LT Spice model for "a delay line."

  • Thread starter .:: Evanescence ::.
  • Start date
E

.:: Evanescence ::.

Guest
Hi.

Do anybody here have a LT Spice model for the "Delay Line" like the "DELAY"
component that exists in PSpice?

Thanks.
 
On 27 Jun 2004 19:08:43 GMT, ".:: Evanescence ::."
<evanescence_br@pop.com.br> wrote:

Hi.

Do anybody here have a LT Spice model for the "Delay Line" like the "DELAY"
component that exists in PSpice?

Thanks.
Click to expand...
The "Delay" element in PSpice is a **digital** device. I don't know
if LTSpice has digital devices, but you could use a transmission line.

...Jim Thompson
--
| James E.Thompson, P.E. | mens |
| Analog Innovations, Inc. | et |
| Analog/Mixed-Signal ASIC's and Discrete Systems | manus |
| Phoenix, Arizona Voice:(480)460-2350 | |
| E-mail Address at Website Fax:(480)460-2142 | Brass Rat |
| http://www.analog-innovations.com | 1962 |

I love to cook with wine. Sometimes I even put it in the food.
 
Um duende verde chamado Jim Thompson <thegreatone@example.com> disse em
news:fv6ud0tvm8hefjbihhrotkmv836h2db5kr@4ax.com:

The "Delay" element in PSpice is a **digital** device. I don't know
if LTSpice has digital devices, but you could use a transmission line.
Click to expand...
I forgot this. LTSpice has basic digital devices (AND/OR/XOR/NOT gates only)
but a transmission line seems the only way to do.

Do you have info on how to make a delay line with a transmission line? I am
new user of LTspice (cool program. sad that I didn't known it before).
 
On 27 Jun 2004 19:21:03 GMT, ".:: Evanescence ::."
<evanescence_br@pop.com.br> wrote:

Um duende verde chamado Jim Thompson <thegreatone@example.com> disse em
news:fv6ud0tvm8hefjbihhrotkmv836h2db5kr@4ax.com:

The "Delay" element in PSpice is a **digital** device. I don't know
if LTSpice has digital devices, but you could use a transmission line.

I forgot this. LTSpice has basic digital devices (AND/OR/XOR/NOT gates only)
but a transmission line seems the only way to do.

Do you have info on how to make a delay line with a transmission line? I am
new user of LTspice (cool program. sad that I didn't known it before).
Click to expand...
Maybe Mikey will jump in here. In *PSpice* you set Zo, length OR
delay. I would imagine that LTSpice is similar. Note, you must load
the transmission line with a resistor of value Zo. otherwise you will
get reflections, just as in real life.

...Jim Thompson
--
| James E.Thompson, P.E. | mens |
| Analog Innovations, Inc. | et |
| Analog/Mixed-Signal ASIC's and Discrete Systems | manus |
| Phoenix, Arizona Voice:(480)460-2350 | |
| E-mail Address at Website Fax:(480)460-2142 | Brass Rat |
| http://www.analog-innovations.com | 1962 |

I love to cook with wine. Sometimes I even put it in the food.
 
Um duende verde chamado Jim Thompson <thegreatone@example.com> disse em
news:v69ud0lm0l1mfh9tgfela4pqu7lg8v69ce@4ax.com:


Maybe Mikey will jump in here. In *PSpice* you set Zo, length OR
delay. I would imagine that LTSpice is similar. Note, you must load
the transmission line with a resistor of value Zo. otherwise you will
get reflections, just as in real life.
Click to expand...
This done the magic!

Many thanks!

--
 
Evanescence,

Maybe Mikey will jump in here. In *PSpice* you set Zo, length OR
delay. I would imagine that LTSpice is similar. Note, you must load
the transmission line with a resistor of value Zo. otherwise you will
get reflections, just as in real life.

This done the magic!

Many thanks!
Click to expand...
Another option is to use a behavioral delay. Below
is an .asc file the illustrates.

--Mike

Version 4
SHEET 1 880 680
WIRE 48 400 48 352
WIRE 48 272 48 224
WIRE 240 224 240 272
WIRE 240 352 240 400
FLAG 240 400 0
FLAG 48 400 0
FLAG 48 224 X
SYMBOL bv 240 256 R0
SYMATTR InstName B1
SYMATTR Value V=delay(V(x),.3m)
SYMBOL voltage 48 256 R0
SYMATTR InstName V1
SYMATTR Value sine(0 1 1K)
TEXT 352 440 Left 0 !.tran 5m
 
Um duende verde chamado "Mike Engelhardt" <nospam@nospam.org> disse em
news:OGMDc.78858$IE.26685@newssvr29.news.prodigy.com:

Another option is to use a behavioral delay. Below
is an .asc file the illustrates.
--Mike

Version 4
SHEET 1 880 680
Click to expand...

