power opamp?

[[email protected]]

What's a good medium-power opamp? I need to source/sink at least 60
mA, with say +-7.5 volt supplies, and swing +-6 maybe. No particular
requirements otherwise.

I suppose I could boost a cheap SOT-23 opamp with a couple of cheap
SOT-23 emitter followers....

                                    v+
                                     |
                                     c
          \               +----Rb---b     npn
            \             |          e
              \           |          |
                \---------+----Rx----+-------
               /          |          |
             /            |          e
           /              +----Rb---b    pnp
                                     c
                                     |
                                    v-

but a single part would be nice; I have to do this 16 times on a
smallish board.

http://www.onsemi.com/pub_link/Collateral/MC33201-D.PDF

The supply voltage is lower than you want - absolute maximum +/-6.5V -
and the guaranteed output current is only 50mA, but it has a
surprisingly potent output stage.

 

[Joerg]

Anybody got decent models for ferrite beads?

For pure frequency domain stuff, yes:

http://www.cadence.com/community/allegro/Resources/resources_pcbsi/mod/tpcoresim_bead_cadence.pdf

However, saturation is another matter. I've not seen much there that I
could really use. In the end you have to design a circuit that can find
the onset and slope and auto-calibrate. Ferrites are like a pasta dish.
When Giuseppe cooks it tastes different than when Antonio is the chef.

 

[[email protected]]

What's a good medium-power opamp? I need to source/sink at least 60
mA, with say +-7.5 volt supplies, and swing +-6 maybe. No particular
requirements otherwise.

I suppose I could boost a cheap SOT-23 opamp with a couple of cheap
SOT-23 emitter followers....

v+
|
c
\ +----Rb---b npn
\ | e
\ | |
\---------+----Rx----+-------
/ | |
/ | e
/ +----Rb---b pnp
c
|
v-

but a single part would be nice; I have to do this 16 times on a
smallish board.

John

ad826 is close I think, and dual

-Lasse

 

[Tam]

What's a good medium-power opamp? I need to source/sink at least 60
mA, with say +-7.5 volt supplies, and swing +-6 maybe. No particular
requirements otherwise.

I suppose I could boost a cheap SOT-23 opamp with a couple of cheap
SOT-23 emitter followers....


v+
|
c
\ +----Rb---b npn
\ | e
\ | |
\---------+----Rx----+-------
/ | |
/ | e
/ +----Rb---b pnp
c
|
v-


but a single part would be nice; I have to do this 16 times on a
smallish board.

John

Have you looked at the audio headphone or speaker drivers? National makes
dozens. If they are stereo, you only need to do it 8X.

Tam

 

[Joerg]

What's a good medium-power opamp? I need to source/sink at least 60
mA, with say +-7.5 volt supplies, and swing +-6 maybe. No particular
requirements otherwise.

Check out if an audio amp might fit. For example, one version of the
LM386 goes to 18V single supply if you really need 15V total:

http://www.national.com/ds/LM/LM386.pdf

Those are really cheap, well under 50c in qties.

[...]

 

[John Fields]

Anybody got decent models for ferrite beads?

 

[gearhead]

I'll trade my entire stock of ferrite beads for one indecent model.

 

[Fred Bartoli]

Joerg a écrit :

You can't use a regular amp with biased npn/pnp into the rail and then
sum the 16 others into its input?



Why not? I once had to cram over 50 parts onto the size of a standard
40c postage stamp. Now wait, that'll be 41c soon. 0201, SC75, oh what
fun to debug.

Doing that right now, but my stamp is round and I have gigaohm
impedances and all that goes 125°C. Fun...

 

[legg]

John Larkin wrote:

If I had trusted SPICE every time many of my design would not exist
today. Can't remember the last time I fired up SPICE. Some time last
year I guess.

It's much easier to get real components to burn.

RL

 

[legg]

John Larkin wrote:

You can't use a regular amp with biased npn/pnp into the rail and then
sum the 16 others into its input?



Why not? I once had to cram over 50 parts onto the size of a standard
40c postage stamp. Now wait, that'll be 41c soon. 0201, SC75, oh what
fun to debug.

Make it small enough, and even microwatt applications will overheat.

RL

 

[MooseFET]

Like this....

I'm going to modify the drawing a little:

v+
| v+
R |
| e
\VCC----------------------b pnp
\ c
\ R1 |
\-------+--/\/\------+-------OUTPUT
/ ! |
/ -/\/\--GND !
/ R2 c
VEE---------------------b npn
| e
R |
| v-
v-

The R1/R2 is made just big enough that the OUTPUT swing is rail to
rail for the op-amp selected.

The parallel combination of R1 and R2 sets the "gm" of the "OUTPUT
stage".

To ensure that it is stable, you may need a capacitor from the output
of the op-amp back to its inverting input. A heavy and capacitive
load can add a pole in the feed back loop.

 

[MooseFET]

Yeah, that's neat but harder to analyze for stability. One would
*really* have to trust the Spice models, or test the heck out of it.

