Genuine STLINK-V3MINIE doesn't work, but clone STLINK-V2 works

[Chupacabras]

I'm trying to make genuine STLINK-V3MINIE to work, but without success.

The target device is STM32F103C8T6.

VCC is connected to T_VCC (pin 3 of STDC14)

SWCLK is connected to T_SWCLK (pin 6 of STDC14)

SWDIO is connected to T_SWDIO (pin 4 of STDC14)

RESET is connected to T_NRST (pin 12 of STDC14)

GND is connected to GND and GNDDetect (pins 5, 7, 11 of STDC14)

STM32CubeProgrammer is properly detecting STLINK-V3MINIE, even shows voltage of the target.

But when I try to connect to target, it gives message:

Error: No STM32 target found! If your product embeds Debug Authentication, please perform a discovery using Debug Authentication

 

This is text from console:

 12:56:46:283 : UR connection mode is defined with the HWrst reset mode
 12:56:46:837 : ST-LINK SN : [redacted]
 12:56:46:838 : ST-LINK FW : V3J16M7
 12:56:46:838 : Board : STLINK-V3MINIE
 12:56:46:838 : Voltage : 3.29V
 12:56:46:839 : Error: No STM32 target found! If your product embeds Debug Authentication, please perform a discovery using Debug Authentication
 12:56:46:876 : Disconnected

 

I tried another PC, but with the same result.

 

When I connect clone STLINK-V2 instead of genuine STLINK-V3MINIE, I'm able to connect to target MCU, and download program to target.

This is the output from console:

 13:00:39:502 : UR connection mode is defined with the HWrst reset mode
 13:00:39:667 : ST-LINK SN : [redacted]
 13:00:39:667 : ST-LINK FW : V2J45S7
 13:00:39:667 : Board : --
 13:00:39:667 : Voltage : 3.18V
 13:00:39:667 : SWD freq : 4000 KHz
 13:00:39:667 : Connect mode: Normal
 13:00:39:667 : Reset mode : Hardware reset
 13:00:39:668 : Device ID : 0x410
 13:00:39:668 : Revision ID : --
 13:00:39:668 : flash loader C:\Program Files\STMicroelectronics\STM32Cube\STM32CubeProgrammer\bin/FlashLoader/0x410.stldr is loaded
 13:00:39:668 : Debug in Low Power mode is not supported for this device.
 13:00:39:668 : Reading data...
 13:00:39:668 : r ap 0 @0x1FFFF7E0 0x00000004 bytes Data 0x00140040
 13:00:39:957 : UPLOADING OPTION BYTES DATA ...
 13:00:39:957 : Bank : 0x00
 13:00:39:957 : Address : 0x4002201c
 13:00:39:957 : Size : 8 Bytes
 13:00:39:958 : Reading data...
 13:00:39:958 : r ap 0 @0x4002201C 0x00000008 bytes Data 0x03FFFFFC
 13:00:39:959 : Bank : 0x01
 13:00:39:959 : Address : 0x1ffff800
 13:00:39:959 : Size : 16 Bytes
 13:00:39:959 : Reading data...
 13:00:39:959 : r ap 0 @0x1FFFF800 0x00000010 bytes Data 0xFFFF5AA5
 13:00:39:960 : UPLOADING OPTION BYTES DATA ...
 13:00:39:960 : Bank : 0x00
 13:00:39:960 : Address : 0x4002201c
 13:00:39:960 : Size : 8 Bytes
 13:00:39:960 : Reading data...
 13:00:39:961 : r ap 0 @0x4002201C 0x00000008 bytes Data 0x03FFFFFC
 13:00:39:961 : Bank : 0x01
 13:00:39:961 : Address : 0x1ffff800
 13:00:39:961 : Size : 16 Bytes
 13:00:39:962 : Reading data...
 13:00:39:962 : r ap 0 @0x1FFFF800 0x00000010 bytes Data 0xFFFF5AA5
 13:00:39:962 : UPLOADING ...
 13:00:39:962 : Size : 1024 Bytes
 13:00:39:963 : Address : 0x8000000
 13:00:39:963 : Read progress:
 13:00:39:963 : Reading data...
 13:00:39:964 : r ap 0 @0x08000000 0x00000400 bytes Data 0x20005000
 13:00:39:964 : Data read successfully
 13:00:39:964 : Time elapsed during the read operation is: 00:00:00.007

 

What is the problem with STLINK-V3MINIE?

