Vout Lo in LM324

The beginning of it, one would hope :)

Tim

Yeah but! if you're looking for a comparator circuit that needs a slight rounded corner signal on the output and both rails to drive, an op-amp serves you very well to remove the need for the extra components.

Which takes me to what I did this weekend at work. I used a LM741 (yes an antique but I have lots of them)Op-amp as a comparator to drive a MosFet which then PWM a solenoid load.

With the compensating cap that is in the op-amp, that gave me the slight slew/skew rates needed to drive the gate avoiding ringing, so no gate R needed there. And I didn't need a pull up because we are not using a conventional comparator..

Yes, it was only 5khz I was using but it worked perfectly for what I needed. The MosFet is switching nicely from what the Lecroy was telling me, because I just happen to have it at my bench at the time.

Jamie

Wow, that's almost like driving it with a 12AT7 tube :-)

But why not if it was there? Just like a former colleague hauls the building materials for all his home remodeling with a 1956 Chevy truck that looks like it was just driven out of the show room.

The question remains, why was the 741 "there" in the first place? If I had a "mint" '56 Chevy, I'd still be driving my '01 beater Ford to the HD.

The 741 has its place. Just because it has some age in design does not mean it's all washed up. :)

For a lot of circuits you'll find that a 741 as you most likely already know, works perfectly.

I most likely have a couple hundred of them around the shop, that and some LM747, LM1458, LM358, LM324, LM339, LM393. etc.. And let us not forget the 555,556 timers in both monolithic and CMOS :)

These are great for conception ideas but when things move into areas of low voltage and power consumption, then we transfer a working circuit over to the appropriate components.

Jamie

Well, Jamie had some in the box so why spend money plus shipping on a LM324 when a uA741 is right in front of you? Heck, you can even still buy them.

As for the Chevy, my former colleague and I see it the same way: If a technical product is nicely kept up or restored then it's not meant to sit in some museum or a garage until the next show. It's meant to be used. I use a lot of tools here that are around 100 years old. To me they'd be no good in some fancy display cabinet.

I can't think of one place I'd use a '741.

So does a 12AT7, I suppose. I'd never design one of them in, either.

I've never had a design go into production with a 555/6, but they're not in exactly the same class as a '741.

...and what is the purpose of the '741, again?

I threw my '741s away *long* ago (probably >20 years). I used a lot of them but see little reason for even the power supplies, now. When I've needed that sort of swing a '741 wasn't going to cut it anyway.

I'd drive it, sure, but not to pick up lumber. There's too much work in restoring such a beauty to whack it with a piece of lumber (my '01 has a nice ding in the tailgate where a solid-core door whacked it in HD's parking lot). No, I don't wear my "Sunday go to meetin'" clothes to the HD, either, even if it is a Sunday.

In a mockup of a physical circuit layout, so you don't waste good parts?

I don't either but if I needed to whip up some circuit and there's a box full of them sitting there, I'd use it.

I did. Well, not the 12AT7 but a long-life Radar IF tetrode when I was a kid. It reduced the number of lightning induced mast preamp deaths to zero.

I've never used that timer either. Always did that with 74HC14,

To amplificate :-)

So would a Philbrick K2-W, but I'll bet that you couldn't use one.

Parts are cheap; not worth the hassle of keeping obsolete junk around.

The power supplies are more hassle than the thing is worth. There's almost always a 3.3V or 5V rail hanging around. Also, as I indicated, all this junk went in the trash three moves ago.

Well, I used a *lot* of uA741s, in the early '70s, too.

I've used '121s and '123s, as well as CMOS gates. Note that I have nothing against a '555 but for product use I've always had a better solution. Some test gear I designed fairly recently had a couple of '555s but it never got built.

poorly. ;-)

Until the day of reckoning cometh. Many years ago I really needed to get the Minolta film camera going again. But we could no longer buy the mercury batteries for the meter in it because they've been outlawed in the whole country. Using a silver-oxide battery was possible with some minor mechanical adaptation. But due to its 200mV in higher voltage it would have required a major disassembly and recalibration of the whole camera. And here comes in what you call "junk". Luckily I had kept Ge-diodes, including some really tiny ones. Fit like a glove, the camera now worked as before but with batteries one could buy at the supermarket.

For my (eBay) Canons, I bought some widgets the size of the mercury batteries but take a Silver Oxide cell and lose the 200mV.

Ge diodes are different (in the class of 555s) since there isn't a decent replacement (though Schottkys come close). There are *so* many replacements for '741s that there is no reason to keep them. OTOH, if you're a hacker who uses those gawd-awful white plastic proto strips and can't deal with SMT, perhaps the junk drawer, from 40 years back, works.

Hi Phil,

Here I can't agree with you. I would allways prefer one quad than two twins.

sun.

Right. and therefor the stock quotes are relative to the Cash in there extreme high, when value is low. so the expensivest stock is with the cheap parts.

Right. Don't foret this. But here it is simply the other way. LM324 are in stock, LM393 and LM358 not. To buy them I have to go to town or to order from webshop. both the costs are much too high for this prototype ;-) And more... It works with just one resistor more. May be I don't need it, when multiplying the resistors and divide the cap.

As far as I see I don't need a PCB I can solder the components directly to the pins of the LM324. This I can never do with two separate ICs ;-)

Sometime at the total oposite side ;-)

Marte

'n Abend Joerg,

Knapp 2 Stunden vor Dir hab ich das hier auch schon geschrieben ;-) Vielleicht nicht ganz so ausführlich ;-)

I'll try this

Tried this. Must be 1 k.

Works pretty fine up to 3.5 V specified and 3.7 V real. Quite enough for lot to do. And it is a student project and LM324 is realy hard to kill. Even they catched it, the LM324 is cheap ;-) So doesn't matter if they plug the power the oposite way.

Marte

It's not yet evening out here, first the barbecue has to be fired up :-)

Yeah, sometimes posts don't trickle in in the order written. Not sure if this has to do with my news server (I am using individual.de from the university in Berlin).

Once the resistors are much larger values than now you can use higher values there as well. Not too high because it has to discharge the (small) gate capacitance in due time.

Oh, ok, for a student project it can be alright. But if these are engineering students teach them to never trust typical data or graphs in datasheets. Only the guaranteed Rdson value at the guaranteed Vgs in the tables and only if not in the "typ" column. So if that says 5V like for some 2N7002 versions then the LM324 off a 5V supply ain't good enough if this was a product. At least not if I am sitting in the design review :-)

ns.

er the sun.

OK that's different. I've stopped using quad's on PCB's. (power pins in the middle is enough of a problem, but why on the 'wrong side'?)

George H.