Maker Pro
Maker Pro

PSpice ABM prob

A

Active8

Jan 1, 1970
0
hi:

i'm feeding an ABM expr part [ EXP1=acos(V(%IN)) ] with a 1Vpp sine
wave. i get errors saying my input is invalid (i.e., >1). something's
wrong here. my input is 2 1Vpp sines multiplied which is 1Vpp (probe
output doesn't even come close to the reported error values) and then
fed to the acos() part.

i see the acos function isn't even listed as an ABM math function in the
PSpice ref, but PSpice hasn't choked on it.

i also can't figure how to get the Average value of a sine wave. not
doable with an ABM expression, is it?

TIA,
mike
 
K

Kevin Aylward

Jan 1, 1970
0
Active8 said:
hi:

i'm feeding an ABM expr part [ EXP1=acos(V(%IN)) ] with a 1Vpp sine
wave. i get errors saying my input is invalid (i.e., >1). something's
wrong here. my input is 2 1Vpp sines multiplied which is 1Vpp (probe
output doesn't even come close to the reported error values) and then
fed to the acos() part.

i see the acos function isn't even listed as an ABM math function in
the PSpice ref, but PSpice hasn't choked on it.

i also can't figure how to get the Average value of a sine wave. not
doable with an ABM expression, is it?

Make a little integrator. I think pspice allows "time" as a variable, so
you can divide the integrated value by time.

Kevin Aylward
[email protected]
http://www.anasoft.co.uk
SuperSpice, a very affordable Mixed-Mode
Windows Simulator with Schematic Capture,
Waveform Display, FFT's and Filter Design.
 
A

Active8

Jan 1, 1970
0
Jim-T@golana- said:
hi:

i'm feeding an ABM expr part [ EXP1=acos(V(%IN)) ] with a 1Vpp sine
wave. i get errors saying my input is invalid (i.e., >1). something's
wrong here. my input is 2 1Vpp sines multiplied which is 1Vpp (probe
output doesn't even come close to the reported error values) and then
fed to the acos() part.

i see the acos function isn't even listed as an ABM math function in the
PSpice ref, but PSpice hasn't choked on it.

From "pspcref.pdf":

ACOS(x) -1 <= x <= +1

So you must be going slightly over. Tweak your amplitudes slightly
lower.

i have to "tweak" by .6 ... .7*Vmultout when over once. that's not
right. 1*1=1 from the mult.

mike
 
J

Jim Thompson

Jan 1, 1970
0
On Sun, 07 Sep 2003 20:10:58 GMT, Active8

[snip]
i have to "tweak" by .6 ... .7*Vmultout when over once. that's not
right. 1*1=1 from the mult.

mike
[snip]

Congratulations! You've found a bug in PSpice. Looks like *both*
ASIN and ACOS are defective.

CHARLIE???

...Jim Thompson
 
A

Active8

Jan 1, 1970
0
Jim-T@golana- said:
On Sun, 07 Sep 2003 20:10:58 GMT, Active8

[snip]
i have to "tweak" by .6 ... .7*Vmultout went over once. that's not
right. 1*1=1 from the mult.

mike
[snip]

Congratulations! You've found a bug in PSpice. Looks like *both*
ASIN and ACOS are defective.

why do you say that? are those functions messing with you?

no. it's telling me the output of the multiplier is 1.2xxx, rather the
input to the acos function is to high. but as i said, the output of the
mult, when probed, never shows anything over 1. maybe a little over, but
not 1.2xxx or even the 1.4xxx i remember seeing on one run.

i can't figure out how the ABM integral works thank's to lame docs. the
integrator part *and* the STD function both return 0V DC for an input
sin/cos which goes from 0V to 1V.

mike
 
J

Jim Thompson

Jan 1, 1970
0
Jim-T@golana- said:
On Sun, 07 Sep 2003 20:10:58 GMT, Active8

[snip]
ACOS(x) -1 <= x <= +1

So you must be going slightly over. Tweak your amplitudes slightly
lower.

i have to "tweak" by .6 ... .7*Vmultout went over once. that's not
right. 1*1=1 from the mult.

mike
[snip]

Congratulations! You've found a bug in PSpice. Looks like *both*
ASIN and ACOS are defective.

why do you say that? are those functions messing with you?

no. it's telling me the output of the multiplier is 1.2xxx, rather the
input to the acos function is to high. but as i said, the output of the
mult, when probed, never shows anything over 1. maybe a little over, but
not 1.2xxx or even the 1.4xxx i remember seeing on one run.

i can't figure out how the ABM integral works thank's to lame docs. the
integrator part *and* the STD function both return 0V DC for an input
sin/cos which goes from 0V to 1V.

mike
CHARLIE???

