OPAx322

B

bitrex

Guest
I've been doing some designs operating off low supply voltages lately,
2-2.5 volts. The OPAx322 series is a really nice amp. 20 MHz, RRIO, 10
v/usec, runs down to 1.8 volts. They come off the rails clean not a
problem at all it's hard to get them to mis-behave.

<http://www.ti.com/lit/ds/symlink/opa2322.pdf>

They're a little too pricey to be called a jelly bean but in performance
per dollar it blows away stuff like the AD8515.
 
On 11/12/19 1:25 PM, John Larkin wrote:
On Tue, 12 Nov 2019 12:54:44 -0500, bitrex <user@example.net> wrote:

I've been doing some designs operating off low supply voltages lately,
2-2.5 volts. The OPAx322 series is a really nice amp. 20 MHz, RRIO, 10
v/usec, runs down to 1.8 volts. They come off the rails clean not a
problem at all it's hard to get them to mis-behave.

http://www.ti.com/lit/ds/symlink/opa2322.pdf

They're a little too pricey to be called a jelly bean but in performance
per dollar it blows away stuff like the AD8515.

That's very nice. Most modern rrio opamps come off the rails fast and
clean, so make good comparators.

They tend to be stable with capacitive loads, too.
Click to expand...

It worked great for my low-voltage inductive transducer/sensor project
to drive current into the inductor; the meander-wire transducer is a
sloppy load with a lot of parasitic C but the amps handle it fine, and
65 mA drive per channel is pretty beefy for a low-voltage CMOS type.
 
On Tue, 12 Nov 2019 12:54:44 -0500, bitrex <user@example.net> wrote:

I've been doing some designs operating off low supply voltages lately,
2-2.5 volts. The OPAx322 series is a really nice amp. 20 MHz, RRIO, 10
v/usec, runs down to 1.8 volts. They come off the rails clean not a
problem at all it's hard to get them to mis-behave.

http://www.ti.com/lit/ds/symlink/opa2322.pdf

They're a little too pricey to be called a jelly bean but in performance
per dollar it blows away stuff like the AD8515.
Click to expand...

That's very nice. Most modern rrio opamps come off the rails fast and
clean, so make good comparators.

They tend to be stable with capacitive loads, too.

--

John Larkin Highland Technology, Inc
picosecond timing precision measurement

jlarkin att highlandtechnology dott com
http://www.highlandtechnology.com
 
On 11/12/19 1:25 PM, John Larkin wrote:
On Tue, 12 Nov 2019 12:54:44 -0500, bitrex <user@example.net> wrote:

I've been doing some designs operating off low supply voltages lately,
2-2.5 volts. The OPAx322 series is a really nice amp. 20 MHz, RRIO, 10
v/usec, runs down to 1.8 volts. They come off the rails clean not a
problem at all it's hard to get them to mis-behave.

http://www.ti.com/lit/ds/symlink/opa2322.pdf

They're a little too pricey to be called a jelly bean but in performance
per dollar it blows away stuff like the AD8515.

That's very nice. Most modern rrio opamps come off the rails fast and
clean, so make good comparators.

They tend to be stable with capacitive loads, too.
Click to expand...

The ST TSV99x is a cheaper alternative, one of the least expensive
low-voltage RRIO CMOS types. But with somewhat worse specs on supply
voltage range and noise. Also not unity-gain stable.

<https://www.st.com/resource/en/datasheet/tsv991.pdf>
 
On 11/12/19 2:53 PM, Winfield Hill wrote:
bitrex wrote...

I've been doing some designs operating off low supply voltages
lately, 2-2.5 volts. The OPAx322 series is a really nice amp.
20 MHz, RRIO, 10 V/usec, runs down to 1. 8 volts. They come off
the rails clean not a problem at all it's hard to get them to
mis-behave.

http://www.ti.com/lit/ds/symlink/opa2322.pdf

They're a little too pricey to be called a jelly bean but in
performance per dollar it blows away stuff like the AD8515.

One unusual feature, they include an internal charge pump
for the input transistors, eliminating a zero-offset-shift
problem affecting most other RRIO op-amps. This adds 28us
of extra startup time. The S version adds an enable pin.
Click to expand...

