There are several "generations" of op-amps
1st generation (late 1960s) - single op-amp per package, not internally
compensated, poor offset voltage, high input bias current, high input
offset current, weak output drive, restricted common mode input range,
restricted output swing
2nd generation (early 1970s) - modern pinout, internally compensated, same
issues with offset, bias current, etc.
3rd generation (late 1970s) - J-FET input op-amps introduced, input bias
and offset current greatly reduced, most of the rest is the same. Low
offset bipolar op-amps introduced.
4th generation (1980s) - workable CMOS op-amps, slower and slightly higher
offset, rail to rail input and output range
5th generation (1990s) - improved CMOS op-amps which have almost ideal
behavior (no input current, full rail to rail input and output voltage
range, reasonably fast, low-ish input offset voltage). Can work down to 2V
supply. Bipolar op-amps still have the absolute best input offset voltage
without resorting to special techniques like auto-zeroing.
5.5th generation (2000 to now) - wide range of op-amps employing special
internal tricks to achieve CMOS performance PLUS microvolt-level offset
voltage. Tricks involve variants of chopper-type amplifiers which work as
servo amplifiers to the main signal path, but many use pseudo-random
dithered switching frequencies to reduce aliasing and other effects.
Standard stability calculations may not work with these.
LM324, LM358, LM741 are all 2nd generation
On Tue, Jul 10, 2018 at 2:03 PM, Jason White <
***@gmail.com> wrote:
> Management was griping over having to pay $0.28/piece @ 2500qty. for the
> -55C rating in a quad op-amp; to use a single $2.00 op amp would probably
> take quite a bit of fight.
>
> The LM124 was chosen because it is a "jelly-bean" op amp.
> (1) it will probably be available in some form the next 30 years (hard
> requirement) - the bar is set *really* low for a pin compatible replacement
> to be put in.
> (2) it is cheap
> (3) it is rated for -55C operation
> (4) it is "HV" rated. We are using a 15V supply that may be subject to
> transients.
>
> The issue is that I can't push too hard against the limitations of the
> LM124, I don't want batch/manufacturer specific behavior. So I will have to
> figure out a way around the slew rate. Probably by reducing the PWM
> triangle wave to 1V peak-to-peak.
>
> Out of curiosity what makes the LM324 family "bad"? Voltage swing? Slew
> rate/bandwidth? Noise/Offset? All the above ;) ?
>
>
> On Tue, Jul 10, 2018 at 11:08 AM, Sean Breheny <***@cornell.edu> wrote:
>
> > Hi Jason,
> >
> > Is there a reason for using such an ancient op-amp? This is part of the
> > LM324 family which is only really ever used now in either ultra cost
> > sensitive devices or legacy designs. Except for cost, there are many
> > op-amps which are better in every regard.
> >
> > I think you are right to be concerned. Slew rate depends on internal
> > capacitances, resistances, and current magnitudes. All of those are
> > temperature dependent AND manufacturing dependent. I would not be
> > comfortable assuming this part can do 0.2V/us unless either the datasheet
> > specified a minimum slew rate of more than that over all temperatures and
> > manufacturing variation OR if the typical value were at least 10x the
> > necessary value (2V/us).
> >
> > Sean
> >
> >
> > On Tue, Jul 10, 2018 at 10:29 AM, Jason White <
> > ***@gmail.com> wrote:
> >
> > > In my current design the operating temperature range is rather large:
> > -55C
> > > to 125C.
> > >
> > > I am planning on using the LM124DT from ST in an analog PWM circuit. My
> > > design uses a 0.5 to 4.5v triangle wave at 25kHz (required slew rate
> > > 0.2V/us).
> > >
> > > The LM124DT has a "typical" slew rate of 0.4V/us at 25C. A minimum slew
> > > rate is not specified and slew is not specified at other temperatures.
> > > Other manufacturers such as TI don't seem to specify this either.
> > >
> > > I would like to get an intuitive sense of how (or if) slew rate might
> > > change over temperature. Additionally I'd like to get some opinions as
> to
> > > whether 0.2V/us could be too fast for the LM124.
> > >
> > > Thanks,
> > > Jason White
> > > --
> > > http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive
> > > View/change your membership options at
> > > http://mailman.mit.edu/mailman/listinfo/piclist
> > >
> > --
> > http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive
> > View/change your membership options at
> > http://mailman.mit.edu/mailman/listinfo/piclist
> >
>
>
>
> --
> Jason White
> --
> http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive
> View/change your membership options at
> http://mailman.mit.edu/mailman/listinfo/piclist
>
--
http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist