Dual supply LTC4446

B

bitrex

Guest
The example circuit for the LTC4446 dual gate driver IC has
the lower fet connected to ground; it seems like it should be possible
to reference it to a negative supply to use as e.g. class D amplifier
gate driver, but it's unclear to me how to do this while keeping Vcc and
the two drive control inputs ground-reference.

Is that possible? Or is there another LT IC better suited for that
purpose in the lib? Thanks

Modified schematic with a negative supply and the LTC4446 inputs in
question referenced to the negative supply:

Version 4
SHEET 1 880 756
WIRE -96 48 -192 48
WIRE -32 48 -96 48
WIRE 112 48 32 48
WIRE 176 48 112 48
WIRE 272 48 240 48
WIRE 496 80 336 80
WIRE 336 96 336 80
WIRE 496 96 496 80
WIRE 112 112 112 48
WIRE -192 128 -192 48
WIRE -96 176 -96 48
WIRE -16 176 -96 176
WIRE 288 176 240 176
WIRE 496 192 496 176
WIRE -16 256 -416 256
WIRE 272 256 272 48
WIRE 272 256 240 256
WIRE 336 256 336 192
WIRE 336 256 272 256
WIRE 384 256 336 256
WIRE 496 256 464 256
WIRE 576 256 496 256
WIRE 496 272 496 256
WIRE 576 272 576 256
WIRE -16 336 -320 336
WIRE 288 336 240 336
WIRE -416 352 -416 256
WIRE -320 352 -320 336
WIRE 336 352 336 336
WIRE 496 368 496 336
WIRE 576 368 576 352
WIRE -192 480 -192 208
WIRE 112 480 112 400
WIRE 112 480 -192 480
WIRE 336 480 336 352
WIRE 336 480 112 480
WIRE -416 560 -416 432
WIRE -320 560 -320 432
WIRE -320 560 -416 560
WIRE 336 560 336 480
WIRE 336 560 -320 560
WIRE 336 608 336 560
WIRE 336 736 336 688
FLAG 496 192 0
FLAG 496 368 0
FLAG 576 368 0
FLAG 336 736 0
SYMBOL voltage -192 112 R0
SYMATTR InstName V1
SYMATTR Value 10
SYMBOL voltage -320 336 R0
WINDOW 123 0 0 Left 2
WINDOW 39 0 0 Left 2
WINDOW 3 -523 74 Left 2
SYMATTR InstName V2
SYMATTR Value PULSE(0 5 20n 10n 10n .96u 2u)
SYMBOL nmos 288 96 R0
SYMATTR InstName Q1
SYMATTR Value IRF7468
SYMBOL nmos 288 256 R0
SYMATTR InstName Q2
SYMATTR Value IRF7468
SYMBOL cap 240 32 R90
WINDOW 0 0 32 VBottom 2
WINDOW 3 32 32 VTop 2
SYMATTR InstName C1
SYMATTR Value .1Âľ
SYMBOL schottky -32 64 R270
WINDOW 0 32 32 VTop 2
WINDOW 3 0 32 VBottom 2
SYMATTR InstName D1
SYMATTR Value 1N5819
SYMATTR Description Diode
SYMATTR Type diode
SYMBOL voltage 496 80 R0
WINDOW 123 0 0 Left 0
WINDOW 39 0 0 Left 0
SYMATTR InstName V3
SYMATTR Value 25
SYMBOL ind 368 272 R270
WINDOW 0 32 56 VTop 2
WINDOW 3 5 56 VBottom 2
SYMATTR InstName L1
SYMATTR Value 1Âľ
SYMBOL cap 480 272 R0
SYMATTR InstName C2
SYMATTR Value 10Âľ
SYMBOL res 560 256 R0
SYMATTR InstName R1
SYMATTR Value .5
SYMBOL voltage -416 336 R0
WINDOW 3 -399 41 Left 2
SYMATTR InstName V4
SYMATTR Value PULSE(5 0 0 10n 10n .99u 2u)
SYMBOL PowerProducts\\LTC4446 112 256 R0
SYMATTR InstName U1
SYMBOL voltage 336 704 R180
WINDOW 123 0 0 Left 0
WINDOW 39 0 0 Left 0
SYMATTR InstName V5
SYMATTR Value 25
TEXT -848 448 Left 2 !.tran 100u startup
 
