VTH changes in DC from Hspice

Hello, everyone.

I’d like to know the threshold of the MOS transistor. And I found the “.print vth()” or “.print lv9()” can give the threshold voltage.
But
1) What does the output value of the .print lv9(Mxx) mean? Is it the same as the VTH0 in the lib?
2) why does the threshold voltage changes in a .DC sweep? when VIN sweep from 0 to 1v, the threshold of NMOS changes from 241.6751m to 421.5062m, for PMOS: from 360.5821m to 140.6713m.

3) the run.lis file gives the threshold from the mos.lib is VTH0p = -0.365, VTH0n = 0.423

Thanks in advance!


The script is:

*.lib './mos.lib' tt
.lib './PTM_bulk_65nm.lib' tt
.global VDD GND
.options list post
.temp 25

MMN OUT IN gnd gnd nmos l=60n w=266n m=1
MMP OUT IN vdd vdd pmos l=60n w=134n m=1

VIN In GND 1v
VVDD VDD GND 1v
VGND GND GND 0v

.DC vin 0v 1v 0.01v

.MEASURE DC VTH WHEN V (IN) = V (OUT)
.MEASURE DC VTGAIN DERIVATIVE V (OUT) AT VTH
.print vth(MMP) vth(MMN)
.print lv9(MMP) lv9(MMN)

Dear phuang
With ".PROBE LV9(M1)", You could measure the current Threshold voltage which has been affected by Drain Induced Barrier Lowerin (DIBL) or Inverse Narrow Width Effect (INWE) or Body Biasing Effect and .... not the Constant Value of Vth0.

2) Variation of Threshold voltage is mostly with DIBL effect.

I think you can find out DIBL effect in VLSI design textbooks.

Best Regards.
Oveis

Oveis.Gharan said:

Dear phuang
With ".PROBE LV9(M1)", You could measure the current Threshold voltage which has been affected by Drain Induced Barrier Lowerin (DIBL) or Inverse Narrow Width Effect (INWE) or Body Biasing Effect and .... not the Constant Value of Vth0.

2) Variation of Threshold voltage is mostly with DIBL effect.

I think you can find out DIBL effect in VLSI design textbooks.

Best Regards.
Oveis

Click to expand...

Dear Oveis,


Thank you for your help, which it more clear for my understanding.

Still three further questions,
1)how do they get the Vth0?
2)Do they consider the short channel effects when they resolve Vth0?
3)Does anyone use the Vth to analyse the aging effect? which Vth should be used, the Vth0 or the .print lv9(Mxx)?


Best,
phuang

Dear Phuang

I saw that your simulation is done with 65nm PTM. this model is in Level 54.
You can refer to BISIM4 documents and find all the equations which applied to gain the simulation results.
Of course that they must consider such SCE in 65nm technology. But if you want to see the simulation result more precisely and see the effects accurately, use foundry models instead of PTM.

Oveis.