TUSB4041I: TUSB4041I Downstream Port Failure Issue

KevinZhang123

Part Number: TUSB4041I
Other Parts Discussed in Thread: TUSB4020BI, TPD1E1B04

Hi, Team

This is Kevin Zhang, analog FAE in South China.

My customer is using TUSB4041I in before-market Infotainment System. The downstream ports are connected to T-Box and Car Camera Recorder. Customer PP 100 sets and test in car, they find 3 TUSB4041I are failure. One of the 4 downstream port can not work. There is chance for each port to fail.

Customer has mass production with TUSB4020BI before, and there is no issue.

Here is schematic of TUSB4041I. Could you help to check it and share your suggestion?

5807.TUSB4041I Schematic.pdf

Thanks

Kevin Zhang

over 8 years ago

0 JMMN over 8 years ago

TI__Guru 60870 points

Hi Kevin,

Can you explain the nature of the port failure? No power on the port? No devices recognized?

The schematic and layout look acceptable.

Regards,

JMNN

0 KevinZhang123 over 8 years ago in reply to JMMN

TI__Expert 5750 points

Hi, JMMN

When the port failure:
1. There is 5V power on the USB port, as TUS4041I does not handle the 5V power for USB.
2. No device can be recognized.
Is any special protection circuit is needed to add in customer schematic?
Do you need to check customer's PCB layout?
Thanks

Kevin

0 JMMN over 8 years ago in reply to KevinZhang123

TI__Guru 60870 points

Hi Kevin,

If these port failures are consistent: same port, same unit with multiple power cycles, then these units needs to be returned for FA.

Please note that battery charging is enabled on the ports, which should not occur unless the TUSB4041 is controlling the USB power switches. Also if USB compliance is required, then the hub should be able to report an overcurrent condition on the downstream ports via the OVERCURxz inputs and the hub should not report as being full power managed unless the hub controls the downstream port power switches. These are minor concerns which would not impact port performance.

Regards,

JMMN

0 KevinZhang123 over 8 years ago in reply to JMMN

TI__Expert 5750 points

Hi, JMMN

All the 3 failure part are pin 33 (USB_DP_DN1) and pin 34 (USB_DM_DN1) failure. We are arranging FA.
As shown in schematic, customer float all the PWRCTLx/BATENx and OVERCURxz pins. The 5V power of USB is always on. Customer just connect T-Box, Camera Recorder, Flash Disk which are not need fast charge in the TUSB4041I downstream ports. Is there any issue?

Thanks

Kevin

0 JMMN over 8 years ago in reply to KevinZhang123

TI__Guru 60870 points

Hi Kevin,

The customer is not floating the PWRCTLx/BATENx pins, they are pulling them up which enables battery charging. It isn't ideal, but it would not cause the damaged ports.

Is the same port damaged on all the units? Is that port always connected to the T-box?

Thanks!

JMMN

0 KevinZhang123 over 8 years ago in reply to JMMN

TI__Expert 5750 points

Hi, JMMN

It is always pin 33 and pin 34 damage. The T-Box maybe replug sometimes.

Here is the FA report:

4370.QEM-CCR-1710-00243.pdf

Thanks

Kevin

0 JMMN over 8 years ago in reply to KevinZhang123

TI__Guru 60870 points

HI Kevin,

Is it always Port 1 that is damaged? Any damage on other ports? This looks like an electrical overstress to the device, it could be caused by a faulty cable. The TUSB4041 silicon is very similar to the TUSB4020B, I would not expect to see much difference in behavior. Are the ESD and EMI protection devices installed?

Regards,

JMMN

0 JMMN over 8 years ago in reply to JMMN

TI__Guru 60870 points

HI Kevin,

Is there any update from the customer? I reread the thread and you indicated that the damage is always on pin 33 and pin 34, can you confirm no other ports are damaged?

Regards,
JMMN

0 KevinZhang123 over 8 years ago in reply to JMMN

TI__Expert 5750 points

Hi, JMMN

Yes, There is always pin 33 and pin 34 damage. No other port damage. They have add ESD protection on the port.

