This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

MSP430F6736: MSP430F

Part Number: MSP430F6736

单板生产176PCS EC570键盘板任务,单板测试发现13PCS键盘上电数码管不亮,故障现象如下。

176 pcs of ec570 keyboard boards were produced on the single board. During the single board test, it was found that the nixie tube of 13 pcs of keyboard was not on, and the fault phenomenon was as follows.

故障现象:

Fault phenomenon:

1.正常键盘板上电后自检显示如下图1(数码管和8个LED灯全亮),随即如下图2上排4个LED全灭,下排LED与数码管一致轮番切换显示频率/电流/电压。

   1. After the normal keyboard board is powered on, the self-test display is shown in Figure 1 below (the nixie tube and 8 LED lights are all on), and then as shown in Figure 2 below, the 4 LEDs in the upper row are all off, and the LEDs in the lower row are consistent with the nixie tube, switching the display frequency / current / voltage in turns.  

 

2.此批次故障板上电显示如下图:8个LED全亮,数码管不良,且一直处于此状态。

 2. The power on display of this batch of fault boards is shown in the figure below: 8 LEDs are all on, the nixie tube is poor, and it has been in this state all the time.

统计此次生产176PCS使用芯片情况如下:

The statistics of 176pcs chips used in this production are as follows:

此芯片代码为:2004.02.0106,其中芯片丝印08AC5JT04A41YT为此次新到物料。散料使用前已烘烤。

The chip code is 2004.02.0106, in which the chip silk screen 08ac5jt and 04a41yt are the newly arrived materials. Bulk materials have been baked before use.

芯片丝印Chip screen printing

生产数量production quantity

测试结果test result

退仓库return to warehouse

芯片周期chip cycle

来料包装incoming packaging

08AC5JT

107

良品OK

 

2032

119PCS整包whole packaging+19PCS散料bulk material

12

不良NG

3片翘脚+16片,3 warped feet + 16

04A41YT

18

良品OK

 

2015

19PCS散料bulk material

1

不良NG

 

13AP4QT

30

良品OK

 

 

以前生产余料Surplus materials from previous production

7CA113T

2

良品OK

 

 

79ALP6T

1

良品OK

 

 

81AC4NT

5

良品OK

2片翘脚

2 warped feet

 

         目前此次新到物料批次为08AC5JT04A41YT的单板共138PCS被品质暂扣,不允许使用,请反馈厂商确认芯片及处理方案。

At present, 138pcs of boards with new material batches of 08ac5jt and 04a41yt have been temporarily suspended and are not allowed to be used. Please feed back to the manufacturer to confirm the chip and processing scheme.

同批次仓库还有2整包共计181PCS未开封使用(20151盘,20321盘)。

There are two whole packages in the warehouse of the same batch, with a total of 181 PCs. unsealed (one in 2015 and one in 2032).

整包来料:

 Whole package incoming materials:

散料1:19PCS,周期2032

 Bulk material 1:19pcs, cycle 2032

散料219PCS,周期:2015

 Bulk material 2:19pcs, cycle: 2015

单板对EC570故障键盘板进行如下验证,故障随芯片转移。

The single board verifies the ec570 fault keyboard board as follows, and the fault transfers with the chip.

1.单板更换电脑重新烧录键盘板程序,故障依旧,数码管不显示。此前已验证烧录旧程序,故障依旧。

1.The board is replaced by the computer, and the keyboard board program is burned again. The fault remains the same, and the nixie tube does not display. The old program has been verified and burned before, and the fault remains.

2.将故障单板拿到技术部,使用张旭梅电脑和烧录器烧录,故障依旧,数码管不显示。

2. Take the fault board to the technology department and burn it with Zhang Xumei's computer and recorder. The fault remains the same and the nixie tube does not display.

3.准备一块EC750键盘板,烧录到最新EC750键盘板程序V2.55.15,更换芯片前测试,开机自检显示正常(见下图1)。将EC570故障键盘板芯片拆下更换到EC750键盘板上,同样烧录到EC750键盘板程序V2.55.15,开机测试自检显示不全(见下图2)。

3. Prepare an ec750 keyboard board, burn the latest ec750 keyboard board program v2.55.15, test before replacing the chip, and the post display is normal (see Figure 1 below). Remove the ec570 faulty keyboard board chip and replace it on the ec750 keyboard board. Similarly, burn the program v2.55.15 on the ec750 keyboard board. The self-test display of power on test is incomplete (see Figure 2 below).

        综上:此批次故障随芯片转移。

To sum up: this batch of faults are transferred with the chip

 

  •   包才信  Jerry

    Shanghai Newlion Electric Technology Co.,Ltd.

    上海新狮电气科技有限公司

    ADD:101,Building NO. 11,130 Lane Qingyu Road,Putuo District,Shanghai 200333,P.R.CHINA

    地址:上海市普陀区清峪路13011101

    Mobile:   13641821965

    Tel:      86-21-62157988

    Fax:     86-21-62157982

  • E-mail:baocaixin@aliyun.com

  • Hi Jerry,

    Where you get these ICs? Contact to your sales to get support.

