The TS appears to be describing a commercial product that they have some responsibility for.

From what I've read so far, the distinction between professional and amateur seems somewhat blurred in this instance.

Sometimes we can't force the TS to show us the detail circuit of commercial product, but since the rules is there, if we just suspect the topic including transformerless power supply, if we need to close the thread then we must be through the discussion.

 

Sometimes we can't force the TS to show us the detail circuit of commercial product, but since the rules is there, if we just suspect the topic including transformerless power supply, if we need to close the thread then we must be through the discussion.

It was my understanding that the TS described a device fed from mains via a dropper resistor.

 

I concur. This is a transformerless, straight from the wall socket, LED driver.
Diagnosing the failure mode for the TS is not the same as instructing a beginner how to make these, but this Thread is in violation of TOS.

 

I concur. This is a transformerless, straight from the wall socket, LED driver.
Diagnosing the failure mode for the TS is not the same as instructing a beginner how to make these, but this Thread is in violation of TOS.

For what its worth - IMO: the most likely cause is stressing the leads during assembly.

The suspicion that the customer is putting the 110V version on 270V merely highlights a newbie mistake of failing to take that eventuality into account during design.

 

For what its worth - IMO: the most likely cause is stressing the leads during assembly.

For what its worth - IMO: the most likely cause is voltage shattering. In 40+ years, I have only seen one TO-92 with a wiggly lead. One might suspect that the mechanically weak transistors were caught in the final inspection process, and so, the end used doesn't see them but, I worked QC in more than one factory. I didn't find wiggly transistors, I found shattered thermoplastic cause by the failure of other components.

This is experience talking. As usual, I might be wrong, but I am repeating myself because I doubt I'm wrong on this one.

 

For what its worth - IMO: the most likely cause is voltage shattering. In 40+ years, I have only seen one TO-92 with a wiggly lead. One might suspect that the mechanically weak transistors were caught in the final inspection process, and so, the end used doesn't see them but, I worked QC in more than one factory. I didn't find wiggly transistors, I found shattered thermoplastic cause by the failure of other components.

This is experience talking. As usual, I might be wrong, but I am repeating myself because I doubt I'm wrong on this one.

IME: a transistor usually blows a crater in the flattest face when that happens, the damage pictured is nowhere near the die, I'd bet money on it being a lead stress case fracture - I've even done it myself once or twice in the past.

 

Perhaps there is more evidence to consider. The lack of a circuit board photo precludes that. The cause of damage might even be apparent from the assembly method which we can not see.

 

It's like someone taking a picture of a piece of tire tread on the side of the road asking what caused the tire to fail.

 

IME: a transistor usually blows a crater in the flattest face when that happens, the damage pictured is nowhere near the die, I'd bet money on it being a lead stress case fracture - I've even done it myself once or twice in the past.

It seems your argument is supported by the fact that the failure is (mechanically) intermittent...

 

Perhaps there is more evidence to consider. The lack of a circuit board photo precludes that. The cause of damage might even be apparent from the assembly method which we can not see.

A patch of vapour-deposited silicon on the top side of the PCB would eliminate any doubt - otherwise I'm going with lead stress.

 

It seems your argument is supported by the fact that the failure is (mechanically) intermittent...

Phutted is phutted - intermittent still works...............sometimes.

 

Phutted is phutted - intermittent still works...............sometimes.

My referance was to the below quoted excerpt:

Further, the transistor is "intermittent" in that we can press on it (i.e. physically move the transistor very slightly) and the circuit will begin to work again

Point being, a catastrophic electrical failure would not likely brook functionality under any circumstance...

IOW: Relax -- I agree with you!

TTFN
HP

 

I think I have been too focused on what might cause mechanical damage to the plastic case and not enough on the intermittent operation, as if intermittent operation was peculiar that that one, single, transistor. Mechanical damage, by the dozens, in exactly one transistor? It's a stress fracture, either from physical forces or electrical forces. Intermittent with a wiggly leg in all of them? I'm gonna need my crystal ball for this one.

