cancel
Showing results for 
Search instead for 
Did you mean: 

fmc operations cause mcu reset repeatly

JJ.Lin
Associate III

Hi everyone,

I met a problem now. I am developing display product using stm32f429zet6, the lib is f41.26.1. In my project, nand flash is used to storage images and fonts, the sdram is used to cache images. Fatfs work upon nand. Obviously nand and sdram use fmc interface. When mcu start, I copy the images from nand to sdram before ui task stark. But in this operation, mcu would reset repeatly reset some times. Times is not fixed. Sometimes is 3 while sometimes is more, but sometimes it works well without reset. There is not mcu reset function in the copy process. I can't express why.

/JJ.Lin

6 REPLIES 6

Scope the power supply rail vs the NRST pin. Might also look for interconnected nets, inconsistent grounding, or circuitous routing of power

Any watchdog running?

Tips, Buy me a coffee, or three.. PayPal Venmo
Up vote any posts that you find helpful, it shows what's working..
TDK
Guru

Determine the cause of the reset by looking at the RCC->CSR register.

If you feel a post has answered your question, please click "Accept as Solution".
JJ.Lin
Associate III

Thank you a lot! If it works well the CSR register show that it is reset by soft reset as is designed, but unexpected reset due to POR/PDR (power on/ down) reset. How can NAND flash or SDRAM read/wirte cause power reset? So how can I do next ?

> How can NAND flash or SDRAM read/wirte cause power reset? So how can I do next ?

They use power, perhaps they are using more than you have available, causing the voltage to drop, causing a reset. Monitor the rail with a scope. Create a power budget to ensure you can power everything off of the LDO you're using (or whatever power generation you have).

If you feel a post has answered your question, please click "Accept as Solution".

Hello,

I scoped the NRST pin and found that the NRST was pulled down some times when operating SDRAM.This is based on the BOR ( brownout reset ) level is off and the PDR_ON pin is float in the air.If I set the level to any level of 1-3, it works well without reset when operating SDRAM. So I guess the SDRAM operation cause the power undervoltage.Is it reasonable? What I don't understand is that only some of boards have this problem in the same batch.

Thank you very much!

Is this a two layer board?​

C​ould be some overly thin traces or poor vias related to plating or construction.

Lack of bulk capacitance?

A​ review of the schematic and board might be more productive.

Tips, Buy me a coffee, or three.. PayPal Venmo
Up vote any posts that you find helpful, it shows what's working..