1 of 1 people found this helpful
For WCO, you have to connect two external load capacitors C1 and C2 as mentioned in the section 8.3 in the following application note. The effective load capacitance should be in the limits (6 pF to 12.5pF).
Once connected, you just have to enable the WCO in the PSoC Creator Clocks tab. The PSoC Creator automatically calls the WCO related APIs to initialize and start the clock.
>>"The software (Modustoolbox and PSoC Creator) indicates that the WCO cannot be started and I'm wondering if this is because the internal circuitry of the device expects a crystal with a higher load capacitance?"
--> Is this the error you are facing? If yes, can you please attach the screenshot? If no, can you please elaborate more?
You can refer the CY8CKIT-062-BLE kit schematics for reference.
Does the above information answers your quetion?
I do have two 10pF capacitors in the normal configuration for a crystal oscillator (to ground on either side of the crystal). By the sound of it, they might be a little low (but only by 1pF - given a bit more from the board traces I might squeak in at 6pF).
I'll update this reply when I get home tonight with the screen shots (in 'debug' I can see the code stalling in a function with a name that indicates the WCO failed to start) if changing the capacitors does not work.
(BTW - I had another thread about the ModusToolbox configuration throwing errors when running a 'release' version but going into the 'debugger' properly. That was with another board of mine - same circuit and components but badly constructed; this is on a proper PCB. However I'm wondering if the 'error' I was really was really the system trying to tell me that the WCO was not starting.)
The function it was stalling in is 'cycfg_ClockStartupError' which was called (indirectly) from 'Cy_CycClk_WcoEnable'. This was why I drew the 'assumption' that the issue was with the WCO.
HOWEVER, all is well now.
I replaced the 10pF capacitors with 22pF and re-checked the soldering and the WCO is now working correctly. I guess the theory is that these might be a little high (making a capacitance of 11pF for a crystal with a load capacitance of 6pF) and so pull the frequency a bit, but given this usage (for the MCWDT while it sleeps) that does not matter.
I have a custom board with the CYBLE-416045-02 and use a crystal Epson FC-135-Q13FC1350000400 as from the BOM of the CYBLE-416045-EVAL board. I use 2 capacitors with 24pF each (the schematics and BOM of CYBLE-416045-EVAL say C2=24pF, C3=47pF, I am not sure if this is correct).
The board works fine with this crystal, no startup error.
Measuring the frequency gives roughly 32,8kHz, i.e. the quartz may be oscillating a little bit too fast.
When you look at the data sheet for that crystal, you will see that it has a load capacitance of 12pF. When you put the 2 capacitors you mention in series, the effective capacitance is just under 16pF.The reason they are different is to slightly unbalance the circuit at startup so as to help triggering the oscillation. This is all fairly standard crystal oscillator design.
My question was about a crystal with a 6pF load capacitance and so the values you specify were not really relevant.