TUSB2046I: USB HUB based on TI USB 2046i Not Detected by computer

Aashiv Vij

Part Number: TUSB2046I

I am implementing a USB HUB using the schematic in datasheet without an EEPROM. I cannot connect this USB to a laptop, one day I got windows notification that USB Malfunctioned, but now I does not appear at all.

Below is the schematic attached as well.

Below is the operating voltages of USB Hub in circuit.

1	3.3	DP0
2	0	DM0
3	3.3	VCC
4	3.3	Rest_L
5	0	EECLK
6	0	EEDATA
7	0	GND
8	0	BUSPWR
9	3.3	PWRON1
10	0	OVRC_L_1
11	0	DM1
12	0	DP1
13	3.3	PWRON2
14	0	OVRC_L_2
15	0	DM2
16	0	DP2
17	3.3	PWRON3
18	0	OVRC_L_3
19	0	DM3
20	0	DP3
21	3.3	PWRON4
22	0	OVRC_L_4
23	0	DM4
24	0	DP4
25	3.3	VCC
26	3.3	EXTMEM
27	0	TSTPLL
28	0	GND
29	3.3	XTAL2
30	0	XTAL1
31	0	TSTMODE
32	0	SUSPND

over 3 years ago

0 Brian Zhou over 3 years ago

TI__Guru**** 161526 points

Aashiv:

please try these procedures

1: add 0.1uf cap on reset pin.

2: add 1.5k PU on DP0 to 3.3v

3: add 15k PD to all DP/DM of downstream ports.

Best

Brian

0 Aashiv Vij over 3 years ago in reply to Brian Zhou

Prodigy 40 points

Reset line is tied to VCC which already has a 0.1uF capacitor, Are you recommending another capacitor?

DP0, I have added the 1.5k, pullup to 3.3V

Next, will add the 15K PD to both data lines and test.

0 Brian Zhou over 3 years ago in reply to Aashiv Vij

TI__Guru**** 161526 points

Generally, a reset with a pulse width between 100 μs and 1 ms is recommended after 3.3-V VCC reaches its 90%. Clock signal has to be active during the last 60 μs of the reset window.

Can you take waveeform between VCC and reset and see if you can meet this requirement?

Regards

Brian

0 Aashiv Vij over 3 years ago in reply to Brian Zhou

Prodigy 40 points

The reset and VCC are tied together, there is a 0.1uF capacitor for Vcc. Is that not correct?

0 Aashiv Vij over 3 years ago in reply to Aashiv Vij

Prodigy 40 points

AS soon as I connect to the computer now, I see a popup saying Power Surge on USB. My design is not USB power driven, I dont use 5V from USB (laptop). So not sure if that message is related to something else.

Also, the USB Hub is also connected to TPS2044D, and I see that the Overcurrent pins are low (Active Low), and PowerON are High (Active Low).

Is it possible that downstream overcurrent shows as a Power Surge on input side when connected to laptop?

0 Brian Zhou over 3 years ago in reply to Aashiv Vij

TI__Guru**** 161526 points

It's possible, can you try per-port overcurrent mode and see?

Best

Brian

0 Aashiv Vij over 3 years ago in reply to Brian Zhou

Prodigy 40 points

All ports currently are overloaded for some reason, and I am operating per-port mode. Its not GANGED.

Please check the schematic attached for connections around TPS2044D

0 Brian Zhou over 2 years ago in reply to Aashiv Vij

TI__Guru**** 161526 points

let's have a webex call.

0 Brian Zhou over 2 years ago in reply to Brian Zhou

TI__Guru**** 161526 points

Hi Brian, David,

Thanks again for the time this afternoon, please see the waveform requested below.

Looks like there is a 200ms delay to bring up the clock.

The yellow is the reset line, which has a rise time of 40ms now, with 15k and 1uF Capacitor.

The pink line is the clock.

The blue line is the Vcc.

0 Bennett Wilson over 2 years ago in reply to Brian Zhou

TI__Prodigy 390 points

Hi Brian,

The suggested solution did worked. Adding the time constant for the reset line fixed the usb communication issue.

Aashiv is now trying to understand the clock if setup correctly.

The measured frequency is very low, probably because of the capacitance in the probes. Is there another way to measure the crystal frequency accurately?