Yet another DC-DC conv. question (+-15V)

[carbonzit]

OK, the idea just popped into my head of making a really simple DC-DC converter to run small battery-operated projects using op amps and such from a small power source. Say 2 AA cells. (The idea is to avoid using 2 fairly bulky 9-volt batteries.) So I'm looking for something like this:

**broken link removed**

Looking through the DigiKey catalog*, I see Murata makes some DC-DC "power modules", like this one that converts 5V--> +-15V @ 33mA for only about eight bucks (100mA will cost a little more); I suppose one could use 4 AAs for a 6-volt supply (you don't think that extra ~1V would hurt this guy, do you?).

Any ideas? I'd like something that's easy to build and would fit on a little scrap of perfboard, say about an inch square. Oh, and it could be unregulated (could use 78xx if needed downstream).

 

[moffy]

Look at the Linear Technology LT1308, but other manufacturers e.g. Texas Instruments, National Semi are also bound to have suitable ic's as it is a very common problem.

 

[carbonzit]

Look at the Linear Technology LT1308, but other manufacturers e.g. Texas Instruments, National Semi are also bound to have suitable ic's as it is a very common problem.

AAAAAARGH ... DigiKey's search function is useless. The part # you gave has a 5V output. However, I'm sure such critters exist, if you can find them. (The max. output I can find for +-15V devices is 66mA, though their paper catalog shows some at 100 mA.)

But I'm still curious about "rolling my own": how hard would this be to do? I'm guessing you'd need, at minimum a PWM oscillator (555?), a gate of some kind, inductor(s), then a rectifier and filter. Surely someone out there can cook up such a circuit?

 

[moffy]

For the LT1308 you just change the FB resistors to change the output voltage. It's good for 30V.

Rolling your own is problamatic because of the low input voltage.

 

[Reloadron]

I have used the Cosel line which have worked well. Starting with a 1.5 watt model @ about $12.50 USD and here is the data sheet working upthrough 3, 6, and 10 watts here is that data sheet with a cost of about $33 USD. I happen to use Allied Electronics here in the US quite a bit but converters ready made plug and play are pretty common animals. The 5 volt versions all run between 4.5 and 9.0 volts.

Ron

 

[carbonzit]

For the LT1308 you just change the FB resistors to change the output voltage. It's good for 30V.

Rolling your own is problamatic because of the low input voltage.

OK, I looked at the LT1308 datasheet. Couple-three things: it seems to be designed to deliver 5 volts, although as you point out it can be modified to produce higher voltages (but it isn't clear just how to do that: I couldn't find any notes that say how to size components properly for voltage boosting). And it has a single-ended output, when I asked for a split supply (+ 0 -). Plus, it requires external components (inductor, etc.). It appears that the Murata and other units are completely self-contained (I could be wrong about that).

Pretty intriguing, though, as it's meant to run off a single Li-ion cell.

 

[audioguru]

Plus and minus 15V is a total of 30v. You want the output to be 100mA so the output power is 3W.

Four AA alkaline cells produce 6V when they are brand new. A voltage stepup circuit might have an efficieny of 80% so the power from battery is 3.75W. The current in the AA cells is 3.75W/6V= 625mA and their voltage will drop to 4.8V in less than 1 hour.

 

[carbonzit]

Plus and minus 15V is a total of 30v. You want the output to be 100mA so the output power is 3W.

Four AA alkaline cells produce 6V when they are brand new. A voltage stepup circuit might have an efficieny of 80% so the power from battery is 3.75W. The current in the AA cells is 3.75W/6V= 625mA and their voltage will drop to 4.8V in less than 1 hour.

Yikes. Thanks for the reality check.

That Linear Technology chip that was suggested is 'spozed to work off a little bitty li-ion cell, though only 33mA output (but that should be plenty for a couple of op amps and such).

More research is needed here ...

 

[audioguru]

The supply current for a TL072 dual audio opamp is typically only 2.8mA if it drives a high resistance load.
The Linear Technology chip is good with a 3.3V output, very good with a 5V output and good with a 12V output. They don't show the efficiency with a 30V output so maybe it is poor.

Some Li-Ion cells are little bitty and others are fairly large. Calculate the current from it then see how long it lasts.

 

[buckysparks]

This circuit may meet your requirement.

100 mA Split Rail SWIFT? Converter Evaluation Module - TPS54060EVM-590 - TI Tool Folder

It generates +/-18V but can be tweaked to +/-15V. Also, the ic works down to 3.5V

 

[4pyros]

But I'm still curious about "rolling my own": how hard would this be to do? I'm guessing you'd need, at minimum a PWM oscillator (555?), a gate of some kind, inductor(s), then a rectifier and filter. Surely someone out there can cook up such a circuit?

The problum is that you will need a transformer.

 

[carbonzit]

The problum is that you will need a transformer.

Not true.

You can make a step-up supply using only inductors and capacitors. The circuit elements are arranged to that they add voltages. Not sure what the limits are with this method, but you can certainly get out more voltage than you put in.

 

[audioguru]

Why do you need a dual polarity supply and why must it be as high as 30V? Simply use newer opamps, not 741 opamps that have poor performance and are 43 years old.

I used TL071, TL072 and TL074 audio opamps for most of my many projects and some are powered from a single 9V battery.

 

[4pyros]

You can make a step-up supply using only inductors and capacitors. The circuit elements are arranged to that they add voltages. Not sure what the limits are with this method, but you can certainly get out more voltage than you put in.

And how do you make it dual polarity?

 

[ParkingLotLust]

My (albeit very crude) solution to this exact same problem was to use a boost converter to step the original voltage up to a little higher than the maximum +ve voltage, then use both a buck and inverting buck regulator to step down, and invert, along with a dual pot to control the "adjust" resistor of both regulators at the same time.

It's easy to create +/- using a rail splitter, but if you want your ground for v+/v- to be the same as common, this wont work...

 

[indulis]

Here's another

Digi-Key - 811-1734-5-ND (Manufacturer - BWR-15/85-D5-C)

It happens to use a 555 for the PWM.