MKW24D512VHA5 work in low temperature -40

[leeyong]

Now my wireless module work in -40, test by my program, I find mcu internal spi wil slower than normal temperature, lead to my program run wrong, I use crystals is NX3225SA-32MHZ-EXS00A-CS02368(load capacitance is 9pF, I used external capacitance 11pf  ) and DS321G(load capacitance is 10pF, I used external  capacitance  11pF),

but I used TWR-MKW24D512 develop board from freescale can work good. My design reference USB-KW24D512 schematic and PCB files. Now  I don't found the solution,please give me some

advices,thanks. By modify the load capacitance value,how much impact? how to solve this porblem?This problem affect my project schedule,so I am very worried,thank you for help me!

[billpoole]

how much slower?

SPI has a clock so it should be synchronous to whatever speed it is sent at.

the temp drift of the 32 MHz XTAL at 32 MHz should be a few ppm (±16 x 10-6 max. ( at -40 to +85 °C )). at 2.4GHZ this is nearly 40 KHz, at a few MHz SPI clock, barely measureable.

your 11pF caps, what temp coefficient do they have? "%" change on the cap causes "ppm" change on the oscillator frequency (as a rough approximation), if you have caps with very large temp coefficient the temp variation of the osc freq can be quite a bit larger, but still not enough to affect SPI. Unless at the other end the SPI link is ignoring the CLK and operating asynchrously, in which case I'd be surprised if it works at all.

You can play with the load caps, larger load caps will cause a smaller frequency.

I will review your other question.

Bill Poole

Freescale Semiconductor

Connectivity and IoT Solutions, Microcontrollers

Applications Engineering