Favorite Cheap and Useful Part? Lemme know.

[Si Ballenger]

74HCT259

 

[John Larkin]

I was using LM324's in the 1977-1987 time frame. I had three basic
ICs that I used, LM324, LM339 and TL084, plus logic. I could build
*anything* ;-)

...Jim Thompson

Anything slow.

John

 

[Jim Thompson]

Anything slow.

John

Speed kills ;-)

...Jim Thompson

 

[John Larkin]

Speed kills ;-)

...Jim Thompson

Speed thrills.

John

 

[analog]

I was using LM324's in the 1977-1987 time frame. I had three basic
ICs that I used, LM324, LM339 and TL084, plus logic. I could build
*anything* ;-)

Yep, me too. Gotta love those jellybeans. Today, of the three, I'd
say that the LM339 seems the least dated (its only really annoying
behavior is its phase inversion weakness).

The crossover distortion of the LM324 never bothered me because one
could just bias it for class-A operation if that were a problem.
Its real Achilles heal, IMO, is its strong propensity to rectify
impulse and RF noise on its inputs. This can be somewhat ameliorated
by putting small capacitors (especially surface mount) physically
right across the inputs and/or to ground, but why bother the 324 at
all when this added cost and trouble sort of defeats the purpose of
using such a cheap opamp in the first place?

 

[Jim Thompson]

Speed thrills.

John

Agreed, with cars !-)

...Jim Thompson

 

[John Woodgate]

<[email protected]>) about 'Singing the praises of the LM324',

The crossover distortion of the LM324 never bothered me because one
could just bias it for class-A operation if that were a problem.
Its real Achilles heal, IMO, is its strong propensity to rectify
impulse and RF noise on its inputs. This can be somewhat ameliorated
by putting small capacitors (especially surface mount) physically
right across the inputs and/or to ground, but why bother the 324 at
all when this added cost and trouble sort of defeats the purpose of
using such a cheap opamp in the first place?

I agree. If you have an r.f. rectification problem, switch to a FET
input device, such as TL084. Instant 20 to 26 dB improvement, if you
haven't done anything silly with the layout.

 

[Keith R. Williams]

Agreed, with cars !-)

and women. ;-) ...unless the wife is near.

 

[N. Thornton]

And there are applications where cost is all that matters. Or does
anyone know of a quad opamp that is cheaper than the LM324?

There's a function generator of far-east brand on my bench that is built
using a fair number of 324s, discretes and CD4051s. It goes up to 10MHz,
although above 1MHz the results aren't very convincing. Still, kudos to
the designers for squeezing this out of such a poor device.

What part does a 324 have in a 10MHz machine? Feedback stabilising? I
cant believe it could do anything at 10MHz.


Regards, NT

 

[Stefan Heinzmann]

What part does a 324 have in a 10MHz machine? Feedback stabilising? I
cant believe it could do anything at 10MHz.

I didn't dig deep enough into the copy of a copy of a schematic sheet
I've got of it to understand it entirely, but it looks as if the
triangular wave is generated by a current source and a current sink,
each made with an LM324 OpAmp and a transistor, feeding the timing
capacitor. The switching is seemingly done by diodes. So I guess the
current sources will at least have to react swiftly enough to regulate
across the switching action.

Of course the output of the OpAmp does not swing noticeably, so Slew
Rate is probably not the limiting factor. The buffer after the cap is
discrete, and the schmitt trigger controlling the diodes is a
construction from ECL comparators, a flipflop, and a discrete driver
stage. So there at least no LM324 is in the path.

So it seems that a current source built with the LM324 is actually
pretty good, even into the MHz range.

There are a lot of OpAmps in the circuit where I don't know what they're
good for. Maybe there's a trick or two I could learn if I studied it
some more...

Cheers
Stefan

 

[John Larkin]

and women. ;-) ...unless the wife is near.

From Master and Commander:

A toast, Gentlemen: to wives and sweethearts,
May they never meet.

John

 

[Russell Shaw]

How do you mean "as intended"? National's original datasheet (9 sheets
of it!) shows lots of examples where the output is DC unloaded, and
will have lots of crossover distortion.

It has a darlington npn for sourcing current and single pnp for
sinking current. It also has a 50uA current sink for keeping the
npn darlington in operation and the pnp off, with no output load.
There is 2V of deadband because the opamp really is *class-A only*.

As long as the instantaneous sink current is less than 50uA (plus
error margin), there is *no* crossover distortion.

With a large-signal frequency range of 5 KHz, it's nowhere near "as
good as any 1MHz GBW opamp".

At 0.5V/us, it's not much worse than any other cheap opamp.

 

[Russell Shaw]

Jim Thompson wrote...


Accckk, LM324 and active filters in the same sentence?!
Have you no shame?

With an output pulldown bias resistor, they're perfectly
adequate as 600Hz audio bandpass filters which i've done.

 

[John Larkin]

As long as the instantaneous sink current is less than 50uA (plus
error margin), there is *no* crossover distortion.

That sort of says that it's an OK opamp as long as its output is never
connected to anything. Agreed.

John

 

[Chris Carlen]

Hopefully in the same package!

I heard it will only be available in BGA soon. ;-)


--
____________________________________
Christopher R. Carlen
Principal Laser/Optical Technologist
Sandia National Laboratories CA USA
[email protected]

 

[John Woodgate]

I read in sci.electronics.design that Chris Carlen

I heard it will only be available in BGA soon. ;-)

..... supplemented by an octal-based version for real men.(;-)

V
V
V
V
V
V
V
8 pins not enough?

It has a base at each end of the package. Or six top-caps.

 

[John Woodgate]

I read in sci.electronics.design that John Larkin <jjlarkin@highlandSNIP
techTHISnologyPLEASE.com> wrote (in <r4e500l0ojrrbnu7602aitl11nd05agj7m@
4ax.com>) about 'Favorite Cheap and Useful Part? Lemme know.', on Mon,
12 Jan 2004:

That sort of says that it's an OK opamp as long as its output is never
connected to anything. Agreed.

Well, perhaps nothing that draws more than 50 uA, which covers a LOT of
things.

I can see I'll have to post some measurements. But I can't do it until
the end of the week. One section used as a DOUBLE Sallen and Key second-
order filter, giving a wide band-pass response 100 Hz to 5 kHz.

 

[Paul Burridge]

I heard it will only be available in BGA soon. ;-)

I'll get me coat. :-(

 

[N. Thornton]

I read in sci.electronics.design that Chris Carlen
.... supplemented by an octal-based version for real men.(;-)

V
V
V
V
V
V
V
8 pins not enough?

It has a base at each end of the package. Or six top-caps.

lol! I've seen one with 4 caps. They were a standard from the 20s
IIRC, used in shortwave radio.


Regards, NT