|
|
Hi, Sean. I send you to 3 pcs failure boards. (2pcs are SKB's samples and 1pcs is mine). Mr Hite Choi will give you them 1. first test with normal board and another my own 12V adapter. 1) power on/off 20 times using multi tap switch (that means the power on/off with power plug pull in/out) : after testing, it is normal 2) power on/off 20 times with dc jack pull in/out : after testing, it is normal. 3) power on and maintain about 20 minutes : after testing, it is normal 2. Second test with normal board and 12V adapter which SKB gave me 1) power on/off 20 times using multi tap switch (that means the power on/off with power plug pull in/out) : after testing, it is normal 2) power on/off 20 times with dc jack pull in/out : after testing, it is normal. 3) power on and maintain about 20 minutes : after testing, the symptom happens (When 5v adapter pull in, there is no booting) |
|
|
Hi SB, Just received the failure boards from Mr. Hite. Hi Jim, Please check the symptom of failure boards which SB provided. Find the root cause, broken componets and try to fix it. |
|
|
Hi SB Please see attached and our command. Thanks. When we receive the 3 failar baords, the eFuse IC alrady damage. Thus, it can not protect our system |
|
|
Hi, Jim thank you for your report. but according to your report, you use only adapter which SKB was gaved. please check again with another 12v adapter. there is a Kaon Settop box in your office. I also checked with another 12v adapter in my office. I also pull in/out dc jack and power plug. there is no failure with another adapter. |
|
|
Dear SB We did 20 times for Kaon adapter, the results is the same. The eFuse can protect our system like previous report. |
|
|
Hi Jim, Sean. I have a question. I understand that EFUSE IC(TI) works abnormally at pulling out ac plug of 12V adapter. but, according to the waveform, the voltage of efuse IC is 5.5V at re turn on period. this 5.5V cannot demage to the DCDC ICs. Can you find another root cause? waveform.gif (41,115 bytes) |
|
|
Dear SB This waveform is for eFuse IC is normal. The eFuse IC alrady damage for 3 failar baords. So the 12V voltage demage the DCDC. We suggest when doing the 12V adapter AC plug out, please wait until the voltage drop to 0V, then plug in the 12V adapter. Or, it will get high risk of eFuse IC damage |
|
|
Hi Jim, Sean according to your report, you mentioned Effuse IC is abnormal below this waveform (see the picture). So I first understand this waveform is abnormal behavior. I already know the eFuse IC already damaged for 3 boards. and I updated my test case. please find the root cause why efuse IC was broken with 12 volt adapter. I saw the TI datasheet. the efuse IC can protect the 20V. waveform_efuse.gif (57,914 bytes) |
|
|
Dear SB We will send these fail IC to TI side, and work with TI to find the IC failure problems |
|
|
Hi Jim 1. How long take the time to receive the result to Vender. In my opinion, the vender usually report IC itself failure because of EOS. I really want to know why EFUSE IC demaged, what is the root cause (This IC protect 20V but, the failure happens 12V) |
|
|
Dear SB We recieved one EVT (without e-Fuse version) failar baord yesterday. First, We find some components already removed from this board. Then we rework the components and find all DCDC (P3V3, P1V2, VDDEE and VDDCPU) already demaged. I think the root cause is P12V damage DCDC IC. |
|
|
Hi Jim. I debugged this board. I wanted to find which point is failed.(I just removed some coils besides DCDC and check pattern) but, I could not find. no schematic, no scope (only multimeter) I heard that TVStorm engineer input 12v adapter for a long time to this board. so many DCDC had demaged. Can you revive the board? |
|
|
Dear SB We already reproduce IC damage case, we use 220V AC as adapter input. When zoom in the scale, we can see SKB adapter has overshoot voltage during DC input. The worst case is 34V. The Kaon adapter only 15.6V during DC input. We can conclusion that the over voltage damage the eFuse IC. SKB_1.png (11,290 bytes) SKB_2.png (12,586 bytes) SKB_3.png (14,439 bytes) Karon.png (13,371 bytes) |
|
|
Hi Jim 1. ch1 is eFuse out, ch2 is input from DC jack? 2. the picture SKB_1, 2, 3 are all over 20V (absolute value) but, according to picture1, 2 efuse out are work well althogh the voltage is spec over. please check that efuse IC can endure the time duration at abnormal voltage state. 3. according to picture SKB_3, the efuse works abnormal. but, the output voltage of CH1 is under 2V. this value cannot demage the DCDC IC (You mentioned that 3 failure board have damage DCDC and BGA IC) |
|
|
1. Ch1 is eFuse out CK2 from DC jack 2. The IC spec is 20V. If the voltage up to 20V, it may damage. The limition of different IC is not the same. 3. The picture 3 is over IC can tolerance. We check that IC alrady damaged. If you plug again, the input P12V=output P12V and DCDC will damage. |
|
|
1. check please the spec of duration to Vender. 2. Make the report with these waveforms please. I'll submit to SKB. 3. Repair 3 boards please. I should return SKB to them |
|
|
Hi SB Can you provide SKB and Kaon adapter spec to us? |
|
|
Hi Jim I already asked SKB to adapter's document. but SKB request to picture of that abnormal adapter's label |
|
|
Dear SB Please refer adapter's label 圖片1.jpg (74,225 bytes) |
|
|
Hi Jim Thank you |
|
|
Hi SB, Attached the updated FA report according to your comment. Please check it and let me know if there is any further questions. Thanks. |
|
|
Dear SB Please refer TI Failure Analysis Report |
|
|
Hi Sean, Jim I'm sorry I'm late. thank you for your support and I'll submit SKB to your report. the SKB manager understood that SKB's adapter own defect. I requested SKB to the adapter's datasheet. but SKB manager didn't provide it but He just gave me that the tolerance is 11.4v~12.6v at normal operating |
