LTspice models

[vk2tty]

Could y'all share your (tr)uSDX LTspice models with me? I think I'm doing OK but lack the experience or training to know either way.

Here's my model for the V12 supply through a random MOSFET and 80m band slot to the antenna. LTspice reports 5.6W out at 95% efficiency in TRAN mode, and a nice notch at 7.1MHz in AC mode:

sharp.png (74.22 KiB) Viewed 196 times

It also used to also show the same curve as my NanoVNA in AC mode when I hooked up a 12V sine wave to IN through a 50Ω resistor, and presumably still does:

09e62c00d86c7557.png (163.76 KiB) Viewed 196 times

That said:

• DL2MAN has warned us not to expect 95% efficiency with the FDT86256, so the model's efficiency report places it under suspicion
• I'm not using a model for the correct MOSFET because I lack one
• I haven't given the parasitic series resistance, parallel resistance, or parallel capacitance for the inductors because I don't know them.
• TRAN mode is showing a 0–44V sine wave on the band slot's input and I don't know why it's not -12–12
• There are probably mistakes I'm not noticing because I don't know any better

Screenshot 2023-12-26 at 9.28.13 pm.png (137.74 KiB) Viewed 196 times

Can you help?

Code: Select all

Version 4
SHEET 1 1764 988
WIRE -192 -944 -272 -944
WIRE -16 -944 -16 -976
WIRE -16 -944 -112 -944
WIRE -272 -880 -272 -944
WIRE -16 -880 -16 -944
WIRE -272 -768 -272 -816
WIRE -192 -768 -272 -768
WIRE -16 -768 -16 -800
WIRE -16 -768 -112 -768
WIRE -16 -672 -16 -768
WIRE -16 -608 -16 -672
WIRE -16 -480 -16 -528
WIRE -16 -480 -64 -480
WIRE 96 -480 -16 -480
WIRE -64 -432 -64 -480
WIRE -112 -352 -160 -352
WIRE -64 -288 -64 -336
WIRE -160 -144 -288 -144
WIRE -32 -144 -80 -144
WIRE 1168 -128 1120 -128
WIRE 1328 -128 1232 -128
WIRE -288 -64 -288 -144
WIRE 1120 -32 1120 -128
WIRE 1216 -32 1120 -32
WIRE 1328 -32 1328 -128
WIRE 1328 -32 1280 -32
WIRE 800 0 784 0
WIRE 912 0 864 0
WIRE -288 48 -288 16
WIRE 208 48 144 48
WIRE 352 48 208 48
WIRE 656 48 352 48
WIRE 784 48 784 0
WIRE 784 48 736 48
WIRE 912 64 912 0
WIRE 992 64 912 64
WIRE 1120 64 1120 -32
WIRE 1120 64 992 64
WIRE 1184 64 1120 64
WIRE 1328 64 1328 -32
WIRE 1328 64 1264 64
WIRE 1408 64 1328 64
WIRE 1520 64 1408 64
WIRE 352 96 352 48
WIRE 352 96 304 96
WIRE 400 96 352 96
WIRE 784 96 784 48
WIRE 832 96 784 96
WIRE 912 96 912 64
WIRE 912 96 896 96
WIRE 0 112 0 64
WIRE 304 112 304 96
WIRE 400 144 400 96
WIRE 1120 160 1120 64
WIRE 1120 160 1072 160
WIRE 1168 160 1120 160
WIRE 1328 160 1328 64
WIRE 1328 160 1280 160
WIRE 1376 160 1328 160
WIRE 1072 176 1072 160
WIRE 1280 176 1280 160
WIRE 1168 208 1168 160
WIRE 1376 208 1376 160
WIRE 0 224 0 192
WIRE 0 224 -240 224
WIRE 304 224 304 176
WIRE 352 224 304 224
WIRE 400 224 400 208
WIRE 400 224 352 224
WIRE 352 288 352 224
WIRE 1072 288 1072 240
WIRE 1120 288 1072 288
WIRE 1168 288 1168 272
WIRE 1168 288 1120 288
WIRE 1280 288 1280 240
WIRE 1328 288 1280 288
WIRE 1376 288 1376 272
WIRE 1376 288 1328 288
WIRE 1120 352 1120 288
WIRE 1328 352 1328 288
FLAG 352 288 0
FLAG 144 48 RF
IOPIN 144 48 In
FLAG 1520 64 ANT
IOPIN 1520 64 Out
FLAG 1120 352 0
FLAG 1328 352 0
FLAG 992 64 PAS
FLAG 1408 64 OUT
FLAG 208 48 IN
FLAG -288 48 0
