Offline 150W LED drivers sometimes blow up at Mains Power up Test....why?

[treez]

10% of our LED drivers are blowing up on Mains Power ON test. Do you know why?
(Schematic and LTspice simulation attached.)

Please bare with me, this situation is a simple one to understand, but lengthy to explain….

We had very many 230VAC, 150W, Offline LED driver PCBs made to which we add (by hand wired Modification to PCB) two 100nF X2 capacitors to help them through EMC.

Aswell as this , we also (as a ‘hand’ modification to PCB) cut through some PCB copper which was connected to the 0.5W Buck converter’s switching node. We did this so it could pass common mode emissions testing. We cut the PCB copper with a Dremel Drill type thing.

Now, the problem is that since doing these two modifications, we now get about 10% or so of manufactured products blowing up at Mains Switch ON test.
What do you think is the cause of these blow-ups? The blow-ups never happened before these two modifications were done.

The blown-up units have the two 1206, 0R56 resistors blown open, the 1A mains input fuse is blown, and the 470uH inductor is badly blasted and goes open…and the downstream 220nF ceramic capacitor is often cracked.
We suspect that the blow-ups happen when the production test staff inadvertently switch the finished LED lamps ON at mains peak. This we think results in a large inrush current.

The thing is, the FET M15 should act somewhat as an inrush limiter. However, we believe that the act of cutting the PCB copper with the Dremel tool is actually ‘rubbing’ the PCB in such a way that the NFET M15 is getting induced ON (Vgs voltage getting induced to >4V). Then when we sometimes inadvertently apply mains at mains peak, there is no inrush limitation and the blow-up happens.

Do you agree that this could be our problem?

Schurter 3403.0166 Fuse datasheet:
https://www.farnell.com/datasheets/...96.333053046.1497723066-1679746183.1489787856

470uH (Wuerth 744045471) inductor datasheet..
https://katalog.we-online.com/pbs/datasheet/744045471.pdf

0R56, 1206 resistor (ERJ8BQFR56V) datasheet:
https://nl.farnell.com/panasonic-el...bqfr56v/resistor-1206-0-33w-1-0r56/dp/1717876

 

[betwixt]

Switching at mains peak obviously means more voltage is applied abruptly but Q5 should be able to react and clamp the G-S voltage of M15 in uS or less so it shouldn't allow long enough for any overcurrent damage to occur. It should limit the current at all times, including switch on, to about (Vb/0.28) = 2.3A.

R63 and R70 will be called upon to dissipate around 1.5W but they are only rated at 0.33W each so there is a risk of them being damaged if overcurrent lasts too long.

Brian.

 

[E-design]

I can think of these things you can try:

1) Bleed any gate charge when the mains supply is removed (bleed resistor in parallel with zener).
2) Pulse the current limit transistor with a spike that occurs when connecting power at the peak of the AC.

 

[treez]

woops...very sorry, i forgot the diode before the 220n capacitor...however, this doesnt affect the suspected diagnosis.
Corrected as attached.

 

[treez]

Switching at mains peak obviously means more voltage is applied abruptly but Q5 should be able to react and clamp the G-S voltage of M15 in uS or less so it shouldn't allow long enough for any overcurrent damage to occur. It should limit the current at all times, including switch on, to about (Vb/0.28) = 2.3A.

Thanks, yes i see what you mean.
It now makes me think that the FET M15 has actually been damaged by static of the dremelling and has just gone to being always ON. (in other words the BJT cant turn it off because its gone short from Drain to Source.

Newer ltspice simulation attached shows that an intiial induced voltage to the fet gate probably wouldnt do anything, due to BJT , as Betwixt said.

 

[betwixt]

That should be easy to measure on a dead board. Is it possible a voltage exists between the Dremmel and the board while cutting the copper?

Does your circuit behave well if the input voltage is lower? I mean, is it possible the damage occurs because the current limiter DID work properly but the resulting lower input voltage to your SMPS made it do something silly?

