Maker Pro
Maker Pro

Arbitrary Voltage boost circuit???

I know the basics of charge pumps but they seem to be limited to
providing integer multiples of the input voltage.

I'm looking for a circuit that given a single available input voltage,
Vin, provides an output voltage, Vout, such that Vout = Vin+10.0volts.
(nice if the circuit actually was tweakable... Vout = Vin + Vadd)

Current requirements for Vout are small. 10-20ma would be
sufficient I think.

Vin should be relatively wide, 12v - 170v would be ideal, 12-48v
would be ok too.

Efficiency doesn't have to be good and output ripple can be pretty
insane. I want to use it to drive high side mosfets on indefinitely.
The gate charge is very tolerant so ripple voltages of a couple of
volts could be handled.

My basic question is: Is such a circuit feasible and simple?
Where should I look for such a beast?

My first thought is since Vin > 10v (or basically Vin > Vadd)
you could use a simple doubling charge pump with the
flying capacitor connected to a current limited
zener diode with a breakdown voltage of 10V (or Vadd).
But at higher voltages this doesn't seem to work well since
the flying capacitor could only be charged with about 10mA
without exceeding the power rating of common zeners.
And of course efficiency would be ultra terrible.

Is there a better way given a Vin to charge a capacitor to
Vadd where Vadd < Vin?

I'm not an electrical engineer so am I missing something
relatively simple?
 
J

Jim Thompson

Jan 1, 1970
0
I know the basics of charge pumps but they seem to be limited to
providing integer multiples of the input voltage.

No. The step-up per section is equal to the P-P clock voltage.
I'm looking for a circuit that given a single available input voltage,
Vin, provides an output voltage, Vout, such that Vout = Vin+10.0volts.
(nice if the circuit actually was tweakable... Vout = Vin + Vadd)

Current requirements for Vout are small. 10-20ma would be
sufficient I think.

Vin should be relatively wide, 12v - 170v would be ideal, 12-48v
would be ok too.

Efficiency doesn't have to be good and output ripple can be pretty
insane. I want to use it to drive high side mosfets on indefinitely.
The gate charge is very tolerant so ripple voltages of a couple of
volts could be handled.

My basic question is: Is such a circuit feasible and simple?
Where should I look for such a beast?

My first thought is since Vin > 10v (or basically Vin > Vadd)
you could use a simple doubling charge pump with the
flying capacitor connected to a current limited
zener diode with a breakdown voltage of 10V (or Vadd).
But at higher voltages this doesn't seem to work well since
the flying capacitor could only be charged with about 10mA
without exceeding the power rating of common zeners.
And of course efficiency would be ultra terrible.

Is there a better way given a Vin to charge a capacitor to
Vadd where Vadd < Vin?

I'm not an electrical engineer so am I missing something
relatively simple?


...Jim Thompson
 
T

Tim Wescott

Jan 1, 1970
0
I know the basics of charge pumps but they seem to be limited to
providing integer multiples of the input voltage.

I'm looking for a circuit that given a single available input voltage,
Vin, provides an output voltage, Vout, such that Vout = Vin+10.0volts.
(nice if the circuit actually was tweakable... Vout = Vin + Vadd)

Current requirements for Vout are small. 10-20ma would be
sufficient I think.

Vin should be relatively wide, 12v - 170v would be ideal, 12-48v
would be ok too.

Efficiency doesn't have to be good and output ripple can be pretty
insane. I want to use it to drive high side mosfets on indefinitely.
The gate charge is very tolerant so ripple voltages of a couple of
volts could be handled.

My basic question is: Is such a circuit feasible and simple?
Where should I look for such a beast?

My first thought is since Vin > 10v (or basically Vin > Vadd)
you could use a simple doubling charge pump with the
flying capacitor connected to a current limited
zener diode with a breakdown voltage of 10V (or Vadd).
But at higher voltages this doesn't seem to work well since
the flying capacitor could only be charged with about 10mA
without exceeding the power rating of common zeners.
And of course efficiency would be ultra terrible.

Is there a better way given a Vin to charge a capacitor to
Vadd where Vadd < Vin?

I'm not an electrical engineer so am I missing something
relatively simple?
Do you have a source of more-or-less steady voltage? If you do you
should be able to drive a transformer primary with a square wave,
reference the secondary to your Vin, and rectify the output to be above Vin.

--

Tim Wescott
Wescott Design Services
http://www.wescottdesign.com

Posting from Google? See http://cfaj.freeshell.org/google/

"Applied Control Theory for Embedded Systems" came out in April.
See details at http://www.wescottdesign.com/actfes/actfes.html
 
G

Genome

Jan 1, 1970
0
I know the basics of charge pumps but they seem to be limited to
providing integer multiples of the input voltage.

I'm looking for a circuit that given a single available input voltage,
Vin, provides an output voltage, Vout, such that Vout = Vin+10.0volts.
(nice if the circuit actually was tweakable... Vout = Vin + Vadd)

This am quite simple stuff unless you am stupid.

All you am need is volt reg thing and make reference voltage Vin + Vadd.

Ov course if you am stupid then education am not wurf bovring wiv.

Not problim EBAY.

DNA
 
G

Genome

Jan 1, 1970
0
Genome said:
This am quite simple stuff unless you am stupid.

All you am need is volt reg thing and make reference voltage Vin + Vadd.

Ov course if you am stupid then education am not wurf bovring wiv.

Not problim EBAY.

