Author Topic: Regulator dies and takes whole circuit down with him  (Read 2463 times)

0 Members and 1 Guest are viewing this topic.

Offline evansgTopic starter

  • Contributor
  • Posts: 14
  • Country: gr
Regulator dies and takes whole circuit down with him
« on: October 25, 2019, 11:28:39 am »
I have seen 3-4 cases so far in similar circuit designs that has the following issue:

For a (strange) reason 5V regulator (I have seen that on LDL1117 from STMicroelectronics) explodes (with a large hole on it) and takes down the whole 5V line as it passes through the 12V DC input. Of course all 5V tolerant components are fried out!

Texas instruments same regulator (LM1117) or similar 3.3V is not affected. Both ST and TI exists in the same circuit. ST dies and destroys whole 5V line. TI 3.3V works flawlessly.

Any idea what can cause that? ESD? I get reports that this  has happened always when the user plugs the power inside the socket. (voltage spike?) and always using a 12-13.8V battery (lifepo or lead acid). Unfortunately I cannot reproduce that Tried many times but coudn't cause a damage to the regulator
Check attached photo of the damage from a board that was returned back.
Thanks

« Last Edit: October 25, 2019, 01:13:28 pm by evansg »
 

Offline SiliconWizard

  • Super Contributor
  • ***
  • Posts: 14899
  • Country: fr
Re: Regulator dies and takes whole circuit down with him
« Reply #1 on: October 25, 2019, 01:59:42 pm »
Would be interesting to see your schematic, at least just the power supply section.

The LDL1117 and LM1117 have different characteristics on a number of points, but their rated input voltage range is the same (20V absolute max).
(The LDL1117 has a much lower dropout, higher max output current, higher PSRR...)
The thing that doesn't seem documented is the ESD rating, whereas it is for the LM1117. I don't know if it's just an omission, or if it was never tested...

ESD is something to consider here. Is there any kind of protection against that on your board?

As a side note, if a LiFePo4 battery is allowed, a typical such "12V" battery will be at 14.4V when fully charged. Still under the max input voltage for your regulator(s), but higher than your rated 12-13.8V... (Probably doesn't matter much here, but just thought it could be worth mentioning.)

One other thing that comes to mind is the possible reverse current when the power source (battery/power switch?) is removed. I haven't read the datasheets thoroughly, but I don't know if those regulators are fully protected against it. You may need to add a diode between IN and OUT for good measure.
 

Offline DBecker

  • Frequent Contributor
  • **
  • Posts: 326
  • Country: us
Re: Regulator dies and takes whole circuit down with him
« Reply #2 on: October 25, 2019, 02:25:09 pm »
Is this an LD1117 or a LDL1117 from ST?
The LDL part has a MOSFET pass transistor, while the LM1117/LD1117 has a NPN.  The MOSFET will conduct in the reverse direction, dumping the capacitor charge through the part if the input jack is shorted.

How certain are you that the part isn't counterfeit?  There are many pin compatible regulators that cost about a penny each in Asia, while a real ST part will cost far more.  Some of the low-end *1117 parts have a datasheet max voltage of 12V or under.  I've blown them out with a 9V supply.  They are really intended for easy duty e.g. regulating 5V down to 3.3V at well under their max current rating.
 

Offline free_electron

  • Super Contributor
  • ***
  • Posts: 8520
  • Country: us
    • SiliconValleyGarage
Re: Regulator dies and takes whole circuit down with him
« Reply #3 on: October 25, 2019, 02:47:44 pm »
Not visible in the picture but .. where are your local anti-oscillation caps ?

paragraph 6.2 : The LDL1117 requires external capacitors to assure the regulator control loop stability.
Any good quality ceramic capacitor can be used but, the X5R and the X7R are suggested
since they guarantee a very stable combination of capacitance and ESR over the
temperature range. The input/output capacitors should be placed as close as possible to
the relative pins. The LDL1117 requires an input capacitor with a minimum value of 1 μF.
This capacitor must be placed as close as possible to the input pin of the device and
returned to a clean analog ground. The control loop of the LDL1117 is designed to work
with an output ceramic capacitor. Other type of capacitors may be used, as long as they
meet the requirements of minimum capacitance and equivalent series resistance (ESR), as
shown in Figure 20: "Stability plan (VOUT = 5 V)" and Figure 21: "Stability plan (VOUT =
1.2 V)".
To assure stability, the output capacitor must maintain its ESR and capacitance in the
stable region, over the full operating temperature range.
The suggested combination of 1 μF input and 4.7 μF output capacitors offers a good
compromise among the stability of the regulator, optimum transient response and total PCB
area occupation.