I've been thinking about inrush limiters and soft-start circuits at high voltages and have an idea but it will take me a while to see if it's feasible. I have in mind that if soft-start is an option it may be possible to use a similar FET circuit but ramp the gate voltage up over a short period, only a few AC cycles so it behaves more like a NTC thermistor but at the same time 'reset' it almost immediately if the AC input ceases. I've done something similar in the past with a phase controlled triac, like an old style light dimmer circuit but automatically going from off to full brightness in a fraction of a second. That wouldn't work in your circuit but it may explain the principle.

Brian.

 

[treez]

Do you think that the attached surge current waveform would be likely to blow up the 1206, 0R56 resistor?
The power waveform is also shown.

Unfortunately the ERJ8BQFR56V datasheet doesn’t allure to its surge handling capability.

ERJ8BQFR56V datasheet (0R56,1206 resistor)
https://www.farnell.com/datasheets/...22.333053046.1497723066-1679746183.1489787856

 

[betwixt]

The graphs suggest the limiter isn't doing anything at all. It should limit the current to <2.5A but the graphs show almost 30A.

As this seems to be a simulation output rather than real measurement, try shorting the FET source and drain together and re-run the simulator to see if it's any different.

Brian.

 

[treez]

The graphs suggest the limiter isn't doing anything at all. It should limit the current to <2.5A but the graphs show almost 30A.

As this seems to be a simulation output rather than real measurement, try shorting the FET source and drain together and re-run the simulator to see if it's any different.

Thanks Betwixt, sorry you're well ahead of me, i forgot to say that i have indeed shorted D-S of the FET for those graphs......i am thinking along the lines that the FET M15 is getting electrostatic damage by the dremelling, such that it is permanenetly short from D-S.

- - - Updated - - -

LTspice sim is as attached, with D-S shorted

- - - Updated - - -

Also, in the real circuit, Q5 is a tiny little SOT323 NON (BC848BW)
BC848BW datasheet.
https://rohmfs.rohm.com/en/products/databook/datasheet/discrete/transistor/bipolar/bc848bw.pdf

 

[treez]

What i also notice is that the Schurter 3403.0166 1A fuse doesnt give any "time to break" data for currents above 10A....and i wonder if this is because it just breaks instantly if it gets more than 10A....(it does see more than 10A if FET M15 is short.)
Schurter 3403.0166 Fuse datasheet:
https://www.farnell.com/datasheets/...96.333053046.1497723066-1679746183.1489787856

- - - Updated - - -

...Also our 1206 resistors are not stated as being surge proof, (ERJ8BQ type)...........
**broken link removed**
....maybe the instantaneous power of 400W over 5us is killing them...

I notice that even the 1206 , SG73-RT resistor is surge rated but over any <10us interval the power goes to 1000W or more it will blow up, as its datasheet tells...
SG73-RT surge resistor
https://www.koaspeer.com/catimages/Products/SG73-RT/SG73-RT.pdf

We are using the ERJ8BQ resistor which isnt surge rated and therefore i am wondering if this is why its blowing when (if) the M15 FET is going short from D-S?

 

[betwixt]

All fuses have a "time to break". They go open circuit because the wire in them melts but there is always a period while it goes from ambient to liquid state.
Perhaps Schurter consider you shouldn't be using that kind of 1A fuse when 10A is expected. They do quote "A^2/S" figures for all their fuses.

I still suspect something less worrying than instantaneous power ratings, maybe the capacitance of the FET and components on the gate is high enough to delay the rise in Vgs for long enough for damage to occur. What concerns me most is why your design blows up at all, even if peak AC is applied. I can understand the perils of excessive inrush current and why avoiding it may improve reliability but you seem to have a more catastrophic failure, suggesting the FET & resistors were perhaps connected into a short circuit. Does the SMPS survive after a blow out?

I have just installed LTSpice XVII on my computer so I may be able to investigate further. I use Linux but it seems to run OK on a Windows 10 emulator.

Brian.

[edit]
an afterthought - what are the voltage ratings of the 470K resistors in the FETs gate feed?

 

[treez]

an afterthought - what are the voltage ratings of the 470K resistors in the FETs gate feed?

