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.
Tool/software:
Hi Michael
I move below E2E post here
We will test your suggestion below and update how it work
Hi David
Pls test the suggestion above. and update here
Hi PYTsai,
I'll take a look tomorrow or Monday and get back to you soon. Thanks for linking the thread above.
Regards,
Michael Ly
Hi Michael,
update some informaion.
Customer's current circuit as below, and they are using DLP5530.
PD cable are seperated from the LED connector and cable
The PD cable customer are using now is a shielded cable.
Can you also clarify/describe the specific test and physical setup they are trying to pass? Ultimately, they may need to consider some type of system shielding of this circuitry? I have not seen this issue with other customers in the past and these customers would have had to pass the EMC compatibility testing. These type of issues may be related to the physical layout and routing of the PCB, cables, and relative grounding.
Dear Jason,
The EMS test status and results are detailed as follows:
1. The vehicle manufacturer requires a test intensity of 140V/m
2. TPS99000 Pin73 TIA_PD1 currently has a 120 ohm Bead, and Pin74 TIA_PD1_FILT currently has a 120 ohm Bead and 0.1uF capacitor. The schematic is as below.
3. Test conditions: Color changes and black screen will appear on the RS test 2.7G-3.9GHz screen. Test result is as below
2. During the test, the screen flickered at 3.04G ~ 3.34G and the brightness would change.
3. During the test, the picture will shake at 2.73G
4. During the test, there were snowflakes on the screen at 2.79G~ 2.87G, 2.96G ~3.38G, 3.68G~ 3.7G
5. During the test, at 3.84G, 3.74G ~3.76G, the screen turned black and you had to manually power on again (the backlight is still there)
6. During the test, at 2.84G~2.87G, 3.22G, 3.64G, the screen turned black and will automatically recover (the backlight is still on)
7. During the test, at 3.24G~3.38G, 3.72G, 3.76G, the screen turned into a black screen, and it will automatically recover after stopping the interference (the backlight is still there)
The test results of countermeasures that have been verified so far are as follows:
1. Pin73 TIA_PD1 and Pin74 TIA_PD1_FILT replace the high frequency with Bead or Choke, verify Fail
2. Change Pin74 TIA_PD1_FILT capacitor to 1000p, verify Fail
3. PD Sensor Cable is changed from original isolation Cable to dual isolation Cable, verification fails
4. The PD Sensor Cable was changed from the original isolated Cable to a double isolated Cable with signal pairing, and the verification failed.
5. Add 10p to Pin73 TIA_PD1 and verify the Pass. But it will cause the brightness of the backlight to be different. EE Function will have problems.
In addition, E-LEAD RnD provides a layout for viewing. There is no problem with Layout.
Thnaks & best regards,
David,
Based on the notes and symptoms provided above, the issue may not be related to the PD feedback circuit.
"screen flicker", "snow flakes", "picture shake" is more indicative of the DLPC230 being impacted by the radiation. This could be related to power supplies of the DLPC230 or the input video patterns. Are they using the system with internal test pattern generator or is video being provided from outside the module?
Are they are able to read back any of the DLPC230 status registers? There are several error codes that might get triggered by the DLPC230, that would help to further identify or narrow down the cause of the issues.
See the ACP tool section to read the error history section:
Thanks,
Jason
David,
Any further follow-up here? I am going to resolve the ticket for now. Let us know if the customer needs more help.
Jason
Dear Jason,
During the EMS test, RnD tried to remove the PD line and the EMS test passed. So RnD thinks that PD is interfered with. Therefore, a small capacitor (10pF) is added to the PD, EMS pass, but this will cause abnormal white balance. RnD also tried adding sheilding to the cable line, but it didn't work. Therefore, EMS RnD is asking if there is any direction to reduce the interference of PD?
Because the E-LEAD test uses a complete machine, it is difficult to read the error code, and it is need to modify the SW to do it. This needs to be discussed with SW RnD. Currently asking RnD to help discuss with SW RnD. If there are any problems in the future, we will post a new post.
Thanks & best regards,
Hi David,
Thanks for the feedback. When the PD line is removed, then there are other things disabled, so this is not necessarily a direct link to the root cause. With that said, E-lead may need to consider stronger shielding of the main PCB (connecting to the PD) or even completely shielding the area of the PD cable ot the main board. Ultimately, they need a working solution. At this late stage, the main mitigation techniques would be to add shielding to the sensitive areas.
When I look at other AR HUD systems, I do not see any special shielding for the PD and/or main board. However, the ones that I have in my lab, do include a large metal plate that serves as a heatsink, but also provides shielding / grounding for the main electronics including the PD. Does their system include a large metal backer plate that is grounded? Maybe they can send pictures of the mechanical setup. If they want to send these privately, then please send by DM or email me directly.
Thanks,
Jason
Dear Jason,
RnD went out to the laboratory to test EMC today. Will discuss with RnD on Monday morning.
Thanks & best regards,
Dear Jason,
Discussed with RnD and they are willing to provide relevant information. And, they want to send these privately, could you give us your email? My Email is david.yang@arrow.com. I will add the relevant people.
Thanks &best regards.
Hi David,
I will send you a direct email. Going to close this thread.
Jason