XTAL crystal for 5LP prototyping kit

Tip / Sign in to post questions, reply, level up, and achieve exciting badges. Know more

cross mob
KeLa_4479261
Level 4
Level 4
First like received

Greetings to All:

I am trying to reduce the ppm number on the 5LP.  I have enabled the USB tab in the clock settings and have got a clock running at 1250ppm.  I was hoping to attach an external crystal with some capacitors and use them to create a more accurate clock.

I was hoping to use a watch crystal with 20ppm.  This is desirable as I am looking to microsecond timers that would be accurate.

Can I use the watch crystal, create and use a clock based off this at 24Mhz?

I am trying to locate documentation on this and my searches do not amount to any thing.

Help is appreciated, regards.

Ken.

0 Likes
1 Solution
MotooTanaka
Level 9
Level 9
Distributor - Marubun (Japan)
First comment on blog Beta tester First comment on KBA

Hi,

I'm sorry hearing that the USB Clock of 5LP does not have enough accuracy.

I tried with CY8CKIT-059. (5LP)

I'm afraid that there is no path from XTAL 32kHz to Master Clock.

The XTAL needs to be 4MHz ~ 25MHz.

000-5LP-clocks.JPG

Then I tried with CY8CKIT-044 (4200M)

If I set wco (32kHz) with 20PPM accuracy

001-4200M-low.JPG

We can trim IMO with WCO, so that IMO accuracy is +/- 0.2%

002-4200M-High.JPG

moto

View solution in original post

0 Likes
8 Replies
MotooTanaka
Level 9
Level 9
Distributor - Marubun (Japan)
First comment on blog Beta tester First comment on KBA

Hi,

I'm sorry hearing that the USB Clock of 5LP does not have enough accuracy.

I tried with CY8CKIT-059. (5LP)

I'm afraid that there is no path from XTAL 32kHz to Master Clock.

The XTAL needs to be 4MHz ~ 25MHz.

000-5LP-clocks.JPG

Then I tried with CY8CKIT-044 (4200M)

If I set wco (32kHz) with 20PPM accuracy

001-4200M-low.JPG

We can trim IMO with WCO, so that IMO accuracy is +/- 0.2%

002-4200M-High.JPG

moto

0 Likes

Hello Moto:

I am going to look at the 8Mhz crystal oscillator, the package from AdaFruit comes with the 20pF caps.

What reference document would be best on how to connect this?  I have located: AN54439, PSoC® 3 and PSoC 5LP External Crystal Oscillators.

I will be using a breadboard as I can connect everything easily.

Does this make sense?

Thanks,
Ken.

0 Likes
MotooTanaka
Level 9
Level 9
Distributor - Marubun (Japan)
First comment on blog Beta tester First comment on KBA

Dear Ken-san,

As I am not a hardware expert,

all I can suggest is somewhat general.

From the application note Figure 2,

I'm hoping that

X1 is the 8.000MHz crystal

CL1 and CL2 are 20pfs come with the kit

And Xi and Xo are P15[0] and P15[1].

Figure2_Pierce_Oscillator.JPG

Since you are planning to use a breadboard, how about trying it first.

And if it does not work, may be you need to post a new question.

But in general, please try to keep the wires between 5LP and these components short.

Best Regards,

9-Mar-2020

Motoo Tanaka

0 Likes

KeLa,

Please note that XTAL frequency is nominal, final frequency depends on XTAL ppm accuracy and capacitors value. 20pf seems high, I believe 12-14pf were recommended in the AN. So buy few more caps. If this is University project, use DigiKey or Mouser, for personal use -AliExpress.

/odissey1

I was just checking the datasheet for the part I got from AdaFruit.

They have a a load capacitance of 18pF and a shunt capacitance of <7pF.

8 MHz Crystal + 20pF capacitors

PRODUCT ID: 2213

Which values should I trust?

Please and thanks.

Ken.

0 Likes

Update:

Converted the formula for determining the two capacitors for the 8Mhz crystal.  The formula is from the AN54430 document section 4.5.1 "Balanced Load Capacitor Example".

Equation 2:   2*CL - (CP + CB)

                 :   2 * 12.5pF - (5 + 1)

                 :   19 => reduced to 18pF

From the datasheet for the AdaFruit 8Mhz crystal, the device is a 18pF crystal.

Using this formula:  2 * 18pF - (5 + 1).

30pF capacitors are to be used on this PSOC 5LP board.

Checked with AdaFruit calculations, they have a page on their website they are subtracting the (CP + CB) twice.

AdaFruit Equation:  2*CL - 2*(CP + CB), this results in 24pF capacitors which they have rounded down to 20pF.

Question:  Which formula is correct?

Regards,

Ken.

0 Likes

Ken,

Did you take into account parasitic capacitance of the board traces? I estimate it about 10p on KIT-059 board plugged into protoboard. Try 18p caps which comes with the XTAL, if PSoC doesn't start - reduce cap values.

/odissey1

0 Likes

Greetings:

  1. I received the 8Mhz crystal from AdaFruit.  This came with 2 - 20pF capacitors
  2. The capacitors were checked using the formula in AN54430 document section 4.5.1 "Balanced Load Capacitor Example".  The calculations in this document are for SMT devices. 
  3. The device I am hooking up is the through-hole XTAL.  The formula in AN54430 requires a small adjustment to add in additional capacitance for the breadboard.  When this is factored in the 20pF capacitors are the correct choice for this through-hole XTAL
  4. Connecting this device to the PSOC 5LP required a little more effort than expected because locating the correct 2 GPIO pins to use for the connection was not straight forward.  The document: PSoC® 5LP: CY8C58LP Family Datasheet, does contain the information required for the 68 Pin version of the board on page: 7 of 139.  Searched through a number of documents before locating this one. 
  5. CY8CKIT-059 PSoC® 5LP Prototyping Kit Guide, Doc. #: 001-96498 Rev. *A.  This document contains the board level layout of the kit with all the GPIO pins labeled.  What is missing is the description for P15[0] and P15[1], these are the Mhz XTAL connections.  The P15[2] and P15[3] are the 32kHz crystal connections.
  6. I was able to use the PSOC Creator 4.3 to verify the correct pins: P15[0] and P15[1]
  7. Hooking up the XTAL on the breadboard was easy, the signal wires to the GPIO pins I kept to a couple of inches and the PPM of +/- 30, is working as expected.
  8. Currently running the XTAL through the PLL and have adjusted the temperature range on the device to > 0 and the devices readily handles 64 Mhz.
  9. Signal playback to the decimal microsecond is now possible.

Thanks to all the people that provided details and information.

Ken.