This is a common mistake when designing with linear regulators. ALWAYS read the datasheet carefully. Some regulators do not need them , some do not want them , some must have them ! If you don't put them on a regulator that needs them , or do put them on a regulator that doesn't want them , or the ESR is too low ( yup, there is such a thing ! everybody is gunning for the lowest esr, but that too is completely wrong. ) the regulator will become unstable and can self destruct.

Simply assuming that all 1117 and 'compatibles' out there behave the same way is wrong. Check your datasheet carefully when selecting alternates.
« Last Edit: October 25, 2019, 02:50:55 pm by free_electron »
Professional Electron Wrangler.
Any comments, or points of view expressed, are my own and not endorsed , induced or compensated by my employer(s).
 

Offline evansgTopic starter

  • Contributor
  • Posts: 14
  • Country: gr
Re: Regulator dies and takes whole circuit down with him
« Reply #4 on: October 25, 2019, 02:53:03 pm »
ST Regulators are always sourced from Mouser (all components were bought from Mouser and Farnell) and moreover the in/out capacitors are Murata X7R 10uF. There is a N mosfet /zener diode at the input which does efficiently  the reverse polarity protection.

Regarding the "reverse current when the power source (battery/power switch?) is removed".  I do not think so as this happens always (based on customer reports) when they plug the socket in the DC IN.
« Last Edit: October 25, 2019, 02:55:43 pm by evansg »
 

Online Kleinstein

  • Super Contributor
  • ***
  • Posts: 14486
  • Country: de
Re: Regulator dies and takes whole circuit down with him
« Reply #5 on: October 25, 2019, 02:58:32 pm »
To get so much heat to get the case to "explode" / lift of the top it usually needs quite some power and current. This often is caused by a latch-up and a powerful supply at the input. ESD is one possible trigger - it was a quite common problem in the old days with computers and printers, when connecting the cable while still powered.

If an MOSFET is used for reverse polarity, this one could also do some current limiting (use a small one, possibly a series resistor) to limit the possible damage from a latch-up.
Starting from 12 V should have plenty of spare voltage.

 

Offline evansgTopic starter

  • Contributor
  • Posts: 14
  • Country: gr
Re: Regulator dies and takes whole circuit down with him
« Reply #6 on: October 25, 2019, 03:54:49 pm »
There are two other TI regulators  (a 3.3V LDO and a 5V/8A switching regulator in the same board and receive 12V input f(passing from same n mosfet). These two never failed. They work absolutely fine in the board that the 5V ST Regulator line was fully damaged. I doubt it is a current issue.

The thing is that regulator has the magic smoke and it passes 12V to its output.. leading to catastrophic results for any other IC.. (It is still passes through the volts to its output... Seems like it is fully melted and bridged inside.. (This thing happened 3 times with exactly the same effect)
« Last Edit: October 25, 2019, 04:16:13 pm by evansg »
 

Offline iMo

  • Super Contributor
  • ***
  • Posts: 4897
  • Country: vc
Re: Regulator dies and takes whole circuit down with him
« Reply #7 on: October 25, 2019, 04:54:42 pm »
Also doublecheck the 100uF capacitor at the 1117's output. While the cap shorts itself the 1117 smokes, and shorts its inp-outp, thus you get 12V into the 5V rail..
 

Offline free_electron

  • Super Contributor
  • ***
  • Posts: 8520
  • Country: us
    • SiliconValleyGarage
Re: Regulator dies and takes whole circuit down with him
« Reply #8 on: October 25, 2019, 04:56:55 pm »
ST Regulators are always sourced from Mouser (all components were bought from Mouser and Farnell) and moreover the in/out capacitors are Murata X7R 10uF. There is a N mosfet /zener diode at the input which does efficiently  the reverse polarity protection.

Regarding the "reverse current when the power source (battery/power switch?) is removed".  I do not think so as this happens always (based on customer reports) when they plug the socket in the DC IN.
too much !. read the appnote.
Professional Electron Wrangler.
Any comments, or points of view expressed, are my own and not endorsed , induced or compensated by my employer(s).
 

Offline evansgTopic starter

  • Contributor
  • Posts: 14
  • Country: gr
Re: Regulator dies and takes whole circuit down with him
« Reply #9 on: October 25, 2019, 04:58:13 pm »
Thanks for this info. Nope, I replaced every time the regulator without changing any capacitor and it was back again working as expected.
I really consider ESD spike or a grounding issue as the cause of this problem. The annoying this is that I cannot reproduce it in the lab... I tried multiple times without a success... to melt down the regulator ...
 

Offline iMo

  • Super Contributor
  • ***
  • Posts: 4897
  • Country: vc
Re: Regulator dies and takes whole circuit down with him
« Reply #10 on: October 25, 2019, 04:58:47 pm »
On that picture above I can see 100uF at the 1117 output..
 

