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.

Prescaler output: Dute cycle not 50%:50%

Status
Not open for further replies.

dd2001

Full Member level 4
Joined
Apr 14, 2002
Messages
236
Helped
1
Reputation
2
Reaction score
1
Trophy points
1,298
Activity points
1,912
Designed Prescaler (P, A, B counter) for 3GHz, found final output from it is not 50% dute cycle.

how can make it equal to 50% dute cycle?
 

design it with two cross coupled DFF in ring configuration and by CML (Current Mode Logic) Full Differential technology for higher speed.
be careful! single ended outputs have not 50 % duty cycle but differential ones have (since only have odd harmonics!)
for this block normally differential logic is used not single ended ones (for to noise immunity).

BEST!
 

You can use a narrow bandpass filter to filter out the fundamental tone and reject the harmonics. Or use a flip-flop as the last divider, if a divide-by-two is feasible within the prescaler chain.
 

As i know there is no need to filter and also it is not possible to implement!
current realizable divider solutions are:
1.miller divide by two (very fast up to several GHz)
2.DFF ring divide by two (same as above when implemented by CML)
3.Injection Locked Frequency Divider, ILFD, (very very high speed even to 20 or 4 GHz sutable for narrow band applications but requires on chip inductor i.e more area but the power consuption is moderate or low!)

may be u mean the LC tank used in ILFD, if so, thats right!

BEST!
 

isaacnewton said:
dd2001 said:
how can make it equal to 50% dute cycle?

May I ask a stupid question? What is 'Dute Cycle'? Thank you.

Duty cycle is the fraction of the time the logic level is high compared to low. The question was about an expected perfect square wave output (high and low half the time each).
 

I think the duty cycle is decided by your architectures, not by the logic styles.
For example, if the pre-scale is divider by 4, it will be 50% duty clycle, if it is divided by 5, it will have 40% duty cycle.

Anyway,why do you need 50% duty cycle? Most of time It is the edge we care.
 

speed up your prescaler
maybe help
 

field_catcher said:
I think the duty cycle is decided by your architectures, not by the logic styles.
For example, if the pre-scale is divider by 4, it will be 50% duty clycle, if it is divided by 5, it will have 40% duty cycle.

Anyway,why do you need 50% duty cycle? Most of time It is the edge we care.

For I/Q generation, 50% duty cycle is required.
 

Usually the used trick is to divide by two the prescale output (of course you have to modify prescaler ratio) to get fifty percent duty cycle.
I hope it can help.
Mazz
 

VSWR said:
You can use a narrow bandpass filter to filter out the fundamental tone and reject the harmonics. Or use a flip-flop as the last divider, if a divide-by-two is feasible within the prescaler chain.


can i ask? to filter out the fundamental tone, we need to spread the frequency before filtering?

and .....why divide by 2 can get 50% duty cycle? we don need to multiply the frequency before dividing?

thanks if anyone can help to answer these question
 

Status
Not open for further replies.

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top