5

I have the following circuit for providing regulated 5V from a 12V power supply: Power input through diode

I have several hundreds of PCB with this circuit, and they work fine. However recently I have had a few of these PCBs stop functioning. I have tracked the problem to diode D1.

When applying 12V to the input I have measured 1.8V on the cathode of the diode, and no appreciable input current (the power supply shows 0.000W output power), so I have discarded a short-circuit down the line as a possibility.

Now comes the interesting part. I tried slowly increasing the voltage to see if there would be any change on the output current. To my surprise, the diode started conducting at around 16V, and when I measured the voltage drop across the diode it was back to its normal 0.7v to 0.8V. After this the PCB was functioning normally again, and I was not able to reproduce the behavior. As matter of fact, now when applying a 12V input, it works as intended.

I have seen this behavior on at least 3 or 4 PCBs out several hundreds identical PCBs.

I am assuming this is a defective diode, but it is the first time I observe this mode of failure and I would like to get more information about it.

EDIT: I would like to add that I made sure that the connector and cable were not at fault here. Also the measurements are on the pins of the diode itself. Furthermore, when increasing the voltage on the power supply, I made sure I was not touching any of the cables or PCB because I suspected that could be one of the problems, but it is not apparently.

Endika
  • 53
  • 5
  • 2
    are you sure the diode isn't soldered in the wrong way around? – Marcus Müller Mar 21 '17 at 14:50
  • 1
    Yap, first thing I checked. Also as I mentioned, the same diode started conducting again after increasing the voltage, and the forward voltage returned to the normal 0.7V. – Endika Mar 21 '17 at 14:53
  • Take a closer look at J1, and its path to D1. Possible oxides? Or fatigued connections? – glen_geek Mar 21 '17 at 14:56
  • I am assuming a loose connection/solder joint that recovered at higher voltages. – 12Lappie Mar 21 '17 at 14:57
  • @glen_geek I measured all the voltages directly on the anode and cathode pins of the diode – Endika Mar 21 '17 at 15:01
  • Are you sure the power connector is the right way around. BtW. I would have wired that centre ground an 12V on pins 1 and 3, then it would not matter if it was backwards. I'm sure you checked the connector though. – Trevor_G Mar 21 '17 at 15:06
  • @Trevor placing the ground in the center is a good suggestion, thank you. And yes, the power connector is properly connected. As I mentioned in another comment, the voltages are measured directly at the pins of the diode, so I think wrong connections and "similar culprits", are not at fault here. – Endika Mar 21 '17 at 15:09
  • Assuming the soldering job was "professional" I'd suspect the connector first. Bad crimps/mating on either the power or the ground would be my first place to look. The fact that it came on as you raised the supply may be a red herring, since you may have inadvertently nudged the cables. – Trevor_G Mar 21 '17 at 15:11
  • @Trevor actually the first thing I have done is wiggling cables and connector, because my first suspect was indeed the connector. I also made sure not to touch any cable as I was rising the voltage on the power supply. – Endika Mar 21 '17 at 15:13
  • SO you are saying you measured 12V on the diode left side and 1.8 on the right side... and I assume it is not glowing in the dark... that probably means your ground is open. Which again.. could be the connector. – Trevor_G Mar 21 '17 at 15:14
  • Or you may just have bought a bad batch of diodes.... – Trevor_G Mar 21 '17 at 15:14
  • @Trevor yes it was not glowing in the dark! :P and the diode is not now blacker than it usually is. It might be a bad batch of diodes, but as I mentioned in the question, I have never seen this mode of failure, and I was asking for more information. – Endika Mar 21 '17 at 15:18
  • 1
    Yes, nor have I. Have you measured the voltage directly across the diode? – Trevor_G Mar 21 '17 at 15:19
  • 1
    What is the diode #? – Trevor_G Mar 21 '17 at 15:21
  • I also ask as Trevor did - what is part number of the diode? Are you *sure* it is the one (having marking on its body and sourced from trusted source)? – Anonymous Mar 21 '17 at 15:23
