EEVblog Electronics Community Forum

Electronics => Projects, Designs, and Technical Stuff => Topic started by: richard.cs on May 14, 2015, 09:55:47 pm

Title: Modelling Roman Black's boost converter in LT spice.
Post by: richard.cs on May 14, 2015, 09:55:47 pm
http://www.romanblack.com/smps/conv.htm (http://www.romanblack.com/smps/conv.htm)

I'm trying to get a spice simulation of this to oscillate, with the intention to tweak it for another purpose. I have tried:
The "startup" option (ramps DC source)
Driving it from a pulsed voltage source to get a sharper startup transient
.ic for inductor current
.ic for the initial voltage on the 1n
Real inductor models from the library
An inductor model that includes saturation (at various currents)
Tweaking various component values hopefully
Combinations of the above.

Whatever I do it settles down to a steady DC state :( - I'm not sure if I've made some silly wiring error or if it's fundamentally un-simulatable because it relies on some component parasitic that isn't modeled. I could use another pair of eyes on it please.

Edit: Not being allowed to upload .asc files on an electronics forum is a bit sucky.
Title: Re: Modelling Roman Black's boost converter in LT spice.
Post by: fmaimon on May 14, 2015, 10:18:31 pm
Try using the BC337-40, as it has higher hfe, or maybe lower R1. In my first try, Q1 (Q3 in your schematic) collector current was limited by it.

I've attached my circuit, renamed to .txt instead of .asc
Title: Re: Modelling Roman Black's boost converter in LT spice.
Post by: Anand on May 14, 2015, 10:22:48 pm
In the http://www.romanblack.com/smps/conv.htm (http://www.romanblack.com/smps/conv.htm) I see the collector of Q2 connected to the base of Q3, not the emmiter.
Title: Re: Modelling Roman Black's boost converter in LT spice.
Post by: c4757p on May 14, 2015, 10:33:00 pm
Quote from: Anand on May 14, 2015, 10:22:48 pm
In the http://www.romanblack.com/smps/conv.htm (http://www.romanblack.com/smps/conv.htm) I see the collector of Q2 connected to the base of Q3, not the emmiter.

Yes!

That transistor resets the switching cycle by pulling Q3's base low. It doesn't make any sense to pull its emitter instead.
Title: Re: Modelling Roman Black's boost converter in LT spice.
Post by: richard.cs on May 14, 2015, 10:44:12 pm
Quote from: Anand on May 14, 2015, 10:22:48 pm
In the http://www.romanblack.com/smps/conv.htm (http://www.romanblack.com/smps/conv.htm) I see the collector of Q2 connected to the base of Q3, not the emmiter.

Thank you! I have no idea how I missed that.  :-[

And lo it springs into life.
Title: Re: Modelling Roman Black's boost converter in LT spice.
Post by: Zero999 on May 15, 2015, 05:06:22 pm
It's more efficient to connect Q2's base to D2 so it shuts off when D2 conducts.

Copy and paste the code into a text editor and save as an .asc.
Code: [Select]
Version 4
SHEET 1 1012 680
WIRE -272 -112 -480 -112
WIRE -80 -112 -272 -112
WIRE -32 -112 -80 -112
WIRE 112 -112 48 -112
WIRE 304 -112 112 -112
WIRE 416 -112 304 -112
WIRE 672 -112 480 -112
WIRE 816 -112 672 -112
WIRE 944 -112 816 -112
WIRE -80 -48 -80 -112
WIRE 304 32 304 -112
WIRE 112 80 112 -112
WIRE -80 128 -80 32
WIRE 16 128 -80 128
WIRE 48 128 16 128
WIRE -480 176 -480 -112
WIRE 944 176 944 -112
WIRE -272 208 -272 -112
WIRE 16 208 16 128
WIRE 400 208 16 208
WIRE 816 208 816 -112
WIRE 672 224 672 -112
WIRE 400 256 400 208
WIRE -480 288 -480 256
WIRE 112 304 112 176
WIRE 192 304 112 304
WIRE 304 304 304 112
WIRE 304 304 272 304
WIRE 336 304 304 304
WIRE 672 304 672 288
WIRE 672 304 336 304
WIRE 112 336 112 304
WIRE 304 368 304 304
WIRE -272 480 -272 272
WIRE 112 480 112 416
WIRE 304 480 304 432
WIRE 400 480 400 352
WIRE 816 480 816 272
WIRE 944 480 944 256
FLAG 112 480 0
FLAG 304 480 0
FLAG 400 480 0
FLAG 816 480 0
FLAG 944 480 0
FLAG -272 480 0
FLAG -480 288 0
SYMBOL npn 48 80 R0
SYMATTR InstName Q1
SYMATTR Value BC337-40
SYMBOL npn 336 256 R0
SYMATTR InstName Q2
SYMATTR Value BC337-40
SYMBOL res -96 -64 R0
SYMATTR InstName R1
SYMATTR Value 10k
SYMBOL res 96 320 R0
SYMATTR InstName R2
SYMATTR Value 6.8
SYMBOL res 288 16 R0
SYMATTR InstName R3
SYMATTR Value 15k
SYMBOL res 288 288 R90
WINDOW 0 0 56 VBottom 2
WINDOW 3 32 56 VTop 2
SYMATTR InstName R4
SYMATTR Value 1k
SYMBOL ind -48 -96 R270
WINDOW 0 32 56 VTop 2
WINDOW 3 5 56 VBottom 2
SYMATTR InstName L1
SYMATTR Value 470µ
SYMATTR SpiceLine Rser=4.9
SYMBOL cap 288 368 R0
SYMATTR InstName C1
SYMATTR Value 1n
SYMBOL cap -288 208 R0
SYMATTR InstName C2
SYMATTR Value 47µ
SYMBOL diode 416 -96 R270
WINDOW 0 32 32 VTop 2
WINDOW 3 0 32 VBottom 2
SYMATTR InstName D1
SYMATTR Value 1N4148
SYMBOL cap 800 208 R0
SYMATTR InstName C3
SYMATTR Value 10µ
SYMBOL zener 688 288 R180
WINDOW 0 24 64 Left 2
WINDOW 3 24 0 Left 2
SYMATTR InstName D2
SYMATTR Value BZX84C12L
SYMATTR Description Diode
SYMATTR Type diode
SYMBOL res 928 160 R0
SYMATTR InstName R5
SYMATTR Value 10k
SYMBOL voltage -480 160 R0
WINDOW 123 0 0 Left 2
WINDOW 39 0 0 Left 2
SYMATTR InstName V1
SYMATTR Value 5
TEXT -512 504 Left 2 !.tran 10m
(https://www.eevblog.com/forum/projects/modelling-roman-black%27s-boost-converter-in-lt-spice/?action=dlattach;attach=152116;image)
Title: Re: Modelling Roman Black's boost converter in LT spice.
Post by: richard.cs on May 15, 2015, 05:53:18 pm
What I'm actually doing with it:
Modifying for 9V in, ~100V output open-circuit
Adding feedback for constant 1mA (might make this switchable to 100 uA)
Combine with cheap ebay DMM module for measuring breakdown voltages of zeners, forward voltages of HV diode stacks, etc. the item under test takes the place of R6.

I can always use a bench supply to do the same thing but it's a bit of a pain when I only want a quick check. I've just realised that it needs a further tweak - I want the current to be limited even for things that breakdown at <9V. Maybe I'll make L1 a transformer so there's no DC path.

Thanks for the help :-)