How to fix LTspice schematic
in [Design]
|
From: Robert Baer on 1 Oct 2006 06:14 John Smith wrote: > "Robert Baer" <robertbaer(a)earthlink.net> wrote in message > news:DRGSg.7187$UG4.3709(a)newsread2.news.pas.earthlink.net... > >> I do not understand the error message and have no clue to fix it. >> The inverter may be the problem; would like it to "emulate" a TPS2818 >>but do not know how. >> >> > > > Robert - > > The inverter is not your problem. Don't worry with trying to simulate a > TPS2818 yet. > > It looks like you are trying to simulate a regulated high voltage power > supply. Except for the effects of diode capacitance, the input to U2+ will > remain low until D3 and D4 avalanche. Is that what you want? How much > voltage across R1 do you desire? > > It would be much better if you posted your requrements rather than a SPICE > netlist. > > Cheers, > John > > I made massive changes and have something working; used NPN/PNP pair as an inverting gate driver. In the process, i have found a classical multivibrator does not work: Version 4 SHEET 1 1444 1224 WIRE -480 624 -512 624 WIRE -368 624 -416 624 WIRE -288 624 -368 624 WIRE -144 624 -208 624 WIRE -512 672 -512 624 WIRE 0 720 -304 720 WIRE 0 752 0 720 WIRE -304 768 -304 720 WIRE -368 816 -368 624 WIRE -336 816 -368 816 WIRE -144 848 -144 624 WIRE -144 848 -192 848 WIRE -112 848 -144 848 WIRE 0 864 0 832 WIRE -496 880 -576 880 WIRE -368 880 -432 880 WIRE -336 880 -368 880 WIRE -576 912 -576 880 WIRE -304 944 -304 928 WIRE -16 944 -304 944 WIRE -16 976 -16 944 WIRE -16 1088 -16 1056 WIRE -576 1104 -576 992 WIRE -464 1104 -576 1104 WIRE -368 1104 -368 880 WIRE -368 1104 -384 1104 WIRE -288 1104 -368 1104 WIRE -144 1104 -144 848 WIRE -144 1104 -208 1104 WIRE -576 1152 -576 1104 FLAG -512 672 0 FLAG -576 1152 0 FLAG -16 1088 0 FLAG 0 864 0 FLAG -112 848 Vo SYMBOL voltage -16 1072 R180 WINDOW 123 0 0 Left 0 WINDOW 39 0 0 Left 0 SYMATTR InstName V2 SYMATTR Value 12V SYMBOL Opamps\\LT6011 -304 848 R0 SYMATTR InstName U2 SYMBOL voltage 0 736 R0 WINDOW 123 0 0 Left 0 WINDOW 39 0 0 Left 0 SYMATTR InstName V1 SYMATTR Value 12V SYMBOL res -192 608 R90 WINDOW 0 0 56 VBottom 0 WINDOW 3 32 56 VTop 0 SYMATTR InstName R1 SYMATTR Value 10K SYMBOL res -192 1088 R90 WINDOW 0 0 56 VBottom 0 WINDOW 3 32 56 VTop 0 SYMATTR InstName R2 SYMATTR Value 1m SYMBOL res -368 1088 R90 WINDOW 0 0 56 VBottom 0 WINDOW 3 32 56 VTop 0 SYMATTR InstName R3 SYMATTR Value 47 SYMBOL cap -416 608 R90 WINDOW 0 0 32 VBottom 0 WINDOW 3 32 32 VTop 0 SYMATTR InstName C1 SYMATTR Value 1? SYMBOL cap -432 864 R90 WINDOW 0 0 32 VBottom 0 WINDOW 3 32 32 VTop 0 SYMATTR InstName C2 SYMATTR Value 0.1? SYMBOL res -592 896 R0 SYMATTR InstName R4 SYMATTR Value 1e15 TEXT -98 650 Left 0 !.tran 0 10s 1U