Here is the ESD diode customer is using. Do you think it is OK?

5481.Bus_Elx_DS__0603ESDA-MLP.pdf

I saw the clamping voltage is 35V and may be too high. I have has customer to try TPD1E1B04.

Any suggestion?

Thanks

Kevin

0 JMMN over 8 years ago in reply to KevinZhang123

TI__Guru 60870 points

Hi Kevin,

Sorry for the delayed response. Yes, you are correct, that is a very high clamping voltage, the TPD1E!B04 is a better choice. Also, they should look for a possible source of the ESD, since it occurs on one specific port. Specifically if there is a noise source or a potential for incorrect connections on that port.

Regards,
JMNN

0 KevinZhang123 over 8 years ago in reply to JMMN

TI__Expert 5750 points

Hi, JMMN

Customer place TPD1E1B04 on DP and DM pins of all 4 downstream ports of TUSB4041I. They use ESD gun to contact discharge 20kV on the D+ and D- pin (which are connected to pin 33 and pin 34) of the USB socket. The pin 33 and pin 34 of TUSB4041I will be broken, but the TPD1E1B04 is still OK.
Is customer's test method correct?
Is there any suggestion to improve the performance?

Thanks
Kevin

0 KevinZhang123 over 8 years ago in reply to JMMN

TI__Expert 5750 points

Hi, JMMN

Customer come across another failure in end-customer's car. Pin 33 and Pin 34 of TUSB4041I short together. As talked before, TUSB4020BI is used in previous project and there is no failure. Customer measure the impedance between downstream D+ / D- pin and GND of TUSB4020BI and TUSB4041. The impedance of TUSB4020BI is about 1.7Mohm, but that of TUSB4041I is only about 500kohm.

Why there is big impedance between TUSB4020BI and TUSB4041I?

Is it cause of this issue? Any work around solution?

The project is already MP and will ramp up soon. It is urgent. Please help on it.

Thanks
Kevin

0 JMMN over 8 years ago in reply to KevinZhang123

TI__Guru 60870 points

Hi Kevin,

The design of the TUSB4020B and TUSB4041 are very similar. I will confirm if there is any difference in the internal ESD structures. Has the failing new failing unit been submitted for FA?

Regards,
JMMN

0 KevinZhang123 over 8 years ago in reply to JMMN

TI__Expert 5750 points

Hi, JMMN

Yes, the failure TUSB4041I has been summitted to FA. Please do help to check the impedance difference of downstream ports between TUSB4041I and TUSB4020BI.

Thanks
Kevin

0 JMMN over 8 years ago in reply to KevinZhang123

TI__Guru 60870 points

HI Kevin,

There will be some differences in the ESD performance of the TUSB4020B vs. the TUSB4041 due to package size, but the designs are similar. Also, I am not seeing the impedance difference between the DP/DM pair on the TUSB4020B va. TUSB4041, in both cases it is ~1.2MOhm. I see that the FA case is open and will be resolved through the standard FA channels.

Regards,
JMMN

0 KevinZhang123 over 8 years ago in reply to JMMN

TI__Expert 5750 points

Hi, JMMN

Customer measured the impedance between D+ pin and GND, and between D- pin and GND. The impedance of TUSB4020BI is about 1.7Mohm, but that of TUSB4041I is only about 500kohm. Could you help to check it?

Here is FA report on the new failure part:

QEM-CCR-1711-01289 FA report.pdf

Thanks

Kevin

0 JMMN over 8 years ago in reply to KevinZhang123

TI__Guru 60870 points

Kevin,

Are they measuring this on the devices or the boards? I see the same impedance values (1M) for both TUSB4041 DP/DM to ground and TUSB4020B DP/DM to ground in my lab.

FA is showing clearing EOS damage which is consistent with a large ESD event or sustained short.

Regards,

JMMN

Interface

Interface forum

TUSB4041I: TUSB4041I Downstream Port Failure Issue