FLAG -32 -144 DRIVE
IOPIN -32 -144 Out
FLAG -240 224 ANT
IOPIN -240 224 In
FLAG 0 64 0
FLAG -160 -352 DRIVE
IOPIN -160 -352 In
FLAG 96 -480 RF
IOPIN 96 -480 Out
FLAG -64 -288 0
FLAG -16 -976 Vin
FLAG -368 -560 0
FLAG -368 -640 Vin
FLAG -16 -672 VDRAIN
SYMBOL Cap 288 112 R0
SYMATTR InstName C50
SYMATTR Value 1n
SYMBOL ind 640 64 R270
WINDOW 0 32 56 VTop 2
WINDOW 3 5 56 VBottom 2
SYMATTR InstName L51
SYMATTR Value 1.3µ
SYMATTR SpiceLine Ipk=50 mfg="   ---" pn="   ---"
SYMBOL Cap 800 16 R270
WINDOW 0 32 32 VTop 2
WINDOW 3 0 32 VBottom 2
SYMATTR InstName C52
SYMATTR Value 2.2n
SYMBOL Cap 832 112 R270
WINDOW 0 32 32 VTop 2
WINDOW 3 0 32 VBottom 2
SYMATTR InstName C53
SYMATTR Value 560p
SYMBOL ind 1168 80 R270
WINDOW 0 32 56 VTop 2
WINDOW 3 5 56 VBottom 2
SYMATTR InstName L52
SYMATTR Value 680n
SYMATTR SpiceLine Ipk=55 Rser=0 Rpar=0 Cpar=0 mfg="   ---" pn="   ---"
SYMBOL Cap 1168 -112 R270
WINDOW 0 32 32 VTop 2
WINDOW 3 0 32 VBottom 2
SYMATTR InstName C56
SYMATTR Value 270p
SYMBOL Cap 1216 -16 R270
WINDOW 0 32 32 VTop 2
WINDOW 3 0 32 VBottom 2
SYMATTR InstName C57
SYMATTR Value 470p
SYMBOL Cap 1056 176 R0
SYMATTR InstName C54
SYMATTR Value 2.2n
SYMBOL Cap 1264 176 R0
SYMATTR InstName C58
SYMATTR Value 2.2n
SYMBOL voltage -288 -80 R0
WINDOW 123 24 118 Left 2
SYMATTR Value2 AC 12 0
SYMATTR InstName VS
SYMATTR Value SINE(0 5 {f} 0 0 0)
SYMBOL Res 16 208 R180
WINDOW 0 36 76 Left 2
WINDOW 3 36 40 Left 2
SYMATTR InstName RL
SYMATTR Value {Rload}
SYMBOL Res -176 -128 R270
WINDOW 0 32 56 VTop 2
WINDOW 3 0 56 VBottom 2
SYMATTR InstName RS
SYMATTR Value {Rsource}
SYMBOL nmos -112 -432 R0
SYMATTR InstName Q5
SYMATTR Value FDS6930A
SYMBOL Res -32 -896 R0
SYMATTR InstName R1
SYMATTR Value {Rshunt}
SYMBOL Ind -32 -624 R0
SYMATTR InstName L15
SYMATTR Value 200µ
SYMBOL voltage -368 -656 R0
SYMATTR InstName VS1
SYMATTR Value 12
SYMBOL Res -96 -960 R90
WINDOW 0 0 56 VBottom 2
WINDOW 3 32 56 VTop 2
SYMATTR InstName R7
SYMATTR Value 10
SYMBOL Res -96 -784 R90
WINDOW 0 0 56 VBottom 2
WINDOW 3 32 56 VTop 2
SYMATTR InstName R8
SYMATTR Value 10
SYMBOL Cap -256 -816 R180
WINDOW 0 24 56 Left 2
WINDOW 3 24 8 Left 2
SYMATTR InstName C2
SYMATTR Value 100n
TEXT 112 -256 Left 5 ;80m Band Low Pass Filter (tr)uSDX
TEXT 200 440 Left 2 ;Shunt Capacitance
TEXT 608 440 Left 2 ;Bandpass
TEXT 1016 440 Left 2 ;Pi Network
TEXT 264 -968 Left 2 !** NOTE: remove comment asterisk (*) from:\n** *.ac if you want to see what your NanoVNA sees\n** *.tran if you want to see how efficiency changes with f\n*.ac dec 1K 3Meg 16Meg\n.tran 0 .00010 .00009\n.options nomarch\n*.step param f list 3400kHz 3500kHz 3600kHz 3700kHz 3800kHz 3900kHz\n.param f 3600kHz \n.param Rsource=10\n.param Rload=50, Rshunt=0.017\n.MEAS AC nadir MIN mag(v(out))\n.MEAS AC zenith MAX mag(v(out))\n.options meascplxfmt=polar\n* vin=10.3 on AC, and I don't trust it:\n.MEAS TRAN Vin RMS(v(vdrain))\n.MEAS TRAN Vout RMS(v(out))\n.MEAS TRAN Ain RMS(i(L15))\n.MEAS TRAN Pin PARAM Vin*Ain\n* https://hackaday.com/2018/08/23/circuit-vr-measuring-with-ltspice/\n.MEAS TRAN Pout PARAM Vout**2/Rload\n.MEAS TRAN Eff PARAM Pout/Pin
RECTANGLE Normal 560 -192 176 416 2
RECTANGLE Normal 976 -192 592 416 2
RECTANGLE Normal 1008 -192 1440 416 2

