Selecting a zener diode for solenoid flyback suppression

Go To Last Post
17 posts / 0 new
Author
Message
#1
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

I'm low-side driving a 24V, 1.2A DC solenoid with a 30V MOSFET. I'd like to protect the MOSFET with a zener diode across the source and drain (cathode to drain, anode to source). If I select, say, a 27V zener, how do I spec the power rating? Is there a way to calculate how much current will be flowing through it when the MOSFET shuts off and the flyback voltage rises above 27V? My MOSFET and (randomly selected) 500 mW zener work fine when driving a 24V, .8A solenoid, but using the 1.2A solenoid fries the zener. Thanks for any tips.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

A picture is work a 1K words, please provide a schematic of your circuit, at least the relay driver portion.

 

Jim

 

 

(Possum Lodge oath) Quando omni flunkus, moritati.

"I thought growing old would take longer"

 

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Here you go.

Attachment(s): 

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Normally one places a diode across the solenoid terminals, the FET already has a body diode.

 

 

jim

 

 

(Possum Lodge oath) Quando omni flunkus, moritati.

"I thought growing old would take longer"

 

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

lautman wrote:
I'm low-side driving a 24V, 1.2A DC solenoid with a 30V MOSFET. I'd like to protect the MOSFET with a zener diode across the source and drain (cathode to drain, anode to source). If I select, say, a 27V zener, how do I spec the power rating? Is there a way to calculate how much current will be flowing through it when the MOSFET shuts off and the flyback voltage rises above 27V? My MOSFET and (randomly selected) 500 mW zener work fine when driving a 24V, .8A solenoid, but using the 1.2A solenoid fries the zener. Thanks for any tips.

 

The peak Zener current is equal to the inductor current, and decays from there - decay time depends on the mH of the inductor.

 

Why do you need a zener ? Most of the time a reverse flyback diode is fine for solenoids, (as above) and only if you have special fast-collapse needs, do you change to an active clamp, & then usually at higher voltage.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Fast fly-back diode is better solution. If you need faster switch off, add some small value resistor in series with fly-back diode. You may also consider to put something like low power 16V zener across gate to source as a protection from possible inducted voltages.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Unless you have a high power coil, nothing is needed....the fet has a built-in zener diode.  If the concern is high (probably not), you can use a beefier avalanche-rated fet, or even better yet, a fet that has an internal gate-clamp (momentarily turns on the fet to quench any overvoltages)

  

 

As far as putting a diode across the coil, a lone zener buys you absolutely nothing---it must be installed to block current & when the coil collapses, the current flows in the arrowed direction, same as a regular diode!  So you get nothing beneficial over using a regular diode.

You can use a zener in series with a regular diode (back to back, opposite directions).  When the coil is on, the diode is blocking.   When the coil is collapsing, the current flows through the diode and the now turned-on zener  (zener Vdrop).

This added voltages drop speeds the coil shutoff droop rate (di/dt =  Vcoil/inductance)  Now the Vcoil=Vzener+Vdiode, hence the speedup (compared with no zener & only Vdiode).  Vdiode typical 0.25 V(Schottky) or 0.7V (std)

 

 

 

 

 

 

When in the dark remember-the future looks brighter than ever.   I look forward to being able to predict the future!

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

If the mosfet is avalanche rated you might not need to add any flyback diode.  More information than you have given about the application and mosfet is needed to make this decision however.

Letting the smoke out since 1978

 

 

 

 

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Who-me wrote:

Why do you need a zener ? Most of the time a reverse flyback diode is fine for solenoids, (as above) and only if you have special fast-collapse needs, do you change to an active clamp, & then usually at higher voltage.

 

    The difference from 24V to 27V is not much, so I don't think this is the reason for the Zener. Sometimes you do not have access to the 24V rail so in this case is easier to just clamp it from the ground. Sometimes the 24V rail is a long wire or noisy and it spikes to 30V and beyond easy.

 

@lautman

    Some datasheet gives the instantaneous power the diode can handle (non repetitive). Here is an example:

    In your case the initial power on the diode is 1.2A x 27V = 32.4W. This decays almost linearly. The exact function is complicated because the clamping voltage is not constant due to the current variation through the diode and the effect of the relays armature movement. Best is take a scope and see how much energy is dissipated. With this, go to the graph and for the average current see how much energy the diode is allowed to take.

