Why did this zener diode explode?

Thread Starter

fablau

Joined May 14, 2020
167
Hello everyone.

Please, help me to understand why D1 of the circuit below exploded when I increased the voltage to 32v. Everything was working perfectly at 22V, then I increased the voltage to 32v and D1 exploded. As you can see from the schematics, that diode drives nine N-Channel MOSFETS when they are activated through the corresponding optocouplers driven by an external controller connected to the Digital Inputs. My guess is that I have exceeded the breakdown voltage, but per specifications, it should be fine and I didn't expect that to happen.

Here are some details about the used components:

Here is the Zener diode I have used:

https://datasheet.lcsc.com/lcsc/1810122016_MDD-Microdiode-Electronics-BZT52C12_C173429.pdf

And here is the N-Channel Mosfet I have used:

https://datasheet.lcsc.com/lcsc/1811141141_HL-Haolin-Elec-5N04_C237240.pdf


The max voltage the circuit should be able to support is max 40v, even though I don't think ever go over 32v. The max current out of the MOSFETs shouldn't be over 0.5 amp each, but of course, they can be all turned on for a max total output of 4.5 amps (but I don't think that should be related to the Zener blowing up, please, correct me if I am wrong).

The circuit's purpose is to drive an array of electromagnets that can either attract (in which case P-channel MOSFETs are used) or repel (in which case N-channel MOSFETs are used instead).

Eager to know your thoughts, let me know if you need more information, and thank you in advance to anyone!

Schematic_Zener Diode Discussion_2021-06-12.png
 

Thread Starter

fablau

Joined May 14, 2020
167
Too much current => Too much heat => Violent outgassing with shrapnel.
Thank you for your reply. Ok, too much current where? As I said, the current draw of the output is not connected to the Zener. The Zener is connected to the MOSFETs gates. Unless you mean that the nine MOSFETs connected to the same Zener draw too much current at the gate level?
 

MrChips

Joined Oct 2, 2009
30,706
Maybe R22 is not 10KΩ or maybe there is a short on the board.
Check with an ohmmeter.
You can also check current flowing across the D1 points. The current should be no more than 5mA. The 500mW zener can handle 40mA.
 

ronsimpson

Joined Oct 7, 2019
2,985
The schematic is a sea if green wires that confuse me.
It looks like a opto-isolator pulls the gate of the MOSFETs to ground?
It looks like a 10K resistor pulls up the gate from (12 volts?).
I finely found the "32V" but can not follow it across the page. Where is the connection from 32V to 12V? How do you make 12V?
 

Papabravo

Joined Feb 24, 2006
21,158
I apologize for may lack of clarity

Too much current thorough the !@#$%^&*() zener diode => Too much heat => Violent outgassing with shrapnel.

If there had been too much current through another component it would have behaved in a similar way. There are many possible root causes from bad design to wiring errors to phases of the moon.
 
Last edited:

Thread Starter

fablau

Joined May 14, 2020
167
Thank you to everyone for your replies to my inquiry.

First of all, I apologize for the complex schematics, I should have simplified it for the sake of discussion. But I can confirm that all the connections and nets are correct, there are no shorts anywhere on the board even because as I previously stated, everything was working correctly at 20-22 volts. The problem appeared at over 30v.

Thank you @MrChips for your suggestion to measure the resistors, and I actually found something weird about the 100k resistor (R24 in the schematics above).

Have a look at this picture of the circuit board:

ExplodedDiode.jpg

As you can see, the exploded diode is right there marked with D1. The R24 nearby is supposed to be a 100k resistor, but when I measure it, it gives me 54k instead. All other resistors are correct, but both the R24 and R23 give a measurement of 54K whereas they should be both 100K.

Is it possible the PCB company used defective resistors? Your thoughts on that?
 

Thread Starter

fablau

Joined May 14, 2020
167
Ok, I have detached the resistor from the board and measured again, and this time it measured correctly 100K. Since I am kind of a newbie here, could you please explain why on board this resistor gave a different value whereas the other resistors appear to be correct even though they also are on board?
 

Thread Starter

fablau

Joined May 14, 2020
167
You where measuring something else in parallel.
Thank you for your explanation. That makes sense.

At this point, I will try to measure the current across the diode and see if that was the problem, then I'll report.

If you have any more ideas, please, let me know.

Thanks again.
 

Thread Starter

fablau

Joined May 14, 2020
167
Ok, I used a new circuit board exactly like the previous one because the one with the blown diode is damaged somewhere else (still to find out what exactly happened!), and I have measured the current across the same diode, and it never goes over 200mA. I have tested voltages up to 25v because I didn't want to blow up this other board as well, but I have noticed that the current across the diode doesn't change at all between 18v and 25v, so... I guess that even at 32v the current would be the same? And if that's a correct assumption, I guess the problem is somewhere else. During my tests, I have noticed that the MOSFETs become very hot, which makes me think that raising the voltage even further, up to 32v, could have made burn them in some way and cause a short somewhere that made the diode explode. Could that be possible? That also would explain why the first board is no longer working even after have replaced the diode (I'll check every single MOSFET to understand if they are still working or what else has been damaged).

Would be anyway weird that the MOSFET would suffer 32v with max 0.5 amp each as I mentioned earlier. Even if the current would go up to 1 amp each @ 32v, they should be still fine.

Here are the MOSFETs I am using:

P-Channel:
https://datasheet.lcsc.com/lcsc/1809291523_Wuxi-NCE-Power-Semiconductor-NCE40P05Y_C163738.pdf

N-Channel:
https://datasheet.lcsc.com/lcsc/1811141141_HL-Haolin-Elec-5N04_C237240.pdf

They all should support 40v and 5 amp max. Eager to know your thoughts on all this.

Thanks again.
 

Thread Starter

fablau

Joined May 14, 2020
167
Is the zenar in the correct orientation?

It might also pay to make a minimal circuit that displays the fault
Yes, the Zener is correct. Please, check the schematics and let me know if you see anything wrong there.

As for making a minimal circuit, do you mean as a schematics or an actual circuit?
 
Top