From: Johnny Boy on 1 Oct 2006 09:05 "Robert Baer" <robertbaer(a)earthlink.net> wrote in message news:IsMTg.1351$Lv3.343(a)newsread1.news.pas.earthlink.net... > Johnny Boy wrote: > > > "Robert Baer" <robertbaer(a)earthlink.net> wrote in message > > news:fplTg.3005$Y24.2825(a)newsread4.news.pas.earthlink.net... > > > >>Johnny Boy wrote: > >> > >> > >>>"Robert Baer" <robertbaer(a)earthlink.net> wrote in message > >>>news:nb3Tg.363$Lv3.147(a)newsread1.news.pas.earthlink.net... > >>> > >>><snip> > >>> > >>>> That might help; how does one specify Vhigh and Trise for A1? > >>>> I was also wondering abuut a power supply for the inverter, but since > >>>>this is an emulator, maybe one does not need to be specified? > >>> > >>><snip> > >>> > >>>Don't know about the inverter supply, but for Vhigh/low and Trise/fall, > >>>right-click the inverter, then copy the following into the "SpiceLine" > >>>attribute, (edit to suit).:- > >>> > >>>Vhigh=12 Vlow=0 Trise=100n Tfall=100n Td=100n > >>> > >>>(Thanks to DNA - I found this in his alternative to your circuit.) > >>> > >>>When I first ran this simulation, I got the same error as you, but after > >>>adding the above line to the inverter attributes, I get a different > > > > error:- > > > >>>Fatal Error: Singular matrix: check nodes n013 and u1:1:va+ > >>> Iteration No. 2 > >>> > >>> > >>>... Johnny > >>> > >>> > >> > >> That circuit he gave is strange: > >>1) "Rl" is not from output to ground, where a practical load would go. > >>2) R1 allows Vout to go to 1.2KV with a small ripple; without Rl, the > >>output only goes to 500V with a lot of ripple. > >>3) Increasing Rf by a factor of 5 has no effect on output voltage. > > > > > > This surprises me - when I manually calculate feedback with the given > > values, the feedback voltage to the inverters is 2.5V as it should be for > > 1kV output. The ripple is determined by the current limit via the 100R > > resistor. > > > > > >>4) Output voltage ramps up in a *linear* fashion. > > > > > > Shouldn't it do this? > > > > > >>5) Have not found what limits/controls the output voltage. > > > > > > See my reply to >3) > > > > > >>6) M1 operates in the linear mode, but "switches between two current > > > > levels. > > > >>7) Looks like the output could be in the 10KV region with minor mods > >>(*not* changing the multiplier). > > > > > > > > I should point out that I'm a complete novice to LTSpice, with 2 days > > experience. > > Perhaps you should try posting this problem to the Yahoo LTSpice group. > > > > ... Johnny > > > > > > > ...and *where* does that "2.5V" value come from????? (1.2kV/(48M + 100K + 1K)) * 101K = 2.5V > Oh, i forgot one other strange item: the lines / wires on the diodes > are *not* at right angles! > How the &#$()#$ was that done??? Dunno - ask DNA (Genome) .... Johnny
From: Johnny Boy on 1 Oct 2006 09:10 "Robert Baer" <robertbaer(a)earthlink.net> wrote in message news:IsMTg.1351$Lv3.343(a)newsread1.news.pas.earthlink.net... > Oh, i forgot one other strange item: the lines / wires on the diodes > are *not* at right angles! > How the &#$()#$ was that done??? I just had a go at this - if you first place a line, then drag the end of it, it's not hard to make it 45 degrees instead of 90. (That's how the &#$()#$ it was done, I guess) .... Johnny