Offline evansgTopic starter

  • Contributor
  • Posts: 14
  • Country: gr
Re: Regulator dies and takes whole circuit down with him
« Reply #11 on: October 25, 2019, 05:09:37 pm »
I see, got the point now. Maybe this capacitor causes problems as the datasheet refers to 4.7uF ceramic. (as already mentioned here)
I'll check to remove this capacitor and use another lower one.
Thanks for the all the help guys! Appreciate it.
 

Offline capt bullshot

  • Super Contributor
  • ***
  • Posts: 3033
  • Country: de
    • Mostly useless stuff, but nice to have: wunderkis.de
Re: Regulator dies and takes whole circuit down with him
« Reply #12 on: October 25, 2019, 07:02:29 pm »
I do not think so as this happens always (based on customer reports) when they plug the socket in the DC IN.

This is a valuable hint to the (possible) root issue:
Plugging the DC IN while the external power brick / battery is already turned on (how would one turn off a battery anyway?) can create an overvoltage spike on your input 12V rail. This spike can cause the regulator to internally break down and you know the rest of the story. There's an application note by former Linear Technology addressing this issue - https://www.analog.com/media/en/technical-documentation/application-notes/an88f.pdf

TLDR:
Add an electrolytic capacitor or any other suitable means to dampen the ringing (in parallel to the DC jack) and your issue will be gone.
« Last Edit: October 25, 2019, 07:04:15 pm by capt bullshot »
Safety devices hinder evolution
 
The following users thanked this post: Siwastaja, Ian.M

Offline DBecker

  • Frequent Contributor
  • **
  • Posts: 326
  • Country: us
Re: Regulator dies and takes whole circuit down with him
« Reply #13 on: October 25, 2019, 07:28:42 pm »
Characterizing the damage as "explodes" suggests that it is a rapid event.  Oscillations would lead to slower overheating which chars the package and board.  A shorted package might later overheat, but a shattered package with minimal charring suggests an over-voltage event.

That regulator should have a bipolar pass transistor.  It should only take a few seconds to check that it doesn't show up as the parasitic diode of a MOSFET.

 

Offline NiHaoMike

  • Super Contributor
  • ***
  • Posts: 9155
  • Country: us
  • "Don't turn it on - Take it apart!"
    • Facebook Page
Re: Regulator dies and takes whole circuit down with him
« Reply #14 on: October 26, 2019, 03:54:00 am »
If you don't actually need the low dropout, add a series resistor on the input to make a filter and current limiter. Or use a standard linear regulator like the 7805.
Cryptocurrency has taught me to love math and at the same time be baffled by it.

Cryptocurrency lesson 0: Altcoins and Bitcoin are not the same thing.
 

Offline KE5FX

  • Super Contributor
  • ***
  • Posts: 1964
  • Country: us
    • KE5FX.COM
Re: Regulator dies and takes whole circuit down with him
« Reply #15 on: October 26, 2019, 04:12:16 am »
Make sure you have less capacitance downstream of the regulator than you do upstream.
 

Offline dmills

  • Super Contributor
  • ***
  • Posts: 2093
  • Country: gb
Re: Regulator dies and takes whole circuit down with him
« Reply #16 on: October 26, 2019, 04:56:15 pm »
Regulator explodes when hot plugged into a LiPo?
Are you by chance using long wires to the battery?

There is a known failure mode associated with ceramic input caps where the inductance of the supply wiring combines with the very low ESR of the input cap to cause the input node to ring up to as much as twice the steady state supply voltage during the startup transient, cure is some ESR for the input cap or a largeish conventional electrolytic across the input ceramic. You only see it when the supply is hot plugged into a stiff power supply.

Regards, Dan.

 

Offline evansgTopic starter

  • Contributor
  • Posts: 14
  • Country: gr
Re: Regulator dies and takes whole circuit down with him
« Reply #17 on: October 26, 2019, 09:40:27 pm »
> Are you by chance using long wires to the battery?

Possible, I don't but the customer cases I had probably they do..
I hope an extra electrolytic capacitor at the input will solve this.
Do you think that an ESD suppressor also like SMAJ15A-Q can help?
Thanks!
 

Online Zero999

  • Super Contributor
  • ***
  • Posts: 19678
  • Country: gb
  • 0999
Re: Regulator dies and takes whole circuit down with him
« Reply #18 on: October 27, 2019, 07:54:26 pm »
An ESD suppressor diode is a good idea. You could also try a fuse and over voltage crowbar on the output of the regulator to protect the rest of the circuit, if it fails, but it depends on whether you're going to be doing component level repairs or not. If you're just going to replace the whole board, due to a regulator failure, then you might as well not bother with the crowbar.
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf