KE02 KE02Z16 in QFN-32 package cannot program/debug

キャンセル
次の結果を表示 
表示  限定  | 次の代わりに検索 
もしかして: 

KE02 KE02Z16 in QFN-32 package cannot program/debug

ソリューションへジャンプ
946件の閲覧回数
Tonkabot
Contributor III

I just built a very small smart sensor board using the KE02Z16 processor in the QFN-32 package.   My populated boards are here and I cannot get the debugger to talk to them with a PEmicro universal multilink.

Looking at the important signals, the Reset line seems to be behaving oddly.   I have a 10k pullup and a 0.1uf cap on the reset line, just like things seem to say I should.

What happens (with or without the debugger attached) is that the reset line is ramping from 2v to 2.5v at 40khz when the part is supplied with 5v.   I tried running it at lower voltages, and the part behaves the same no matter the voltage - oscillating around Vdd/2 at 40khz.  [ once the Vdd drops below 2.62 volts, the reset remains grounded]

If I remove the .1uF cap on the reset line, then goes from 5v to gnd, stays at gnd for 2.5us, and then goes back to 5v for 22.5us before repeating.

The debugger is able to ground the signal and toggle stuff out on swd_DIO and swd_CLK just like I would expect on a working device.

I added my exact 6 pin debug connection to a Freedom KE02 board [which has a 64 pin CPU], and my PE micro talks to it just fine, and there is no V/2 crud on the reset line.

[I have an adapter cable to get from the PE Micro port A 14 pin connector to my 6 pin connector - works great on the freedom board]

Can some one tell me what Is wrong?

The Vreg voltage is shut off (0v) until turned on by software to a value of 4.5 volts [which I can't do until I can program the micro]

KE02.png

ラベル(2)
0 件の賞賛
1 解決策
685件の閲覧回数
Tonkabot
Contributor III

Okay, it turns out that the NMI and SWD_DIO pullup business is all a red herring.

The actual problem is that I was trying to turn on and off the voltage VReg, that powers all my analog circuitry.

It defaulted to OFF.    And apparently various parts of the CPU are powered by the VddA/VRefH pin, which must be within .3v of Vdd.

So if you try to power up the CPU with Vdda at GND it will not run or program and sinks 6mA at the NMI pin.   The BOD and POR circuits on the CPU are powered by the VddA, and are very unhappy until you give them nearly the same voltage as Vdd.

So it is all solved now.

元の投稿で解決策を見る

0 件の賞賛
4 返答(返信)
686件の閲覧回数
Tonkabot
Contributor III

Okay, it turns out that the NMI and SWD_DIO pullup business is all a red herring.

The actual problem is that I was trying to turn on and off the voltage VReg, that powers all my analog circuitry.

It defaulted to OFF.    And apparently various parts of the CPU are powered by the VddA/VRefH pin, which must be within .3v of Vdd.

So if you try to power up the CPU with Vdda at GND it will not run or program and sinks 6mA at the NMI pin.   The BOD and POR circuits on the CPU are powered by the VddA, and are very unhappy until you give them nearly the same voltage as Vdd.

So it is all solved now.

0 件の賞賛
685件の閲覧回数
kerryzhou
NXP TechSupport
NXP TechSupport

Hi Brynn Rogers,

      Actually, to the new kinetis chip without the code, the reset pin have the repeating ramp wave is correct, After you program the chip, the reset voltage will be high.

     Now, please add an 10K pullup resistor in the SWD_DIO pin.

     PTB4, it is NMI pin, don't connect with your sensor, and add an 10K pull up resistor, this is very important!

     I don't know why you use PE Micro port A 14 pin to 6 pin connector, and how you connect the signal. As I know FRDM-KE02 is the 10 pin SWD interface, then how you connect the PORTA to your 6 pin? Please give me your  connector wire order.

    Here I suggest you use the PE Micro PORTG or Port B,  if you use PORTB, the standard ARM port, you can connec it like the following KL25 SWD pin order:

jtag-swd.jpg

     TMS  = SWD_DIO

     TCK  = SWD_CLK

     TCC=5V

    Please check your board SWD wire order again.

    Besides, how many board you are soldered? Please solder more than 2 boards, and check it, whether all the board have the same problem. Please take care, the solder temperature can't exceed 260 °C. Higher solder temperature will damage the chip.

Wish it helps you!

If you still have question, please let me know!

Have a great day,
Jingjing

-----------------------------------------------------------------------------------------------------------------------
Note: If this post answers your question, please click the Correct Answer button. Thank you!
-----------------------------------------------------------------------------------------------------------------------

0 件の賞賛
685件の閲覧回数
Tonkabot
Contributor III

Okay,  I am able to debug now.    The problem is that I randomly choose PTB4 to be the IO controlling a voltage regulator.  Didn't realize it was NMI and

its default function is NMI.

Once I put a pullup on PTB4 the problem is solved.  Didn't actually need the pullup on the swd_DIO - I'll add it for the next rev anyway.

I am using a custom 6 pin header because there is no space on my board for other options,   The board is 0.483 x 0.898 inches and fits inside a 1/2" ID pipe fitting.

using a 6-pin running off the end of the board I can actually debug when the board is in the pipe fitting.

cable_adapter.png

0 件の賞賛
685件の閲覧回数
kerryzhou
NXP TechSupport
NXP TechSupport

Ok, that's very good!

SWD_DIO add 4.7K to 10K pull up is for good design, because the internal pullup is weak, then for stable design, we always recommend our customer to add the external pull up!

Wish it helps you!

Jingjing

0 件の賞賛