From: Robert Baer on 1 Oct 2006 23:43 Johnny Boy wrote: > "Robert Baer" <robertbaer(a)earthlink.net> wrote in message > news:IsMTg.1351$Lv3.343(a)newsread1.news.pas.earthlink.net... > >>Johnny Boy wrote: >> >> >>>"Robert Baer" <robertbaer(a)earthlink.net> wrote in message >>>news:fplTg.3005$Y24.2825(a)newsread4.news.pas.earthlink.net... >>> >>> >>>>Johnny Boy wrote: >>>> >>>> >>>> >>>>>"Robert Baer" <robertbaer(a)earthlink.net> wrote in message >>>>>news:nb3Tg.363$Lv3.147(a)newsread1.news.pas.earthlink.net... >>>>> >>>>><snip> >>>>> >>>>>> That might help; how does one specify Vhigh and Trise for A1? >>>>>> I was also wondering abuut a power supply for the inverter, but > > since > >>>>>>this is an emulator, maybe one does not need to be specified? >>>>> >>>>><snip> >>>>> >>>>>Don't know about the inverter supply, but for Vhigh/low and Trise/fall, >>>>>right-click the inverter, then copy the following into the "SpiceLine" >>>>>attribute, (edit to suit).:- >>>>> >>>>>Vhigh=12 Vlow=0 Trise=100n Tfall=100n Td=100n >>>>> >>>>>(Thanks to DNA - I found this in his alternative to your circuit.) >>>>> >>>>>When I first ran this simulation, I got the same error as you, but > > after > >>>>>adding the above line to the inverter attributes, I get a different >>> >>>error:- >>> >>> >>>>>Fatal Error: Singular matrix: check nodes n013 and u1:1:va+ >>>>> Iteration No. 2 >>>>> >>>>> >>>>>... Johnny >>>>> >>>>> >>>> >>>> That circuit he gave is strange: >>>>1) "Rl" is not from output to ground, where a practical load would go. >>>>2) R1 allows Vout to go to 1.2KV with a small ripple; without Rl, the >>>>output only goes to 500V with a lot of ripple. >>>>3) Increasing Rf by a factor of 5 has no effect on output voltage. >>> >>> >>>This surprises me - when I manually calculate feedback with the given >>>values, the feedback voltage to the inverters is 2.5V as it should be > > for > >>>1kV output. The ripple is determined by the current limit via the 100R >>>resistor. >>> >>> >>> >>>>4) Output voltage ramps up in a *linear* fashion. >>> >>> >>>Shouldn't it do this? >>> >>> >>> >>>>5) Have not found what limits/controls the output voltage. >>> >>> >>>See my reply to >3) >>> >>> >>> >>>>6) M1 operates in the linear mode, but "switches between two current >>> >>>levels. >>> >>> >>>>7) Looks like the output could be in the 10KV region with minor mods >>>>(*not* changing the multiplier). >>> >>> >>> >>>I should point out that I'm a complete novice to LTSpice, with 2 days >>>experience. >>>Perhaps you should try posting this problem to the Yahoo LTSpice group. >>> >>>... Johnny >>> >>> >>> >> >>...and *where* does that "2.5V" value come from????? > > > (1.2kV/(48M + 100K + 1K)) * 101K = 2.5V > > > >> Oh, i forgot one other strange item: the lines / wires on the diodes >>are *not* at right angles! >> How the &#$()#$ was that done??? > > > Dunno - ask DNA (Genome) > > ... Johnny > > I ment, where is that 2.5V reference?
From: Robert Baer on 1 Oct 2006 23:44
Johnny Boy wrote: > "Robert Baer" <robertbaer(a)earthlink.net> wrote in message > news:IsMTg.1351$Lv3.343(a)newsread1.news.pas.earthlink.net... > > >> Oh, i forgot one other strange item: the lines / wires on the diodes >>are *not* at right angles! >> How the &#$()#$ was that done??? > > > I just had a go at this - if you first place a line, then drag the end of > it, it's not hard to make it 45 degrees instead of 90. (That's how the > &#$()#$ it was done, I guess) > > ... Johnny > > Ahhh.... Thanks! |