An arc blow-out kills almost all of them instantly, with the rare occurrence of a "leg to an external conductor" arc path. A physical stress with a missing chunk lets a lot more of them survive. A whole row of intermittent transistors with exactly one wiggly leg? Color me stumped.

 

I think I have been too focused on what might cause mechanical damage to the plastic case and not enough on the intermittent operation, as if intermittent operation was peculiar that that one, single, transistor. Mechanical damage, by the dozens, in exactly one transistor? It's a stress fracture, either from physical forces or electrical forces. Intermittent with a wiggly leg in all of them? I'm gonna need my crystal ball for this one.

An arc blow-out kills almost all of them instantly, with the rare occurrence of a "leg to an external conductor" arc path. A physical stress with a missing chunk lets a lot more of them survive. A whole row of intermittent transistors with exactly one wiggly leg? Color me stumped.

Perhaps the OP's got hold of a 'bad run'?

 

My referance was to the below quoted excerpt:



Point being, a catastrophic electrical failure would not likely brook functionality under any circumstance...

IOW: Relax -- I agree with you!

TTFN
HP

I wasn't picking on you - just putting it in more colourful terms.

 

I think I have been too focused on what might cause mechanical damage to the plastic case and not enough on the intermittent operation, as if intermittent operation was peculiar that that one, single, transistor. Mechanical damage, by the dozens, in exactly one transistor? It's a stress fracture, either from physical forces or electrical forces. Intermittent with a wiggly leg in all of them? I'm gonna need my crystal ball for this one.

.

My guess is the transistor in question has badly positioned holes in the PCB, careless assembly is also possible - but could affect any transistor on the board.

 

I wouldnt have given much reply without seeing the whole thing + schematic

 

I wouldnt have given much reply without seeing the whole thing + schematic

Guessing games can be an amusing pastime.

 

It's like someone taking a picture of a piece of tire tread on the side of the road asking what caused the tire to fail.

The CSI TV drama often shows the team identifying a car make and model from the tread prints on a victim.

 

The CSI TV drama often shows the team identifying a car make and model from the tread prints on a victim.

Yeah -- and notice how they can take a picture of a tire tread and identify the make and model and year of the vehicle that is 30 years old. Because, of course, we all know that everyone always puts on the exact same type of tires that originally came on the vehicle from the factory and we all know that the tire companies make a different tire for each year/make/model of car.

But, then again, just by using the magic phrase "image enhancement" they can take the photo from a security camera several blocks away and several stories up and zoom in on the reflection off some guy's watch face to get the VIN number from the dashboard of a van with darkened windows that was parked around the corner.

CSI, NCIS, Law and Order, and other similar shows have so many patently absurd things that you have to think of them as being comedy shows much of the time.

I always love how they solve some bizarre case and prove the guilt of someone by finding "dry ice residue" or "liquid nitrogen residue" on the killer's boots or some similarly asinine thing.

The best (worst) one though I think was during the first season of NCIS when Abby was thwarting a cyber attack and because the attack was so aggressive she couldn't type on the keyboard fast enough to stay ahead of it so Tim jumped in to help and Abby typed on the left side of the keyboard while Tim typed on the right side and, between them, they saved the day!

Overall, I can just shake my head at all the crap, but what really bugs me is when the writers and directors go on and on and on about how thoroughly the investigative techniques are researched and how they have these experts that vet all of the science to get it all correct. Reminds me of the show MacGyver where, before the show premiered, they had this huge publicity campaign touting how everything in the show was accurate and tested and all they did was change one little thing in order to make it impossible for some poor kid to hurt themselves at home. So I waited eagerly for the pilot episode and almost immediately realized that the one thing they changed was the laws of physics. They had glaring error after glaring error. It was so horrible that it was impossible to willingly suspend disbelief.