I want a single supply op amp that will operate near the positive rail. Supply voltage will be about 15 volts. Doesn't have to work within millivolts of the positive rail; two tenths of a volt will do. I looked at the LM741 datasheet, but to me it's unclear how close to the positive rail it will go. But since I already have it, it would be nice if I could use one.
It won't work; take a look at the "output voltage swing" spec at
You'll need something rated as "rail to rail output".
But anyway, even for casual hobbyist use, consider throwing that 741 in the trash and getting something modern. The 741 was a great chip, in 1970; the technology has come a long, long way since then. Better opamps are widely available and cost pennies apiece.
Can you say more about what you're trying to do?
Yeah I kind of knew it was wishful thinking. Take a look at my post "current sensing" on sci.electronics.basics and .design, a few days ago. I built a voltage regulator prototype without current limiting and it works, but now I have to implement current limiting. Can't risk burning up an antique generator. The original mechanical voltage regulators had current limiting for a reason. I think I'd actually rather use discretes than an op amp. If I do it with discretes though, I think I will have to trim the circuit each time I build a voltage regulator, and I may be building a number of these (lots of friends with antique bikes and cars). So I'm keeping the option of using an op-amp, build it and accurate with no trimming. I've only had time briefly to experiment with the current sensing circuit using discretes on a breadboard, but it looks kind of fiddly.
I just realized I don't need ouput to the positive rail. I need output to within a diode drop of the negative rail, and input to the positive rail.
The LM301 suggested in your earlier thread would probably work. (Even though it's ancient as well.) It's not rated to go that close to the negative rail, but for current-limiting you don't usually need to bring the output voltage all the way to zero; get it within a volt or two and nothing upstream or downstream will be hurt. If you did need the opamp output to go to one diode drop above ground, one possibility is to put a couple of diodes in series with the output, to drop the voltage.
You might want to use LM101 instead of LM301, to get the extended temperature range, for automotive application.
But, glancing at that thread, I'm not sure which solution you eventually chose. None of them look that great, to me: for instance, MOSFETs were briefly mentioned as pass elements but then dismissed, but it seems to me that with your 10A limit, an NPN is not going to cut it. (How come? Because at Ic = 10A, to get a junction transistor into saturation you'll need base current of more than 500mA, and no ordinary opamp is going to manage that.) Unless you used a high-current Darlington NPN; which would introduce an extra diode drop, to your benefit. Hmm.
You might want to open a new thread, on sci.electronics.design, and post your proposed schematic and the requirements. I think the old thread may have drifted a bit off topic.
Oh, one thing to clarify would be whether you want a device that *limits* current to 10A, versus something like an "electronic fuse" that, if it detects current over 10A, completely shuts everything off until it's reset. Given your application I assume it's the former, but maybe I'm wrong.
No, the way the circuit works I only need on the order of 100 uA. This will turn on a npn transistor that grounds the base of the mosfet (1k pullup resistor) that conducts field current.
Yes, the former.
Now, how can I post ASCII diagrams in Google Groups? I don't have an ISP (do most of my internet use at cafes) to provide me with news service, so I have to go with something web-based. It would be much easier for me if Google had an option for posting in a non-proportional font. I hate not being able to post diagrams.
You want to use a Rail to Rail Input/Output Op amp. Something like the LMC6482 (dual) or LMC6484(quad). This is rated for 15V operation, and is a reasonable device if you don't need a speed demon or more then a few ma output drive M Walter
I think it would scare me to use a CMOS opamp in an automotive application. I'd worry about ESD. Maybe that's not a valid concern?
Regardless of that, though, the CMOS rail-to-rail amps seem to be very prone to oscillation with even a hint of capacitive load. I don't remember kell's circuit offhand - IIRC there's the Miller capacitance of the NPN to worry about, but maybe that's all.
ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.