Counterfeit Capacitor Investigation

Hi everybody,

Shitfoot recently spotted some counterfeit tantalum caps on some EXXXX digital motherboards in assembly. He spotted some Vishay 0180-XXXX “P4” date code parts, a type of countefeit tantalum that we originally addressed in a line shutdown in mid-March of ’04. I confirmed that they were indeed counterfeit parts. They are in location C312 on the EXXXX-60201 board. They are used on a +7V power supply line. All of the motherboards that I looked at with counterfeit parts are 0409 date code motherboards. This date falls before March 11, 2004, when Venture was originally notified of the problem. Venture cleared their stock of this part probably by mid March.

Attached below are the notes that I sent out after the risk assesment in mid-March. In that meeting, we had decided to only rework the counterfeit parts on power supply lines of +8V and higher voltages. I am going to recommend that we hold to that same criteria, which means that I recommend that we not do any rework for the C312 parts on the +7V lines on the motherboard.

Thanks!

Scott H.

———-

Hi everybody,

Here are the EXXXX risk assesment meeting notes addressing the companywide counterfeit Vishay 0180-XXXX part issue with the EXXXX line’s conclusions. Other lines may want to use this as a reference while deciding what their own course of action will be in addressing this issue.

Problem:

Counterfeit Vishay 0180-XXXX 47uF 20V date code “P4” parts have been loaded onto many of the boards in each of our signal source instruments. 10 suspect counterfeit “P4” parts were taken off of EXXXX rear panel boards and analyzed by the Rel Physics Lab and Vishay and found to be 47uF 16V parts of varying ages. A couple of the parts that were analyzed by Vishay were found to have been produced in the early 1990s. Vishay has commented that they have found that most counterfeit Vishay parts are made from scrapped out parts that Vishay sent to China as part of Vishay’s metal recycling process. Vishay commented that parts sre scrapped for various reasons: process defects, high DC leakage, etc….. The EXXXX line has only seen one Vishay 0180-XXXX P4 date code failure. It occurred in November of 03 on a +8V line. Since this failure occurred in November, we know that we have been shipping counterfeit parts since November. In addition to that one confirmed failure, we suspect that we probably had at least one more in production and one in warranty due to comments from EXXXX technicians.

Corrective Action Decisions:

* All instruments that have finished the assembly process will not be reworked. We will let them flow through the line, but with one added step. All un-reworked instruments that ship out the door must have at least a 120 hour on time. This is to burn the instruments in to try and screen out the failures. Looking at our history of tantalum failure. A 120 hour on time should catch about half of all total would-be field failures. This process will continue until all suspect boards have been flushed out.

* All loose boards and chassis that have not been assembled that meet the following criteria will need to be inspected for 0180-XXXX parts of the P4 date code and reworked if P4 is found. The criteria for inspection and rework are:

1. The board must have 0180-XXXX parts that are being used on voltage lines >= 8V. (In the meeting we decided > 8V, but I then realized that EXXXX’s 1 failure was on an 8V line, so…..I changed it to “>=” ) If a board has at least one >= 8V part, all of the 0180-XXXX parts on that board must be reworked regardless of the voltage line that they are on.

2. The board must be either unshielded or a shielded board that is already being modded or inspected for something else.

3. If the board comes to the line installed in the chassis, it must be easily accessible for rework in the chassis.

All P4 date code 0180-XXXX parts found during inspection must be reworked with new non P4 date code 0180-XXXX parts. We will need to buy a bunch of these parts, and the parts we use need to be inspected by Brian O. or Scott H. Attached below is a spreadsheet that details all source boards that use the 0180-XXXX part along with all of the reference designators of the 0180-XXXX parts and the voltage line that those parts are on. For EXXXX, the list of boards that must be reworked are:

EXXXX-60203
EXXXX-60079
EXXXX-60046
EXXXX-60205
EXXXX-60044 GTLS

Refer to the attached spreadsheet for the reference designator info.

* Someone needs to coordinate a massive order of 0180-XXXX and make them readily available to PIMO and SOCO – Brian can you coordinate this?

* Venture needs to clear out all of their P4 date code stock of the 0180-XXXX Vishay part – DONE

* Detective work needs to be done by Venture and our purchasing department to figure out how to not get counterfeit parts again. Our line and others keep running into this problem. We need to buy from only vendor qualified suppliers. Someone needs to be held accountable for this – Brian O. and Tom have this action item.

* PIMO needs to work on getting a solder techinician up to expert status, so that they will be properly equipped for line rework in the future – Teik W.

* Have Tom M. analyze the one failed EXXXX Vishay part – Scott

* Brian O. to coordinate action with other lines

Thanks,

Scott H.

caps-id-slide1a.jpg

———-

Wait a minute, you mean no shutdown, no rework???

Why in the hell do you wait until you leave the company to get calibrated?

Thanks Scott,

Bossman

———-

I guess that, in the end, I am neither Mr. Warranty nor Mr. Reliability. In the end, I am Mr. Softie.

Scott H.

———-

Well, don’t tell that to teh ladies!

Factory_Peasant

———-

LOL

Bossman

———-

I stepped right into that one, huh. ; )

Scott H.

Advertisements

~ by factorypeasant on June 21, 2007.

2 Responses to “Counterfeit Capacitor Investigation”

  1. the supplyer of these componants should be shot as an enamy of the state!

    dt

  2. dt- one of the reasons i decided to write about the counterfeit component epidemic in the electronics industry is because i feel companies such as Vishay, AVX, and Kemet should be publicly ridiculed for their substandard business practices. since they were too lazy to handle defective part recycling on their own Vishay helped a growing counterfeit problem become far worse.

    we also were at fault for putting into place a procurement system that is by it’s nature set up for failure. when you task your materials buyers with keeping inventory levels as low as possible and offer them monetary bonuses for locating supplies as cheap as they possibly can, your company will no doubt be shipping in many types of fake/defective components. you will have no spare stock on-hand to replace fake components once they are discovered, causing further schedule slips in production and suffer increased short term cost as a direct result. real dumb.

    after a couple years of dealing with this shit you’d think somebody in management would have wised up but noooooooooooo. we kept getting ourselves burned.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

 
%d bloggers like this: