what I would use for a low current 100Volt supply.

[Jamie]

This is my contributions to the already going thread on a 100 volt
capable DC-DC buck boost using a 555 and a zener for regulation that
uses the minimum parts. I don't even have a timing cap in there.

Set R1 for the Max I you need as the load and the zener will suck
up what you don't use to maintain a constant voltage.. Of course this
type of boost is only good if your load constraints are not all over the
place.

Replace R1 with a PNP current sink that can be reversed bias with a
feed back regulator and you can have a large output unit with low stand
by current, instead if using a Zener like I do with low current demands

Version 4
SHEET 1 880 680
WIRE -128 -80 -256 -80
WIRE -112 -80 -128 -80
WIRE 320 -80 -112 -80
WIRE -256 -48 -256 -80
WIRE 432 -32 0 -32
WIRE -112 32 -112 -80
WIRE 320 64 320 -80
WIRE 320 64 304 64
WIRE 336 64 320 64
WIRE 432 64 432 -32
WIRE 432 64 416 64
WIRE 464 64 432 64
WIRE 464 96 464 64
WIRE -112 128 -112 112
WIRE -112 128 -208 128
WIRE 0 128 0 -32
WIRE 80 128 0 128
WIRE -112 144 -112 128
WIRE -32 192 -48 192
WIRE 80 192 48 192
WIRE 432 192 432 64
WIRE 432 192 304 192
WIRE 464 192 464 176
WIRE 528 192 464 192
WIRE 608 192 592 192
WIRE 464 208 464 192
WIRE 608 208 608 192
WIRE -112 256 -112 240
WIRE 80 256 64 256
WIRE 400 256 384 256
WIRE -208 320 -208 128
WIRE 384 320 384 256
WIRE 384 320 -208 320
WIRE 464 320 464 304
WIRE 608 320 608 272
FLAG -256 32 0
FLAG 80 64 0
FLAG 608 320 0
FLAG -128 -80 P+
FLAG 64 256 P+
FLAG 464 320 0
FLAG -112 256 0
SYMBOL Misc\\NE555 192 160 R0
SYMATTR InstName U1
SYMBOL voltage -256 -64 R0
WINDOW 123 0 0 Left 2
WINDOW 39 0 0 Left 2
SYMATTR InstName V1
SYMATTR Value 10
SYMBOL res 432 48 R90
WINDOW 0 0 56 VBottom 2
WINDOW 3 32 56 VTop 2
SYMATTR InstName R1
SYMATTR Value 250
SYMBOL ind 448 80 R0
SYMATTR InstName L1
SYMATTR Value 1mh
SYMATTR SpiceLine Rser=0.1
SYMBOL zener 528 208 R270
WINDOW 0 32 32 VTop 2
WINDOW 3 52 118 VBottom 2
SYMATTR InstName D1
SYMATTR Value DFLZ33
SYMATTR Description Diode
SYMATTR Type diode
SYMBOL cap 592 208 R0
SYMATTR InstName C1
SYMATTR Value 1µf
SYMBOL npn 400 208 R0
SYMATTR InstName Q1
SYMATTR Value 2N5550
SYMBOL npn -48 144 M0
SYMATTR InstName Q2
SYMBOL res 64 176 R90
WINDOW 0 0 56 VBottom 2
WINDOW 3 32 56 VTop 2
SYMATTR InstName R3
SYMATTR Value 1200
SYMBOL res -128 16 R0
SYMATTR InstName R4
SYMATTR Value 1200
TEXT -240 360 Left 0 !.tran 0 0.010 0 0.01 startup

Jamie

 

[petrus bitbyter]

This is my contributions to the already going thread on a 100 volt
capable DC-DC buck boost using a 555 and a zener for regulation that
uses the minimum parts. I don't even have a timing cap in there.

Set R1 for the Max I you need as the load and the zener will suck
up what you don't use to maintain a constant voltage.. Of course this
type of boost is only good if your load constraints are not all over the
place.

Replace R1 with a PNP current sink that can be reversed bias with a feed
back regulator and you can have a large output unit with low stand by
current, instead if using a Zener like I do with low current demands

