Welcome to EDAboard.com

Welcome to our site! EDAboard.com is an international Electronics Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.

32768 hz quartz oscillator

Status
Not open for further replies.

nikemoto

Newbie level 4
Joined
Jul 28, 2007
Messages
6
Helped
0
Reputation
0
Reaction score
0
Trophy points
1,281
Activity points
1,332
32768 hz oscillator circuit

hi this my 1st post here.

topic is a watch crystal oscillator for PIC (microcontroller).

i do not want to use logic gates.

here the circuit (a thread on the microchip forum)
https://forum.microchip.com/tm.aspx?m=271212
there are pictures, and small movies of the circuit in action.

it works for 4mhz crystal, using a BSS138 MOSFET. also for ceramic resonators.
but why it does not work for watch crystal?

any hints? please explain. every information helps.
 

nxtech

Full Member level 5
Joined
Jun 13, 2007
Messages
258
Helped
17
Reputation
34
Reaction score
2
Trophy points
1,298
Location
NO HIDDEN LINKS
Activity points
2,772
32.768 khz oscillator mosfet

Have you checked the drive strength of the crystals you are using. The 4MHz crystal and ceramic resonators most likely are generating enough current to turn the FET on where as the 32khz crystals were not designed to that.

If you truly need to run a FET at 32khz you are going to have to do something to boost the drive strength to turn that FET on.

E
 

nikemoto

Newbie level 4
Joined
Jul 28, 2007
Messages
6
Helped
0
Reputation
0
Reaction score
0
Trophy points
1,281
Activity points
1,332
quartz 32768

i understand what you mean. however, the MOSFET is voltage controlled- all what's driven are internal capacities (and they are in the 10pF range).

i do not understand MOSFET, or quartz oscillator completely, true.

is there are better explanation than the equivalent circuit (L-C)?

maybe it would help to add a series resistor?

also i figure out, the circuit can not startup from a programmer board 7805, only from USB power supply (switched by software).
i should also test with 6 volts battery (OK i will hook-up a 9v Eblock later on).

the circuit is here: https://i25.photobucket.com/albums/c69/nikemoto2511/electronics/bss138_oscillator_circuit.png
 

Fom

Advanced Member level 2
Joined
Mar 10, 2004
Messages
633
Helped
84
Reputation
168
Reaction score
31
Trophy points
1,308
Location
Taiwan
Activity points
4,456
bss138 quarz

I don't know about BSS138, but the most of modern MOSFET switches have about 1nF gate capacitance not 10 or 20 pF.
 

nxtech

Full Member level 5
Joined
Jun 13, 2007
Messages
258
Helped
17
Reputation
34
Reaction score
2
Trophy points
1,298
Location
NO HIDDEN LINKS
Activity points
2,772
32,768hz quarz

nikemoto, adding a series resistor is not going to help. You need to increase the drive strenght (i.e. voltage) on the gate of the FET to get passed the Rdson resistance.

Look for a FET driver to put inbetween the FET and your oscillator circuit. Personally I would use either a micro or a CPLD to drive the FET.

E
 

jcpu

Full Member level 4
Joined
Dec 17, 2005
Messages
215
Helped
18
Reputation
36
Reaction score
6
Trophy points
1,298
Activity points
2,968
4069 32khz

hello, nikemoto, please post the circuit again here.
That one at microchip site is not easily read.
What makes 32768 special is it's high impedance (>10 Mohm)
due to much lower freq and that calls for extreme care of driver design.

have fun.
 

chemelec

Full Member level 4
Joined
Jul 21, 2007
Messages
219
Helped
46
Reputation
92
Reaction score
42
Trophy points
1,308
Location
BC. Canada
Activity points
2,368
32.768 khz oscillator using logic gate

Using the 32.768khz crystal in an oscillator with a 4069 will produce enough voltage to drive the Mosfet.
 

nikemoto

Newbie level 4
Joined
Jul 28, 2007
Messages
6
Helped
0
Reputation
0
Reaction score
0
Trophy points
1,281
Activity points
1,332
32768

bss138_oscillator_circuit.png


here the circuit again.

yes i know CMOS gates work for 32768 hz crystal. however, i could also use a PIC directly.

maybe the circuit produces waveforms, but for the LF crystal they are not enough to drive the clock input?

10MOHM*10pf=
10KOHM*10nf=
10OHM*10uf

this should not be impossible for 32khz to oscillate but maybe it is for some reason.
 

angara

Newbie level 1
Joined
Aug 1, 2007
Messages
1
Helped
0
Reputation
0
Reaction score
0
Trophy points
1,281
Activity points
1,285
32,768 oscillator

nikemoto said:

maybe the circuit produces waveforms, but for the LF crystal they are not enough to drive the clock input?

10MOHM*10pf=
10KOHM*10nf=
10OHM*10uf

this should not be impossible for 32khz to oscillate but maybe it is for some reason.

Measure output voltage using some high impedance tool. The voltage should be half of power supply voltage. If not add some MegOms resistor in parallel with the crystal.
 

nikemoto

Newbie level 4
Joined
Jul 28, 2007
Messages
6
Helped
0
Reputation
0
Reaction score
0
Trophy points
1,281
Activity points
1,332
testeur quartz 32khz

here also a photography of the circuit. the SMD transistor should be visible. works fine for 4Mhz, but also no startup from transformer based power supply.
https://www.flickr.com/photos/nike6/923812447/

from computer USB startup easily.
i will test with E-Block battery (5 to 6 volts) later on today. the design is not critical, i am just interested in the subject.

you can share the flickr link freely for the purpose to discuss the circuit, or maybe even link to it for reference purpose. however, i do not give a guarantee.
 

Status
Not open for further replies.

Similar threads

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Top