Thanks, the resistors Vishay Draloric RCV1206470KFKEA
https://www.farnell.com/datasheets/2049090.pdf

...that is , of course, presuming that the PCB assemblers bothered to use these ...since once mounted, there is no way of telling if its just a standard 1206 resistor.

- - - Updated - - -

Does the SMPS survive after a blow out?

I repaired a unit that had blown up by replaing the downstream 220nF, the fuse, the 0R56 resistors...but i didnt have a choke so i just put in a 10R power resistor in its place.
When i switched it back on (and i made sure i did it at mains peak), it worked fine, and the 0.5W SMPS worked fine too. The non-SMPS 150W led driver worked, though i shoudl really have checked the current waveform to see if it was working properly. But I got moved on to another job, and these blown PCBs are all piling up on the rest room coffee table as weve nowhere else to put them.

- - - Updated - - -

Thanks yes the i^2.t is useful. The attached inrush into the X2 capacitor is up to 36A but only lasts for 3us , so that is wel below the i^2.t of 2.8, so we shoudl be ok with that.

 

[treez]

Do you think the wuerth 744045471 470uH inductor can take the following (attached) surge current without blowing up?
This is the current surge it gets if the FET M15 has already failed short drain-source

744045471 470uH inductor datasheet
https://katalog.we-online.com/pbs/datasheet/744045471.pdf

 

[betwixt]

It's a very small inductor and your graph shows it passing as much as ~700mA when it is rated at 105mA but that wouldn't necessarily blow it up. Like resistors and fuses, they 'blow up' because of over heating so a time element as well as current is involved. It looks like the surge only lasts about 250uS which is probably too short for it to heat appreciably. I can't tell what the residual current through it is but I assume it is less than the 105mA rating.

What does happen though is the core saturates at the higher current, especially at 7x rating and instead of behaving like an inductor it starts looking more like a resistance. It's effectiveness as a filter is diminished but I would guess it is there to filter outbound SMPS noise which probably isn't even present (Buck converter start up time will be > 5.25ms) so the surge would be well over before its real purpose comes into play.

Given that you mentioned cracked 220nF capacitors I would think the inductor might be a victim rather than a cause of the problems. A bigger one might not blow up but it wouldn't fix the underlying problem. It would indicate there was a significant current path through the inductor though so maybe the Buck converter is causing the excess current at start up.

I'm not at a simulator right now but please do this experiment and try simulating: Add a capacitor (100nf) across D108 and a diode from the top of D108 to the junction of R66 and R67 with cathode towards the resistors. The diode needs to be rated > 500V PIV and NOT a Schottky device but the capacitor can be a low voltage type. It should delay the rise of gate voltage so the FET turns on slowly but still allow a reasonably fast discharge path through the new diode to discharge it when power is removed. Strictly speaking you could take the new diode to the top of R67 but I would feel happier with it at the bottom end, even if it introduced a very small risk of the limiter not fully working if the AC input is pulsed rapidly.

Brian.

 

[mtwieg]

The thing is, the FET M15 should act somewhat as an inrush limiter. However, we believe that the act of cutting the PCB copper with the Dremel tool is actually ‘rubbing’ the PCB in such a way that the NFET M15 is getting induced ON (Vgs voltage getting induced to >4V). Then when we sometimes inadvertently apply mains at mains peak, there is no inrush limitation and the blow-up happens.

I'm not sure what you mean by "rubbing" the PCB but I don't see any way for mechanical disturbances to affect the FET's electrical properties. It could be that some joints are cracking during the process though. If the failures really are related to your hacking, then your only hope of finding out how is by identifying a bad unit before it's destroyed, by devising a more clever QA test. Perhaps at first you supply them with a DC PSU whose output is slew rate limited.

I'm not clear on what the purpose of the FET+BJT current limit circuit is. Your DC link capacitors are so small that you should only expect very brief pulses of current. Is the circuit even fast enough to work on them? Have you measured what the circuit is doing in real life?

Consider that this circuitry may not be the primary point of failure. If something downstream fails and causes an increase in load current, M15 would see most of the stress. It would probably fail short, followed by the rectifier diodes or the fuse. On the bad units, have you verified that the rest of the circuitry is still working?