LSE crystal doesn't work on low temperatures

[oscargomezf]

Posted on June 26, 2018 at 10:03

Hi everyone,

I'm working on an STM32F1 platform. I have a problem with the LSE crystal. I'm using this one ABS07-120-32.768kHz-T by Abracon. The problem is with temperatures below 5�C. I made some calculations according to the Application Note AN2867:

According to the  datasheet ABS07-120-32.768kHz-T:

C0=1.20pF, ESR=55kohm (Max 60kohm), CL= 6pF and F= 32.768KHz

And according to the datasheet (page 62/136) del STM32F103RF:

gm = 5uA/V

gmcrit = 4 * ESR * (2?F)^2 * (C0 + CL)^2 = 4 * 60000*(2 * pi * 32768)^2 * (1.2*10^(-12) + 6 *10^(-12) )^2 = 0.0000005273947973 A/V = 0.5273947973 uA/V

gain margin = gm/gmcrit = 5 / 0.5273947973 = 9.48 (60k)

So we meet that the gain margin = gm / gm_crit >= 5.

Considering Cin=Cout=3pF (STM32 processor), CSTRAY=1pF, C1=C2=x and CL=6pF, we get C1=C2:

CL = (Cin + C1)*(C2 + Cout)/(Cin + C1 + C2 + Cout) + CSTRAY

6 = (3 + x)*(3 + x)/(6 + 2*x) + 1x=7pF

C1=C2=6.8pF

So I think that the selection of the crystal and C1/C2 is right, but when I tried to init the platform inside a thermal chamber the crystal it seems to be that it wasn't able to oscillate because I was checking this function and it's blocked in the while:

while(RCC_GetFlagStatus(RCC_FLAG_LSERDY) == RESET)

Does anyone help me out?

Best regards.

#lse-startup #stm32f1-clock

[STOne-32]

Posted on June 26, 2018 at 22:54

Dear Gentleman,

We have an errata on our STM32F1 series and especially for the XL density. The LSE_OSC_IN pin is very sensitive to leakage that is amplified at low temperature. See Section 2.7 

https://www.st.com/content/ccc/resource/technical/document/errata_sheet/ce/77/87/11/49/3c/41/03/CD00273032.pdf/files/CD00273032.pdf/jcr:content/translations/en.CD00273032.pdf

 

Can you add this high value parallel resistance and let me know if it works ?

This is behavior is not linked to a crystal compatibility or not.and that Abracon part is tuned to match our STM32 MCUs and designed in 2011 so enough mature and in production for > Millions  units so far.

Cheers,

STOne-32.

[oscargomezf]

Posted on June 27, 2018 at 18:23

Hi STOne-32,

I tried to place 2 resistors of 10MOmh in series, because I didn't have one of 20 MOhm, but It doesn't work. I was waiting for more than 3 minutes in order to start the oscillation.

Do you have any suggestion?

Best regards.

[STOne-32]

Posted on June 28, 2018 at 01:09

Hi,

So it seems another issue may linked to your PCB or board. If you can clean the flux residue , the. To swap another MCU to see it is following the Board or our micro, if it is the micro, we may need to investigate it and get the sample.

regards,

STOne-32

[oscargomezf]

Posted on June 28, 2018 at 07:55

Hi STOne-32,

This is the footprint:

And the track length:

For the 32.768KHz crystal: 6.731 mm and 4.786 mm

For the 8.0MHz crystal: 8.176 mm and 9.874 mm

Do you think the PCB design is good enough?

Best regards.

[STOne-32]

Posted on June 28, 2018 at 10:46

Hi,

The design seems good.  Can you please share all Layers seperatly to check the Vias and Ground connections.  I fear there is a ground loop creating a leakage somewhere, but to be confirmed.

Cheers,

STOne-32.

[Russ Heibel]

Posted on June 28, 2018 at 12:34

Very similar problem experienced on STM32F32x series processor. The problem was resolved to be leakage caused from moisture on the board. The LSE circuits in the STM parts are tremendously sensitive. Our solution was to conformal coat the board encompassing this entire circuit. Haven't had the problem recur since then.

Good luck...

[Nicholas Yunker_2]

Posted on June 28, 2018 at 15:24

If this is indeed the problem then placing the board in an environmental chamber and lowering the temperature could cause condensation on the PCB.  This would typically not happen unless some humidity also made it into the chamber, but it is possible.

[Nicholas Yunker_2]

Posted on June 28, 2018 at 15:25

I concur, I've been doing hardware design for the last 10 years and this appears to be exactly how I would have done it.

[oscargomezf]

Posted on July 03, 2018 at 08:03

Hi STOne-32,

I'm sorry for the delay. I send you all layers seperatly.

TOP:

L1:

L2:

Bottom:

1. Do you think it is right the design?

2. I'm working on a new design based on the STM32L4. I was doing the calculation and it seems to be that I can use the same crystals. Could you check my calculations?

LSE: ABS07-120-32.768kHz-T

HSE: ABMM2-8.000MHZ-E2-T

According to this application note by ST Microelectronics AN2867:

https://www.st.com/content/ccc/resource/technical/document/application_note/c6/eb/5e/11/e3/69/43/eb/CD00221665.pdf/files/CD00221665.pdf/jcr:content/translations/en.CD00221665.pdf

CALCULATIONS FOR ABS07-32.768KHZ-7-T WITH STM32L431 

According to the datasheet 

https://abracon.com/Resonators/ABS07-120-32.768kHz-T.pdf

C0=0.9-1.20pF, ESR(max)=60kohm, CL=6pF and F= 32.768KHz

And according to the datasheet (page 117/118) 

https://www.st.com/resource/en/datasheet/stm32l431cb.pdf

gm [LSEDRV[1:0] = 00]= 0.5 uA/V

gm [LSEDRV[1:0] = 01]= 0.75 uA/V

gm [LSEDRV[1:0] = 10]= 1.7 uA/V

gm [LSEDRV[1:0] = 11]= 2.7 uA/V

gmcrit = 4 * ESR * (2?F)^2 * (C0 + CL)^2 = 4 * 60000 *(2 * pi * 32768)^2 * (1.2 * 10^(-12) + 6 * 10^(-12))^2 = 0.0000005273947973 A/V = 0.5273947973 uA/V

gain margin = gm / gmcrit = 0.5 / 0.5273947973 = 0.9480563755269339

So we don't meet the requirements gain margin = gm / gm_crit >= 5.

CALCULATIONS FOR  ABMM2-8.000MHZ-E2-T  WITH STM32L431

According to the datasheet 

https://abracon.com/Resonators/ABMM2.pdf

C0=7pF, ESR(max 8.0MHz)=72ohm, CL=18pF and F= 8.0MHz

And according to the datasheet (page 116/118) 

https://www.st.com/resource/en/datasheet/stm32l431cb.pdf

gm = 1.5 mA/V

gmcrit = 4 * ESR * (2?F)^2 * (C0 + CL)^2 = 4 * 72 *(2 * pi * 8 * 10^6)^2 * (7.0 * 10^(-12) + 18 * 10^(-12))^2 = 0.00045479137080222 A/V = 0.4547913708022 mA/V

gain margin = gm / gmcrit = 1.5 / 0.4547913708022 = 3.2982

So we don't meet the requirements gain margin = gm / gm_crit >= 5.

Do you know If I'm doing well the calculations?

Can you give me any part number that ST microelectronics has checked?

Best regards.

[Nicholas Yunker_2]

Posted on July 03, 2018 at 14:11

Just giving this a quick look this looks good.  I have not personally used the ABMM2 series crystals from Abracon but that doesn't mean they won't work.