Zener Diode 1N746A problems

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

Hello guys, I've recently bought a bunch of 3.3V zener diodes to protect my accelerometer inputs from an accidental surcharge (I use a 5v arduino for testing it). However upon connecting them, my voltage was trimmed to around 2V. I tested all the 5 I got in my batch and they show voltages from 1.92 to 2.08V (used 6v battery with 10k resistor). Those are 1N746A as shown on the package and were marked 3.3v both in the store and on the datasheet. I've tested the forward voltage to make sure I didn't connect them wrong somehow and it was 0.7v as expected...

Did something like that ever happen to you? Or am I missing something?

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

Handi3 wrote:
Hello guys, I've recently bought a bunch of 3.3V zener diodes to protect my accelerometer inputs from an accidental surcharge (I use a 5v arduino for testing it). However upon connecting them, my voltage was trimmed to around 2V. I tested all the 5 I got in my batch and they show voltages from 1.92 to 2.08V (used 6v battery with 10k resistor). Those are 1N746A as shown on the package and were marked 3.3v both in the store and on the datasheet. I've tested the forward voltage to make sure I didn't connect them wrong somehow and it was 0.7v as expected...

Did something like that ever happen to you? Or am I missing something?

Low voltage zener diodes have a very gradual turn on and those diodes will have a reverse breakdown voltage of about 2V @ 1mA. It is only when you get up to 20mA that they will have 3.3v breakdown. They will also have very high capacitance (several hundred pF). If you wish to clamp some signals I would recommend using one zener with biased with ~10mA (180 Ohm to +5V) and then use diodes such as 1N914 between the zener and each of the points you wish to protect. A single zener could protect multiple signals. Since it is not connected directly to the signals the capacitance won't matter.

kevin

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

Ah, thanks for the quick reply. I didn't know this behaviour of zener diodes. I think I will just trust my level converter and test the signals carefully before I hook up the accelerometer, since I don't have any switching diodes like the 1N914.

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

I've actually realized my level converter doesn't have enough outputs for SPI... So yeah, I've connected the 1n746 through a 240ohm resistor to the 5v output pin and it seems to produce a very clean 3.1V signal (<100ns rise time and about 100ns fall time).

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

is surcharge anything like surge in charge?

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

Handi3 wrote:
Ah, thanks for the quick reply. I didn't know this behaviour of zener diodes.

It reads on the component datasheet, you get the specified Zener voltage at the specified test current which is 20mA for all components in this series with different Zener voltages.

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

Quote:
If you wish to clamp some signals I would recommend using one zener with biased with ~10mA (180 Ohm to +5V) and then use diodes such as 1N914 between the zener and each of the points you wish to protect.

Sounds like a pretty cool trick. Have to try it sometime

hj

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

No trick. See what happens with a zener diode on an analog input then you'll see the reason for the alternate method.

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

Jepael wrote:
Handi3 wrote:
Ah, thanks for the quick reply. I didn't know this behaviour of zener diodes.

It reads on the component datasheet, you get the specified Zener voltage at the specified test current which is 20mA for all components in this series with different Zener voltages.


This is right, I should have payed better attention to the Voltage / Current curve.

ignoramus wrote:
is surcharge anything like surge in charge?

Nah. What I meant is just that if by accident I'd connect it somehow to 5v, I wouldn't burn the device.

ford2go wrote:
Quote:
If you wish to clamp some signals I would recommend using one zener with biased with ~10mA (180 Ohm to +5V) and then use diodes such as 1N914 between the zener and each of the points you wish to protect.

Sounds like a pretty cool trick. Have to try it sometime

hj


I guess, that require a 2.6-2.7V zener to account for the voltage drop of the diode?

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

Handi3 wrote:

ford2go wrote:
Quote:
If you wish to clamp some signals I would recommend using one zener with biased with ~10mA (180 Ohm to +5V) and then use diodes such as 1N914 between the zener and each of the points you wish to protect.

Sounds like a pretty cool trick. Have to try it sometime

hj


I guess, that require a 2.6-2.7V zener to account for the voltage drop of the diode?

That would depend entirely on how much current you want to tolerate before voltage rises too much and smoke happens.

The 1N914 curves seem to have something like 0.6V@1mA, 0.8V@10mA and 1.2V@100mA.

So to protect a 5V input, and to limit it to 5.3V@10mA, the voltage at zener diode needs to be 4.5V max. Now the zener diode needs to be carefully selected, because the zener diode has a given impedance at the zener current. For instance the 1N749A is a 4.3V zener at 20mA test current, at which point it has 22 ohms impedance. So add 10mA of current to 30mA and the voltage will rise 0.22V to 4.52V, where the voltage after diode is 5.32V. Not necessarily smoke yet.

But when the input has 1mA limiting current, it means there is 4.3+0.022=4.322V at zener, and the diode drop of 0.6 makes pin voltage limit to 4.922V when already 1mA is flowing in. This will load down the source and has to be taken into account.