50% duty cycle using 555 timer

EDAboard.com Forum Index -> Electronic Elementary Questions -> 50% duty cycle using 555 timer

Author

Message

KaiZX

Joined: 06 Oct 2006
Posts: 18

29 Dec 2006 17:03 555 timer duty cycle

I need an oscillator to generate a clock for a project I'm building. From what I've read from data sheets, it seems like I cannot get 50% duty cycle with a 555 timer configured in astable mode. Any suggestions as to alternative ICs?

FYI, I'm looking to generate a 2MHz clock for a logic circuit.

Last edited by KaiZX on 29 Dec 2006 18:27; edited 1 time in total

E-design

Joined: 01 Jun 2002
Posts: 947
Helped: 68

29 Dec 2006 18:24 555 duty cycle

It is very difficult to get perfect symmetry with a 555 but you can get near 1% difference. You need to get the ratio of your timing resistors as high as possible. You will be limited by the minimum resistance (Rt) values for the 555

Google
AdSense
Google Adsense

29 Dec 2006 18:24 Ads

KaiZX

Joined: 06 Oct 2006
Posts: 18

29 Dec 2006 18:32 555 timer 50% duty cycle

E-design wrote:

I neglected to mention this in my original post: I'm trying to make my project work for several different Vcc values (for 2V, 3V, and 5V). So according to you, I should be a little concerned with the resistor limitations. According to what I've read for the Motorola MC1455, I'm limited to a resistor value:

RA = Vcc/0.2

Since duty cycle = RB / (RA + 2 * RB), this means that my RB value is also going to be limited to Vcc as well, and RB would also change with Vcc changes.

Are there any alternative IC packages I can use to the 555 timer? Or any circuit topology I can build myself?

E-design

Joined: 01 Jun 2002
Posts: 947
Helped: 68

29 Dec 2006 18:33 555 timer 50 duty cycle

2MHz is not practical for the 555. You better off to use a few gates and a 2MHz crystal to do the job. If you need 50% duty cycle you can do 4MHz and put it through a flip-flop

Diagram shows a typical circuit. U2 provides 50% duty cycle at 2MHz. You may want to make R2 variable to determine the correct value experimentally.

Sorry, but you need login in to view this attachment

Last edited by E-design on 29 Dec 2006 20:04; edited 1 time in total

KaiZX

Joined: 06 Oct 2006
Posts: 18

29 Dec 2006 19:04 50% duty cycle oscillator

Thank you for the quick reply.

Just curious: what if I'm doing like an ASIC design vs. breadboard prototyping like I'm doing? I'm assuming ASIC designers do not necessarily use crystal oscillators? I've learned ring oscillator topologies, but from some simulation work I've done in classes we've learned that their frequencies vary too much with changes in Vdd. What other topologies are there?

nguyennam

Joined: 06 Sep 2006
Posts: 459
Helped: 28

30 Dec 2006 5:10 555 50 duty cycle

E-design is right when using 4013 as this flip-flop will give you the perfect 50% duty cycle pulses, and avoiding using 555 as it can work for frequency up to 2 or 3 hundred thousand hertz only.

However, for the oscillator, to choose CMOS for working with crystal, the best device is 4069UB. In principle, any NAND or NOR can be used, but only 4069 will not make you troubled in any case. Consult 4069 datasheet for oscillation circuits.

nguyennam

unkarc

Joined: 28 Apr 2001
Posts: 287
Helped: 36

30 Dec 2006 6:22 555 duty cicle

Hi,

I would like to draw your attention to the CMOS version of the 555 circuit, the LMC555 from National Semi. or TLC555 from Texas Inst. and the astable mode working frequency is claimed to be 2MHz typically at Vcc of 5V. Its data sheet includes a diagram for 50% duty cycle operation too. See this link that includes it:
http://www.williamson-labs.com/480_555.htm and look for DATA SHEETS around the middle of the page.

It works from 1.5V supply voltage too. So it would be worth trying I suppose. I know the MC1455 is ALSO a CMOS version of this timer family but its claimed upper frequency limit is defined as higher than 500kHz and its recommended lower supply limit is 5V.

The only drawback of the LMC/TLC555 could be a possible frequency instability at 2MHz due to temperature, if this is important, than you may use this interesting variable duty cycle circuit with the diodes for temperature compensation:
http://www.elecdesign.com/Articles/Print.cfm?ArticleID=6235
Due to the diodes forward drop voltage this solution may need at least 3-3.5V supply voltage or a bit higher, a test/experiment is needed.

rgds,
unkarc