Which to buy, TL082 or LM393?

Discussion in 'The Projects Forum' started by wayneh, Jul 6, 2012.

  1. wayneh

    Thread Starter Expert

    Sep 9, 2010
    12,093
    3,031
    I'm off to the Shack; I need an amplifier to convert an AC signal (maybe 20kHz or more, ±<200mV) to clock pulses for counting on a 4017. I know that the LM393 will work, but I think the TL082 may be a bit faster and that could be better for my application.

    The LM393 datasheet says it has a response time of 1.3µs, whereas the TL082 datasheet says it has a slew rate of 13V/µs. I'm using a single 5V supply (maybe 9v soon), so I believe the TL082 will switch a bit faster than the LM393, maybe under 1µs?

    Any advice or comments? I may just buy and try both of them.
     
  2. upand_at_them

    Active Member

    May 15, 2010
    246
    29
    The LM393 is a comparator. RadioShack carries the LM324, get that.
     
  3. Wendy

    Moderator

    Mar 24, 2008
    20,765
    2,535
    20Khz may be too fast for the LM324. It is little more than a souped up 741.

    The LM393 is a digital gate. A single bit A/D convertor. I would go with that instead.
     
  4. bountyhunter

    Well-Known Member

    Sep 7, 2009
    2,498
    507
    The LM393 is a comparator which can not be used as an op amp. The TL-082 is a JFET input op-amp which could be used as a comparator.
     
  5. Wendy

    Moderator

    Mar 24, 2008
    20,765
    2,535
    If the OP is driving a digital gate such as a 4017 he doesn't need an op amp, he needs a gate. A comparator is a type of gate.

    A op amp may or not be able to imitate digital level signals close enough to work. Without a schematic your guess is as good as mine.

    It is not a matter of switching speed, that does not matter. What matters, and it is the only thing, is the voltage levels of the high and the low. Meet the criteria the input of the 4017 (standard CMOS) and it will work.
     
  6. wayneh

    Thread Starter Expert

    Sep 9, 2010
    12,093
    3,031
    The schematic I'm working with is in my thread regarding an inductor ring tester, which I'm building to test a flyback transformer discussed in a thread about my kid's TV, ;) Like layers on an onion.

    A commercial device I'm following sends a pulse through the inductor under test and uses a LM393 comparator to turn the LC ringing pulses into clock pulses for the 4017 counter - that's why I knew the LM393 should work.

    But FWIW, I tried the TL082 first and it's working great. My usual go-to LM358 was too slow. It took me a while to figure that out, because the ringing worked great with big inductors, which result in slow oscillations, giving good counts at the 4017 and lit LEDs. (I never would have suspected the LM358 if it weren't for tips offered by others here.) Small coils just didn't register because the LM358 couldn't move far enough quickly enough to clock them. With the TL082 even small coils give big counts!

    In fact, the counts are too big. Now I need to make my 4017 stop and not count past 10 until it's reset. Even a small coil will count 12 rings or more, which looks like just 2. Is there a trick to stop it at 10?

    [update] Got it. Just connect "9" output to the enable pin. Shuts off counting at 9.
     
    Last edited: Jul 7, 2012
  7. wayneh

    Thread Starter Expert

    Sep 9, 2010
    12,093
    3,031
    It matters if the switching is too slow to clock the signal. The TL082 works great because it is about 20X faster than the LM358, which fails most of the time. The TL082 might also be 2X faster than the LM393. It may not get as close to the rails as the LM393, but it's enough to clock the 4017 on a 5V single supply.
     
  8. Audioguru

    New Member

    Dec 20, 2007
    9,411
    896
    The opposite. Since the LM324 is low power (slow performance) then it is a souped-DOWN 741 opamp.
     
  9. wayneh

    Thread Starter Expert

    Sep 9, 2010
    12,093
    3,031
    The TL082 is working for me, but one issue I'm having is that the input voltage range on a 5V supply is more narrow than the LM358, which I got because it senses down to the ground rail. The TL082 certainly does not.
     
  10. Sensacell

    Well-Known Member

    Jun 19, 2012
    1,128
    266
    When it comes to opamps, the newer ones are SOOOO much better than what we started with back in the day. Why do we still insist on torturing newbies with the LM741? why pass that headache on to the next generation?

