Aluminum Electrolytic Capacitor Short circuit failure and Infant mortality

[Ashraf Almubarak]

I have produced 1000 units of our new developed energy meter. Almost 20 aluminum electrolytic capacitors (2%) have failed short when we test them during the production processes. The Capacitor is a 470uF 35V (UPW1V471MPD) made by Nichicon.
The failure itself varies between total short circuit to small variable resistance. One of the caps spit out its electrolyte.
I have 3 other different aluminum electrolytic caps from the same manufacturer that didn’t fail. A year ago, I have produced 300 units (same design and part no.) and they running perfectly and didn’t find any failed capacitors.
My question has 3 parts:
1-   How this occurred?
Is it normal to have an infant mortality of 2% with short circuit failure for aluminum electrolytic caps?
Or, may be the cap is counterfeited and the con manufacturer didn’t do enough quality testing. (I bought them through a PCB and components supplier in Shenzhen). If they are counterfeit, how to verify that?
2-   What should I do with the rest of the batch?
If I release the batch, would I guarantee that there will be no more short circuit failure? If it didn’t die during the manufacturing, it will not die in the near future! (U shape/bathtub graph).
Or it is better to run my meters for a period of time (24 hours) if didn’t fail it will not fail in the future.
Or I should go the hard way and change them all?
3-   How to avoid this in the future, because I will be going for large quantities > 100K.
I didn’t find very useful notes about electronics parts quality and reliability other than MIL-HDBK-978B. But it is out dated and didn’t have the electrolytic caps.

[MagicSmoker]

Nichicon makes very good quality caps and, aside from the capacitor plague issues in the early 2000's, I have never seen one fail. So, are you sure you got genuine Nichicon parts here? Can you post the link to the actual website you purchased them from?

[Ashraf Almubarak]

I didn't buy them from internet.
I have a supplier bought them and my other parts from Shenzhen. He did the PCB and PCBA
The chance to buy a counterfeit one from Shenzhen is high?

[Ashraf Almubarak]

These photos of the same capacitor (presumably)
One we purchased 2 years ago and the other one month ago. can you notice any notable difference?

[OwO]

No, parts should be the very last thing you blame when troubleshooting failures. Please first give us the important details, like what voltage the cap sees, what inrush and ripple current it sees, what process and temperature they were soldered at, and how the failures were diagnosed (did you remove the caps from the circuit to test? how were they removed?)

[OwO]

If used correctly and with proper amount of derating, even off-brand or rumored "bad brand" capacitors (chengx, capxon etc) should not have anywhere near 2% failure rate. If you really suspect the caps are bad, it wouldn't take too long to test 500 samples on a breadboard (put 10-20 caps in parallel at a time) and look for leakage current. If they are indeed bad, the most likely explanation still isn't "counterfeit" parts (they could be rebranded ones but even rebranded ones should not fail), but rather abuse/incorrect handling at some point in the supply chain.

[wraper]

Text font looks suspicious also positioning of temp rating against series (PW) is different. Also did heatshirink deform during removal or was it such from the beginning? In second case I say it's counterfeit.

The chance to buy a counterfeit one from Shenzhen is high?

Very high. Electrolytic capacitors failing short is very rare. I guess they were defective from the beginning.

[Ashraf Almubarak]

No, parts should be the very last thing you blame when troubleshooting failures. Please first give us the important details, like what voltage the cap sees, what inrush and ripple current it sees, what process and temperature they were soldered at, and how the failures were diagnosed (did you remove the caps from the circuit to test? how were they removed?)

I have produced 300 units with the same design (very minor change in other parts of the circuit). These 300 units are working for 2 consecutive years with 0% fail in this cap.

The voltage cap sees is 20V (the rated is 35V !). There is no high currents. The power supply is a capacitive power supply maximum supply a 17mA. The capacitor it self don't operate continuously. Once it is charged it will not operate (discharge) until certain event occur. The capacitor is used to store energy to operate a latch (main switch in a prepaid energy meter) when the power went off or the credit is 0.
The latch/relay has a resistance of 60 ohm. There is a freewheeling diode for the back e.m.f.

I removed the capacitors from the board and I tested them.
The temperature of the wave soldering machine is 300 C.

[OwO]

Ok that makes it much clearer, should have provided this information in the beginning. I would confront the PCBA vendor because they are using crap parts. If I had to guess I'd say they are used caps salvaged from computer mainboards. Personally I've had better luck specifying generic values on my BOM ("470uF >=35V electrolytic") because the PCBA vendor likely knows what's easy and low risk to source, but if you put in a specific brand it puts them in a difficult situation because they may have no prior experience sourcing that brand and will likely turn to less familiar suppliers). If you really need Nichicon caps I recommend sourcing them yourself (LCSC has genuine ones at good prices) and shipping it to the PCBA house.

[Ashraf Almubarak]

I thought if I provide them with specific part no. I would guarantee the original one. And If I gave them generic values I will not get any good one. It worked the opposite to me.

For sourcing the material. I don't how to do this by my self this time. We are new company in a country that is very new to electronics industry. I need to have a supplier to arrange the receiving of the parts and arrange with assembly factory.

Any way thanks for your precious answer.

[Ashraf Almubarak]

Yes I noticed that too.
However, those have two years difference age (presumably). Would the manufacturer make difference in the packaging printing?

I believe I was looking for your answer in the probability to purchase a counterfeited one. thanks

[Conrad Hoffman]

Aluminum electrolytic caps almost never fail shorted. I suspect counterfeit caps, but also check the solvent resistance and how the boards are cleaned. Easy to shorten the life of a cap with bad cleaning. Also watch out for water-washable flux that isn't being fully washed out.