Version 4
SHEET 1 880 680
WIRE -128 -80 -256 -80
WIRE -112 -80 -128 -80
WIRE 320 -80 -112 -80
WIRE -256 -48 -256 -80
WIRE 432 -32 0 -32
WIRE -112 32 -112 -80
WIRE 320 64 320 -80
WIRE 320 64 304 64
WIRE 336 64 320 64
WIRE 432 64 432 -32
WIRE 432 64 416 64
WIRE 464 64 432 64
WIRE 464 96 464 64
WIRE -112 128 -112 112
WIRE -112 128 -208 128
WIRE 0 128 0 -32
WIRE 80 128 0 128
WIRE -112 144 -112 128
WIRE -32 192 -48 192
WIRE 80 192 48 192
WIRE 432 192 432 64
WIRE 432 192 304 192
WIRE 464 192 464 176
WIRE 528 192 464 192
WIRE 608 192 592 192
WIRE 464 208 464 192
WIRE 608 208 608 192
WIRE -112 256 -112 240
WIRE 80 256 64 256
WIRE 400 256 384 256
WIRE -208 320 -208 128
WIRE 384 320 384 256
WIRE 384 320 -208 320
WIRE 464 320 464 304
WIRE 608 320 608 272
FLAG -256 32 0
FLAG 80 64 0
FLAG 608 320 0
FLAG -128 -80 P+
FLAG 64 256 P+
FLAG 464 320 0
FLAG -112 256 0
SYMBOL Misc\\NE555 192 160 R0
SYMATTR InstName U1
SYMBOL voltage -256 -64 R0
WINDOW 123 0 0 Left 2
WINDOW 39 0 0 Left 2
SYMATTR InstName V1
SYMATTR Value 10
SYMBOL res 432 48 R90
WINDOW 0 0 56 VBottom 2
WINDOW 3 32 56 VTop 2
SYMATTR InstName R1
SYMATTR Value 250
SYMBOL ind 448 80 R0
SYMATTR InstName L1
SYMATTR Value 1mh
SYMATTR SpiceLine Rser=0.1
SYMBOL zener 528 208 R270
WINDOW 0 32 32 VTop 2
WINDOW 3 52 118 VBottom 2
SYMATTR InstName D1
SYMATTR Value DFLZ33
SYMATTR Description Diode
SYMATTR Type diode
SYMBOL cap 592 208 R0
SYMATTR InstName C1
SYMATTR Value 1µf
SYMBOL npn 400 208 R0
SYMATTR InstName Q1
SYMATTR Value 2N5550
SYMBOL npn -48 144 M0
SYMATTR InstName Q2
SYMBOL res 64 176 R90
WINDOW 0 0 56 VBottom 2
WINDOW 3 32 56 VTop 2
SYMATTR InstName R3
SYMATTR Value 1200
SYMBOL res -128 16 R0
SYMATTR InstName R4
SYMATTR Value 1200
TEXT -240 360 Left 0 !.tran 0 0.010 0 0.01 startup

Jamie

It's not clear to me what current you need but the voltage does not even
comes close to the 100V you say you want to achieve and that's without load.

If you really want 100V with some mA you better go for an old MC34063. It's
not state of the art and has some drawbacks but you'll get what you want.

petrus bitbyter

 

[Nico Coesel]

This is my contributions to the already going thread on a 100 volt
capable DC-DC buck boost using a 555 and a zener for regulation that
uses the minimum parts. I don't even have a timing cap in there.

Set R1 for the Max I you need as the load and the zener will suck
up what you don't use to maintain a constant voltage.. Of course this
type of boost is only good if your load constraints are not all over the
place.

In the past I've used the NE555 several times for these sort of
step-up converters. Its works well and doesn't need -to quote Joerg-
boutique parts. At a point one of my former employers was so short on
money I had to design a step-up converter from parts that where on
stock. Well... that included an NE555.

 

[Jamie]

It's not clear to me what current you need but the voltage does not even
comes close to the 100V you say you want to achieve and that's without load.

I know, I used a 33V zener that was sitting in LTSPICE.. The point is,
with a 100V zener, it will do it just as well as it does at 33v.

If I knew more about making models in Spice I would of simply done
that how ever, I do more bench testing than I do spice sims. I find
that many
times live experiments yields different results how ever, I do know this
one happens to work on the bench and in spice the same way.


Jamie

 

[Jamie]

It is nice, can it be done without the inverter transistor?

I did that because it needed to be inverted, as you can see. If you were
looking for a (-) output then the circuit could be altered and yes, only
one transistor is needed, since then, you'd be using a PNP on the top
side and the output of the timer would not need to be inverted.

Using a P-channel mosfet as an follower instead sounds attractive how
ever, it'll just drop into linear mode when the output on the source
gets to the point where there isn't enough gate voltage from the timer.

I've done this also with a dual comparator (8 pin package) and set up
a window trigger and threshold, like the timer is. With that, you can
simply alter the inputs that are used and you only need one transistor
on the buck switch.

It's just with a timer, the parts count is low and it has a hefty
output.

Something to think about I guess

Jamie

 

[Nico Coesel]

I couldn't find the link to the previous thread, but wouldn't a PIC10
or any other small micro be simpler. You get more features, softstart
etc

Cycle-by-cycle current limiting is a piece of cake with an NE555. This
is one of my designs to create 55V for a few hundred uA (post
regulators not shown):

http://img840.imageshack.us/img840/3929/ne555stepup.jpg

 

[Tim Williams]

Too lazy to use LTSpice and look at your circuit but it likely pales in
comparison to the discrete simplicity of mine,
http://myweb.msoe.edu/williamstm/Images/Discrete_Flyback_HV_Schem.jpg
Was using a PCB of it earlier today with breadboarded avalanche
experiments. Short circuit, current limit, whatever, it works fine. A
couple mA output, low efficiency, unbreakable*. Hottest component is the
toroid core, because powdered iron sucks at any frequency.

Tim