# Zener Minimum Current?

Discussion in 'Electronic Basics' started by Brian, Mar 1, 2005.

1. ### BrianGuest

In a simple voltage regulator circuit, using a resistor and zener diode. The
formula for the resistor in series with the zener diode is:

R = [Vin(max) - Vz] / [Iz(min) + IL(max)]

My question is, how do you calculate or find what the minimum zener current
should be? I don't find this in the data sheets for zener diodes.

Thanks,

Brian

2. ### Robert MonsenGuest

There is a 'test current' given in the datasheets. That's the best one
to use if you want the zener to be near it's rated value.

--
Regards,
Robert Monsen

"Your Highness, I have no need of this hypothesis."
- Pierre Laplace (1749-1827), to Napoleon,
on why his works on celestial mechanics make no mention of God.

3. ### BrianGuest

So in a worst case situation, if say the input voltage varied from 10
volts to 20 volts and the zener volts was 9 volts (with a load current of 20
milliamps), you would use the zener test current for the Iz(min)? I know,
there is a better way to do this than with a zener diode, but for
information sake I would like to know the right way to do this.
Thanks,

Brian

4. ### john jardineGuest

The minimum zener current is that current which you want to still flow
through the zener when a maximum load current is being taken.
Zeners are pretty crap at low currents anyway and work better with at least
1/2 ma through them at all times.
It's easiest just to make the minimum current about (say) 10% of the maximum
This way, the load can increase a bit without the zener voltage falling
over.
i.e. Resistor is = [Vin(max) - Vz] / [1.1 x IL(max)]
regards
john

5. ### John LarkinGuest

Resistor-zener regulators are numerically challenged. In your example,
the drop across the input resistor varies over an 11:1 range, and it
has to be able to supply the load on the low end, never mind a min
zener current. So the worst-case input current will be at least 11x
the max load current, or 220 mA. At 20 volts in, the zener dissipates
almost 2 watts and the resistor more. Efficiency is insane and things
will fry.

Oh, 1 mA is a good min zener current.

John

6. ### Robert MonsenGuest

I think your forumla is wrong.

You need to think about what's happening. A zener diode is a shunt
regulator. It will gobble up as much current as it can if the voltage
across it is above it's rated value. Thus, your resistor is there to
prevent a virtual short circuit between Vin and Vz.

One main problem with shunt regulators is that they don't work very well
if the input voltage drops below their design value (I guess this is
true of all regulators)

Because of this, if you want regulation, you want to make sure the
voltage is always above Vz, given your required load current. If the
current is maximum (IL + Iz(min)) then the voltage drop across the
resistor is R * (IL + Iz(min)). Since the zener must have at least it's
rated voltage across it to work, so the input voltage must be *at least*
Vin = Vz + R * (IL + Iz(min)).

What this means is that the formula should really call for Vin(min).

As an example, suppose your input voltage ranges between 10V to 20V, and
you have a 5.1V zener. It's Iz(min) is 10mA, and your load is 20mA.

R = (20 - 5.1)/(30mA) = 496 ohms.

However, if you use that with your 20mA load at 10V, then, even if the
zener stops drawing any current (which it will do), your resistor will
drop 9.93V! This only leaves you 0.07V. So, you aren't regulating.

However, if you use

R = (10 - 5.1)/(30mA) = 163.3 ohms

then at 20V, your zener will still be regulating (although the voltage
across your load will be a bit higher, since the zener voltage will rise
a bit with more current). It will suck up the excess current. If your
load is using 20mA, then it'll take in the extra 70mA.

Thus, using Vin(min) in the forumla, your regulator will be regulating
when input is between 10V and 20V.

One other parameter you need to complete your design of the zener
regulator is the power rating of the zener. You need to ensure that the
power rating of the resistor and zener are within their maximums. And
you need to consider temperature effects on the zener. However, you

Usually, using a linear regulator that has temperature compensation and
current limiting built in is a better bet. However, there are times when
a shunt regulator is the only way to go.

--
Regards,
Robert Monsen

"Your Highness, I have no need of this hypothesis."
- Pierre Laplace (1749-1827), to Napoleon,
on why his works on celestial mechanics make no mention of God.

7. ### ChrisGuest

Good morning, Brian. The minimum zener current is Iz(k), or the zener
knee current. It is specified on zener data sheets.

For instance, my Fairchild data sheet for the 1N4740 (10V zener) lists
an Iz(k) for the 1N4740 of 0.25mA. If I was using this zener, I would
need real justification to go below 1mA minimum current -- zeners just
don't work very well near the knee.

Good luck
Chris

9. ### JamieGuest

i normally tried to have the Zener at 50% rated current while the load
is on the circuit.
there should be a Knee current spec'ed some where, this is the lowest
suggested current to operate the diode in. I(k) or something like that.
i think it depends on who makes the docs, i have also seen it referred
to as valley current.

10. ### BrianGuest

You are right about the formula being wrong. I meant to put Vin(min).
After seeing all the different answers (which I really appreciate), I did
some more digging into data sheets for zener diodes. I did find one data
sheet for 5 watt zeners, which had curves for zener volts versus zener
currents. On these curves, it showed that there wasn't a lot of change in
zener voltage (for zener currents above 1 milliamps), on zener diodes from 6
volts and above. On zener diodes below 6 volts, zener current was very
important. On zener diodes below 6 volts, using the test current shown was
very important (which I found to be, the zener current required for it to
dissipate 1/4 of its' wattage rating).
Thanks,

Brian

11. ### peterkenGuest

One better and more stable way is to set the zener using a current source
since when Iz varies Vz also will vary a bit
For the best current in a reference....
Usually I aim for a zener 5V1 @ 10mA since these are most stable if set this
way

12. ### mikeGuest

A Zener is a poor choice for this application.
You've picked a set of numbers that are a challenge for most simple
regulators. IF you can tolerate a minimum of 2V or so between min in
and out, use a 3-terminal regulator. Or a more expensive LDO one that
will run on 1V differential.

You don't give clues to the application. There are some common issues
ignored by newbies.

Zeners have a resistive term.
Zeners have a temperature coefficient...sometimes significant.
Zeners below 5.1V have a very soft knee, often.
Input voltage can have ripple that puts the minimum trough voltage
well below what you read on a meter.
Sometimes you can reduce the min/max current ratio by using a PTC
resistor. An incandescent light bulb makes an interesting PTC.
Plot some V/I curves for 12V-28V light bulbs and see if that helps.
mike

--
Return address is VALID but some sites block emails
..
Wanted, PCMCIA SCSI Card for HP m820 CDRW.
FS 500MHz Tek DSOscilloscope TDS540 Make Offer
Wanted, 12.1" LCD for Gateway Solo 5300. Samsung LT121SU-121
Bunch of stuff For Sale and Wanted at the link below.
MAKE THE OBVIOUS CHANGES TO THE LINK
ht<removethis>tp://www.geocities.com/SiliconValley/Monitor/4710/

13. ### jsmithGuest

First of all, most low voltage 1/2W zeners have a test current of 20ma.
Above 10-12V the test current begins to decrease. The value is easily
checked out in most catalogs or on line. Since you know the B+ use ohms law
to select a resistor value that will allow the test current to flow through
the zener. Whammo, you've got it. To verify the stability, test your circuit
by varying the voltage across the desired range using a bench power supply.
Remember the rule, if you double the voltage across the resistor you have
selected, know that the current will also double, and the wattage will
increase by four times. Useful rule of thumb to always be aware of.