  • I bought it on your official website ti.com. The order numbers are t01433316 and t01794145. I fed back the problem to your company. The problem number is: case number: cs0725914. There, let me send a post on Ti E2E English forum to verify whether the product has quality problems. Once the E2E engineer confirms that the product has quality problems, I can contact the customer support center again through the forum link for after-sales service.

  • Hi Caixin,

    We will discuss how to support you quickly.

    Thanks!

    Best Regards

    Johnson

  • Hi Caixin,

    It seems some problem with the device. Some of the LCD pin is multiplex. Could you test whether other function of the pins is worked? Does only the LCD function of these pins work abnormally. Another test is that could you compare with the waveform of the LCD pins to find any difference?

  • Hi   Allen,

    You suggest to test whether other functions of the pin work normally, and compare the waveform with that of the LCD pin to find out any differences. These two tests are in progress and will be fed back to you in time when the results are available. At present, the MCU has produced 280pcs, and 31pcs are defective. Take down the 31 defective products and replace them with good products, and the equipment can work normally. After removing the good products and replacing them with the 31 defective products, the equipment can not work again. Has your chip been upgraded in software or hardware? If so, please let me know. Thank you.

  • Hi,

    All the devices are the same in the same batch. There is no upgrade. If you have any feedback of the test. Please update to me.

  • Hi Allen

    The chip we use is not multiplexed,The test waveform is as follows

    Measure the waveform of pin 62 and GND. The normal chip waveform is shown in Figure 2, and the bad chip waveform is shown in Figure 3

  • Hi caixin,

    P62 is also a GPIO P5.3. Can it output high and low. I want to know that if this pin is shooted. If it can output a high/low signal. It seems that something wrong with the LCD module. You may looking for customer support center for further help.

  • Our LCD is normal. After replacing the good chip with the bad chip, the LCD can work normally.

  • Hi Caixin,

    I mean that some happened with the LCD peripheral of the device. Could it output low and high as a GPIO? I want to confirm that the internal circuit of the pin is OK. As you shown, abnormal device pin62 always output high? And could you list out all the abnormal pins like pin62?

  • The GPIO 62 ~ 74 of the failed chip to the LCD and nixie tube drive signal is always high, and the display status is as follows:

    Ec570 keyboard board using fault chip      Ec750 keyboard board using faulty chip

    The GPIO 62 ~ 74 of the normal chip to the LCD and nixie tube drive signal has high and low, and the low-level nixie tube displays, as shown in the figure below is the display effect of the normal product:

    Ec570 keyboard board using normal chip    Ec750 keyboard board using normal chip

    The normal chip is exchanged with the faulty chip, and the fault phenomenon is transferred with the chip, which is independent of PCBA.

    Defective rate of test chip 17.3%

  • Hi Caixin,

    Can you provide the result of the test I mentioned before : toggle these pin as a GPIO function. 

    You can refer to the example code to output high and low.

    dev.ti.com/.../node

  • Hi Allen

    At present, the program with abnormal chip is not running. Use the emulator to check the program operation online. It is found that the abnormal chip is stuck in the program shown in the figure below during initialization.

    The normal chip runs with the same program without any problems; Why does the same program cause this phenomenon on different chips???

  • Because the program can't run now, the technology says it can't do the test you want?

  • Hi Caixin,

    As your description, the program stack at waiting for clock stable. There is some fault on clock.  You can test the XT2 to see whether the crystal oscillate normally.

  • Hi Allen

    We use the internal clock of the chip instead of the external clock, and the code is also written according to the braking file

    Could you tell me if the chip has quality problems according to these tests? If not, can you tell me what needs to be tested so that you can know what the problem is faster? In order to determine that there is a problem with the chip quality, I can feed back to the customer support center according to the results and ask them to return or exchange goods for us. Thank you

  • Can you give me your phone number? I let our technology communicate with you directly, so it's more direct and fast

  • Hi Caixin,

    Go into debug mode.  Check the register UCSCTL6 to make sure you turn off the external crystal. You need to check the UCSCTL7 register to see which fault bit is set. If you use REFO for source and turn off the external crystal, the XT2 and XT1 fault bit should not be set. If the DCO fault bit can't be cleared, you run a clock example code at below link. It uses the internal REFO as source to configure the clock at 8MHz.

    https://dev.ti.com/tirex/explore/node?node=ANSXXA54TyfP9i3x6yqm-A__IOGqZri__LATEST&search=6736

    You can also do the corresponding test on the chip that can work normally to make a compare.

    If it still don't work on the abnormal device but work on the normal device at the same environment. You feed back to customer support center for help.

  • Hi Allen

    Good news, thank you for your advice. According to the chip modification method you provided, the product has been tested and can be used normally. Should your product be operated in this way since 13 years ago? Or did it start in the last two years? Because we never encountered such a similar problem when using this product two years ago.

  • Hi Caixin,

    Glad you solve your problem. The  ticket will be closed.  The operation of this device is always be like this.

    Thanks.

**Attention** This is a public forum