High Current negative voltage from a positive one

Thread Starter


Joined Jan 7, 2020

For a project I am working on I need to generate a negative voltage from a single positive one. More precisely, I need -18V from +18V with at least 1A.

For the first prototype, I am using an LT1054 (http://www.ti.com/lit/ds/symlink/lt1054.pdf) powered @+15V and everithing works as expected. Unfortunately, the LT1054 can supply only 100mA and I cannot load the circuit as per requirements.

I am not able to find any Voltage Inverter IC which can provide the required current.

My questions:

1) How can I generate the negative voltage (-15V@ >=1A from 15V or even better -18V@ >=1A from 18V) without using either an IC or a transformer with two secondary windings?

2) Can you suggest me any voltage inverter with the required specs?

Thank you in advance.


Joined Aug 21, 2008
Note: The schematics above are missing the feedback voltage divider.

MC34063 / KA34063 is made by more manufacturers than I can remember. This is good for your application when used with an external bipolar switching transistor. It only needs a suitable choke, not a transformer provided you don't need isolation.

At the link below you can download the datasheet with some applications information.

This application note tells all about and designing with it.

Thread Starter


Joined Jan 7, 2020
Hi Dick,
Thank you for your answer. I already saw that IC. If I am not wrong, it is not able to provide the required current though.
I didn't go through all the calculations myself but I used some of the calculators available on line (e.g. https://www.electronicproducts.com/DC-DC_Circuit_Calculator.aspx). Entering all the data:

Vin = 18V
Vout = -18V
Output current: 1000mA
Output Voltage ripple = 100mVpp
Frequency = 100KHz

I get this error message: Switch peak current 4164mA exceeds 1500mA limit!

According to the calculators, the actual maximum output current for my input / output voltages is only 360mA or 500mA@Vin=40V.


Joined Oct 7, 2019
Here is a 3A switch version of what you Dick and FB64 was talking about. Change the feed back resistor for 15V not 12V and it should work.


Joined Aug 21, 2008
The two circuits in post #2 have external transistors to boost the current. The question becomes: How many amps can you get though those transistors with up to 1.5 amps of base drive?


Joined Aug 21, 2017
To answer correctly some indications of HOW MUCH mighty the PS must be.
Generally, it is very very very in-vise to produce the negative strong out of positive weak. The main PS instead must be arranged let it gives out those polarity demanding a largest current, and smallest current then may be organized out of this.

However, mostly, at least in my understanding what is much, some 100A up to 1000A and more may be called "much". Thus the imminent problem about output cascade paralleling is spanking out. The most excellent solution for this problem ever I seen is applied in first generation of table PC, where core needs to have a 1 Volt 300 Amp PS, where out of ATX comes only 5V and some 25...75 Amps. The solution was based on one-two-three-four phases pulse generator and identical mosfet driver let the each transistor is used with overload current but very short time, thus the average power of each is inside the norm. Works perfect. The IC~s have a wondeful AppNotes, just make as it is drawn there. Names - Renesas HiP6301 and HiP6601. All the topologies from halfbridge, fullbridge, boost-up, boost-down and so on are available for these most brilliant ever, wonderful but nowadays little obsolet IC. However still I get a chance to purchase some hundred of it and am not regretting it nor a single moment.