Minimum current for zener diodes
Sonnich Jensen
What is the minimum current for a zener, such as BZX55/C5V6 to work
properly?
I tried using 500uA, but it does not seem good. I looks like less than
1mA is not enough, more like 2mA.
Sonnich
Globemaker
most zeners need more than 50nA. 1 uA is plenty
Grant
>Hi
>
>What is the minimum current for a zener, such as BZX55/C5V6 to work
>properly?
It's in the datasheet, did you miss it?
>I tried using 500uA, but it does not seem good. I looks like less than
>1mA is not enough, more like 2mA.
Bad zener? "not seem good" in what way? 500uA is enough, C5V6 should
get between 5.2 to 6V according to one datasheet I looked at.
>
>Sonnich
Bill Sloman
It's an interesting question, and the answer is complicated and varies
with the breakdown voltage.
Low voltage "zener" diodes break down by the Zener mechanism, while at
higher voltages, avalanche breakdown takes over
http://people.seas.harvard.edu/~jones/es154/lectures/lecture_2/breakdown/breakdown.html
Your 5.6V diodes sit right on the border between the two mechanisms
and both are likely to be active.
Avalanche breakdown tends to be noisy at low currents - the actual
areas in which the avalanches occurs tends to be tiny, so at "low"
currents (less than a milliamp) there may be only one electron in
transit at any one moment. The process by which this electron creates
new charge carriers is statistical, and if that one electron doesn't
succeed in generating at least one new charge carrier as as it makes
its way across the junction, the current stops for a bit until thermal
noise eventually promotes another charge carrier into the conduction
band, giving you are rather jagged voltage-versus time curve.
Search on "micro-plasmas" and note that glass-packaged zener diodes
tend to behave different when exposed to light (and have been known to
emit light when carrying current).
The Zener mechanism seems to be quieter.
There was a nice - tolerably long - thread on the subject - "Zener
Diode Oscillation" - here back in 1997. Winfield Hill and the late
Tony Williams posted a lot of interesting measurements and some
interesting references from the semiconductor physics literature.
--
Bill Sloman, Nijmegen
John Fields
<sonnic...@gmail.com> wrote:
---
For a Zener to "work properly" the current through it should be as
specified on the data sheet:
http://www.rectron.com/data_sheets/bzx55c-bs.pdf
--
JF
George Herold
There's some nice graphs in the General Semi data sheet here.
http://www.datasheetcatalog.com/datasheets_pdf/B/Z/X/5/BZX55-C5V6.shtml
The low voltage zeners have really mushy corner's, the dynamic
impedance is pretty poor if you just barely turn them 'on'.
At 2mA it looks like the C5V6 is still in the corner of the I-V. You
can run the higher voltage zeners at lower current... 'course as Bill
point out they are noiser. Why do you want to run at the minimum
current?
George H.
Globemaker
> On May 12, 5:14 am, Sonnich Jensen <sonnichjen...@gmail.com> wrote:
>
> > Hi
>
> > What is the minimum current for a zener, such as BZX55/C5V6 to work
> > properly?
> > I tried using 500uA, but it does not seem good. I looks like less than
> > 1mA is not enough, more like 2mA.
>
> > Sonnich
>
>
> The low voltage zeners have really mushy corner's, the dynamic
> impedance is pretty poor if you just barely turn them 'on'.
>
> At 2mA it looks like the C5V6 is still in the corner of the I-V. You
> can run the higher voltage zeners at lower current... 'course as Bill
> point out they are noiser. Why do you want to run at the minimum
> current?
>
> George H.
5 milliamps is on the spec sheet
Vladimir Vassilevsky
George Herold wrote:
> On May 12, 5:14 am, Sonnich Jensen <sonnichjen...@gmail.com> wrote:
>
>>Hi
>>
>>What is the minimum current for a zener, such as BZX55/C5V6 to work
>>properly?
>>I tried using 500uA, but it does not seem good. I looks like less than
>>1mA is not enough, more like 2mA.
>
> There's some nice graphs in the General Semi data sheet here.
> http://www.datasheetcatalog.com/datasheets_pdf/B/Z/X/5/BZX55-C5V6.shtml
>
> The low voltage zeners have really mushy corner's, the dynamic
> impedance is pretty poor if you just barely turn them 'on'.
The TempCo of Zeners depends on the current, too. No wonders, TempCo is
minimal at the recommended current. TempCo is typically positive for
lower currents and negative for higher currents.
Vladimir Vassilevsky
DSP and Mixed Signal Design Consultant
http://www.abvolt.com
Nico Coesel
In my experience you'll need at least several mA (say about 4mA)
through a zener to get it operating at its specified voltage.
