In Xschem when using such a symbol (the 'nf' one) and then having

W=12 L=0.5 nf=3

The W is the total width right ? Or the per-finger one ? I was pretty sure it was the total one given you have stuff like

ad="'int((nf+1)/2) * W / nf * 0.29'"

in the param box ( the

W/nf

being the clue ).

@tnt the *_nf.sym mosfets take the single finger width as the W parameter. If you have a

nfet_01v8_nf.sym

device instantiated with: W=2 nf=3 L=0.15 you get the following netlist:

XM1 net1 net2 net3 net4 sky130_fd_pr__nfet_01v8 L=0.15 W='2 * 3 ' nf=3 ad='int((nf+1)/2) * W / nf * 0.29' as='int((nf+2)/2) * W / nf * 0.29'
+ pd='2*int((nf+1)/2) * (W / nf + 0.29)' ps='2*int((nf+2)/2) * (W / nf + 0.29)' nrd='0.29 / W ' nrs='0.29 / W ' sa=0 sb=0 sd=0 mult=1
+ m=1

you see the expression

W='2 * 3'

, total width is calculated as product of finger width * nf The expressions for ad, as, pd, ps use this calculated W value. The only reason for the *nf.sym symbols was user request. Some users asked to have devices where the finger width is specified in transistor attributes, instead of the total width, as default in xschem_sky130

Damn, I wasn't aware of that, I though the

nf

thing was just cosmetic.

@tnt a W=2 L=0.15 transistor drawn as a single gate is very similar but not identical to a 2 finger device with dual gate W=1 L=0.15. Hopefully model files take account for the specified number of fingers and refurn the correct I/V characteristic. *_nf.sym and no _nf symbols in xschem are identical, only the W value has a different meaning. When netlisting the *nf.sym devices the W is immediately evaluated as total width (fingerW * nf), this is what spice model files expect for transistor W

Yeah, I had in memeory that W was the total length and set the params like that in xschem and when doing the sims. But when doing the layout I used that as the total W ... so in my layout my transistor are 4 to 7 times too small vs what I had designed in the schematic ...

the *_nf.sym symbols can generate confusion (I was not very happy for adding them, but there was some pressure from more than 1 users). Do not mix *_nf.sym and and normal fet symbols. I have added notes in the top intro schematic and also graphically these symbols look different to try to minimize confusion.