### Author Topic: Zener diode accuracy, not getting expected reference voltage.  (Read 4214 times)

0 Members and 1 Guest are viewing this topic.

#### David97

• Contributor
• Posts: 46
##### Zener diode accuracy, not getting expected reference voltage.
« on: August 21, 2016, 06:22:18 am »
Hi everyone. A very noobie question here.

Bought a bunch of zeners from ebay. Just to use for various projects.
http://www.ebay.com/itm/322146150045?_trksid=p2057872.m2749.l2649&ssPageName=STRK%3AMEBIDX%3AIT

I wanted to start by making a low voltage alarm for may car. When I'm charging rc batteries while flying, I dont want to drain my car battery flat so I cant go home. Idea was to use a zener as a reference voltage, and then scale the batteries voltage through a resistor divider. Use a opamp to compare the voltages, and when the battery voltage fell below the reference, a beeper would sound. Pretty simple.

So I started by wiring a 5.1V zener in reverse bais in series with a 10k resistor to see weather I could make a reference voltage. However using multiple 5.1V zeners I got 4.31, 4.14, 4.24, 4.35, 4.44, 4.39, 4.20, 4.13V None of them were close to the 5.1V that they claimed to be.

Looking more into it, I need to make sure that the zener is getting enough current to fully saturate it before making a stable supply. currentley on the bench, i'm just using a battery pack giving 7.9V. So in series with a 10k resistor gives...
i=v/r = (7.9-4.44)/10000 = 346microamp

Which seems plenty for most zeners. I dont have the datasheet, so cant be sure. But doing some reading tells be most zeners are saturated at 100microamp.

Wondering why I'm not getting a good reference voltage. Any ideas? The zeners are marked 5C1 5T

Any help is appreciated. Thanks.
« Last Edit: August 21, 2016, 06:25:45 am by David97 »

#### Ammar

• Regular Contributor
• Posts: 154
• Country:
##### Re: Zener diode accuracy, not getting expected reference voltage.
« Reply #1 on: August 21, 2016, 06:40:59 am »
Do you have a data sheet for them?

#### crazyguy

• Regular Contributor
• Posts: 97
• Country:
##### Re: Zener diode accuracy, not getting expected reference voltage.
« Reply #2 on: August 21, 2016, 06:44:22 am »

#### rob77

• Super Contributor
• Posts: 1848
• Country:
##### Re: Zener diode accuracy, not getting expected reference voltage.
« Reply #3 on: August 21, 2016, 07:38:52 am »
definitely increase the current to few milliamps... most zeners need at least 1mA and some need as high as 5-10mA to behave nicely

but i would suggest something different for your low voltage alarm, use a TL431 shunt regulator instead of a zener diode. the price is in the same ballpark but the results are order of magnitude better.

#### T3sl4co1l

• Super Contributor
• Posts: 14583
• Country:
• Expert, Analog Electronics, PCB Layout, EMC
##### Re: Zener diode accuracy, not getting expected reference voltage.
« Reply #4 on: August 21, 2016, 05:25:26 pm »
Note that you'll spend a lot of bias current this way, whether it's putting a zener in the sweet spot (where its noise is also manageably low) or powering a hungry TL431 (which will, however, deliver a nice and stable output voltage).

Comparators with internal reference are available, or you can use an external micropower reference device.  These use much less power (~uA) than a TL431 (~1mA) or TLV431 (~100uA), and are only somewhat noisier.

Tim
Seven Transistor Labs, LLC
Electronic design, from concept to prototype.
Bringing a project to life?  Send me a message!

• Super Contributor
• Posts: 4906
• Country:
• A qualified hobbyist ;)
##### Re: Zener diode accuracy, not getting expected reference voltage.
« Reply #5 on: August 21, 2016, 05:31:52 pm »
Most 0.4W and 0.5W Zener diodes need about 3mA for a stable voltage.

#### w2aew

• Super Contributor
• Posts: 1746
• Country:
• I usTa cuDnt speL enjinere, noW I aR wuN
##### Re: Zener diode accuracy, not getting expected reference voltage.
« Reply #6 on: August 22, 2016, 01:17:32 pm »
Note that you'll spend a lot of bias current this way, whether it's putting a zener in the sweet spot (where its noise is also manageably low) or powering a hungry TL431 (which will, however, deliver a nice and stable output voltage).

Comparators with internal reference are available, or you can use an external micropower reference device.  These use much less power (~uA) than a TL431 (~1mA) or TLV431 (~100uA), and are only somewhat noisier.

Tim

My favorite device with internal reference and opamp/comparator is the oldie-but-goodie LM10.  Very flexible device.
FAE for Tektronix
Technical Coordinator for the ARRL Northern NJ Section

#### David Hess

• Super Contributor
• Posts: 10515
• Country:
• DavidH
##### Re: Zener diode accuracy, not getting expected reference voltage.
« Reply #7 on: August 23, 2016, 05:28:46 am »
The LM10 is an amazing part.  Linear Technology even makes an uprated pin compatible version in the LT1635.  Robert Widlar wrote an application note which describes its operation.

#### Cliff Matthews

• Supporter
• Posts: 1800
• Country:
##### Re: Zener diode accuracy, not getting expected reference voltage.
« Reply #8 on: August 23, 2016, 11:53:30 pm »
The LM10 is an amazing part.  Linear Technology even makes an uprated pin compatible version in the LT1635.  Robert Widlar wrote an application note which describes its operation.

#### nuno

• Frequent Contributor
• Posts: 559
• Country:
##### Re: Zener diode accuracy, not getting expected reference voltage.
« Reply #9 on: August 24, 2016, 01:52:50 am »
I can add some real world measurements of some zener diodes (from the interesting quadrant, rotated to be more explicit)

And one from a 3.3V zener, which is... unusable (I don't know why do they even take the trouble of producing them); it's a
BZX85C3V3 (I have characterized one other different 3V3 model and it's also unusable)

#### T3sl4co1l

• Super Contributor
• Posts: 14583
• Country:
• Expert, Analog Electronics, PCB Layout, EMC
##### Re: Zener diode accuracy, not getting expected reference voltage.
« Reply #10 on: August 24, 2016, 01:41:55 pm »
Most zener series go down to 2.4V!  They're practically resistors...

This, by the way, is why you never see TVS diodes below 5V rating -- or, if you do, the peak surge voltage isn't any lower than 5V parts.

If you need a zener at 1.5-3V, consider an LED!  They're not well controlled, but they are fairly stable.  Some even have complementary or negative tempcos relative to a Vbe, so you can make a very stable current source/sink using a, I think, red or yellow LED.

Tim
Seven Transistor Labs, LLC
Electronic design, from concept to prototype.
Bringing a project to life?  Send me a message!

Smf