We have using LM4871 at our factory without problem, but recently about 30% ones at reels has issue. Amplifier stay still at shutdown stage and does not work. Our PCB, components and technology are not changed. It's serial product and we manufacturing it about 1K every month.
Our schematic the same as reference schematic at datasheet, I did test fixture with ZIF connector and have had the same result. About 30% of IC are not work. I see that it was National product but TI's now.
I can not image that this very reliable tested tested product has 30% defects. But what can I do now?
Thanks and Regards,
Sergey Skriblovsky
Hi Sergey,
Could you please elaborate on the specific test conditions of the system before the issue occurs? ie initial system power-up or controlling the shutdown through a standby switch?
Also, how are you controlling the Shutdown pin? What is driving/controlling this pin? Are you using a microcontroller or a physical switch through a pullup resistor?
Have you verified by probing that the voltage on the shutdown pin can be controlled to low, but the IC remains in shutdown mode? Can you force the pin to low externally and get the IC to come out of shutdown?
I look forward to your responses.
Best Regards,
JD
Hello Sergey,
What is the end application for this device? Would it be possible to send me your schematic, I know you mentioned it is the same as the reference schematic but I would like take a look anyway. When the device is powered on do the outputs bias to VDD/2? Does it draw any current (IDDQ)? Also can you send me the top marking on the failed parts so I can track them through our manufacturing flow?
Nisha
Hello, Nisha!
I'm trying to answer for your questions and for questions at previous from John.
This is GSM/GPS Registrator for automotive. I've attached part of its schematic.
OUT_A, OUT_B goes to connector and opened. UNC_ON connected to processor, SP_GSM connected to SIM900D GSM module (audio output). 4.5V powered by TPS5430 DC/DC converter.
My test fixture very similar, but shutdown pin connected manual switch to GND. For good partss bias set to Vdd/1, but for broken ones bias stay zero. Draw current for broken parts 3 mA (it's at test fixture with 16 Om output load and SHDN=zero).
I've attach picture files with reel labels and IC case.
Then, John asked what test condition - power up with shutdown level equal zero. I did check voltage level at all pins and for broken parts all are zero (except Vdd). Then I did try to set shutdown pin pulled up at the power up start and any another variants but nothing more occur.
BTW, could you please let your opinion. Is it possible to select 70% parts as good part and use it at factory? Do you think that we must discard all reels without selection?
All parts that leave our facility are tested per datasheet specifications. Can you let me know what region you are located so I can put you in touch with our local sales team? I think the best option is to send the failed devices through our Failure Analysis (FA) channel and your local team can assist you with that process.
we are from Russia, Nizhny Novgorod, Micro LIne Ltd.
But firstly I did call to Moscow TI's office and explained this issue. After that official TI's manager at our region (Agata Sinkevich) called me and said that if we bought parts from non- authorized dealer then TI can not help us. That is all.
Actually we bought couple reels from small local dealer and now we working to clearing this supply chain.
Let me talk to the Product Engineer to see if he/she know of any failures on the LM4781. I will also find someone who can directly help you out, possibly find out where you can send those failed units.
Gerardine LyAudio ApplicationsTexas Instruments-Silicon Valley Analog
Hello Gerardine!
I appreciate your help very much! We will be ready to send broken parts if it will be possible.