--
Failure does not prove something is impossible, failure simply
indicates you are not using the right tools...
nico@nctdevpuntnl (punt=.)
--------------------------------------------------------------
Jim Thompson
wrote:
>Sonnich Jensen <sonnic...@gmail.com> wrote:
>
>>Hi
>>
>>What is the minimum current for a zener, such as BZX55/C5V6 to work
>>properly?
>>I tried using 500uA, but it does not seem good. I looks like less than
>>1mA is not enough, more like 2mA.
>
>In my experience you'll need at least several mA (say about 4mA)
>through a zener to get it operating at its specified voltage.
We seem to have a "random rattling" issue here.
Did _anyone_ pay attention to Field's post showing that the I VALUE is
specified on the data sheet?
And ALL "zener's" have an Ohmic term. To get really steep slope you
need to use an active device, such as a TL431.
...Jim Thompson
--
| James E.Thompson, CTO | mens |
| Analog Innovations, Inc. | et |
| Analog/Mixed-Signal ASIC's and Discrete Systems | manus |
| Phoenix, Arizona 85048 Skype: Contacts Only | |
| Voice:(480)460-2350 Fax: Available upon request | Brass Rat |
| E-mail Icon at http://www.analog-innovations.com | 1962 |
Obama is not the solution, Obama is the problem!
John Fields
---
Zener current only affects the _slope_ of the tempco in a minor way,
while Zener voltage affects both the sign _and_ the slope in a major
way.
Below about 5 volts the diode is in "Zener" mode and exhibits a
negative tempco, while above about 5 volts it's operating in
"avalanche" mode and exhibits a positive tempco.
Here are some excerpts from the 1980 Motorola Zener Diode Manual for
your perusal:
news:ck6os61esk28lnkh1...@4ax.com
--
JF
Wimpie
Hello Sonnich,
The minimum current depends on what is acceptable for you (so for you
this answer isn't useful).
If you want to go a step further below mA range, you may try a low
voltage avalanche diode like PLVA2600A. This series is specified at
current well below 1 mA.
Kind regards,
Wim
PA3DJS
www.tetech.nl
Vladimir Vassilevsky
John Fields wrote:
> On Thu, 12 May 2011 09:03:51 -0500, Vladimir Vassilevsky
> <nos...@nowhere.com> wrote:
>
>
>>
>>George Herold wrote:
>>
>>
>>>On May 12, 5:14 am, Sonnich Jensen <sonnichjen...@gmail.com> wrote:
>>>
>>>
>>>>Hi
>>>>
>>>>What is the minimum current for a zener, such as BZX55/C5V6 to work
>>>>properly?
>>>>I tried using 500uA, but it does not seem good. I looks like less than
>>>>1mA is not enough, more like 2mA.
>>
>>>There's some nice graphs in the General Semi data sheet here.
>>>http://www.datasheetcatalog.com/datasheets_pdf/B/Z/X/5/BZX55-C5V6.shtml
>>>
>>>The low voltage zeners have really mushy corner's, the dynamic
>>>impedance is pretty poor if you just barely turn them 'on'.
>>
>>The TempCo of Zeners depends on the current, too. No wonders, TempCo is
>>minimal at the recommended current. TempCo is typically positive for
>>lower currents and negative for higher currents.
>
>
> ---
> Zener current only affects the _slope_ of the tempco in a minor way,
> while Zener voltage affects both the sign _and_ the slope in a major
> way.
Yes, you are right.
TTman
"Nico Coesel" <ni...@puntnl.niks> wrote in message
news:4dcc04f5.4147738640@news.kpnplanet.nl...
> Sonnich Jensen <sonnic...@gmail.com> wrote:
>
>>Hi
>>
>>What is the minimum current for a zener, such as BZX55/C5V6 to work
>>properly?
>>I tried using 500uA, but it does not seem good. I looks like less than
>>1mA is not enough, more like 2mA.
>
> In my experience you'll need at least several mA (say about 4mA)
> through a zener to get it operating at its specified voltage.
LOL, I had a board today where a 220K res had been fitted instead of 220R.
Instead of +15V, I got -1.5V.( +/- 18V5 to +/- 15V via zeners)
Michael A. Terrell
You stated "most zeners need more than 50nA. 1 uA is plenty" in a
previous message.
--
You can't fix stupid. You can't even put a Band-Aid™ on it, because it's
Teflon coated.
Sonnich Jensen
> On May 12, 5:14 am, Sonnich Jensen <sonnichjen...@gmail.com> wrote:
>
> > Hi
>
> > What is the minimum current for a zener, such as BZX55/C5V6 to work
> > properly?