DNA

See, so now I have solved your problim what's this stuff about bacon in a
fring pan along wiv toasty rolls and delicatlye tosted bunt cheese wiv
crispy stuff plus softlee frid egg.

DNA
 
G

Genome

Jan 1, 1970
0
Genome said:
See, so now I have solved your problim what's this stuff about bacon in a
fring pan along wiv toasty rolls and delicatlye tosted bunt cheese wiv
crispy stuff plus softlee frid egg.

DNA

There you go. YOU ABSOLUTE BASTARD YOU.

Come along and nance about with your nancy bollocks shite from Google and
then you bugger off.

FUCKING SCUM PIECE OF SCUM ****.

YOU SELL LOANS TO YOUR MATES AND THEN **** THEM OVER FOR A JOB AT HAMBURGERS
R US TO APPEAR ON JUDGE JUDY.

Noncy boi.

DNA
 
P

Paul E. Schoen

Jan 1, 1970
0
I know the basics of charge pumps but they seem to be limited to
providing integer multiples of the input voltage.

I'm looking for a circuit that given a single available input voltage,
Vin, provides an output voltage, Vout, such that Vout = Vin+10.0volts.
(nice if the circuit actually was tweakable... Vout = Vin + Vadd)

Current requirements for Vout are small. 10-20ma would be
sufficient I think.

Vin should be relatively wide, 12v - 170v would be ideal, 12-48v
would be ok too.

Efficiency doesn't have to be good and output ripple can be pretty
insane. I want to use it to drive high side mosfets on indefinitely.
The gate charge is very tolerant so ripple voltages of a couple of
volts could be handled.

My basic question is: Is such a circuit feasible and simple?
Where should I look for such a beast?

My first thought is since Vin > 10v (or basically Vin > Vadd)
you could use a simple doubling charge pump with the
flying capacitor connected to a current limited
zener diode with a breakdown voltage of 10V (or Vadd).
But at higher voltages this doesn't seem to work well since
the flying capacitor could only be charged with about 10mA
without exceeding the power rating of common zeners.
And of course efficiency would be ultra terrible.

Is there a better way given a Vin to charge a capacitor to
Vadd where Vadd < Vin?

I'm not an electrical engineer so am I missing something
relatively simple?

I think what you are looking for is a bootstrap charge pump doubler, often
used to drive high-side IGBTs in motor controllers. Here is a good app
note:

http://www.ixys.com/tmosign1.pdf

A rough sketch of the circuit is:

ZD1
/
+-->|----+
| / |
Vin-----------+---------+---)|---+------Vout
| | C2 |
+---+ | |
|Q1 | V D1 |
+---+ --- |
| | |
+----)|---+-->|----+
| C1
+---+
|Q2 |
+---+
|
GND-----------+--------

The idea is that C1 is pumped to Vin when Q2 conducts. When Q1 conducts,
you get nearly 2*Vin at Vout, except it is limited by ZD1. You may need a
resistor across Q1 to get the pump primed. Choose C1 and C2 according to
switching frequency and output current requirements. It is not highly
efficient, but it only loses perhaps one or two watts out of a motor
control that could be thousands of watts.

BTW, the schematic in the link from IXYS shows ZD1 connected to GND, which
I believe is incorrect. I think it must go to Vin to get Vin + VZD1 on
Vout.

Paul
 
This does not seem that hard. Generate a variable regulated voltage
(i.e same as your 10V or whatever you want to add to Vin) with an LDO
or op-amp. Power a clock or squarewave generator, about 100kHz (freq
not important but higher f means smaller caps can be used) from that.
Some CMOS gates should do (in parallel if you need more current) for
the clock output. From the clock output connect a cap (maybe 0.1uF and
rated at the max Vout). Connect the other side of the cap to the center
of 2 series-connected diodes (also rated for max Vout). The diodes
point in the same direction. Connect the cathode (diode arrow pointing)
end of the diode pair to another 0.1uF cap (that is your output).
Connect the anode end of diode pair to your Vin. Vout will now be Vin +
variable reg output - two diodes. You may have to mess with the freq
and cap values to get output current and ripple to your liking.
 
Yes. The high side rail of the high side mosfets is a stable source.
The problem is that it is not fixed from application to application.

With some motors, 24V is connected to the high side while others
require 48V and so on. So even though the step-up is not a
multiple but rather fixed this strikes me as not working. Whatever
I build to add 10V ontop of 24V will add 20V on top of 48V when
different motors are used and this will eventually lead to a gate
to drain voltage that exceeds maximum and breaks down the
insulation layer.

The trick I think is to get a wide range DC-DC power supply
that given some input 10 - 180V (180V let's me use a very simple
rectifier bridge and large cap. as an AC-DC power supply. Ripple
in the motors isn't much of a problem.)

Anyhow. I would think if I had such a DC-DC converter (none
of the simple switcher national parts go this high though) then I
could repeatedly charge a cap from the 10V regulated source and
pump it on top of the high rail to get a constant step-up of 10V
regardless of the available power supply.

Thanks,
 
Fantastic!!! Tie ground to the high rail primary voltage that I
already have, tie the output (which will be Vcc+Vadd) to the
gate logic like I want and tie the voltage regulator's input to...
what?

If I had a secondary voltage high enough to drive a voltage regulator
then I wouldn't need the voltage regulator would I? The problem is
getting
the secondary voltage in the first place and also a secondary voltage
that is of a fix value relative to Vcc and not proportional.

Was your account highjacked and abused or are you just naturally
this unfriendly?
 
Top