STM32F103C8T6 always shows “SLEEP” in CubeProgrammer, cannot perform full chip erase—no user code loaded

[Melih--]

Hi everyone,
I'm working with an STM32F103C8T6. Even though I haven't programmed any code onto it myself, when I connect with STM32CubeProgrammer, the Registers panel immediately shows “SLEEP” (while on other ST boards it would show “HALTED”). Because of this, I can't perform a Full Chip Erase — the option appears but doesn't work.

Here’s what I’ve tried:

• Connected only SWDIO, SWCLK, GND, and 3.3 V (no NRST).

• CubeProgrammer shows “SLEEP,” and erase fails.

• On other boards, “Connect under reset” works fine and allows erasing.

Possible issues I suspect:

• No NRST pin connection → ST-Link can’t halt the MCU before it runs.

• Option Bytes might have RDP (Read-Out Protection) enabled.

• BOOT0 pin wrongly tied (stuck in System Memory mode).

• Some factory test code or previous user code (making MCU go to sleep on reset).

What I tried from forum suggestions:

• Tried “Connect under reset” with NRST — this works on other boards but seems not to work here yet.

Any ideas how to force-erase or regain access?
I’d appreciate any advice—including Option Bytes configuration, hardware wiring, or CubeProgrammer settings.
Thanks!

[AScha.3]

Hi,

What st-link you use?

And set connect; software reset, and enable low power modes.

Try to connect, then show, what Cube programmer shows, screen shot.

[Tesla DeLorean]

Connect Under Reset from the debug pod expects NRST to be connected to the Target MCU

You can use the BOOTx pins to cause a power-up / reset into System Memory rather than a user program in FLASH sleeping the device or otherwise breaking the debug interface.

Consider have a delaying loop in System Reset to allow debugger time to connect initially

[Melih--]

ı use stm32f446re nucleo board as stlink,ı can not connect stmf103 with cubeide ,only with cubeprogrammer in hot plug mode,

this is the screenshoot ,only see stm in sleep mode , ı try my nucleo board and ı see halted,is there any chance my stm32f103c8t6 is clone 

[AScha.3]

>is there any chance my stm32f103c8t6 is clone 

Everything could be a "clone" , but it is very unlikely , as cpu ID (and revision) show same, as on a F103 here:

So connect with :  software reset, nrst not needed;

Then look in OptionBytes:   rdp not set:

next see  in OptionBytes: all (!) sectors have write protection OFF (= checked).

down to 31:

Then full/mass erase should work:

Try.

[Melih--]

hello again

so when ı checked all option bytes click apply programmer stop working 

 

 

these are screenshoot when ı try to full erase chip 

 

[AScha.3]

And you looked at all sectors in OptionBytes  ?

: all (!) sectors have write protection OFF (= checked).

btw

I also got some new chips from distributor, that had some write protected sectors - no full erase possible.

Until you set all to unprotected.