> > I tried using 500uA, but it does not seem good. I looks like less than
> > 1mA is not enough, more like 2mA.
>
> > Sonnich
>
>
> The low voltage zeners have really mushy corner's, the dynamic
> impedance is pretty poor if you just barely turn them 'on'.
>
> At 2mA it looks like the C5V6 is still in the corner of the I-V. You
> can run the higher voltage zeners at lower current... 'course as Bill
> point out they are noiser. Why do you want to run at the minimum
> current?
I run on battery power, and need to limit my current. I read below,
that TL431 is another option - I will give that some toughts, as that
might be the solution, more exact and only slightly more expensive.
If it takes 5mA, then the question is - is a 78L05 or LM317LZ better?
I need the 5V for powering a panelmeter from a 12V battery. It takes
1mA. The LM317LZ has a minimum load of 3-5mA, which will force me to
fry it off.
The loss will be the same, but I will get a more stable power.
Sonnich
>
> George H.
Sonnich Jensen
Checked it out. Seems like the 79L05 (for my need) will work with a
load of 1mA, fine for me, and I-bias @ max 6mA. I will check hot much
is it really it tomorrow. Otherwise LM317/337 will work with a minimum
load of ~2mA, app the same. Considering the resistors needed, it will
work as well.
Which brings me to protection - overvoltage protection (the charger
has fried 3 ICs as of now)(even that it can be fixed, protection is
always a good thing).
I was thinking of doing this by a transistor, zener and resistor,
which would work. But an 78L12 or 15 can do the same, and it should
let the voltage through and still work.
Using a 78L15 and 79L05 would do it, it is cheap, and require fewer
components, and the power used is the same (or less)
How much is there to discuss?
Sonnich
TTman
I run on battery power, and need to limit my current. I read below,
that TL431 is another option - I will give that some toughts, as that
might be the solution, more exact and only slightly more expensive.
If it takes 5mA, then the question is - is a 78L05 or LM317LZ better?
I need the 5V for powering a panelmeter from a 12V battery. It takes
1mA. The LM317LZ has a minimum load of 3-5mA, which will force me to
fry it off.
The loss will be the same, but I will get a more stable power.
Sonnich
>
> George H.
Plenty of 5V LDO regulators that only take 2-3 microamps.... take a look at
Microchip.com
TTman
I run on battery power, and need to limit my current. I read below,
that TL431 is another option - I will give that some toughts, as that
might be the solution, more exact and only slightly more expensive.
If it takes 5mA, then the question is - is a 78L05 or LM317LZ better?
I need the 5V for powering a panelmeter from a 12V battery. It takes
1mA. The LM317LZ has a minimum load of 3-5mA, which will force me to
fry it off.
The loss will be the same, but I will get a more stable power.
Sonnich
>
> George H.
http://ww1.microchip.com/downloads/en/DeviceDoc/22008E.pdf
Bill Sloman
If you want a low current voltage reference, don't use a zener diode.
Miinimal goggling gave me this
http://www.national.com/mpf/LM/LM4121.html#Overview
and Analog Devices have got parts with supply currents down to 12uA
http://www.analog.com/ps/psthandler.aspx?pstid=10201&la=en
The ADR293 offers 5V at 15uA
http://www.analog.com/static/imported-files/data_sheets/ADR293.pdf
I've not used any of these parts, so I don't know what the downsides
are. but there does seem to be a lot of stuff available that would do
your job better than a zener diode.
--
Bill Sloman, Nijmegen
Jim Thompson
wrote:
The quiescent current is quite low, plus there's a CMOS version, but I
can't remember the part number off the top of my head :-(
Peter Bennett
<sonnic...@gmail.com> wrote:
>If it takes 5mA, then the question is - is a 78L05 or LM317LZ better?
>I need the 5V for powering a panelmeter from a 12V battery. It takes
>1mA. The LM317LZ has a minimum load of 3-5mA, which will force me to
>fry it off.
>The loss will be the same, but I will get a more stable power.
>
>Sonnich
>
The "minimum load 5 mA" spec on the data sheet just means that the
manufacturer doesn't guarantee that the device will meet all specs
below that current. It does not mean that the thing won't work
acceptably below that current.
I've tested many 78xx regulators with no load other than a DVM, and
have always found the output voltage to be close enough.
--
Peter Bennett, VE7CEI
peterbb4 (at) interchange.ubc.ca
GPS and NMEA info: http://vancouver-webpages.com/peter
Vancouver Power Squadron: http://vancouver.powersquadron.ca
John Larkin
wrote:
>Sonnich Jensen <sonnic...@gmail.com> wrote:
>
>>Hi
>>
>>What is the minimum current for a zener, such as BZX55/C5V6 to work
>>properly?