How to make it work?

It's pretty frustrating that clone works and genuine debugger not.

 

[gbm]

Check the setting of SWD clock frequency, set it to 1 MHz or less,

[Chupacabras]

It was on default frequency, 8000 kHz.

I changed it to 1000 kHz, or 200 kHz.

Still the same problem.

[AScha.3]

Try with setting: connect mode normal + software reset ; (nrst not needed)

if V3miniE not finding target, you puzzled the swd/swc lines, or no contact on a line to target.

Or (too) long wires to target - V3 is much more sensible for stray-in , no reliable connection then.

I tried all : the V2 ("clones" , with STM32 cpu, but even with GD32 or Geany cpu ), V3MiniE, V3mods , J-link.

All working fine.

Just dont forget: the V2-sticks give 3v3 power out, the original STM not : only 3v3 in , to check target supply.

[Chupacabras]

I tried "normal mode + software reset", no effect.

I have checked continuity directly from the pins of MCU to pins on STLINK-V3MINIE connector. All good.

Wires are 32cm long (~ 1 foot).

Board is powered by its power supply. I know STLINK-V3MINIE doesn't provide power.

[AScha.3]

Ok, so try : to your target no other connection to mains or mains ground ;

also supply for target from PC (USB) or if needs extra supply from mains: connect it to same outlet, as PC is connected. And cables close together, to avoid any magnetic loop.

[Carl_G]

Settings look good. Connections Look good. I'd try a different reset mode as suggested. I can't see your schematic so I don't know how this is connected to the MCU and what's on the reset line.

[Chupacabras]

I have added full schematic (PDF). I could upload even kicad project if necessary.

There's only a button on reset line.

[Carl_G]

If the knock off works the genuine should definitely work in the same configuration. Unless there is something wrong with the connectivity between the STLINK and the MCU. Have you tested that V3MINNIE on any other boards?

On my board I put a 100n to GND on the reset line as well. ST does that on their demo boards too. But on my board without it, it works fine.

Check your actual board pinout. Make sure the pins are going where you think they should be on the MINNIE.

[Andrew Neil]

---

@Chupacabras wrote:

The target device is STM32F103C8T6.

---

Are you sure that this device is genuine?

Does the V3MINIE work with any other targets?

[Chupacabras]

I'm not sure about origin of that device.

Is it possible to verify genuineness of that device?

I would expect different message in CubeProgrammer if it was counterfeit part.

 

I tried 2 different boards/targets.

1. genuine STM32L031F6P6, I was able to connect with V3MINIE

2. bluepill from aliexpress (with STM32F103). I do not know whether it has genuine or clone STM32. I was able to connect with V3MINIE

 

[Andrew Neil]

---

@Chupacabras wrote:

I'm not sure about origin of that device.

Is it possible to verify genuineness of that device?

---

You could try: https://mecrisp-stellaris-folkdoc.sourceforge.io/bluepill-diags-v1.640.html 

via: https://community.st.com/t5/others-stm32-mcus-related/black-pill-debugging-stm32f103rct6-in-keil-fails-to-enter-main/td-p/789441

 

---

@Chupacabras wrote:

2. bluepill from aliexpress  

---

So vanishingly small chance of genuine STM32

[Carl_G]

Talking with the reset line asserted isn't going to work. Seems like the Minnie has a broken reset line or you had it on the wrong reset setting. Try other reset settings and see if the line behavior changes. If not looks like reset is shorted on the Minnie.

[Carl_G]

Interesting. So now it looks like the MCU isn't genuine? Even so, why would the debugger refuse to talk to it? So the static GND value isn't really stuck at 1 value?

[Andrew Neil]

---

@Carl_G wrote:

So now it looks like the MCU isn't genuine? Even so, why would the debugger refuse to talk to it? 

---

• ST might choose to have their software not connect to non-genuine parts.
• The clone might be marginal - so it just manages to connect to one debug probe, but fails to connect to another.

 

That's the thing with fakes - you have no idea what you're getting!

[Carl_G]

But ST isn't the only user of ARM Cortex cores so it makes sense not to connect to a different M3.

What made you think this was an ST Chip? Was the number STM32F103C8T6 on the chip or package somewhere?