APT32F172K8T6 - Chinese 32-bit MCU recently seen in Transistor Testers

[elecdonia]

Who can educate me about the APT32F172K8T6 MCU?

This MCU chip has recently turned up in “ Transistor Tester “ aka “ Component Tester “ devices.
     https://www.ebay.com/itm/275434129848

For more than 10 years the Transistor Tester OSHW project has relied almost exclusively on one Atmel MCU: The ATmega328. Yes, this is the same MCU found on many Arduino boards and clones.

If you aren’t yet familiar with the Transistor Tester Project, then please visit this gigantic (324 pages!) EEVblog topic:
     https://www.eevblog.com/forum/testgear/$20-lcr-esr-transistor-checker-project/

Recently many Asian sources are selling Transistor Tester units with the APT32F172K8T6 MCU substituted for the Atmel ATmega328.

But these vendors aren’t disclosing the substitute MCU. In fact some APT32F172K8T6 MCU chips found in these Transistor Tester units have been expertly and deviously relabeled as Atmel ATmega328.

But the only thing in common between these two MCU chips is they are housed in 32-pin TQFP packages. Their pinouts, including the location of their ground and VCC pins, are totally different. The ATmega328 is a very popular 8-bit MCU which has been in production for many years. The APT32F172K8T6 has a 32-bit architecture and seems to have been in mass production for only a few years. The chip manufacturer, APT also known as Shenzhen Aipute Microelectronics Co. Ltd. has a website:
     http://en.aptchip.com/

Their product datasheets are in Chinese language only. They do not appear to be seeking business in USA/Canada/UK.However, information about the APT32F172K8T6 does appear at JLCPCB:
     https://jlcpcb.com/partdetail/Aptchip-APT32F172K8T6/C448572

Not surprisingly these two MCU chips are totally incompatible with each other. None of the huge body of existing open-source Transistor Tester application software will run on the APT32F172K8T6. Rather the manufacturers of these Transistor Tester and Component Tester products did a cheap-and-dirty proprietary port of the existing open-source Transistor Tester software. This was done without any contact whatsoever with the large community of folks who are active users and contributors for the OSHW Transistor Tester Project. This is causing great consternation among hobbyists and electronics enthusiasts who purchase a new Transistor Tester with the expectation that it will perform as expected.

I’m really curious to see if the Asian manufacturers will soon start offering clone Arduino boards containing the APT32F172K8T6 MCU. In order for this to happen a bootloader and the core Arduino software must be created and introduced into the Arduino tool chain. I don’t think this has happened yet.

So I’m very interested in hearing from anyone who has actually used the APT32F172K8T6 in a project.

[bingo600]

IMHO the name is a bit of a giveaway  - The 32F172 "smells of a STM32 ARM lookalike"

A quick glance in the DS seems to agree - (i don'r read chinese) , see attached

Does the RiscV have a NVIC ??


/Bingo

[PCB.Wiz]

IMHO the name is a bit of a giveaway  - The 32F172 "smells of a STM32 clone/lookalike"

Well, it's 32 bit, but the core is not Cortex, but a 3rd tier Chinese 32 bit core from C-Sky Microsystems Co., Ltd

It may be better to look at RISC V cores  ( or maybe Cortex) , as the tool support is likely to be more solid ?
eg the CH32V203 series from WCH, comes in TSSOP20 and LQFP32, tho that variant is not wide Vcc.
The ideal part here would be wide Vcc in QFP32.

Something like PY32F030K18T6 ? - 48MHz 64kF in LQFP32 and 1.7~5.5V  and 12b ADC 
 

[GromBeestje]

NVIC? Why would there be a ARM interrupt controller in a RISCV chip? RISCV has its own interrupt controller, and there exist implementation similar to NVIC (eg ECLIC mode on the GD32VF103)

But as you point as SWD, I guess you're thinking about a CoreSight DAP. While the RISCV Debug Specification specify how to implement debugging over JTAG, it should be possible to implement a RISC-V behind a CoreSight DAP.

However, be aware some Chinese manufacturers refer to  their proprietary 2-wire debugging protocol as "SWD" (eg. WinChipHead CH32V series)

[SiliconWizard]

NVIC? Why would there be a ARM interrupt controller in a RISCV chip? RISCV has its own interrupt controller, and there exist implementation similar to NVIC (eg ECLIC mode on the GD32VF103)

RISC-V doesn't define much in terms of an interrupt controller. It's extremely barebones and as such, always needs writing a specific controller which is not specified in any RISC-V document.
There have been some example implementations, such as what SiFive has done, but they are not part of RISC-V specifications. Yet. To be fair, the PLIC specs are underway (AFAIK) but not even ratified yet - why would you expect any vendor to use them at this point rather than come up with their own stuff?

And even it if were, RISC-V is just a set of specs, not any kind of implementation.

So all this to say that if you want to design a RISC-V-based MCU, you need to design your own controller (or copy one). To that end, why not make a clone of NVIC if that fits the bill? I might have a different opinion when the PLIC spec is ratified and if it looks complete enough to be a good fit. At this point I personally don't know, and I happen to have designed a RISC-V core and my own interrupt controller completely custom. I'm not surprised that vendors would do that as well.