Unstable USB on EVMK2H

Wojciech Piechowski

Other Parts Discussed in Thread: TPS54620

Hello

On my EVMK2H V4 the USB port is unstable. Sometimes it works but usually it does not. When it does not work, it means that the connected device is not enumerated. Sometimes even the presence of the device is not detected. It happens no matter what kind of a device I connect. I did some investigation and I have some observations:

- we have 7 K2H EVM's, USB is working fine only on 3 of them. Some of the failed ones were just unboxed to test the USB, so nothing could have happened to them before

- on one of EVM's (maybe others too, I don't know) it seems like there is a problem with VCC5V. Even when nothing is connected to the USB port, after a few seconds from power-on the VCC5V becomes unstable and gets voltage waves reaching 6V. During these waves VCC5_PGOOD drops and 5V converter gets disabled. I have measured it on TP3 and TP4, scope screenshots available.

over 10 years ago

0 Raja over 10 years ago

TI__Guru* 81335 points

Welcome to the TI E2E forum. I hope you will find many good answers here and in the TI.com documents and in the TI Wiki Pages (for processor issues). Be sure to search those for helpful information and to browse for the questions others may have asked on similar topics (e2e.ti.com). Please read all the links below my signature.

We will get back to you on the above query shortly. Thank you for your patience.

0 Titusrathinaraj Stalin over 10 years ago

TI__Guru** 116100 points

Hi,
Could you share the EVM board revision of all the 7 boards.

0 Wojciech Piechowski over 10 years ago in reply to Titusrathinaraj Stalin

Prodigy 60 points

all of them are REV 40 and were bought in a single order

0 David (ASIC) Liu over 10 years ago in reply to Wojciech Piechowski

TI__Guru**** 176281 points

Hi, Wojciech

Have you compared the VCC5 behavior between the good board and the bad board? Do you have the schematic of the EVMK2H? If this is an issue associated with VCC5, we will need to take a look at the components associated with TPS54620 that provides the VCC5V power. Can you please send one of the board back for me to debug as well?

Thanks

David

0 Wojciech Piechowski over 10 years ago in reply to David (ASIC) Liu

Prodigy 60 points

Problem solved!

VCC5 has 3 caps on schematics: C399, C39 and C403. On all our boards only C39 was populated, two others were missing. We solderd C399 on all boards and now USB works on all of them.

0 Titusrathinaraj Stalin over 10 years ago in reply to Wojciech Piechowski

TI__Guru** 116100 points

Hi Wojciech,
Sounds good.
Thanks for your update.

0 Raja over 10 years ago in reply to Wojciech Piechowski

TI__Guru* 81335 points

Thank you for the update.

I have an question,
Have you de-populated those caps or factory out EVM's missing those caps?

0 Wojciech Piechowski over 10 years ago in reply to Raja

Prodigy 60 points

The caps were unpopulated on brand new boards, we did not desolder anything

0 Tom Johnson 16214 over 10 years ago in reply to Wojciech Piechowski

TI__Mastermind 46460 points

Wojciech,

Please confirm that your new rev4.0 EVMs only contained C39 and did not contain C399 or C403. The schematic shows that C399 is supposed to be installed and C39 and c403 are the NL parts. C399 is a 100uf bulk capacitor and I would expect it is required. If you confirm this, I will ask our assembler about your observations.

Tom

0 Morganne Lerch over 10 years ago in reply to Wojciech Piechowski

Prodigy 10 points

Hi Wojciech,
I'm having the exact same problem that you described with one of my two boards. However, I don't see any difference between the two boards with the capacitors you've named - it doesn't seem to be a problem with C399 for me. Did you come across anything while troubleshooting your boards that might point to a different solution?
Thanks,
Morganne

0 Tom Johnson 16214 over 10 years ago in reply to Morganne Lerch

TI__Mastermind 46460 points

Morganne,

Did you install all 3 caps on your board? Does it still fail after making sure all 3 are installed?

Tom

0 Victor Prieto over 10 years ago in reply to Tom Johnson 16214

Prodigy 30 points

Hi Tom,

I'm working on the same team as Morganne. We are ECE students at Rice, and may not have the proper equipment to add the components. Heating up the solder on the pads with a soldering iron did not seem to have any effect on the solder itself. We also tried to solder the necessary components with other solder on top of the solder itself and had no luck (it would not stick).

To add more information about our problem, we measured the voltage across the pads with a multimeter and found that the voltage on our good model stayed at a nice 5.07V but that the defective board would start at 5.02V and subsequently (within a matter of 10s) drop down to 0V. Would this be indicative of insufficient capacitance?

Thank you,

Victor (AKA Forty-two)

0 Victor Prieto over 10 years ago in reply to Victor Prieto

Prodigy 30 points

To clarify further, the only capacitor in both boards is C399, the ceramic 100 uF cap.

0 Tom Johnson 16214 over 10 years ago in reply to Victor Prieto

TI__Mastermind 46460 points

Victor, Morganne,

The simple solution is explained in this tread. You will need to find someone with proper soldering skill and equipment to add these capacitors.

The problem is power supply stability on the 5V supply required for USB operation. It has almost no load at power-on. Unfortunately, this makes it overshoot. The over-voltage protection detects the overshoot and then shuts down the 5V power supply.

You might also dampen the power supply overshoot by adding a load resistor like 100 ohms across the 5V supply. It will need to be a 1/4W resistor to prevent overheating.

Tom

0 Victor Prieto over 10 years ago in reply to Tom Johnson 16214

Prodigy 30 points

Hi Tom,

After adding liberal flux and using extra solder (the solder would not yield) we were able to form reasonable connections. Having made these, the supply works as intended! Thank you for your extra suggestion about a load resistor. Thankfully adding the missing two capacitors sufficed.

Thanks again,

Forty-two

0 Tom Johnson 16214 over 9 years ago in reply to Victor Prieto

TI__Mastermind 46460 points

The root cause of this problem is an unstable VCC5 supply at power-up since it has no load at that time. We have not found any missing components on the boards that we examined. The ideal solution is addition of a resistor load on VCC5. We have been installing 510 ohm 0805 resistors on the pads of C403 on the backside.

Tom

Processors

Processors forum

Unstable USB on EVMK2H