## 555 polarity inverter circuits

12 posts / 0 new
Author
Message

I found a couple of schematics for voltage inverters using a 555 timer:

http://www.ecelab.com/circuit-po...

http://www.aaroncake.net/circuit...

The first website says that the circuit only provides a limited amount of current. I need to convert +5V from my supply (a USB cable) to -5V and power three op-amp ICs in dual-supply mode: two LM324s and one LM339. All 12 op amps will be used. Will this draw too much negative current from the 555 circuit?

Are there better options for converting +5V to (roughly) -5V?

Well, it's always fun to play with a 555, but if this is for a serious application then go with something else.

You can get pre-made DC/DC converters, or you could build an inverter with one of the many SMPS ICs out there (e.g. linear.com makes a heap of ICs).

There would be other ways, too, like using a PWM signal from an AVR, but like the 555 circuit I wouldn't use it for 12 OPAmps.

BTW: Did you do your homework and calculated/estimated the power consumption of the OPAmps?

Stealing Proteus doesn't make you an engineer.

Take a look at some of the max232 chips. (RS232 driveres).
They double and then invert.

When in this forum why not use a AVR and a FET.

sparrow2 wrote:
Take a look at some of the max232 chips. (RS232 driveres).
They double and then invert.

When in this forum why not use a AVR and a FET.

Thanks, I'll look into those.

Also, I have no free I/O pins on my AVR, so I can't use its PWM to generate the square wave.

ArnoldB: I haven't built the circuit yet, so I don't know how much current will be drawn. The LM324 datasheet says '0.18 mA supply current per channel' and the LM339 datasheet says '0.2 mA supply current per channel.' So would the op amps be drawing (0.18*8 + 0.2*4) = 2.24mA total?

From what I read, the 555 can supply 200mA of output current. I'd assume that the converter circuit could supply at least a few milliamps, which is enough for me.

This circuit to convert 9V to -5V apparently supplies 12mA of output current:
http://ourworld.compuserve.com/h...

Last Edited: Wed. Feb 18, 2009 - 08:30 PM

autorelease wrote:
Thanks, I'll look into those.

Stealing Proteus doesn't make you an engineer.

ArnoldB wrote:
autorelease wrote:
Thanks, I'll look into those.

See my edited post above.

Wouldn't it be easier to do this with two LDO's??

Just some guy

I built Bill Bowden's circuit, using +5V connected to the 555 instead of +9V, and it works, almost. It puts out roughly -3.4V, but the op-amp causes an additional drop, and the actual voltage at the op-amp's negative supply is about -2.4V. (not negative enough for my application)

The 7660 IC looks promising and I can get one at a local store. The datasheet doesn't say how much current it can output, but posts in various newsgroups suggest tens of milliamps, which should be fine for me.

Look at the datasheet 7660 !

You have some Uout versus Iout diagrams. With them

Ah, now I see...the output voltage is affected by the load current. The highest voltage I'd be able to tolerate is -3V, so it looks like I can draw just under 35mA. Should be enough for me, I'll get the chip tomorrow and be on my way :)

Careful... Got a scope to play with?

The LM324 is a great general purpose op amp, and I've got a bunch on my bench, but it is NOT a rail-to-rail op amp, and it doesn't operate well near its V+ and V-, (or Gnd if unipolar supply) limits. It works well with a significant unipolar or bi-polar power supply so the signal does not have to approach either V+ or V- (or Gnd if unipolar supply is used).

I think you may encounter some difficulties using this chip on such a low supply voltage. Been there, done that.

JC