Blank MKL28Z / no FW - what internal HW consumes power?

取消
显示结果 
显示  仅  | 搜索替代 
您的意思是: 

Blank MKL28Z / no FW - what internal HW consumes power?

1,119 次查看
mci
Contributor III

Hi,

As the title/question asks - even before there's any FW that will "intentionally" put the MKL28Z to VLLS0 lowest leakage current stop mode through firmware setup - does anyone know which among the internal HW of the MCU is "turned-on" and consuming any power?    We have a backup battery circuitry in our hardware design and for some reason the MCU is drawing some current which consumes the battery. 

Our target is, if flashed with FW to put it in VLLS0 if there's battery backup to draw current at 331 nA or less if I turn off even more.

 I understand if not flashed, there is base ROM bootloader and of course all the 90+ GPIO pins that are connected to external circuitry that may be part of the power consumption there.

But does anyone know which INTERNAL peripheral(s) inside the MCU is running without flashed firmware? 

And in what power mode of sort will it be?

Thanks,

MI

标签 (1)
0 项奖励
回复
3 回复数

999 次查看
mjbcswitzerland
Specialist V

Hi

If the ROM loader is operating when there is no application loaded you can see which of the peripherals (and ports) are enabled in the ROM loader description in the user's manual.

There are some ROM loader fields that can be configured in Flash that configure which of the loader peripherals are used, but this would only be valid when there is something loaded in flash to control it. I expect the ROM loader is always in RUN mode but don't know for sure.

You can also disable the ROM loader in the flash configuration but then, if no application loaded, the processor will be continuously booting at about 20MHz CPU speed and resetting again, but with no peripherals enabled. If you are discussing the Kinetis state when un-programmed in production with battery connected the ROM loader will not be controllable since there are no flash parameters set just yet.

It may make sense if you manufacture with battery immediately connected to the board to immediately program the chips, or to pre-program them with the FW to avoid current drain.

Regards

Mark

0 项奖励
回复

999 次查看
mci
Contributor III

Yes, Mark.   What you say makes sense.   Yes it is advisable to only put on the battery, plug in to outlet to power-up and then flash the FW to program the MCU. Then enter manufacturing test right after programming.

There are a few pin ports that may be pulled-up high, for example, the PTA4 which is currently connected to an LED.  For some reason, with or without FW, that LED is constantly on which means it is drawing current from the MCU, but this is dependent on the new circuit design.

It seems that some GPIO's "have a life of their own" (no one has pointed out to me which internal peripheral module they are connected to could do this) based on that PTA4 LED remaining ON even though application FW did NOT do that.

So I have to plan ahead which among all the GPIO and internal peripheral modules needs to right away be turned OFF or disabled right after manufacturing test has used them.

I guess I have to spend some time studying your so-called 16-byte "flash configuration parameters" which may be the key to "controlling" how the MCU HW will be setup at the very beginning.

Thanks,

MI

0 项奖励
回复

999 次查看
mjbcswitzerland
Specialist V

Hi

An interesting test would be to load just the flash configuration to disable the ROM loader (without any actual FW) and then you can see what the difference is between the processor booting into no code and the processor operating in the ROM loader. If ports are different it can ten be assumed that they are configured by the ROM loader.

Generally to ensure lowest power consumption ensure that GPIO do not float. Floating inputs can cause unexpectedly high current consumption when their level is not well defined.

Regards

Mark

0 项奖励
回复