    My personal favorite is the MCP6024 - R-R in and out, fast, low offsets and quiet.
     
  11. wayneh

    Thread Starter Expert

    Sep 9, 2010
    12,093
    3,031
    Who's WE? I don't recall ever reading a single recommendation here for the 741, only advice to not use it. But I know what you mean - it's surprising how many newbs come here asking for help with some 741 circuit.

    Thanks for the tip on the MCP6024. If only I could go buy it locally.
     
  12. Wendy

    Moderator

    Mar 24, 2008
    20,765
    2,535
    The OP specified Radio Shack. This limits options, a lot. I feel his pain.
     
  13. SgtWookie

    Expert

    Jul 17, 2007
    22,182
    1,728
    The TL082 really isn't a good choice for the project, as it's common mode range is from -V + 3v to +V - 1.5v; so with a 5v supply, you only have a 0.5v input range - pretty small.

    If you're limited to Radio Shack for a supplier, the LM339 quad comparator will be adequate; it'll be much faster than the TL082 and far faster than the LM324.. As with the LM393, the outputs are open-collector, and the input goes down to the negative rail (in your case, ground).
     
  14. wayneh

    Thread Starter Expert

    Sep 9, 2010
    12,093
    3,031
    Yup, the range is narrow enough that I've been battling with it. But I'm getting results working below -V+2, so I'm seeing a bit wider range than above. Taking the circuit up to 9V will help.

    Curious about your comment regarding speed. The 13V/µs slew rate of the TL082 as quoted in the datasheet suggests it should swing 5V in under 1µs, putting it near the 1.3µs transition quoted for the LM393. Am I way off in my understanding of that? Thankfully the TL082 is fast enough that it does work in this application. But I won't be buying another one, even if it is convenient to buy.

    Why the heck doesn't the Shack sell a really good op-amp in stores, instead of the 741 and this one? Why would you put the crappiest old op amp you can find into a brick and mortar store? It'd make more sense to a sell a premium amp than a gutter amp. Seems like they're just begging us to go elsewhere.
     
  15. SgtWookie

    Expert

    Jul 17, 2007
    22,182
    1,728
    Your mileage may vary, considerably. The -V +3v is the guaranteed spec over temperature. If you want to be certain that the circuit will work, you go with that number. You're more or less going on the bleeding edge of the specs; don't count on it working over the full temperature range.

    Here's a link to TI's datasheet for the LM139/239/339/2091:
    http://www.ti.com/lit/ds/symlink/lm339.pdf
    Look at the plots on page 9. Initially, there is a response delay of ~0.3uS to ~1.2uS, but once it starts changing, the transition between states is very fast; faster than the TL082. This is because the TL082 is internally compensated, and comparators are not internally compensated.
    If you look at the simplified internal schematic of the TL082 in the datasheet:
    http://www.ti.com/lit/ds/symlink/tl082-n.pdf
    you will see the compensation capacitor between the base and collector of the output driver transistor; this improves the linear characteristics of the opamp for continuously variable signals, but limits the peak bandwidth of the opamp. Also, this opamp has a totem-pole output instead of an open-collector output like the LM339 does.

    The TL082 and LM324 aren't terrible opamps, even though they're pretty ancient. The 741 is probably only carried because people who took physics in college learned opamps with the 741, and they ask for it because they really don't know any better.

    The Radio Shack stores of the 1960's and 1970's carried LOTS of parts. You could buy most of the 4000 series CMOS and the 74 TTL series, along with lots of interesting analog ICs.

    Then video games came out, and most people started playing the games instead of taking up electronics as a hobby. Radio Shack's sales in their electronics really dropped like a rock; and then their market share of computer sales largely dried up. Most of the mom & pop electronics stores went out of business.

    Radio Shack is actually trying to respond at least a bit to hobbyists' desires. Nowadays, hobbyists are asking more for microcontrollers than other components, as you can do so much more with them.
     
  16. Audioguru

    New Member

    Dec 20, 2007
    9,411
    896
    Radio Shack in Canada went out of business about 8 years ago. Their biggest stores were bought by Circuit City and called "The Source".
    Bell Canada bought The Source stores from Circuit City in 2009 to concentrate on selling cell phones. They might have a few parts left from when Radio Shack was there.
     
Loading...