-5V PSU out of +5V or +9V

[neazoi]

Hello,
I am trying to find a schematic for a -5V PSU that can be derived from +5V or +9V input voltage. Current consumption is about 300-500mA. My transformer has not gor a center tap. I would like to use a +9v from a jack wall

I am going to use this to feed some opamps
The simplest the better, I would prever the use of linear regulators.

Thank you

 

[kapilp]

You can use MAX660 by National Semi. Works good. Check the datasheet for details. Are you sure your op-amps will consume 300-500mA? Try to verify that with a table top dc power supply.

 

[neazoi]

I am trying to make a circuit to replace this one. I would make this one if I could find the details of the coil used.

I avoid MAX chips because maxim stops the production of specific types afrer some years.

 

[kapilp]

the series that i gave you is easily replaceable by other manufacturers like Maxim/National/Linear/Intersil, with varying performance for same pinout ... you should have no issues in finding a replacement. In any case the manufacturer is National and the part is very popular

 

[neazoi]

is there a military ceramic package version of it?

 

[kapilp]

i dont think so .... check the datasheets and websites of manufacturers quoted above ... you should be able to find something of interest

 

[Syncopator]

The easiest method would be to use a 9Vdc wall adapter just for the -5V regulator. They're cheap too.

 

[saikat36]

Easiest way to get -5V from +9V is to use Voltage regulator 7905.

 

[betwixt]

Ignore Saikat36's last reply, it's nonsense!

The 7660 and it's clones has been available for years and will be around for a long time yet. The schematic you show is to generate +12V as well as -5V so it is more complicated than necessary. Your current requirement is quite high so the schematic wouldn't manage to cope anyway. I suggest you look at DC-DC converters which you buy as small plastic packages and have isolated outputs. Either you could get one with "+5 0v -5V" which has the advantage that the + and - outputs track each other, this might help to keep op-amp center rails at 0V, or you can get a single +5V output and reverse the output pins so it gives you -5 out. The outputs are isolated from the input supply so it's quite safe to do that.

Brian.

 

[kapilp]

The DC-DC converters mentioned by Brian usually are described as being 'awesome'/'great' depending on manufacturer. You will get them for all sorts of specifications for +/-5V output. But they also tend to cost a little more than other options.
I still recommend that you try out the circuit with a lab power supply (opamps using 800mA current seems a little too much), get the optimum current requirement and then evaluate the options given in this forum.

 

[neazoi]

Easiest way to get -5V from +9V is to use Voltage regulator 7905.

You cannot. the 7905 requires a negative input voltage.

---------- Post added at 13:16 ---------- Previous post was at 13:15 ----------

Thanks a lot guys

 

[saikat36]

Sorry for unintentionally giving wrong information.

 

[BradtheRad]

Here's a clever method to obtain a negative supply from positive pulses.

www.edaboard.com/attachments/8893d1115569736-555_179.gif

Also:

newelectronicdesings2011.blogspot.com/2009/11/negative-supply-from-single-positive.html

Provides low power only. If you want more than a few mA...

Then you'll send the oscillator pulses to drive a mosfet. It will chop the pure + supply DC. Then you send that through the conversion stage. Use higher value capacitors depending.

 

[saikat36]

Your procedure is too messy to implement mate.
Can anybody suggest atleast simpler process than that?

 

[neazoi]

Here's a clever method to obtain a negative supply from positive pulses.

www.edaboard.com/attachments/8893d1115569736-555_179.gif

Also:

newelectronicdesings2011.blogspot.com/2009/11/negative-supply-from-single-positive.html

Provides low power only. If you want more than a few mA...

Then you'll send the oscillator pulses to drive a mosfet. It will chop the pure + supply DC. Then you send that through the conversion stage. Use higher value capacitors depending.

I have seen that before. It works, but only for low current. I do not believe it can provide 300-500mA of current as it is.
I am interested to see the final though

 

[BradtheRad]

I have seen that before. It works, but only for low current. I do not believe it can provide 300-500mA of current as it is.
I am interested to see the final though

Now that I refresh my memory using the circuit simulator...

To get more power it won't work to simply chop the supply with a mosfet.

The device that provides pulses has to both source and sink current.

More like a hi power op amp. It must have very low ON resistance at the output pin.

 

[memarian]

Hi,

Take a look at MAX 774 from Maxim , Inverting Switching Regulator (without Transformer) , -5 V 1A out , 3~16.5V input.

Datasheet Link : **broken link removed**

I used MAX764 (0.26A Version) and it works like a charm.

Good Luck!

 

[neazoi]

These integrated solutions all work I am pretty sure and thank you.

Has anyone seen a way todo it using discrete components and avoiding the tapped coil?

 

[betwixt]

Look up the 8 pin IC called MC34063A, they are very inexpensive and can be wired to provide a regulated step-up, step-down or voltage inversion with a negative output in excess of 100mA. The data sheet shows how to do it and you only need one single coil inductor which you can buy or wind yourself.

Brian.

 

[memarian]

These integrated solutions all work I am pretty sure and thank you.

Has anyone seen a way todo it using discrete components and avoiding the tapped coil?

Hi,

For Discrete implementation , take a look at Buck-Boost topology. All of these integrated circuits use this.Take a look at these :

Wikipedia Buck-Boost Converter
**broken link removed**
http://www.fairchildsemi.com/onlineseminars/regularTopologies_implementations.pdf
**broken link removed**