>>I tried using 500uA, but it does not seem good. I looks like less than
>>1mA is not enough, more like 2mA.
>
>In my experience you'll need at least several mA (say about 4mA)
>through a zener to get it operating at its specified voltage.
Higher voltage zeners, like 12v maybe, have sharp knees and very low
leakages below the zener voltage. You can do micropower regulators
with a small zener and an emitter follower, running the zener at a few
uA.
John
George Herold
>
> - Show quoted text -- Hide quoted text -
>
> - Show quoted text -
Oh, if you're trying to get the best efficiency from a battery, then
you've gotta look at switching power supplies. (It's not a subject, I
know much about.) If you want to make a 5V supply from zener's or
other linear IC's then you are throwing away more than 1/2 your
energy. (assuming a 12 V battery.)
George H.
Sergey Kubushyn
If you are on battery power you need a micropower reference. TI REF33xx is a
nice series reference (like 3-terminal regulator,) LM4041xx-xx is kinda like
TL431 but they work at microamps. There are others, these ones are what I
personally use for battery-powered designs. Also LM4040 for a series one and
others around LM404x for series and shunt (zener-like, or TL431) mode, fixed
and adjustable voltage.
---
******************************************************************
* KSI@home KOI8 Net < > The impossible we do immediately. *
* Las Vegas NV, USA < > Miracles require 24-hour notice. *
******************************************************************
John Fields
<sonnic...@gmail.com> wrote:
---
You might want to try using something like an LM4040-5, which is a
shunt voltage reference/regulator which has an output voltage which
remains [fairly] constant over an input current range of from 75µA to
15mA.
http://www.national.com/ds/LM/LM4040.pdf
Assuming a dead battery at 10V, 100µa into the regulator and 1mA into
your panel meter load, the circuit would look like this: (View in a
fixed pitch font)
+5
/
+10>--[R1]--+---------+
|K |
[LM4040-5] [5K]
| |
GND>--------+---------+
Since the drop across R1 will be 5V and the current through it will be
the load current plus the shunt regulator's quiescent current,
10V - 5V 5V
R1 = -------------- = --------------- ~ 4545 ohms
Ireg + Iload 1e-4A + 1e-3A
The closest available 1% value is 4530 ohms, so that'll increase the
current into the regulator from 100µA to about 4µA.
Assuming your battery's lead acid, fully charged it'll be at about
13.8 volts, so at that point the current into the regulator will be:
Vbat - Vreg
Ireg = ------------- - Iload
Rs
13.8V - 5V
= ------------ - 1e-3A
4530R
= 9.4e-6A = 946 microamperes
So, fully charged, the drain on the battery would be less than 2mA
and, at the recharge point, the drain would be about 1.1mA.
--
JF
John Fields
. /
. Oops...104µA
>
>Assuming your battery's lead acid, fully charged it'll be at about
>13.8 volts, so at that point the current into the regulator will be:
>
> Vbat - Vreg
> Ireg = ------------- - Iload
> Rs
>
> 13.8V - 5V
> = ------------ - 1e-3A
> 4530R
>
>
> = 9.4e-6A = 946 microamperes
. \
. Oops... 9.46e-4A
Fred Bloggs
>
> - Show quoted text -
Your application is non-critical. A simple little regulator sufficient
for your panel meter using junk box parts would be like so- uses a
current starved zener under the assumption that 10% IZT gets 80% VZT:
Please view in a fixed-width font such as
Courier.
.
.
.
.
. PN2222A
. BATT --+--------c e -----+-->
. | b |
. | | |
. '--[15K]---+ |
. | |
. c |
. PN2222A b -----'
. e
. |
. +
. 5.6Vz
. -
. |
. |
. -------------+------------
. |
. ---
. ///
.
.
Fred Bloggs
I believe the most effective conservation strategy would be a
momentary switch push-to-take reading...
Sonnich Jensen
> Sonnich Jensen <sonnichjen...@gmail.com> wrote:
> > On May 12, 4:36špm, George Herold <gher...@teachspin.com> wrote:
> >> On May 12, 5:14šam, Sonnich Jensen <sonnichjen...@gmail.com> wrote:
>
> >> > Hi
>
> >> > What is the minimum current for a zener, such as BZX55/C5V6 to work
> >> > properly?
> >> > I tried using 500uA, but it does not seem good. I looks like less than
> >> > 1mA is not enough, more like 2mA.