[dl6sez]

Hello OM,

nice simulation, but does't help much to get the real behaviour
We know the boundries with ideal components since Dan Tayloe N7VE developped these filters in the 1990s, but nothing in electronics especially in RF is an ideal L or C or FET.... nor a simple drawn line in your schematics.
So we know ideal components lead to best efficieny and output power and you need a highest possible speed switching device. You can for sure make the circuit better, but not with the same amount of money and switchable for several bands.
All is already known about class-E PAs.

http://www.norcalqrp.org/files/Class_E_Amplifiers.pdf

73 and Merry Christmas de Chris

[vk2tty]

nothing in electronics especially in RF is an ideal L or C or FET....

Oh, for sure: that's why I'm asking about the parasitics on the inductors. I keep tripping over advice from homebrewers to try out designs in LTspice, so figured it'd be fun to reproduce the (tr)uSDX RF chain. I don't expect to improve on it.

Any advice on how to improve the model?

73 and Happy New Year, VK2TTY

[dl6sez]

Hello OM,

no i have no better toroid inductors model
Playing with LTspice is fun okay but don't think you can get a 100% fitting of a real RF circuit and therefore it is mostly, i say mostly, not possible to get a clue of how to adopt it to the next circuit.
What i want to say is that if you get a better fit of your 80m class-E PA you will not be able to predict how the 20m class-E PA acts exactly in your simulation.... at a good clue it will work but not exactly... you already have seen it
I simulated HEMTs (150GHz ft) and we also had a partner which tried to simulate the full HEMT from the structure on Wafer and the layers on GaAs. So both simulations had every time to be fitted to the real HEMT and couldn't predict when we change something at the HEMTs!
Simulations in RF is mostly for developing and if you need it exact you only have one choice to measure the behaviour and fit your models to show the behaviour.
But with this fit one can not really conclude exactly what happens at lower or higher frequencies precisely.

That's what i have learnt in my Semiconductor engineering job in former years.

73 de Chris and hope you slide very good into 2024