I noticed that and I also wondered what black magic they use to get the
input voltage noise density so low despite there being a freaking
switcher on the chip. It's still almost twice as good as some other amps
around the same price point.
 
bitrex wrote...
I've been doing some designs operating off low supply voltages
lately, 2-2.5 volts. The OPAx322 series is a really nice amp.
20 MHz, RRIO, 10 V/usec, runs down to 1. 8 volts. They come off
the rails clean not a problem at all it's hard to get them to
mis-behave.

http://www.ti.com/lit/ds/symlink/opa2322.pdf

They're a little too pricey to be called a jelly bean but in
performance per dollar it blows away stuff like the AD8515.
Click to expand...

One unusual feature, they include an internal charge pump
for the input transistors, eliminating a zero-offset-shift
problem affecting most other RRIO op-amps. This adds 28us
of extra startup time. The S version adds an enable pin.


--
Thanks,
- Win
 
bitrex wrote...
On 11/12/19 2:53 PM, Winfield Hill wrote:

One unusual feature, they include an internal charge pump
for the input transistors, eliminating a zero-offset-shift
problem affecting most other RRIO op-amps. This adds 28us
of extra startup time. The S version adds an enable pin.

I noticed that and I also wondered what black magic they use
to get the input voltage noise density so low despite there
being a freaking switcher on the chip. It's still almost
twice as good as some other amps around the same price point.
Click to expand...

Black magic is the right word, they aren't telling.


--
Thanks,
- Win
 
On 11/12/19 7:00 PM, John Larkin wrote:
On 12 Nov 2019 15:01:35 -0800, Winfield Hill <winfieldhill@yahoo.com
wrote:

bitrex wrote...

On 11/12/19 2:53 PM, Winfield Hill wrote:

One unusual feature, they include an internal charge pump
for the input transistors, eliminating a zero-offset-shift
problem affecting most other RRIO op-amps. This adds 28us
of extra startup time. The S version adds an enable pin.

I noticed that and I also wondered what black magic they use
to get the input voltage noise density so low despite there
being a freaking switcher on the chip. It's still almost
twice as good as some other amps around the same price point.

Black magic is the right word, they aren't telling.

One old rrio, I think a Burr-Brown, with an internal power supply for
the front end, ran the supply a some hundreds of MHz.

Hang a spectrum analyser probe near it!
Click to expand...

A spectrum analyzer and impedance analyzer are next on the lab-to-buy list.

I could probably budget a HP 4192A or Wayne Kerr 6425 from the
80s-boatanchor-series for the latter. Don't know about the former
 
On 12 Nov 2019 15:01:35 -0800, Winfield Hill <winfieldhill@yahoo.com>
wrote:

bitrex wrote...

On 11/12/19 2:53 PM, Winfield Hill wrote:

One unusual feature, they include an internal charge pump
for the input transistors, eliminating a zero-offset-shift
problem affecting most other RRIO op-amps. This adds 28us
of extra startup time. The S version adds an enable pin.

I noticed that and I also wondered what black magic they use
to get the input voltage noise density so low despite there
being a freaking switcher on the chip. It's still almost
twice as good as some other amps around the same price point.

Black magic is the right word, they aren't telling.
Click to expand...

One old rrio, I think a Burr-Brown, with an internal power supply for
the front end, ran the supply a some hundreds of MHz.

Hang a spectrum analyser probe near it!

--

John Larkin Highland Technology, Inc
picosecond timing precision measurement

jlarkin att highlandtechnology dott com
http://www.highlandtechnology.com
 
bitrex wrote...
I could probably budget a HP 4192A or Wayne Kerr 6425 from the
80s-boatanchor-series for the latter. Don't know about the former
Click to expand...

Aha, HP 4192A, a real workhorse. So many invaluable measurements.


--
Thanks,
- Win
 
On 11/12/19 8:29 PM, Winfield Hill wrote:
bitrex wrote...

I could probably budget a HP 4192A or Wayne Kerr 6425 from the
80s-boatanchor-series for the latter. Don't know about the former

Aha, HP 4192A, a real workhorse. So many invaluable measurements.
Click to expand...

looks like it weighs as much as a horse, too. Gonna need a sturdier bench
 
You must log in or register to reply here.

Welcome to EDABoard.com

Sponsor

Back
Top