four-quadrant smps, LT8714

On a sunny day (23 Jan 2016 04:07:57 -0800) it happened Winfield Hill wrote in :

Looking at the datasheet, if you clamp the flyback pulse at the top, then the output goes negative.. Nothing new I think.

Looks like an interesting part. For real four-quadrant action it'll need a nice beefy zener on the input! Or maybe a matching four-quadrant offline PSU. ;)

Cheers

Phil Hobbs

Heh, Cuk with a different output voltage reference!

Tim

-- Seven Transistor Labs, LLC Electrical Engineering Consultation and Contract Design Website:

Looks like a variation on a Cuk converter. Watch the gate drives and you should be able to sort it out. ...Jim Thompson

Sim is extremely sensitive to source impedance. Dig those current spikes.

RL

It says "patent pending."

A full-bridge synchronous switcher can drive an ungrounded load, like a TEC, with the same two inductors, but needs four fets... and a much cheaper control chip!

The latter is one of the key problems I often run into. It is very helpful to have LTSpice models and the great support that LTC provides but on some designs $4-5 for the switcher chip alone cannot be justified.

Absolutely correct. Synchronous switchers can also source and sink loads. But sometimes it's nice to be able to go negative.

That is where class-D chips come in handy. Generate a stiff positive and a negative rail, then a class-D. One client was quite impressed when they let the circuitry "step on it" ... *POP* ... phsssss ... it destroyed the load and the driver didn't even get warm.

At first blush it's a class-D amplifier: the kicker is, it takes a single positive power supply. That calls for some finesse, which apparently (in the shown app circuit) means huge ripple currents on that little (22uF x2) capacitor pair, if you want a negative output voltage. The output filter (100 uF x4) is shown as a nonpolar electrolytic, though one might cheat with connecting the (+) terminal of a polar 400 uF to Vcc (as long as Vcc is 12V and output goes from +5 to -5V).

I predict difficulties getting suitable ripple current for the blocking capacitors, and some confusion about what 'suitable' means. It's also a little odd that you apply

It must be a leftist power supply >:-} ...Jim Thompson

Not in my instruments, you don't!

Ripple on the TEC power can completely destroy the performance of a stabilized diode laser or any sort of sensitive detector mounted on the cold plate.

Something like this is a big win, as long as one side is grounded.

Cheers

Phil Hobbs

====================== Version 4 SHEET 1 880 680 WIRE 16 -48 -32 -48 WIRE 64 -48 16 -48 WIRE 112 -48 64 -48 WIRE 192 -48 112 -48 WIRE 224 -48 192 -48 WIRE 352 -48 320 -48 WIRE 496 -48 352 -48 WIRE 544 -48 496 -48 WIRE -32 16 -32 -48 WIRE 64 16 64 -48 WIRE 112 16 112 -48 WIRE 192 48 192 -48 WIRE 224 48 192 48 WIRE 352 48 352 -48 WIRE 352 48 320 48 WIRE 544 48 544 -48 WIRE 272 112 272 16 WIRE -32 128 -32 96 WIRE 64 160 64 80 WIRE 112 160 112 80 WIRE 112 160 64 160 WIRE 144 160 112 160 WIRE 272 160 272 112 WIRE 272 160 224 160 WIRE -32 240 -32 208 WIRE 272 256 272 224 WIRE 544 256 544 128 FLAG -32 240 0 FLAG 272 256 0 FLAG 544 256 0 FLAG 496 -48 Vload FLAG 16 -48 Vraw SYMBOL voltage -32 112 R0 WINDOW 0 40 59 Left 2 WINDOW 123 0 0 Left 2 WINDOW 39 0 0 Left 2 SYMATTR InstName V1 SYMATTR Value SINE(0 5 10) SYMBOL voltage -32 0 M0 WINDOW 3 55 48 Left 2 WINDOW 123 39 80 Left 2 WINDOW 39 0 0 Left 2 SYMATTR InstName V2 SYMATTR Value PULSE(0 0.1 1u 100n 100n 500n 1.2u) SYMATTR Value2 AC 1 SYMBOL npn 224 16 R270 WINDOW 0 100 -23 VRight 2 WINDOW 3 94 34 VRight 2 SYMATTR InstName Q1 SYMATTR Value 2N4401 SYMBOL pnp 224 112 R270 WINDOW 0 38 114 VRight 2 WINDOW 3 -6 78 VRight 2 SYMATTR InstName Q2 SYMATTR Value 2N4403 SYMBOL cap 256 160 R0 SYMATTR InstName C1

SYMBOL res 240 144 R90 WINDOW 0 0 56 VBottom 2 WINDOW 3 32 56 VTop 2 SYMATTR InstName R1 SYMATTR Value 1k SYMBOL res 528 32 R0 SYMATTR InstName R2 SYMATTR Value 50 SYMBOL diode 80 16 M0 WINDOW 3 21 89 Left 2 SYMATTR InstName D1 SYMATTR Value BAT54 SYMBOL diode 96 80 M180 WINDOW 0 24 64 Left 2 WINDOW 3 24 0 Left 2 SYMATTR InstName D2 SYMATTR Value BAT54 TEXT 304 320 Left 2 !.tran 0.2

I recall seeing four quadrant DC/DC converters from PowerDSine used as POTS ring voltage generators for small phone exchanges in the mid-'90s.

Microsemi bought PowerDSine 2006, and they no longer have those products so I can't find a datasheet, but the schematic in my head looks like the one on the LT8714 page.

Regards, Allan