WARMRSTN Stuck in pull down

[Muhammad_Aladdin]

Dear all,
We have designed a single board computer based on OSD3358-512M-BSM chip and according to schematic check list “https://octavosystems.com/app_notes/osd335x-schematic-checklist/#_Toc382081429” and Octavo consultation, we have the following circuit for the reset function in img1. After manufacturing the board, it was functioning just fine at the beginning, though after couple successful operations, the board stopped responding to the SD card, UART debug interface or any processing function. As far as we know now it has a problem with the reset pin, for the power rails we tested the voltage level for the LDO 3.3V and 1.8V VDDRTC and they were functioning just fine, also the PMIC_PGOOD and PWRONRSTN node had 1.8V on it. The WARMRSTN pin has zero voltage level on it without connecting anything! what was so odd as the internal pull up should keep it at 3.3V. We are not sure if it’s the rest issue causing the damage or the reset is just a side effect of another issue. However we don’t know the main cause the leaded to the reset behavior.

Attachments:

1. IMG.zip

[Muhammad_Aladdin]

Hello,
I need to mention that we didn’t add the clamping circuit, so I wounder if that my be a possible cause?!

[Neeraj Dantu]

Hi Alladin,

If PMIC_PGOOD and PWRONRSTN are both at 1.8V, it means that the PMIC has successfully brought up all the power rails(See Figure 6 in http://www.ti.com/lit/ug/slvu551i/slvu551i.pdf). The output of the buffer IC C1(RESET#) should be 3.3V(pull up). If you are seeing a 0V on this signal, it could mean that the buffer IC C1 is at fault. If C1 is faulty, replacing it should fix the problem. Another possible reason would be a short on the RESET# signal to GND (caused by layout).

Also, if R4 is not populated, it needs to be populated as the output of the buffer is open drain and WARMRSTN pin has a weak pull-up.

Let us know if that helps and if you have more questions.

Best,

Neeraj

[Muhammad_Aladdin]

Dear Neeraj,
Thanks for your response, we have checked for the #RESET short with ground and it’s not shorted, otherwise the pin should be internally pulled up, so ever if the buffer is damaged I should find 3.3V on it, though there is 0V level on it, on the other hand we have removed the buffer IC totally from the circuit and measured the voltage on the reset pin and still 0V! we doubt that it’s damaged internally, though we are not sure of the reason that my lead to such an issue.

Regards,
Aladdin

[Neeraj Dantu]

Hi Alladin,

Can you check whether both your crystal oscillator(24MHz) as per this post: https://e2e.ti.com/support/processors/f/791/p/404843/1439658. Also regardless of internal pull-up, can you populate R4 and see if that fixes the issue?

Do you have more than 1 board that is exhibiting this behavior?

Best,

Neeraj

[Muhammad_Aladdin]

Hello Neeraj,
We don’t use th RTC crystal as we don’t need the RTC, also this is the same in beaglepocket board. We have 2 boards with the same issue. We tried several times with R4 and without it and still the same!

Regards,
Aladdin

[Neeraj Dantu]

Hi Alladin,,

Just to make sure, you have verified 24MHz oscillating crystal on the main OSC0 clock input?

As the issue is persisting, please send us your schematic and layout files for us for further debug. You can send them to our Business Development Manager at martin.burgos(at)octavosystems.com. You have communicated with him before.

Best,

Neeraj

[AlexO]

We have same issue with our board.
Did you found reason that cause this problem?
Can some one help me?

[Neeraj Dantu]

Hey Alex,

Can you give us more information on the issue?

1. Do all the power rails have the correct voltage? Please verify the voltages and power sequencing as described in the power application note: https://octavosystems.com/app_notes/osd335x-sm-power-application-note/. Does WARMRSTN go HIGH at any point during power up?

2. Do you have the 24MHz(OSC0) and 32KHz RTC(OSC1) clocks functioning?

3. Are there other components is the WARMRSTN signal connected to that could be driving this signal LOW?

Best,

Neeraj

[Author]

Posts