|
|
Hi SB, Got it. Thanks for your updating. So, let's say both of ALT & SKB understand that this issue is caused by defected adapter, right? And I believe that If the DC output of adapter is working in normal operation range, e.g. 11.4~12.6V, the Eagle board will not be damaged. If so, shall we close this issue? |
|
|
Hi Sean. 1. Your opinion is correct. this issue is cuased by failure adapter. 2. SKB manager request to 1) check the current waveform from booing start (plug in) to booting end (display home menu) 2) check the current waveform at booting start (plug in time) |
|
|
Hi SB, The SKB is reqesting to use our 5V adapter for waveform measurement? Also, can you describe in more detail about the difference for booting start(plug in) and booting start(plug in time) |
|
|
Hi Sean 1. That's correct (5V editech adapter's current waveform). 2. It is same "plug in" to "plug in time" |
|
|
Hi SB, According to your comment, we will measure the item 1 for ALT/SKB. But for the booting end (display home menu), it's hard to catch the timing for dispaly home menu, we may measure the booting end as HDMI output the signal. 1) check the current waveform from booing start (plug in) to booting end (display home menu) Hi Jim, Please follow it up asap. |
|
|
Hi Sean. Just power on when current probe is applied to 5v dc. and capture the waveform. Do you have any problems? |
|
|
Hi SB, Based on your comment on 8/12, it seems SKB want to measure two kind of waveforms, but it looks that we only need to measure one waveform, the 5V waveform when plugged in the DC plug, right? ==================================================================================== 2. SKB manager request to 1) check the current waveform from booing start (plug in) to booting end (display home menu) 2) check the current waveform at booting start (plug in time) |
|
|
Hi Sean that's correct. the current means not "present" but using current probe. |
|
|
Hi SB please refer DC plug in report. Thanks |
|
|
Hi Sean, Jim I mentioned current waveform with current probe not voltage. In your report, there is only voltage waveform |
|
2019-08-23 16:33 developer |
current_probe.gif (37,418 bytes) |
|
|
Hi SB Update inrush current in the report |
|
|
Hi SB Update inrush current in the report |
|
|
Hi Jim. thank you for your report but I also request to check the current waveform with current probe frome booting start to booting end. |
|
|
Add the current waveform from booting start to booting end |
|
|
According to your block diagram, there in no connection of Ethernet. In your test, no ethernet? or wrong block diagram? |
|
2019-08-26 10:58 developer |
blockdiagram.gif (67,961 bytes) |
|
|
Hi SB No ethernet for the test. |
|
|
Hi Jim, the test should be full load condition. please connect the ethernet and test again |
|
|
Hi SB, What kind of condition you are asking us to do the current waveform measurement? Would you please provide us in details? The current waveform we are measuring now is kind of like inrush current measurement which is the same condition as we did for the 12V adapter that SKB provide to us before. |
|
|
Hi Sean. SKB request to : 1. the current waveform from booting start to booting end. 2. the inrush current waveform at booting start they request SGN to debug that failure 12V adapter. first they want to know the current waveforme with normal 5v adapter. (they may want to the report with full load condition.) If you think that the test condition is enough with only adapter and TV, confirm please And in my opinion, all test should be tested with full load condition. If you don't agree, confirm please. |
|
|
Hi SB Update the report as attached. |
|
|
Hi Jim, Sean Thank you for your report, Good job |
|
|
SKB manager accept this report |
|
|
SKB request to another document |
|
|
Hi Sean SKB manager request to the schematic from dc jack to TI ic |
|
|
Hi SB, Attached the SCH from DC jack to TI IC. Please let us know the conclusion between ALT and SKB about this issue. DC input circuitry_SKB.png (93,330 bytes) |
|
|
Hi Sean Thank you for your support and I have a question.' According to your schematic, there is only on cap C338(100nF) at the net "DC_5V" Does the net "DC_5V" has more cap in other pages? let me know the total cap value at DC_5V net. SKB manager worry that the cap value of DC_5V is small. |
|
|
Hi SB, We put a total 44uF(22uF+22uF) at PP_PWR_IN net to reduce the output peak voltage of TI eFuse and stabilize the main system voltage when the eFuse turns on. |
|
|
Hi Sean. SKB ask me to change higher capacitance in C338 with same size 0402 1. what is the maximum capacitance of C388 0402size 2. If the c388 is changed maximum capacitance, how much is the cost up? SKB will diside to change capacitor if cost up is reasonable. but they think cost up is higher, they will not change. |
|
|
Hi SB, 1. The maximum capacitance of 0402 size is 1uF/50V 2. Actually, we have awared of this experiment during we found this issue then. Attached is the test result for your reference. After changing the capacitance from 0.1uF to 1uF, we don't see any significant changes. So, we don't recommended to change the capacitance of C388. Please kindly check attached file and let us know SKB's comment. |
|
|
Hi Sean Thank you for your information. please let me know the price of each cap 0.1uf and 1uf If the price information is confidential in your company. please give me the gap of cost between 0.1uf and 1uf. I can explain and persuade easily with the exact cost up information. |
|
|
Hi Sean thank you for supporting this issue. I discussed SKB to this issue with your report and cost gab information SKB manager accept to No change C338 |
- yk.jung ( (ALTech) Younkwang Jung ) developer
.gif)
.gif)
.png)
.png)
.png)
.png)
.jpg)
.gif)
.gif)
.png)