 

    Easier is take a 1W diode and actuate the relay at least twice the max rate in the field and see if survives. If fails, add more or thing of other ways to fix this problem. 30V as for the MOSFET is tight. One other way is to add a zener diode in between source and gate, with an extra normal diode in series. But this complicates things.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

The MOSFET holds up in testing without any external protection (because of the body diode).  I just thought adding the zener would provide more margin.  Seems to be getting more complicated than it's worth.  I'll just stick with the bare MOSFET.  Thanks for the tips, though.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

lautman wrote:

The MOSFET holds up in testing without any external protection (because of the body diode).  I just thought adding the zener would provide more margin.  Seems to be getting more complicated than it's worth.  I'll just stick with the bare MOSFET.  Thanks for the tips, though.

 

No, it is not the reverse body diode that is taking the energy, it is the MOSFET in avalanche.  Check that it is rated for that, or you will get field failures.

 

I recall killing a TO248 power mosfet, with just a ferrite bead on the drain lead. 

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Anyhow, you're not going to use a 27V zener to protect a 24v relay, that is way too tight..the supply might actually be 26v, the zener might really be 24.6V (which means the zener shorts the supply & blows).

 

If you had a 12V ckts you might use an 18V zener to clamp transients, or probably even higher than 18V & simply design things so that larger transients are ok...there should be a huge margin, since activating on a non-transient is fatal(because  it is then clamping the full-time power supply).

 

With your fet & zener both having the same trip voltage (30V/27V), you wouldn't know who is doing the clamping anyway.  That would be alleviated using a common as dirt 60V FET, since it would not self-clamp anywhere in the vicinity of 28 volts.   You're not going to have a problem anyhow, unless you are PWMmin'g the coil at KHz, if even then (assuming a sturdy power FET is in use).  I PWM 10amps all day at 30KHz, no issue with a TO-220 FET. 
 

When in the dark remember-the future looks brighter than ever.   I look forward to being able to predict the future!

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

By not putting the clamp across the inductor, you're relying on the impedance of your supply to soak up the energy - how successful this is depends on your supply. You may also be inviting EMC issues. With the clamp across the inductor, then a loop is formed. You know where the current flows.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 1

Kartman wrote:

By not putting the clamp across the inductor, you're relying on the impedance of your supply to soak up the energy - how successful this is depends on your supply. You may also be inviting EMC issues. With the clamp across the inductor, then a loop is formed. You know where the current flows.

 

Not entirely. 

The power supply has to already be able to deliver the inductor energy, and a Zener/avalanche actually has softer current edges, because the supply load decays slowly to zero.

The ground currents are also softer, as the Zener/avalanche current flows to ground, while it decays.

 

The down side is as the OP noted, you do 'need a bigger zener' if using largish solenoids.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Who-me wrote:

No, it is not the reverse body diode that is taking the energy, it is the MOSFET in avalanche.  Check that it is rated for that, or you will get field failures.


I’ve always thought it was the body diode that avalanches. Oh, well, I stand corrected. On researching this issue I find it’s difficult to get the necessary information from the data sheet. Several of the MOSFETs I’ve been looking at don’t even quote EAS or UIS. As this is for a personal project, rather than a commercial one, it’s not so critical and I’m hoping that stress testing a single part beyond what it will experience in my application will be sufficient. So far, two different parts have passed, without external protection.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Well, the FET is really one device, not  a "separate" body diode....it is parasitic to the design.

 

All semiconductor devices contain parasitic components intrinsic to the physical design of the device. In power
MOSFETs, these components include capacitors due to displaced charge in the junction between p and n regions,
resistors associated with material resistivity, a body diode formed where the p+ body diffusion is made into the nepi-layer, and an NPN (bi-polar junction transistor henceforth called BJT) sequence (BJT) formed where the n+
source contact is diffused.

In avalanche, the p-n junction acting as a diode no longer
blocks voltage. With higher applied voltage a critical field is
reached where impact ionization tends to infinity and carrier
concentration increases due to avalanche multiplication.
Due to the radial field component, the electric field inside the
device is most intense at the point where the junction bends.

 

https://www.vishay.com/docs/90160/an1005.pdf

 

Avalanche rugged MOSFETs are designed to contain no

single consistently weak spot, so avalanche occurs

uniformly across the device surface until failure occurs

randomly in the active area.

 

 

 

When in the dark remember-the future looks brighter than ever.   I look forward to being able to predict the future!

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

> Anyhow, you're not going to use a 27V zener to protect a 24v relay ...

Good point, though much as I would like to use a 60V MOSFET, I haven’t been able to find one that meets my other requirements, including a SOT-23 package. That left me looking at 30V devices (due to the variety of choices in this range), and I was trying to find a zener between 30V and 24V, since Inwas worried about subjecting the MOSFET to a voltage higher than its rating. I now realize they routinely tolerate 38V - 40V in avalanche, so I could have picked a zener closer to that range. In any event, it looks like I can ditch the zener and just count on the MOSFET’s avalanching. The ones I’ve tested so far don’t even get warm after several thousand 400 msec. tests.