[Bud]

If you have access to samples of both capacitors, cut them open and examine the guts for obvious differences.

[RoGeorge]

You should send a few failed caps to Nichicon, together with a few good ones from the same lot, so Nichicon can tell if those are fake, or just out of specs caps that were sold on grey market instead of being destroyed.

Most probably those are fake Nichicon.  Shenzen is full of very cheap components for a reason.

[Jwillis]

I don't want to start a heated debate but here,s what I found out.
 As far as I know there are no Nichicon facilities in  Shenzen . But There is one in Wuxi that is approved by Nichicon  for some electrolytic and foil capacitors for power supplies. Which specific ones I couldn't find information on. Nichicon does have a trading office in Shenzen. There are some other facilities in Malaysia but didn't research those. As far as I can tell ,most of  Nichicon capacitors are manufactured in Japan.
There are probably hundreds of possible theories that iI won't get into. But to be fair ,I can't see  Nichicon exporting product to be sold cheaper, unless it's old stock or factory seconds. That's just my own opinion of course. Also the print just doesn't seem right .But again it could be old stock to be fair.

[senso]

Print is much lower resolution, heat-shrink has a different texture and colour, font is different, positioning of text is different.

No manufacturer would change a whole line to make crappier heatshrink and printing, so, you have bought 1000 fake caps, most likely rebadged xxuF at 16V.

Measure the capacity of the still not failed ones..

[graybeard]

If you put a small constant current (say 100µA) on the caps they should ramp up in voltage until they breakdown.  If they are from a reputable source that breakdown should occur at greater than twice the rated voltage, but usually less than 4x.  If they are new and breakdown at less than 2x they are crap and the lot should be rejected.

[Ashraf Almubarak]

These photos of the same capacitor (presumably)
One we purchased 2 years ago and the other one month ago. can you notice any notable difference?

id say the 2 are different
maybe the unit on the R is fake?

so whats the correct answer?

Yes you are correct probably the one to the right is fake. I contacted my supplier and he didn't proof that they are purchased from Nichicon, he said he will need time. Mean while he will buy me 1000 pcs again from mouser.

I thought if I provide them with specific part no. I would guarantee the original one. And If I gave them generic values I will not get any good one. It worked the opposite to me.

For sourcing the material. I don't how to do this by my self this time. We are new company in a country that is very new to electronics industry. I need to have a supplier to arrange the receiving of the parts and arrange with assembly factory.

Any way thanks for your precious answer.

you say you gave a bom to a assembler? but how do they prove to you things are original?
if you search for a video by bunnie huang on fake chips etc, even he as a professional builder, get cheated by authorized dealers. according to his statistical study on certain products, distributors in china could mix in 3% of fake chips intentionally with originals to sell in bulk. and this is authorized distributors, not those down at assembler levels.

[wraper]

id say the 2 are different
maybe the unit on the R is fake?

Cap on the right looks exactly like genuine nichicon. Cap on the left have several dodgy things.

[Ashraf Almubarak]

If you put a small constant current (say 100µA) on the caps they should ramp up in voltage until they breakdown.  If they are from a reputable source that breakdown should occur at greater than twice the rated voltage, but usually less than 4x.  If they are new and breakdown at less than 2x they are crap and the lot should be rejected.

It is interesting test to do.
Do capacitors manufacturer perform this test? It is my first time to know about it.

[Ashraf Almubarak]

id say the 2 are different
maybe the unit on the R is fake?

so whats the correct answer?
Yes you are right the suspecious one is on the right.
I contacted my supplier and they claim it is original but they could give a proof to trace back to Nichicon.
Now they will buy 1000 pcs from mouser. 

[Ashraf Almubarak]

you say you gave a bom to a assembler? but how do they prove to you things are original?
if you search for a video by bunnie huang on fake chips etc, even he as a professional builder, get cheated by authorized dealers. according to his statistical study on certain products, distributors in china could mix in 3% of fake chips intentionally with originals to sell in bulk. and this is authorized distributors, not those down at assembler levels.

This is very shocking info. Authorized dealers mix 3% with the original one??
It will be a different topic, however, if you are experienced in manufacturing electronics in bulk. How to supply 100K pcs of each component?? I don't believe Mouser & Digikey would have this amount. I should contact the manufacturer directly and bypass the dealers?? Or I will find the stock in the dealers warehouse??

[Ashraf Almubarak]

I don't want to start a heated debate but here,s what I found out.
 As far as I know there are no Nichicon facilities in  Shenzen . But There is one in Wuxi that is approved by Nichicon  for some electrolytic and foil capacitors for power supplies. Which specific ones I couldn't find information on. Nichicon does have a trading office in Shenzen. There are some other facilities in Malaysia but didn't research those. As far as I can tell ,most of  Nichicon capacitors are manufactured in Japan.
There are probably hundreds of possible theories that iI won't get into. But to be fair ,I can't see  Nichicon exporting product to be sold cheaper, unless it's old stock or factory seconds. That's just my own opinion of course. Also the print just doesn't seem right .But again it could be old stock to be fair.

If you are experienced with electronics industry. what is your suggestions to supply large quantities of such components?? 20K, 50K and 100K.

[Ashraf Almubarak]

Very good information.
Our company is just in its beginning so we learn the hard way.

I will consider this in the future.

[Kerlin]

I have had substitution problems in China.
The factory was owned by the company I worked for.
I found and proved the problem was a supervisor.
When  I asked him about it he reminded me where I was and challenged me to do anything about it.
My wife had the same problem. The imported Japanese clay had been substituted with local rubbish which had dirt in it.
Why go there? It's cheap ? Not when it costs you your business.

So what do you intend to do - They bet, and I bet, you and others will go back.!