>
> >> > Sonnich
>
> >> There's some nice graphs in the General Semi data sheet here.http://www.datasheetcatalog.com/datasheets_pdf/B/Z/X/5/BZX55-C5V6.shtml
>
> >> The low voltage zeners have really mushy corner's, the dynamic
> >> impedance is pretty poor if you just barely turn them 'on'.
>
> >> can run the higher voltage zeners at lower current... 'course as Bill
> >> current?
>
> > I run on battery power, and need to limit my current. I read below,
> > that TL431 is another option - I will give that some toughts, as that
> > might be the solution, more exact and only slightly more expensive.
>
> > If it takes 5mA, then the question is - is a 78L05 or LM317LZ better?
> > I need the 5V for powering a panelmeter from a 12V battery. It takes
> > 1mA. The LM317LZ has a minimum load of 3-5mA, which will force me to
> > fry it off.
> > The loss will be the same, but I will get a more stable power.
>
> If you are on battery power you need a micropower reference. TI REF33xx is a
> nice series reference (like 3-terminal regulator,) LM4041xx-xx is kinda like
> TL431 but they work at microamps. There are others, these ones are what I
> personally use for battery-powered designs. Also LM4040 for a series one and
> others around LM404x for series and shunt (zener-like, or TL431) mode, fixed
> and adjustable voltage.
>
> ---
> ******************************************************************
> * KSI@home KOI8 Net < > The impossible we do immediately. *
> * Las Vegas NV, USA < > Miracles require 24-hour notice. *
> ******************************************************************
I will check that. I put the "final" solution together today, it is
here:
http://hot.ee/sonnich/meter3a.JPG
I found a 79L05 and 78L15 and just used them as of now. However,
current jumped from 4-5mA to 25. I will have to check the new meter,
and try without U4, the 78L12 (which is 15 as that is what I found in
my drawer). These drawer components are some 20 years old, that might
affect it too. Anyway, shops are closed until monday so tomorrow I
will look and think at improvements.
The U4 is simply for protection as mentioned before. I have seen
spikes kill my LM224 before :)
WBR
Sonnich
Fred Bloggs
>
> - Show quoted text -
I'm pretty sure that panel meter is not liking your signal common at
7V below what it thinks is power ground...
Sonnich Jensen
> > On May 12, 4:36špm, George Herold <gher...@teachspin.com> wrote:
> >> On May 12, 5:14šam, Sonnich Jensen <sonnichjen...@gmail.com> wrote:
>
> >> > Hi
>
> >> > What is the minimum current for a zener, such as BZX55/C5V6 to work
> >> > properly?
> >> > I tried using 500uA, but it does not seem good. I looks like less than
> >> > 1mA is not enough, more like 2mA.
>
> >> > Sonnich
>
> >> There's some nice graphs in the General Semi data sheet here.http://www.datasheetcatalog.com/datasheets_pdf/B/Z/X/5/BZX55-C5V6.shtml
>
> >> The low voltage zeners have really mushy corner's, the dynamic
> >> impedance is pretty poor if you just barely turn them 'on'.
>
> >> can run the higher voltage zeners at lower current... 'course as Bill
> >> current?
>
> > I run on battery power, and need to limit my current. I read below,
> > that TL431 is another option - I will give that some toughts, as that
> > might be the solution, more exact and only slightly more expensive.
>
> > If it takes 5mA, then the question is - is a 78L05 or LM317LZ better?
> > I need the 5V for powering a panelmeter from a 12V battery. It takes
> > 1mA. The LM317LZ has a minimum load of 3-5mA, which will force me to
> > fry it off.
> > The loss will be the same, but I will get a more stable power.
>
> If you are on battery power you need a micropower reference. TI REF33xx is a
> nice series reference (like 3-terminal regulator,) LM4041xx-xx is kinda like
> TL431 but they work at microamps. There are others, these ones are what I
> personally use for battery-powered designs. Also LM4040 for a series one and
> others around LM404x for series and shunt (zener-like, or TL431) mode, fixed
> and adjustable voltage.
I replied yesterday, but my posting got lost :(
I will take a look at this, as of now I have an 78L15 and 79L05, found
in my drawer. The meter takes 2mA, and the system with 79L05 and
without the 78L15(12) consumes 7,4mA. With it all it takes a bit more
than 10mA, in other words 50% is used for the regulators.
I found that the LM4040 and a 1,8K resistor with a stable 10V supply
will free up 2mA. The 78L15(12) is for stopping spikes and I need to
rethink that part. My charger has once fried ICs.
My schematic is here as of now: http://hot.ee/sonnich/meter3a.JPG
[now I will be off for a few days]
