# Step source for TRAN and HB

Question asked by TNAKA on Jun 5, 2003
Latest reply on Jun 6, 2003 by nacim1
In spectre, I use a step signal as voltage source to kick start oscillator.
For example, I use following step voltage source.
V0 (vcc! 0) vsource dc=5 type=pulse val0=0.0 val1=5 rise=1n
This voltage source works as step signal with rise of 1nsec for transient-ana
and constant 5V for DC-ana.

#uselib "ckt", "rfdeVf_Pulse"
rfdeVf_Pulse:V0 "vcc!" 0 Vpeak=5 Vlow=0.0 Vdc=5.0 Rise=1n Weight=0 \
Harmonics=16 SaveCurrent=1

This signal works as step signal for tran-ana,
but for HB-ana it is not appropriate.
The above signal works:
5V for DC-ana
Step signal with trise=1nsec, Vlow=0V and Vhigh=5V for TRAN-ana
0V for HB-ana

For HB-ana I expect and hope constant 5V DC-source.
Properly speaking, it is impossible to express step signal
as fourier summation(spectral source).
So I don't think this replacement is not useful for HB-ana.

I expect constant 5V DC-source for HB-ana and step signal for transient-ana.
To satisfy these two needs, I can use the following signal.
V_Source:V0 "vcc!" 0 V_Tran=pulse(time,-5,0,0n,1n) Vdc=5 Vac=0 \
SaveCurrent=1

The above signal works:
0V for DC-ana
Step signal with trise=1nsec, Vlow=0V and Vhigh=5V for TRAN-ana
constant 5V for HB-ana

Netlister should treat spectre'e step signal like the above, I think.
To tell the truth, I hope 5V also for DC-ana.
How do you think of this ?

[Note] Following signal is not appropriate.
V_Source:V0 "vcc!" 0 V_Tran=pulse(time,0,5,0n,1n) Vdc=5 Vac=0 \
SaveCurrent=1

This signal results in:
5V for DC-ana
Step signal with trise=1nsec, Vlow=5V and Vhigh=10V for TRAN-ana
constant 5V for HB-ana