low current inverter

HC14s are okay as charge pumps, if you parallel several sections and run them at 5V.

Try a BAT54S dual Schottky.

I'm doing exactly that in my boxcar lock-in gizmo--the charge pump provides reverse photodiode bias, but turns off during the actual measurement.

Cheers

Phil Hobbs

That works.

I like LM2662 type charge pumps to flip V+ to V-. Run them in "high frequency" mode.

if you have that many inverters kicking around, then get a small toroid transformer and do a push pull, the sec side will be much simpler to set up.

I have some small donuts here i use for things like that via some enamel wire..

Jamie

I had used a TI charge pump for a negative rail, they work better than I expected.

Cheers

I recall doing something vaguely similar to generate a negative bias voltage for a LCD driver. I actually made a three phase ring oscillator with three inverters of one 74HC04. I put a 39k, 1nF low pass RC between each inverter and the next one in the ring oscillator. I then buffered the three phase outputs of the ring oscillator with the other three inverters of that package, and then capacitively coupled it to another

Chris

OK thanks, I was trolling digikey for dc-dc converters but it was a fire hose. I add some serious filters... but then miss things.

George h.

Hmm push pull with inverters? I guess I'd like to keep it simple.

Yeah I don't want to wind transformers.. at least not for this.

George H.

Wow, I'm glad you added a picture. I don't have that many 'spare' inverters :^)

George H.

The SIP things, CUI and Murata and such, are cool for about $4, but they are unregulated, noisy, and usually blow up if the output is shorted.

LM2664 is a low current, 40 mA, charge pump. We pay about 40 cents for that one. We've used 21,000 so far, no problems.

The SIP things, CUI and Murata and such, are cool for about $4, but they are unregulated, noisy, and usually blow up if the output is shorted.

LM2664 is a low current, 40 mA, charge pump. We pay about 40 cents for that one. We've used 21,000 so far, no problems.

Yeah, Great! Push pulls are good for the higher power SMPSs. That way you'll get as much as twice mW out of this.

Ain't it great?

Obviously. Don't you already have some SMPS on board? Then highjack the switching node with the charge pump.

The LM2664 looks real good as a full bridge driver. Pins 5+8 at Vcc, load from pins 2 to 4, pin 3 to gnd. At 150KHz (75KHz load), driving a low E*T primary transformer and the secondary could have many uses. Is there anything better for this application?

Cheers, Harry

Here's an isolated supply using an IR part:

The LM2664 looks real good as a full bridge driver. Pins 5+8 at Vcc, load from pins 2 to 4, pin 3 to gnd. At 150KHz (75KHz load), driving a low E*T primary transformer and the secondary could have many uses. Is there anything better for this application?

Cheers, Harry

Here's an isolated supply using an IR part:

yes but the LM2664 is a Full bridge with internal drivers.

Harry.

Yes, that would be an interesting way to use that part. You might test it, to make sure the switches work properly in the quadrants required that way.

Den fredag den 25. april 2014 23.16.39 UTC+2 skrev John Larkin:

you could probable use the ir2153* without the fets the drivers are good for a few hundred mA; Vb to Vcc, Vs to gnd, load from HO to LO

-Lasse

Den fredag den 25. april 2014 23.16.39 UTC+2 skrev John Larkin:

you could probable use the ir2153* without the fets the drivers are good for a few hundred mA; Vb to Vcc, Vs to gnd, load from HO to LO

-Lasse

Let me repeat, the LM2664 is a FULL bridge with 4.0R switches and operates at

hey

pdf

for

max256,

Den fredag den 25. april 2014 23.16.39 UTC+2 skrev John Larkin:

you could probable use the ir2153* without the fets the drivers are good for a few hundred mA; Vb to Vcc, Vs to gnd, load from HO to LO

-Lasse

Oops, I need a drink! Your IC are full bridges, my mistake.

Sorry, Harry

A 74HC14 comes close!