...Jim Thompson

I ran ACOS with just a single-sine input... crashed PSpice v9.2.3 :)

Reduced amplitude until it ran... 0.6V peak... output was absurd, but
*was* centered on 1.57 which smells like 1.57 radians = 90°.

Further testing reveals a cumulative problem. Works right up to input
amplitude of 0.99. Amplitude of 1.0 locks up PSpice, trying a smaller
value then crashes PSpice.

...Jim Thompson
 
A

Active8

Jan 1, 1970
0
Active8 said:
hi:

i'm feeding an ABM expr part [ EXP1=acos(V(%IN)) ] with a 1Vpp sine
wave. i get errors saying my input is invalid (i.e., >1). something's
wrong here. my input is 2 1Vpp sines multiplied which is 1Vpp (probe
output doesn't even come close to the reported error values) and then
fed to the acos() part.

i see the acos function isn't even listed as an ABM math function in
the PSpice ref, but PSpice hasn't choked on it.

i also can't figure how to get the Average value of a sine wave. not
doable with an ABM expression, is it?

Make a little integrator. I think pspice allows "time" as a variable, so
you can divide the integrated value by time.

thought about that.

i can't figure out how the ABM integral works thank's to lame docs. the
integrator part *and* the SDT function both return 0V DC for an input
sin/cos which goes from 0V to 1V.


thanks,
mike
 
A

Active8

Jan 1, 1970
0
Jim-T@golana- said:
Jim-T@golana- said:
On Sun, 07 Sep 2003 20:10:58 GMT, Active8

[snip]

ACOS(x) -1 <= x <= +1

So you must be going slightly over. Tweak your amplitudes slightly
lower.

i have to "tweak" by .6 ... .7*Vmultout went over once. that's not
right. 1*1=1 from the mult.

mike
[snip]

Congratulations! You've found a bug in PSpice. Looks like *both*
ASIN and ACOS are defective.

why do you say that? are those functions messing with you?

no. it's telling me the output of the multiplier is 1.2xxx, rather the
input to the acos function is to high. but as i said, the output of the
mult, when probed, never shows anything over 1. maybe a little over, but
not 1.2xxx or even the 1.4xxx i remember seeing on one run.

i can't figure out how the ABM integral works thank's to lame docs. the
integrator part *and* the SDT function both return 0V DC for an input
sin/cos which goes from 0V to 1V.

mike
CHARLIE???

...Jim Thompson

I ran ACOS with just a single-sine input... crashed PSpice v9.2.3 :)

Reduced amplitude until it ran... 0.6V peak... output was absurd, but
*was* centered on 1.57 which smells like 1.57 radians = 90°.

Further testing reveals a cumulative problem. Works right up to input
amplitude of 0.99. Amplitude of 1.0 locks up PSpice, trying a smaller
value then crashes PSpice.

...Jim Thompson

that's what i got for a DC input... .99V max. the error for 1VDC says
the acos input is 1.001

that .6V sin wave seems to work here. returns a value in radians for the
given input. acos(.595) = .933, acos(-.593) = 2.2, acos(0) = 1.5708...
why we can't input 1V is beyond me.

mike
 
J

Jim Thompson

Jan 1, 1970
0
On Sun, 07 Sep 2003 22:44:17 GMT, Active8

[snip]
[snip]
that's what i got for a DC input... .99V max. the error for 1VDC says
the acos input is 1.001

that .6V sin wave seems to work here. returns a value in radians for the
given input. acos(.595) = .933, acos(-.593) = 2.2, acos(0) = 1.5708...
why we can't input 1V is beyond me.

mike

I have reported the problem to PSpice support.

...Jim Thompson
 
Top