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.

interview qstn on comb ckt to remove less time pulse

Status
Not open for further replies.

Guru59

Full Member level 4
Joined
Jul 10, 2006
Messages
217
Helped
7
Reputation
14
Reaction score
3
Trophy points
1,298
Activity points
2,812
hello all,i faced an interview yesterday and was stuck with one particular question of digital design.
the question is as design a Combinational logic which takes a train of pulses with different pulse width (1ns - 5ns) and produces pulses of only 5ns as output.it should discard pulse widths less than 5ns.
help appreciated.

i have attached a reference figure

 

not combinational.....synchronous.
 

No...the interviewers were keen to know a synthesizable combinatinal DUT
 

can we use an RS latch?
 

I believe you can build a comb logic to monitor the input every 1ns and have a counter running. If for 5 consecutive sampling if the input is sensed as HIGH then the output should be driven HIGH for 5ns and reset the counter. Otherwise the output should be driven LOW and reset the counter.

Obviously the method I described above has a delay of 5ns from input to output as the system built is a causal system(i.e. input has to be observed completely before the output is driven. We can't make any assumption about the input in real time.)

Any objections or good ideas?
 

well i said combinational circuit...
how can counter be a combinational.....?
 

Two latches, 1st latch has input data and enable is clock of 5ns, 2nd latch has input from out put of 1st latch and enable is inverted clock of 5ns.
see below figures



HTH
 

shitansh, the question required a combinational circuit.
 

Rely on gate capacitance to do RC filtering?
 

So can be a flop for that matter

**broken link removed**


Is a ring oscillator combo logic?
 

i think a generally accepted definition of combinational logic is logic that doesn't maintain state, so a latch wouldn't count.
 

Be interested to know the answer...if there is one...

It would have to rely on combo delay, which would be very PVT sensitive would it not? I'm sure you could come up with a pulse-rejection cct, but how accurate to 5 ns it would be I don't know.

Unless it's a trick question where they want you to explain why such a cct would not be that reliable or accurate.
 

hi friends,

how abt the following combo ckt...?
very optimistic solution....



its a thought but need some development and discussion on this.

Sunil Budumuru
 

May be this circuit also gives the required solution. I have attached the picture.

Comment are welcome
 

Dear Reddy,

are these ( 1ns, 2ns, 3ns, 4ns, 5ns ) delay buffers or something else?


Dear Sunil,
Your solution has one drawback:
We can give a pulse of max 5ns but here you are giving a pulse of 6 ns to get output pulse of 5 ns.

can we play with number of levels(delay buffers and AND gates) to get proper output?

your approach is very good, but it need some modification.


thanks & regards
jit
 

Status
Not open for further replies.

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top