This is other way to do it. Thanks for both (Jim and you) that reply to me.

--
"I must be dreaming..."
 
Just curious. Does this "delay" method run faster, slower or the same as
delay line?

George

Mike Engelhardt wrote:

Evanescence,

Maybe Mikey will jump in here. In *PSpice* you set Zo, length OR
delay. I would imagine that LTSpice is similar. Note, you must load
the transmission line with a resistor of value Zo. otherwise you will
get reflections, just as in real life.

This done the magic!

Many thanks!

Another option is to use a behavioral delay. Below
is an .asc file the illustrates.

--Mike

Version 4
SHEET 1 880 680
WIRE 48 400 48 352
WIRE 48 272 48 224
WIRE 240 224 240 272
WIRE 240 352 240 400
FLAG 240 400 0
FLAG 48 400 0
FLAG 48 224 X
SYMBOL bv 240 256 R0
SYMATTR InstName B1
SYMATTR Value V=delay(V(x),.3m)
SYMBOL voltage 48 256 R0
SYMATTR InstName V1
SYMATTR Value sine(0 1 1K)
TEXT 352 440 Left 0 !.tran 5m
Click to expand...
 
George,

They use slightly different timestep control algorithms, so
it should be possible to contrive simulations that will
demonstrate each as faster. Neither is ever used in the
SMPS macro models because these IC's don't contain high
fidelity analog delay lines. Analog delays in LTspice's
SMPS macro models are usually RC time constants. They
better mimic the delay of something like a comparator
because they will switch state faster given more over-
voltage and switch slower if the input slowly approaches
the threshold. Real comparators act somewhere between
integrators and discrimators. Anyway, RC time constants
in this case are not only more accurate, but simulate much
faster than any delay device. When looking for a delay,
one should ask if they really mean a delay that preserves
the whole analog waveform. A critter like that is much
less common then some gate that gives some delay.

--Mike

"George" <steber@execpc.com> wrote in message news:40E01A24.7C406953@execpc.com...
Just curious. Does this "delay" method run faster, slower or the same as
delay line?

George

Mike Engelhardt wrote:

Evanescence,

Maybe Mikey will jump in here. In *PSpice* you set Zo, length OR
delay. I would imagine that LTSpice is similar. Note, you must load
the transmission line with a resistor of value Zo. otherwise you will
get reflections, just as in real life.

This done the magic!

Many thanks!

Another option is to use a behavioral delay. Below
is an .asc file the illustrates.

--Mike

Version 4
SHEET 1 880 680
WIRE 48 400 48 352
WIRE 48 272 48 224
WIRE 240 224 240 272
WIRE 240 352 240 400
FLAG 240 400 0
FLAG 48 400 0
FLAG 48 224 X
SYMBOL bv 240 256 R0
SYMATTR InstName B1
SYMATTR Value V=delay(V(x),.3m)
SYMBOL voltage 48 256 R0
SYMATTR InstName V1
SYMATTR Value sine(0 1 1K)
TEXT 352 440 Left 0 !.tran 5m
Click to expand...
 
Mike,

Thanks for explanation. I've used a delay line to provide accurate phase shifts in a
bridge simulation but it always simulates slowly. Now I'll try the bv model and see if
there is a difference. For this case a simple RC delay wouldn't work well, as you
note, since it doesn't preserve the waveform.

George

Mike Engelhardt wrote:

George,

They use slightly different timestep control algorithms, so
it should be possible to contrive simulations that will
demonstrate each as faster. Neither is ever used in the
SMPS macro models because these IC's don't contain high
fidelity analog delay lines. Analog delays in LTspice's
SMPS macro models are usually RC time constants. They
better mimic the delay of something like a comparator
because they will switch state faster given more over-
voltage and switch slower if the input slowly approaches
the threshold. Real comparators act somewhere between
integrators and discrimators. Anyway, RC time constants
in this case are not only more accurate, but simulate much
faster than any delay device. When looking for a delay,
one should ask if they really mean a delay that preserves
the whole analog waveform. A critter like that is much
less common then some gate that gives some delay.

--Mike

"George" <steber@execpc.com> wrote in message news:40E01A24.7C406953@execpc.com...
Just curious. Does this "delay" method run faster, slower or the same as
delay line?

George
Click to expand...
snip
 
You must log in or register to reply here.

Welcome to EDABoard.com

Sponsor

Back
Top