    Post Thanks / Like
    Likes (Given)
    1753
    Likes (Received)
    2946

    Default

    As mentioned, the current at the instant of opening the circuit is the same current that exists just after that.

    Then the current decays, as energy is used up in the resistance of the coil, and the forward voltage of the diode. The diode holds the voltage to the supply voltage plus a diode drop (whatever it is at that current). This is because "X" amount of "volt seconds" of energy is stored in the inductance (1/2 x L x I^2 )

    The diode needs to easily handle the current for the time it takes to decay. They are cheap enough that in most cases it costs more to spend time optimizing to find the smallest diode that can be used. Decay time is important for the relay release time, which may be an important issue, however.

    To improve decay time, a small resistor can be added.... size that so that (R x I) plus the diode drop does not exceed your safe voltage. The resistance will speed the opening of the relay etc, by removing energy faster. (it is volt-seconds again.....The volts are made higher during the dissipation time, so seconds are smaller, the energy is removed faster)

    The capacitor idea is actually good for ignition, but bad for relays. First, it increases the turn-on surge of the switching transistor, which has to charge the cap and supply coil current. Also, it stores even more energy, and that needs to be dissipated before the relay opens.

    It can be done by combining the capacitor with a resistor, forming a standard arc damper. But BOTH are needed, and you need to consider the "Q" of the resonance, to make sure the energy is removed quickly. The resistor does the dissipation, adjusting the "Q" (sharpness of resonance and time to dissipate energy) to allow a quick relay release. Diodes are usually less trouble and often cheaper, both in cost and in PC board space. Years ago, capacitors were cheaper than the rectifiers, and were commonly used. (Still are where diodes are not appropriate, as in AC circuite). MOVs are good if you can tolerate the voltage, because they increase the volts, and so decrease the seconds, the time to release the relay.

    In an ignition, you want to KEEP the energy IN the coil core, and redirect it to the secondary to make a spark. The capacitor does that, and allows a re-strike spark, since there is a resonance, so there is a chance to have multiple sparks. In the process, the voltage at the points can reach several hundred volts in a 12V system, repeatedly. This is NOT what you want in a relay coil, where you want to quickly dissipate the energy (volt seconds) that exists in the relay coil, and limit the voltage.

  26. #20
    Join Date
    May 2017
    Country
    FINLAND
    Posts
    1,554
    Post Thanks / Like
    Likes (Given)
    442
    Likes (Received)
    721

    Default

    Quote Originally Posted by steve-l View Post
    I think you guys have it all wrong. The purpose of a clamping diode is to block back EMF from exceeding the the reverse bias voltage of the driver transistor. Since there is only a switch, this doesn't fix the problem of burning out contacts. The real answer is a very large cap. This gives the switch contacts adequate time to get enough dielectric air (space)between the contacts to prevent arcing.
    Maybe I figured out the misunderstanding.

    You are probably thinking of placing the diode parallel to switch contacts. (not working or at least there is good risk of toasting the diode with high voltage spike.)
    And "everyone" else assumes its where it should be, parallel with the load/inductance. (works)

    Capacitor would "sort of" work in both cases but diode is different animal.
    If the diode is placed across the switch contacts it has to be zener diode with suitable ratings.

  27. Likes JST liked this post

Tags for this Thread

Bookmarks

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •