Continue to Site

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.

How to get 4MHz square wave from 20MHz crystal & inverte

Status
Not open for further replies.

kibrun

Junior Member level 1
Joined
Feb 11, 2002
Messages
18
Helped
0
Reputation
0
Reaction score
0
Trophy points
1,281
Location
msa
Activity points
102
help:design clock 4MHz

hi, i hope that somebody can help me/ give suggestion how can i achieve a 4MHz square wave from a 20MHz crystal and CMOS inverters.. or if you have a better way to do it, that is most welcome :D

i've tried to use 4MHz crystal but the produced waveform is not exactly square. below is the circuit...


---------/\/\/\-----------|
| |
|--------|>o-------------|----------->o----------
| |
|----------|crystal|-----|
| |
gnd gnd

i hope somebody could help .. thanks.
 

4mhz

use e.g. 8 mhz and dived it by 2 using e.g. 4020. or other freq with binary div. after 4020 or other logic you should have good square ..

regards

cancel
 

may be this can help

Hi
Just take a look to the 4045 and 4060 CMOS integrated circuits.
But also look at this link:

h**p://www.farnell.com/datasheets/9094.pdf

CRYSTAL OSCILLATOR,4.000000MHZ: Temperature operating max 70 °C Frequency output 4 MHz Voltage supply 5 V Duty cycle lower limit 45 % Duty cycle 55 % Case style DIL Crystal case type DIL Capacitance load 15 pF Time rise 15 ns.
regards,
 

Maybe you should use another crystal. Something like 8 MHz or similar.
If you MUST use 20 MHz xtal (I cannot imagine why), you can either use PLL or you can build the oscillator with transistors, isolate 2nd harmonic and divide by 2 and by 10 (7474 and 7490 will do this job).
 

Ooops, divide only by 10 (7490)
 

Hmmm,

when I am reading my previous posts, I was either under influence of some drugs. Or maybe, problem is that today is Friday night and I am a bit tired.

The right solution is of course the use of PLL.
Use PLL to multiply the 20 MHz by 2 (40 MHz) and subsequently divide it by 10 with one 74HCT90.
 

If you multiply by 2 you will have 40 Mhz, next if you divide by 10 (7490) you will have 4 Mhz. But the wave will have a non symetrical form. The work cycle will be 10% aprox.
I think you should change the crystal to 4 MHz, or multiple... The simple the better....
 

Alternatively, it is possible to divide by 5 (with 7490) and subsequently by 2 (one FF from 7474 which is used for divison by 2 in PLL loop).
But it is only theoretical exercise :) Better is to change the crystal.
 

PLL

Just take a IDT2305. This component have some PLL's. With it you can devide the 20 MHz!

Phytex
 

Re: help:design clock 4MHz

Use a 4Mhz and 1 or 2 cmos inverters if you could'n obtain a square wave maybe you need a capacitor between Vdd and Vss
 

Re: help:design clock 4MHz

kibrun said:
hi, i hope that somebody can help me/ give suggestion how can i achieve a 4MHz square wave from a 20MHz crystal and CMOS inverters.. or if you have a better way to do it, that is most welcome :D

i've tried to use 4MHz crystal but the produced waveform is not exactly square. below is the circuit...


---------/\/\/\-----------|
| |
|--------|>o-------------|----------->o----------
| |
|----------|crystal|-----|
| |
gnd gnd

i hope somebody could help .. thanks.
Do you need 4Mhz clk 50% duty? If it is not, it is very easy, you can use state machine to divide 20Mhz clk by 5 then get 4Mhz clk. Please email me if you need more detail help.
 

mutoucanada.. i do need the 50% duty cycle.

i've tried to use hystheresis (using different circuit than above) and it can produce the 50% duty cycle (almost perfect 90 degree up/down) but the top/ bottom portion of the wave have jitters. does anybody know how to cut-off this portion?

the circuit can produce up to 9V whereas i just need to operate at about 5V. i've even tried to use a limitter circuit (consists of only 2 zener diode) to get rid the jittering portions but the duty cycle was affected. any other suggestions?

thanks for the reply guys/ gals :) i really do appreciate them..
 

Status
Not open for further replies.

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top