I can imagine way too many oscillation modes!

We recently did something similar, using opamp supply currents to
drive transistors sort of like that, but current-mirror mode, to turn
the opamp into a fairly precise current source. The sim we did didn't
work at all, apparently because the opamp model doesn't actually pull
current from the rails!

All of the Linear models can't de trusted for the supply currents.

The LT1498's model can be used to power up your circuit. Each one
makes about 0.2V IIRC.

 

[Joerg]

It sounds like, pretty soon, you and Jim are going to have to
collaborate -- your scales of construction are constantly converging!

A surprising number of circuits are done discrete because of assembly in
Asia. The prohibitive placement costs of SMT parts are long gone. So the
crossover point where a chip design makes since can be in the tens of
thousands of units per year because you must amortize the desigs costs
and NRE over a reasonable (small) number of years.

About ten years ago I did a chip design including proto fab for under
$200k and this was on an automotive HV process. I don't think that's
possible anymore.

 

[Joerg]

Make it small enough, and even microwatt applications will overheat.

This catches many engineers by surprise. 50mW might not sound much but
for a resistor the size of a grain of salt that's toasty.

 

[Joerg]

It's much easier to get real components to burn.

That's what I keep saying, the good old days are now. When I was a kid
we didn't have Digikey and a 74LS90 or an RF transistor would cost a
week's allowance. You wouldn't dare to burn one up back then.

Unfortunately many of the new grads have never built stuff with real parts.

 

[Jim Thompson]

A surprising number of circuits are done discrete because of assembly in
Asia. The prohibitive placement costs of SMT parts are long gone. So the
crossover point where a chip design makes since can be in the tens of
thousands of units per year because you must amortize the desigs costs
and NRE over a reasonable (small) number of years.

About ten years ago I did a chip design including proto fab for under
$200k and this was on an automotive HV process. I don't think that's
possible anymore.

I regularly do chip designs that come in at around $100K total. You
just have to shop carefully for a foundry.

If you think TI, Motorola or Zarlink are good places to go, I have
some seaside property in Arizona to sell you.

...Jim Thompson

 

[Joerg]

I regularly do chip designs that come in at around $100K total. You
just have to shop carefully for a foundry.

But I bet that does not include your Syrah and Merlot

The 200k were including my time and that of the layouter.

If you think TI, Motorola or Zarlink are good places to go, I have
some seaside property in Arizona to sell you.

Oh, we did shop around. Problem is, 45V or 60V processes aren't found
seaside in Arizona ;-)

One place I was always quite happy with is AMI, for non-HV stuff. Just
not much fun to travel up there because it's so freaking cold in the winter.

 

[Jim Thompson]

But I bet that does not include your Syrah and Merlot

The 200k were including my time and that of the layouter.



Oh, we did shop around. Problem is, 45V or 60V processes aren't found
seaside in Arizona ;-)

X-Fab DMOS?

One place I was always quite happy with is AMI, for non-HV stuff. Just
not much fun to travel up there because it's so freaking cold in the winter.

I'll say. Of course I have an old project, resurrecting for an
upgrade, that uses AMI.

Pocatello is so-o-o-o much fun in the Winter.

Last time I was there I made the car rental guy walk me to my car...
for fear I'd get lost in the parking lot due to the blizzard.

And those sideways landings in little two-engine prop planes are so
thrilling too ;-)

BTW, OnSemi just bought AMIS. I don't whether that will be good or
bad.

...Jim Thompson

 

[Joerg]

X-Fab DMOS?

We used Delco. ABB-Hafo was another contender but they didn't really
want to, AFAIR because we had to seriously "bend" a lot of design rules.
It was for a disposable and didn't have to live more than a few hours.
That tends to give old school semiconductor folks goose bumps.

I'll say. Of course I have an old project, resurrecting for an
upgrade, that uses AMI.

Pocatello is so-o-o-o much fun in the Winter.

Last time I was there I made the car rental guy walk me to my car...
for fear I'd get lost in the parking lot due to the blizzard.

And those sideways landings in little two-engine prop planes are so
thrilling too ;-)

I still remember my wife's words: "Let's get outta Idaho!"

BTW, OnSemi just bought AMIS. I don't whether that will be good or
bad.

Yes, ON Semi.

<swallowing hard>

But IMHO ON Semi is a pretty good company. Do you know if Christine King
will stay on?

 

[Jim Thompson]

We used Delco. ABB-Hafo was another contender but they didn't really
want to, AFAIR because we had to seriously "bend" a lot of design rules.
It was for a disposable and didn't have to live more than a few hours.
That tends to give old school semiconductor folks goose bumps.



I still remember my wife's words: "Let's get outta Idaho!"



Yes, ON Semi.

<swallowing hard>

But IMHO ON Semi is a pretty good company. Do you know if Christine King
will stay on?

Don't know who is staying. I sent an E-mail to my last device model
contact... no response. So I'll call the rep and find out what's
going on.

...Jim Thompson