  • It is a S1D-13-F from Diodes Inc., but I am pretty sure our manufacturer has replaced it with a knock-off with no recognizable print. – Endika Mar 21 '17 at 15:27
  • Then in parallel to your electronics endeavor start business investigation why unrecognizable (and most probably wrong) part was used by the contractor. I hope you have related clauses in the contract. – Anonymous Mar 21 '17 at 15:32
  • I'm starting to suspect it has something to do with the initial forward voltage when you first apply power, which, because of the caps, will be the full supply voltage. What is the power source normally connected to that connector. I think it might be more prudent to put the diode on the other side of that bulk cap. – Trevor_G Mar 21 '17 at 15:33
  • Or better use a different reverse wiring protection method, such as a up diode between ground and 12V and a polyfuse. Assuming reverse polarity is a real possibility. Though if you are accepting a.c. adapters forget all that ;) – Trevor_G Mar 21 '17 at 15:40
  • If your company has design control and Mfg substitutes a part without notice, that's grounds for batch rejection, field notice/recall and supplier retribution, otherwise, it's a routine QA problem – Tony Stewart EE75 Mar 21 '17 at 15:48
  • @Trevor the S1D-13-F should be able to handle the full supply forward voltage and inrush current to the caps, if that is what you are suggesting. But you have a point, and this "unknown" diode might not be able to handle it. On the reverse polarity, the connector has a poka-yoke to avoid a wrong connection, but the reverse polarity can happen somewhere else during the assembly process unfortunately, but that is another battle. The polyfuse and diode solution would definitely work but the current solution is inexpensive and it has worked for a long time until the current batch. – Endika Mar 21 '17 at 15:51
  • 1
    "should" is the operative word here though. And you stated *"I have had a few of these PCBs stop functioning"* which suggests they were working for some time then failed. That suggests something is getting stressed, and the peak forward voltage and current would be by prime suspect. IF it IS overstressing it, then failure mode could be weird. As I say, moving the diode to the right of C1 would solve that issue. Easy enough rework to the cap connection too.. – Trevor_G Mar 21 '17 at 15:57
  • 1
    @Trevor That is actually a very valid point, I will consider it for the next batch. Thank you very much! – Endika Mar 21 '17 at 16:05
  • 2
    If substitute Diode is rated for <30A pk and ESR of 10uF cap and voltage source is < 0.4 Ohms (possible) then it will exceed Diode rating on turn on, resulting in exceeding absolute max rating. – Tony Stewart EE75 Mar 21 '17 at 16:20
  • 2
    @Trevor Don't forget to write that up in an answer! – W5VO Mar 21 '17 at 17:46

1 Answers1

2

I suspect it has something to do with the initial forward voltage when you first apply power, which, because of the caps, will be the full supply voltage.

Although the S1D-13-F should be able to handle the full supply forward voltage and inrush current to the caps, "should" is the operative word here though.

You stated "I have had a few of these PCBs stop functioning" which suggests they were working for some time and then failed. That suggests something is getting stressed, and the peak forward voltage and current would be by prime suspect. If it IS overstressing it, then failure mode could be weird.

Moving the diode to the right of C1 would solve that issue. Easy enough rework to the cap connection too.. Though it does leave the cap open to being reversed, it at least limits the threat.

Ultimately, if it is a real and present danger of reverse connection, a better method is to use a different technique, such as a up diode between ground and 12V and a polyfuse. That method gets rid of that unnecessary diode voltage drop in the line too.

BTW. I would have wired that connector with the centre ground and 12V on both pins 1 and 3. That way it will not matter if it gets plugged in backwards.

Trevor_G
  • 46,364
  • 8
  • 68
  • 151
  • 1
    You could also do [reverse-voltage protection using a P-Channel MOSFET](http://www.youtube.com/watch?v=IrB-FPcv1Dc) with minimal losses. – tangrs Mar 21 '17 at 18:57