On Fri, 14 Feb 2020 21:27:24 -0500, bitrex <user@example.net> wrote:

The example circuit for the LTC4446 dual gate driver IC has
the lower fet connected to ground; it seems like it should be possible
to reference it to a negative supply to use as e.g. class D amplifier
gate driver, but it's unclear to me how to do this while keeping Vcc and
the two drive control inputs ground-reference.
Click to expand...

One interesting feature of the single-ended class-D amp is that it can
push power *into* one of the supplies.



--

John Larkin Highland Technology, Inc

The cork popped merrily, and Lord Peter rose to his feet.
"Bunter", he said, "I give you a toast. The triumph of Instinct over Reason"
 
On 2/14/20 10:52 PM, bitrex wrote:
On 2/14/20 10:35 PM, jlarkin@highlandsniptechnology.com wrote:
On Fri, 14 Feb 2020 21:27:24 -0500, bitrex <user@example.net> wrote:

The example circuit for the LTC4446 dual gate driver IC has
the lower fet connected to ground; it seems like it should be possible
to reference it to a negative supply to use as e.g. class D amplifier
gate driver, but it's unclear to me how to do this while keeping Vcc and
the two drive control inputs ground-reference.

One interesting feature of the single-ended class-D amp is that it can
push power *into* one of the supplies.




Well, to be honest I don't really want to use it as a class D amp
either. I found the requirements on this power-booster I need are
relatively loose so while designing a high-voltage booster with op amp
that can do several MHz at 50V P2P why bother if a rough sine is all
that's needed on the output.

Have two gate driver ICs with four fets two connected +25 and -25, and
two to +25/2 and -25/2, use ye olde "stepped sine" to fire
comparators/level shifters and trigger those FETs in the appropriate
sequence. LC filter the output, run it open-loop.
Click to expand...

only thing that needs to be tightly controlled is frequency so sine is
digitally-generated. Doesn't need to do arbitrary waveforms so a full
class D is overkill
 
On 2/14/20 10:35 PM, jlarkin@highlandsniptechnology.com wrote:
On Fri, 14 Feb 2020 21:27:24 -0500, bitrex <user@example.net> wrote:

The example circuit for the LTC4446 dual gate driver IC has
the lower fet connected to ground; it seems like it should be possible
to reference it to a negative supply to use as e.g. class D amplifier
gate driver, but it's unclear to me how to do this while keeping Vcc and
the two drive control inputs ground-reference.

One interesting feature of the single-ended class-D amp is that it can
push power *into* one of the supplies.
Click to expand...

Well, to be honest I don't really want to use it as a class D amp
either. I found the requirements on this power-booster I need are
relatively loose so while designing a high-voltage booster with op amp
that can do several MHz at 50V P2P why bother if a rough sine is all
that's needed on the output.

Have two gate driver ICs with four fets two connected +25 and -25, and
two to +25/2 and -25/2, use ye olde "stepped sine" to fire
comparators/level shifters and trigger those FETs in the appropriate
sequence. LC filter the output, run it open-loop.
 
bitrex wrote...
The example circuit for the LTC4446 dual gate driver IC
has the lower fet connected to ground; it seems like it
should be possible to reference it to a negative supply...
Click to expand...

If you use an isolated part like the UCC21520, you can
place the high and low side FET drivers wherever you like.


--
Thanks,
- Win
 
bitrex wrote...
Well, to be honest I don't really want to use it as a class D amp
either. I found the requirements on this power-booster I need are
relatively loose so while designing a high-voltage booster with
op amp that can do several MHz at 50V P2P why bother if a rough
sine is all that's needed on the output.

Have two gate driver ICs with four fets two connected +25 and -25,
and two to +25/2 and -25/2, use ye olde "stepped sine" to fire
comparators/level shifters and trigger those FETs in the
appropriate sequence. LC filter the output, run it open-loop.
Click to expand...

