10pS delay genarators

R

RA

Guest
Hi everyone,
I have used in the past the AD9501 programmable
delay generator for achieving 10pS delays.
Unfortunately those devices are obsolete now, so
I'm left with nothing to replace them. I can still
source some from Rochester Electronics, but sooner
or later they are going to run out of them.
Does anyone here know of a possible replacement, it
doesn't have have to be pin compatible or something.
Just some programmable delay generator that allows
to achieve picoseconds delays and can be programmed.
I hate to think of rolling my own sawtooth generator
with comparators, current sources yada, yada,...
Any directions are greatly appreciated.
Regards
Reinaldo Alvares
 
Thanks for your reply,
I checked it the TI home page but couldn't find
what I was looking for. I have no experiences
whatsoever with FPGA's, perhaps it is now the time
to start :) You seem to know about these FPGA's,
what's about the embedded delay lines in them?
I thought of using two signals very close in frequency
and derive the small delays from the phase difference.
I thought of obtaining them from the same master clock.
Programming the frequency of one signal would give me
the wanted delay, using some DDS or something.
It sounds too good to be true, I mean with regards to
the jitter it may have etc. But I guess I will have to
give it a try if nothing else appears.
Regards
Reinaldo Alvares

"Martin Riddle" <martinriddle@hotmail.com> wrote in message
news:AEN8c.862$NL4.343@newsread3.news.atl.earthlink.net...
TI has some clock delay ic's. You could also look int FPGA's. Some have
PLL and embedded delay lines that might be programmable to
the extent of ps. However,they will suffer errors with temperature.

Cheers
 
"RA" <reinaldoalvares????@telia.com> wrote in message
news:EWQ8c.54116$mU6.225567@newsb.telia.net...
Thanks for your reply,
I checked it the TI home page but couldn't find
what I was looking for. I have no experiences
whatsoever with FPGA's, perhaps it is now the time
to start :) You seem to know about these FPGA's,
what's about the embedded delay lines in them?
I thought of using two signals very close in frequency
and derive the small delays from the phase difference.
I thought of obtaining them from the same master clock.
Programming the frequency of one signal would give me
the wanted delay, using some DDS or something.
It sounds too good to be true, I mean with regards to
the jitter it may have etc. But I guess I will have to
give it a try if nothing else appears.
Regards
Reinaldo Alvares

"Martin Riddle" <martinriddle@hotmail.com> wrote in message
news:AEN8c.862$NL4.343@newsread3.news.atl.earthlink.net...
TI has some clock delay ic's. You could also look int FPGA's. Some
have
PLL and embedded delay lines that might be programmable to
the extent of ps. However,they will suffer errors with
temperature.

Cheers



On Semi has some ECL delay chips that do 20psec, there may be
some finer resolution ones there ISTR.

Regards
Ian
 
"RA" <reinaldoalvares????@telia.com> wrote in message news:<jNC8c.87355$dP1.254518@newsc.telia.net>...
Hi everyone,
I have used in the past the AD9501 programmable
delay generator for achieving 10pS delays.
Unfortunately those devices are obsolete now, so
I'm left with nothing to replace them. I can still
source some from Rochester Electronics, but sooner
or later they are going to run out of them.
Does anyone here know of a possible replacement, it
doesn't have have to be pin compatible or something.
Just some programmable delay generator that allows
to achieve picoseconds delays and can be programmed.
I hate to think of rolling my own sawtooth generator
with comparators, current sources yada, yada,...
Take a look at the 5V Motorola MC100E196

http://www.onsemi.com/pub/Collateral/MC10E196-D.PDF

There is also a 3.3V/5V MC100EP196.

the delay is temperature dependent, so you might want to use two chips
in good thermal contact to set up a programmable delay difference.

When I contemplated using them, a few years ago, I went for frequent
auto-calibration against a crystal-controlled 500MHz clock, but while
that should have been fine in the system I was designing, not every
system can accomodate a few milliseconds of auto-calibration from time
to time.

------
Bill Sloman, Nijmegen
 
Bill Sloman wrote:

"RA" <reinaldoalvares????@telia.com> wrote in message news:<jNC8c.87355$dP1.254518@newsc.telia.net>...

Hi everyone,
I have used in the past the AD9501 programmable
delay generator for achieving 10pS delays.
Unfortunately those devices are obsolete now, so
I'm left with nothing to replace them. I can still
source some from Rochester Electronics, but sooner
or later they are going to run out of them.
Does anyone here know of a possible replacement, it
doesn't have have to be pin compatible or something.
Just some programmable delay generator that allows
to achieve picoseconds delays and can be programmed.
I hate to think of rolling my own sawtooth generator
with comparators, current sources yada, yada,...


Take a look at the 5V Motorola MC100E196

http://www.onsemi.com/pub/Collateral/MC10E196-D.PDF

There is also a 3.3V/5V MC100EP196.

the delay is temperature dependent, so you might want to use two chips
in good thermal contact to set up a programmable delay difference.

When I contemplated using them, a few years ago, I went for frequent
auto-calibration against a crystal-controlled 500MHz clock, but while
that should have been fine in the system I was designing, not every
system can accomodate a few milliseconds of auto-calibration from time
to time.
Bill,
did you try to forward correct it with a temperature
measurement ?

Rene
--
Ing.Buero R.Tschaggelar - http://www.ibrtses.com
& commercial newsgroups - http://www.talkto.net
 
Rene Tschaggelar <none@none.net> wrote in message news:<40645323$0$710$5402220f@news.sunrise.ch>...
Bill Sloman wrote:

"RA" <reinaldoalvares????@telia.com> wrote in message news:<jNC8c.87355$dP1.254518@newsc.telia.net>...

Hi everyone,
I have used in the past the AD9501 programmable
delay generator for achieving 10pS delays.
Unfortunately those devices are obsolete now, so
I'm left with nothing to replace them. I can still
source some from Rochester Electronics, but sooner
or later they are going to run out of them.
Does anyone here know of a possible replacement, it
doesn't have have to be pin compatible or something.
Just some programmable delay generator that allows
to achieve picoseconds delays and can be programmed.
I hate to think of rolling my own sawtooth generator
with comparators, current sources yada, yada,...


Take a look at the 5V Motorola MC100E196

http://www.onsemi.com/pub/Collateral/MC10E196-D.PDF

There is also a 3.3V/5V MC100EP196.

the delay is temperature dependent, so you might want to use two chips
in good thermal contact to set up a programmable delay difference.

When I contemplated using them, a few years ago, I went for frequent
auto-calibration against a crystal-controlled 500MHz clock, but while
that should have been fine in the system I was designing, not every
system can accomodate a few milliseconds of auto-calibration from time
to time.

Bill,
did you try to forward correct it with a temperature
measurement ?
No. Amongst other things, the MC100E196 dissipates quite a lot of
heat when it is active, so it isn't all that easy to measure the
temperature of the MC100E196.

-------
Bill Sloman, Nijmegen
 
"RA" <reinaldoalvares????@telia.com> wrote in message news:<jNC8c.87355$dP1.254518@newsc.telia.net>...
Hi everyone,
I have used in the past the AD9501 programmable
delay generator for achieving 10pS delays.
Unfortunately those devices are obsolete now
If its a fixed 10ps you're after how about using stripline PCB
techniques (a bit of wiggly track) to get the delay you want. Less
temperature dependant than silicon,
Sprow.
 

Welcome to EDABoard.com

Sponsor

Back
Top