TPS25980: TPS259807O Retry Delay Problem

Hi Kerem,

What are you getting at output when 5V rail is not performing the 5second delay. Can you probe Vin, Vout, Input current for scope shots. Check if it is stable at 5V, if it drops below 4.48V then it will basically reset the EN pin which will turn the device on again. You can do this for 12V rail also, so we have some data for comparison. 

Best Regards,

Arush

Hi Arush,
Thank you for your message.
When the eFuse does not perform the 5second delay, it starts conducting after the short circuit is released.
I get the power from a PC power supply, so I don't expect voltage dropping on the rail.
And I forgot to mention that I shorted the EN pin to the 5V rail. So the the UV of 4.48V threshold does not apply.
For testing, I short out output pins to ground. The main task of the circuit is to avoid any accidental short circuits.

It looks strange to me that I see oscillation on the RETRY_DLY pin sometimes with the 5V eFuse, but always for the 12V eFuse.
Could there be something else that I am missing?

Best regards

Kerem 

Hi Kerem,

Can you provide me with some waveform of Vin, Vout, Input current (if possible). When you create a short at output, the inrush current can be huge and will cause drop in power supply voltage. Let's try this please. I want to first eliminate other possibilities before thinking about the oscillations you mentioned.

You also mentioned that 5V sometimes works. I feel that when do manual hard short, it is not same during every try. Sometimes the supply drop is less, sometimes it is more.

Best Regards,

Arush 

Hi Arush,
I checked the voltage drop on short circuit event. 
You are correct the voltage drop is quite large. I can not provide any unfortunately.
Then I added a 22uF cap at the input terminal.
The following is the 5V input and the PG output. the DC output is the RETRY_DLP output.

Best regards

Kerem

Hi Kerem,

• Please label the waveforms. Can you tell me color to signal legend in the above waveform.
• Do you any other power supply. You can try using external power supply at EN to prevent power reset and you should get 5s delay after this.
  • This will remove UVLO but you will still get UVP (under voltage protection at around 2.42-2.53V)
  • 
• What is the end application. At what stage of development are we at?

Best Regards,

Arush

Hi Kerem,

Were you able to resolve the issue by using external power supply for EN pin? 

Please reply here if any further questions. 

Best regards,

Arush

Hi Arush,
I am sorry for the late reply. Thank you very much for your message.
Our application is a test measurement equipment. And the power from the PC power supply is distributed via a non functional PCIe (no PCIe connections but is plugged into the PCIe slot) board to the whole test measurement system.
Sometimes the power connector coming from this PCIe board is accidentally short circuited.
The eFuse that I use serves this purpose.

On the wafeform image:
The yellow trace is Power Good
Orange trace: the input Voltage. Therefore ENABLE pin. I shorted EN pin to VIN.
Green is the REPLY_DLY pin.

When I look at the traces the 5V power rail does not drop at all below EN pins threshold.
This makes me believe an issue not related to ENABLE pin dropping too low.
As I mentioned the ENABLE pin is directly connected to the power rail 5V.

Unfortunately I can not use any other power supply, because there are 2 power rails 12V and 5V and I need to protect both from short circuits from downstream devices.
I need to connect the ENABLE pins to the power rails directly.
Kind regards

Kerem

Hi Kerem,

I will get back to you regarding this by Friday.

Best Regards,

Arush

Hi Kerem,

I have few questions and observations.

• Why is your input/enable signal (Orange) clamped at some level lower than 5V before the trigger and then going back to 5V. What was the test you did for this waveform capture. 
• Your PG (yellow) signal is going low, so that means issue is in Vin.
• Can you please provide Vin and Vout in the waveforms. RTRY_DELY is not helping much. 
• Can you try probing IMON so that we can figure out the current.

Best Regards,

Arush

Hi Arush,
Thank you very much for your email.
I tried to capture the waveforms that you requested.
This board has a DSub connector, I connect a 1 meter long cable to this board. And at the other side of the cable I short the 5V rail to GND.
The EN pin is again connected to the 5V rail directly.

Yellow: Vin
Green: PG
Orange: IMON
Purple: VOUT

I connected a 47u and 10u Electrolic cap on the power rail input connector. And this is really close the eFuse Chip.
IMON output shows different behaviours like this:

It is sometimes quite high, sometimes low on short circuit trigger.

Hi Kerem,

I will reply back to you by Thursday EOD. I have multiple bench testing which are taking all of my time for now. 

Thank you for understanding,

Arush

Hi Arush,
Happy new year. I haven't heard from you related to the problem that I was having.
Were you able to look at the issue?
Kind regards

Dear Arush,
Thank you very much for your message and all your efforts in providing a solution to my problem.
The PCB is located inside the PC and it gets power directly from the PSU of the PC with a relatively long cable around 30cm.
To be honest, I didn't expect cable length to create a problem.
I will follow your advice and add relevant parts to my circuit for both rails +12V and +5V.
I hope this resolves my problem.
Best regards

Kerem