# Zener Diode to Shunt Coil

#### Artbuc

Joined Jun 8, 2014
16

It talks about picking a zener based on voltage required to protect the downstream circuit. However, it does not mention zener current. In my case I want a 20VDC zener to protect a MPSA06 transistor. The relay coil will be passing about 170mA. I understand that when the coil is de-energized, the initial voltage from the collapsing magnetic field will be high but the current will never be higher than the 170mA. In looking at the zener data sheets, I do not see a spec for max reverse bias current. How do I know if a 20V DC zener will handle the 170mA transient current? Thanks.

#### ericgibbs

Joined Jan 29, 2010
9,570
hi A,
The zener is always used in serial connection with a regular diode.

The zener will in effect be wired in its zener working forward state, any reverse current will be very small, equal to the reverse current of the diode.

E

EDIT: the working zener diode current can be calculated from its rated wattage and zener voltage

example: a 10Vz 5Watt zener could conduct 5/10 = 500mA

#### Artbuc

Joined Jun 8, 2014
16
hi A,
The zener is always used in serial connection with a regular diode.

The zener will in effect be wired in its zener working forward state, any reverse current will be very small, equal to the reverse current of the diode.

E

EDIT: the working zener diode current can be calculated from its rated wattage and zener voltage

example: a 10Vz 5Watt zener could conduct 5/10 = 500mA
Thanks. Yes, that is what I meant. The zener will be forward biased when dissipating the de-energized coil. It is this initial "de-energizing" current I am concerned about. Looks like I got the wrong zener. It is rated for 1.3 watts which allows only 65 mA. I assume that means my relay will fry it when it de-energizes? I assume the initial current of 170mS will last for just a fraction of a second but they may be enough to kill the zener?

Last edited:

#### t_n_k

Joined Mar 6, 2009
5,447
This question touches on the issue of device non-repetitive ratings.

Consider a BZX79-20V zener diode with a nominal rating of 500mW.

This would suggest an absolute maximum continuous current rating of 25mA. However, the same diode can carry a current of 1.5A for 100usec without damage - provided there is sufficient time between current pulses to allow the junction temperature to return to near ambient [notionally 25 C] condition after any transient.

So one needs to know the highest repetition rate of the current transient and the likely transient duration. Zener manufacturers will publish transient rating curves which will allow one to ensure the selected diode meets the anticipated circuit parameters.

In the case of relay coil suppression, you would need to know or "guesstimate" the coil inductance profile during the relay release phase to be able to predict the anticipated transient current behavior. Presumably you already know the coil resistance, which would also have some influence on the transient behavior.

#### ericgibbs

Joined Jan 29, 2010
9,570
Thanks. Yes, that is what I meant. The zener will be forward biased when dissipating the de-energized coil. It is this initial "de-energizing" current I am concerned about. Looks like I got the wrong zener. It is rated for 1.3 watts which allows only 65 mA. I assume that means my relay will fry it when it de-energizes? I assume the initial current of 170mS will last for just a fraction of a second but they may be enough to kill the zener?
hi,
Do you have a part number for your zener or datasheet you could post, we could check the spec'

EDIT: for the small extra cost and peace of mind, as its a one off app, I would fit a 5W zener, that will cover the worst case.

E

#### Artbuc

Joined Jun 8, 2014
16
hi,
Do you have a part number for your zener or datasheet you could post, we could check the spec'

EDIT: for the small extra cost and peace of mind, as its a one off app, I would fit a 5W zener, that will cover the worst case.

E
I got the 78-ZPY20. I agree with the 5W zener. Crap, the postage will be 10x the cost of the zener.

http://www.mouser.com/ds/2/427/zpy3v9-108163.pdf

#### alfacliff

Joined Dec 13, 2013
2,458
using a zener to snub an inductor isnt rally a good idea. the snubber diode is to absorb the reverse voltage as the coil field colapses. for this, the diode is connected oposite to the polarity of the power applied to the coil. a zenner connected that way would limit the voltage applied to the coil.

#### Artbuc

Joined Jun 8, 2014
16
using a zener to snub an inductor isnt rally a good idea. the snubber diode is to absorb the reverse voltage as the coil field colapses. for this, the diode is connected oposite to the polarity of the power applied to the coil. a zenner connected that way would limit the voltage applied to the coil.
I would configure the shunt as shown in the TE document linked in my first post. Unless I am missing something, it would not have any effect on the voltage applied to coil. It would ensure 20 volts of "back" emf when the coil de-energizes to extend relay contact life. Since Vce max for my transistor is 60 volts, all should be well. Of course it is entirely possible that I am misunderstanding the concept.

#### Alec_t

Joined Sep 17, 2013
10,916
Can't you just use a regular silicon diode (e.g. 1N400x), rather than a zener?

#### ericgibbs

Joined Jan 29, 2010
9,570
using a zener to snub an inductor isnt rally a good idea. the snubber diode is to absorb the reverse voltage as the coil field colapses. for this, the diode is connected oposite to the polarity of the power applied to the coil. a zenner connected that way would limit the voltage applied to the coil.
This image from the OP's pdf is the usual way, diode and zener, it gives a faster relay release than a single diode.

E

#### Attachments

• 42.7 KB Views: 54

#### Alec_t

Joined Sep 17, 2013
10,916
it gives a faster relay release than a single diode.
True, but I saw no indication that release time was an issue here.

#### Artbuc

Joined Jun 8, 2014
16
Can't you just use a regular silicon diode (e.g. 1N400x), rather than a zener?
Yes, that is what I am doing now and what you typically see when you google relay flywheel diodes. I just read the TE document and thought I would get fancy. Probably not very important as the first relay lasted many years without any diode protection and so did the transistor. I say many years because I cut apart the old relay and saw it did contain some short of shunt but it had burned away.

#### ericgibbs

Joined Jan 29, 2010
9,570
True, but I saw no indication that release time was an issue here.
hi alec,
I would agree, its just because the OP is quoting from an earlier post pdf which states a diode/zener combination is faster, which the OP may consider necessary.?

E

#### Artbuc

Joined Jun 8, 2014
16
hi alec,
I would agree, its just because the OP is quoting from an earlier post pdf which states a diode/zener combination is faster, which the OP may consider necessary.?

E
Actually, I thought the purpose of the zener was to reduce arcing and extend relay contact life?

#### ericgibbs

Joined Jan 29, 2010
9,570
hi,
When the current flowing in an inductor is interrupted, ie: the relay coil a back EMF is generated.

The value of this EMF is calculated using Lenz's Law EMF = - L di/dt.
Which basically means the faster the given current flowing in the inductor is interrupted the higher the back EMF.
If this EMF is not limited in some way this very high EMF will possibly damage the driving transistor as well as radiating a strong EMF field from the inductor which can cause problems in other parts of the circuit.
To clamp/limit the EMF peak its possible to use a series resistor and capacitor across the inductor or use a clamp diode.

The diode conducts when the EMF exceeds the supply voltage + 0.7V.
The downside is that relay release time is increased due to the current flowing around the diode and inductor circuit.

In some circuits its important that the release time of the relay or solenoid is not increased by the clamping, so a zener and diode combination is used.
The back EMF will exceed the supply voltage by the zener voltage + the diode voltage. Usually a zener and diode clamp is less effective in reducing radiated EMF. Also the driving transistor must be chosen that the Vce maximum voltage is not exceeded.

E