Any zippy quad opamps under 50 cents?

J

Joerg

Guest
Hallo All,

Have to do another DC-DC converter, amongst other stuff. Like
usual the voltage range, cost etc. doesn't allow for one of those
nice PWM chips. Except for the ol' LM324 I rarely use opamps in
my designs because of cost, it's all done in discrete plus logic.
So I looked around again (last time was in 97) whether there is
anything new in the opamp arena and I was not enthused. Either
too slow or too expensive.

Is there any quad opamp that can do 30MHz unity gain or more, has
a slew rate above 50V/usec and retails under 50 cents in large
qties? No precision or anything but usually these apps need GND
in the common mode range. Just curious, I can do it the classical
way again without opamps.

Regards, Joerg
http://www.analogconsultants.com
 
Joerg wrote:
Hallo All,

Have to do another DC-DC converter, amongst other stuff. Like
usual the voltage range, cost etc. doesn't allow for one of those
nice PWM chips. Except for the ol' LM324 I rarely use opamps in
my designs because of cost, it's all done in discrete plus logic.
So I looked around again (last time was in 97) whether there is
anything new in the opamp arena and I was not enthused. Either
too slow or too expensive.

Is there any quad opamp that can do 30MHz unity gain or more, has
a slew rate above 50V/usec and retails under 50 cents in large
qties? No precision or anything but usually these apps need GND
in the common mode range. Just curious, I can do it the classical
way again without opamps.

Regards, Joerg
http://www.analogconsultants.com
Click to expand...
What function in a DC to Dc converter could need an opamp with that
kind of bandwidth? Are you sure you are not looking for a cheap quad
comparator (like the LM339)?

--
John Popelish
 
Hi John,

Pretty much all of them, oscillator, error amp, inductor current sense.
That needs three to four amps. The LM339 could be used but it is way on
the slow side.

To be able to use small inductors these design have to run at several
hundred KHz to a MHz when under load.

Regards, Joerg
http://www.analogconsultants.com
 
On Tue, 27 Apr 2004 23:57:55 GMT, the renowned Joerg
<notthisjoergsch@removethispacbell.net> wrote:

Hallo All,

Have to do another DC-DC converter, amongst other stuff. Like
usual the voltage range, cost etc. doesn't allow for one of those
nice PWM chips. Except for the ol' LM324 I rarely use opamps in
my designs because of cost, it's all done in discrete plus logic.
So I looked around again (last time was in 97) whether there is
anything new in the opamp arena and I was not enthused. Either
too slow or too expensive.

Is there any quad opamp that can do 30MHz unity gain or more, has
a slew rate above 50V/usec and retails under 50 cents in large
qties? No precision or anything but usually these apps need GND
in the common mode range. Just curious, I can do it the classical
way again without opamps.
Click to expand...
Power supply range?

Maybe not.

http://cache.national.com/ds/LM/LMH6642.pdf $1.30 in 1K , ~12V max

or ST TSH95 1.35 in 5K (GND not in the CM range) ~12V max



Best regards,
Spehro Pefhany
--
"it's the network..." "The Journey is the reward"
speff@interlog.com Info for manufacturers: http://www.trexon.com
Embedded software/hardware/analog Info for designers: http://www.speff.com
 
On Tue, 27 Apr 2004 23:57:55 GMT, Joerg
<notthisjoergsch@removethispacbell.net> wrote:

Hallo All,

Have to do another DC-DC converter, amongst other stuff. Like
usual the voltage range, cost etc. doesn't allow for one of those
nice PWM chips. Except for the ol' LM324 I rarely use opamps in
my designs because of cost, it's all done in discrete plus logic.
So I looked around again (last time was in 97) whether there is
anything new in the opamp arena and I was not enthused. Either
too slow or too expensive.

Is there any quad opamp that can do 30MHz unity gain or more, has
a slew rate above 50V/usec and retails under 50 cents in large
qties? No precision or anything but usually these apps need GND
in the common mode range. Just curious, I can do it the classical
way again without opamps.

Regards, Joerg
http://www.analogconsultants.com
Click to expand...
Not to change the subject, but why quads? I prefer to stock duals, so
placement is easier and you can use just 2 if that's all you need. Two
duals take about the same area as one quad.

Single SOT-23 opamps are cool too.

John
 
Hi John,

Yes, duals would work as well if the price is right. For a PWM circuit you
normally need more than two if it is to be current mode feedback. That
type of feedback I found to be most suitable when you have a really wide
load range and the inductor has to be as tiny as possible. I found that
duals cost nearly the same as quads and four SOT23-5 would really blow the
budget in most cases.

Comparators are also an option as John Popelish has hinted. But in such
apps it would have to be push-pull outputs and fast, you can't use those
1.3usec varieties. Pretty slim pickings in comparators here.

What surprises me is that while uC, DSP and ASIC have enjoyed a tremendous
increase in bang for the buck over the last five to ten years this does
not seem to be quite the case for analog parts.

Regards, Joerg
http://www.analogconsultants.com
 
On Wed, 28 Apr 2004 19:22:18 GMT, Joerg
<notthisjoergsch@removethispacbell.net> wrote:

Hi John,

Yes, duals would work as well if the price is right. For a PWM circuit you
normally need more than two if it is to be current mode feedback. That
type of feedback I found to be most suitable when you have a really wide
load range and the inductor has to be as tiny as possible. I found that
duals cost nearly the same as quads and four SOT23-5 would really blow the
budget in most cases.

Comparators are also an option as John Popelish has hinted. But in such
apps it would have to be push-pull outputs and fast, you can't use those
1.3usec varieties. Pretty slim pickings in comparators here.

What surprises me is that while uC, DSP and ASIC have enjoyed a tremendous
increase in bang for the buck over the last five to ten years this does
not seem to be quite the case for analog parts.
Click to expand...
As I keep suggesting (without the proper cheering throngs) an LVDS
line receiver is a dynamite 1.5 ns, high-drive comparator. You can
make a great LC or crystal oscillator with one, too.

I'm pretty happy with analog progress (18 GHz gbw opamps, fabulous
flash ADCs, PHEMTs, lots of fun stuff) but the supply voltages keep
going down, and that is sometimes annoying.

John
 
Hi John,

Yes, LVDS stuff is pretty neat. The reason why I cannot use much of this
technology is that they are mostly restricted to a narrow supply voltage range
which, as you said, keeps going down. My designs are often battery operated,
extremely low power and a regulated voltage isn't really in the cards on many.

I agree that some progress has been made. But unfortunately that is more in the
Rolls Royce class where a few quarters more don't make a difference. Stuff like
that new ADS5500 is really impressive, but then again, that's a $100 device.

When I look at the digital world they enjoy ever smaller, cheaper, fatter FPGA,
great new uC designs such as some MSP430 devices at under $2 and so on. Well, I
guess I should not be drooling over the grass being greener there.

Regards, Joerg
http://www.analogconsultants.com
 
You must log in or register to reply here.

Welcome to EDABoard.com

Sponsor

Back
Top