Yes, you can have intermediate levels, with isolated drivers and
back-to-back MOSFETs, as shown in Fig 3x.115.


--
Thanks,
- Win
 
lørdag den 15. februar 2020 kl. 04.56.46 UTC+1 skrev bitrex:
On 2/14/20 10:52 PM, bitrex wrote:
On 2/14/20 10:35 PM, jlarkin@highlandsniptechnology.com wrote:
On Fri, 14 Feb 2020 21:27:24 -0500, bitrex <user@example.net> wrote:

The example circuit for the LTC4446 dual gate driver IC has
the lower fet connected to ground; it seems like it should be possible
to reference it to a negative supply to use as e.g. class D amplifier
gate driver, but it's unclear to me how to do this while keeping Vcc and
the two drive control inputs ground-reference.

One interesting feature of the single-ended class-D amp is that it can
push power *into* one of the supplies.




Well, to be honest I don't really want to use it as a class D amp
either. I found the requirements on this power-booster I need are
relatively loose so while designing a high-voltage booster with op amp
that can do several MHz at 50V P2P why bother if a rough sine is all
that's needed on the output.

Have two gate driver ICs with four fets two connected +25 and -25, and
two to +25/2 and -25/2, use ye olde "stepped sine" to fire
comparators/level shifters and trigger those FETs in the appropriate
sequence. LC filter the output, run it open-loop.

only thing that needs to be tightly controlled is frequency so sine is
digitally-generated. Doesn't need to do arbitrary waveforms so a full
class D is overkill
Click to expand...

class-E ?
 
On 2/15/20 10:05 AM, Winfield Hill wrote:
bitrex wrote...

Well, to be honest I don't really want to use it as a class D amp
either. I found the requirements on this power-booster I need are
relatively loose so while designing a high-voltage booster with
op amp that can do several MHz at 50V P2P why bother if a rough
sine is all that's needed on the output.

Have two gate driver ICs with four fets two connected +25 and -25,
and two to +25/2 and -25/2, use ye olde "stepped sine" to fire
comparators/level shifters and trigger those FETs in the
appropriate sequence. LC filter the output, run it open-loop.

Yes, you can have intermediate levels, with isolated drivers and
back-to-back MOSFETs, as shown in Fig 3x.115.
Click to expand...

Thanks, is that reference for AoE III or AoE X-treme?
 
On 2/15/20 10:05 AM, Winfield Hill wrote:
bitrex wrote...

Well, to be honest I don't really want to use it as a class D amp
either. I found the requirements on this power-booster I need are
relatively loose so while designing a high-voltage booster with
op amp that can do several MHz at 50V P2P why bother if a rough
sine is all that's needed on the output.

Have two gate driver ICs with four fets two connected +25 and -25,
and two to +25/2 and -25/2, use ye olde "stepped sine" to fire
comparators/level shifters and trigger those FETs in the
appropriate sequence. LC filter the output, run it open-loop.

Yes, you can have intermediate levels, with isolated drivers and
back-to-back MOSFETs, as shown in Fig 3x.115.
Click to expand...

In my brief time fiddling with some two-stage linear power-booster ideas
e.g. something like this:

<https://tubecad.com/2017/10/02/FET-PNP%20Compound%20Amplifier.png>

but class-AB complimentary with a grounded-gate JFET appropriately
biased and complimentary MOSFETs driving the output from their drains,
wrapping feedback around to a fast op-amp, 1-2MHz 50V P2P is easy in a
simulation, but getting to 5MHz delivering a couple watts into a load
eludes me it would require much more careful thought on component
selection particularly the Cgs of the output fets. Annoying.

If a linear power booster of that spec was needed it might be more
plausible to drive the output devices directly from tapping the supply
like the classic:

<https://vignette3.wikia.nocookie.net/howto/images/2/27/Op-amp_voltage_booster_schematic_30Vpp.png/revision/latest?cb=20091028192317&path-prefix=en>
 
You must log in or register to reply here.

Welcome to EDABoard.com

Sponsor

Back
Top