| View previous topic :: View next topic | 
	
	
		| Author | Message | 
	
		| Regular Guy 
 
 
 Joined: 04 Jun 2017
 Posts: 110
 
 
 
			    
 
 | 
			
				| PIC 16F84A 32khz Crystal Capacitors 68 pF |  
				|  Posted: Sat Jul 03, 2021 11:23 am |   |  
				| 
 |  
				| Our PIC in Practice book calls for 68 pF capacitors with the 32khz crystal. 
 The Microchip capacitor guide only goes up to 33 pF for using with crystals.
 
 What will happen when we stair step down with the capacitors from
 
 68 to 56 then 47 then 39?
 
 Thanks!
 
 We are keeping the config word set a LP oscillator.
 _________________
 PIC Hobbyist
 |  | 
	
		|  | 
	
		| newguy 
 
 
 Joined: 24 Jun 2004
 Posts: 1924
 
 
 
			    
 
 |  | 
	
		|  | 
	
		| Ttelmah 
 
 
 Joined: 11 Mar 2010
 Posts: 19966
 
 
 
			    
 
 | 
			
				|  |  
				|  Posted: Sun Jul 04, 2021 1:09 am |   |  
				| 
 |  
				| The book will probably have a 'part number' for the crystal they intend to be used. Now the commonest 32K crystals today are the so called 'watch'
 crystals and are designed for low load capacitances like 6 to 15pF. These
 are normally referred to as 'tuning fork cut types (XY). However there are
 other crystal cuts available (though very uncommon), so you can get AT
 and BT cut 32KHz crystals, and these commonly need much higher load
 capacitances.
 So it may be that the particular crystal the the book actually used, was a
 type needing a much higher capacitance than normal.
 
 Now the effect of the loading being 'wrong' will be first to cause inaccuracy
 in the frequency, then at larger errors, it may even stop the crystal from
 oscillating. As Newguy says, you need to start with the part number of the
 crystal you actually 'have', Then make an estimate for the track
 capacitances, add the pin capacitances (data sheet), and calculate the
 correct value to use. The 'estimate' should be based on the width and
 length of the tracks (formulae online for the capacitance of a track).
 |  | 
	
		|  | 
	
		| Regular Guy 
 
 
 Joined: 04 Jun 2017
 Posts: 110
 
 
 
			    
 
 | 
			
				|  |  
				|  Posted: Sun Jul 04, 2021 9:48 am |   |  
				| 
 |  
				| Thank you newguy and Ttelmah 
 Let me give our input.
 
 1 The data sheets seem to sometimes have info on capacitors for crystals
 
 and then other times have info for resistor and capacitors for the
 
 RC circuits.
 
 2 We did blowout one 32khz crystal.It may have been on a development
 
 board. Most of our boards seem to have 22 pF capacitors soldered on.
 
 3 We did get some hard to find 32khz 4 pin canned oscillators. We blew
 
 one of those up too.
 
 We are pretty set on doing the 'stair step' down experiment.
 
 See how the circuit oscillates. See how long they last.
 
 Thanks!
 _________________
 PIC Hobbyist
 |  | 
	
		|  | 
	
		| Ttelmah 
 
 
 Joined: 11 Mar 2010
 Posts: 19966
 
 
 
			    
 
 | 
			
				|  |  
				|  Posted: Sun Jul 04, 2021 11:21 am |   |  
				| 
 |  
				| It is very difficult to make a crystal fail in a circuit. Mechanical shock (dropping it or hitting it).
 Overheating when soldering.
 Both much more likely to make a crystal fail.
 Making it fail electrically would require applying quite high voltages.
 |  | 
	
		|  | 
	
		| Regular Guy 
 
 
 Joined: 04 Jun 2017
 Posts: 110
 
 
 
			    
 
 | 
			
				|  |  
				|  Posted: Sun Jul 04, 2021 11:47 am |   |  
				| 
 |  
				| Thank you Ttelmah 
 The 4 pin 32khz oscillator can may have been the one was on our
 
 PICDEM 2 Plus.
 
 We left the jumper for the RC 200khz circuit on.
 
 When we had a 4 pin can in the socket.
 
 They need to put a big red sticker on the board about that!
 
 Noticed something on RC circuit was really hot.
 
 By then it was too late.
 
 The 32khz crystal was just on a breadboard.
 
 Just stopped  working.
 
 Thanks!
 _________________
 PIC Hobbyist
 |  | 
	
		|  | 
	
		|  |