Continue to Site

Welcome to EDAboard.com

Welcome to our site! EDAboard.com is an international Electronics Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.

Shoddy Quality Control?

Status
Not open for further replies.

E-design

Advanced Member level 5
Joined
Jun 1, 2002
Messages
2,014
Helped
507
Reputation
1,016
Reaction score
408
Trophy points
1,363
Activity points
14,193
This specific Chinese supplied power adapter had much more output ripple than the rest of the batch.
On closer inspection, the series inductor in the 2nd-stage ripple filter was shorted out.
On the rest, this was not shorted. It makes one wonder how this slipped through? It is almost like a reject pcb made its way into the production line.

The ripple level returned to normal when the track section was removed.
 

Attachments

  • shorted_filter.png
    shorted_filter.png
    2.3 MB · Views: 92

My observation is that the goal of companies in any of the
cut-throat market segments (including low end power
supplies, but also the ICs that feed them) is to get to
a "inspect nothing, believe the statistics" production
footing. Which maybe works until the first thing changes.
Such as PCB vendor or artwork. That doesn't look like a
flaw to me, it looks like deliberately printed. Wonder if
you see any board version ID anywhere?
 

Here is what a normal functioning pcb looks like. I will look for any signs of a pcb version difference next time this problem shows up. Good idea. I don't have the suspect pcb available any more.
 

Attachments

  • filter_normal.jpg
    filter_normal.jpg
    129.3 KB · Views: 62

Small boards like that are usually panelized with maybe 20 fabricated on a single board then 'snapped' out. Most likely,there is an artwork error on one or more of the of the PCB masks and there are lots of other boards around with the same error. As it wouldn't stop it working, the manufacturer either didn't notice the error or just ignored it.

Quantity over quality!

Brian.
 

The quality control is needed at several stages. If the quality control is done at the last step, a fully populated but non-functional or defective board may have to be discarded or repaired. If the quality control is implemented at *every* step, the loss at the last step (this is the most expensive) will reduce significantly but you will need to invest in the quality control.

Consider an example- purely hypothetical- in which error crops up at a rate of 1%. If the final product has 100 components, then the probability of getting a *good* product finally is considerably low. If the bad components are weeded out at every step, the rejection rate at the final step will be only 1%. The overall cost of quality control will increase but it is worth if the final product is relatively complex. If the product is relatively simple, it is very tempting to avoid any kind of quality control.

Sad but true.
 

The end of the line test, in these cheap modules, may be nothing fancier than a cheap DMM and some resistors.
If the output voltage range falls within certain limits, the unit is packed for shipment.
With a DMM in DC mode, one will not see ripple, unless it is very severe.
 

Small boards like that are usually panelized with maybe 20 fabricated on a single board then 'snapped' out. Most likely,there is an artwork error on one or more of the of the PCB masks and there are lots of other boards around with the same error. As it wouldn't stop it working, the manufacturer either didn't notice the error or just ignored it.

Quantity over quality!

Brian.
Another thought I had about this, someone may have copied someone else's artwork and made a small error in the process.
As the resulting product appeared to work, the error may not have been noticed.
That looks like a deliberate track, not a scratch, or big lump of grunge landing on a photographic negative.

I have in the past been handed circuit boards by some "shifty businessman types" and asked if I would copy it so he can start his own back yard manufacturing business. I am sure this dishonest practice is pretty widespread.
 

As a follow up, we received a new shipment of these power adapters and so far everyone we tested in the batch had this pcb flaw. We notified the manufacturer and are awaiting a response.
 

It's very likely they made a new PCB revision which can function without the inductor, thus saving a few cents per unit. But then someone was stupid enough to not tell the assembly floor to not populate the inductor.

So basically they're going to fix the "flaw" by not populating the inductors in the future. Don't expect them to admit any responsibility for the high ripple current.
 

They company is a small one and they are only considering the short term profit and not the long term reputation.

Perhaps they themselves are not sure whether they will be there tomorrow or not.

It is entirely possible that they will agree to replace the whole batch. China is at the same point where Japan was 50 years back.

Quality control becomes indispensable once you reach a certain level. It is no more a choice but a compulsion.

If they refuse a full replacement, they are essentially asking you to change the vendor.
 

Here is part of their reply:
"We are very sorry, factory confirmed, PCB production error. Our engineering department has contact the factory to modify"

At least, they acknowledge that it is a problem. We are waiting to see what they offer as a remedy.
 

Status
Not open for further replies.

Similar threads

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top